Understanding the Use of Ecosystem Service Knowledge in Decision Making: Lessons from International Experiences of Spatial Planning
McKenzie, E., Posner, S., Tillmann, P., Bernhardt, J. R., Howard, K., Rosenthal, A.
Citation: Environment and Planning C: Government and Policy: doi:10.1068/c12292j
http://www.envplan.com/abstract.cgi?id=c12292j
- See more at: http://wwfscience.org/resources/wwf-literature-digest/2605#sthash.kwfXnW7K.dpuf
McKenzie, E., Posner, S., Tillmann, P., Bernhardt, J. R., Howard, K., Rosenthal, A.
Citation: Environment and Planning C: Government and Policy: doi:10.1068/c12292j
Publication Year: 2014
The
limited understanding of how ecosystem service knowledge (ESK) is used
in decision making constrains our ability to learn from, replicate, and
convey success stories. We explore use of ESK in decision making in
three international cases: national coastal planning in Belize; regional
marine spatial planning on Vancouver Island, Canada; and regional
land-use planning on the island of Oahu, Hawaii. Decision makers,
scientists, and stakeholders collaborated in each case to use a
standardized ecosystem service accounting tool to inform spatial
planning. We evaluate interview, survey, and observation data to assess
evidence of ‘conceptual’, ‘strategic’, and ‘instrumental’ use of ESK. We
find evidence of all modes: conceptual use dominates early planning,
while strategic and instrumental uses occur iteratively in middle and
late stages. Conceptual and strategic uses of ESK build understanding
and compromise that facilitate instrumental use. We highlight attributes
of ESK, characteristics of the process, and general conditions that
appear to affect how knowledge is used. Meaningful participation,
scenario development, and integration of local and traditional knowledge
emerge as important for particular uses.
Please contact author for a copy of this article: mckemily@gmail.com
Please contact author for a copy of this article: mckemily@gmail.com
Global water resources affected by human interventions and climate change
- Ingjerd Haddelanda,1,
- Jens Heinkeb,c,
- Hester Biemansd,
- Stephanie Eisnere,
- Martina Flörkee,
- Naota Hanasakif,
- Markus Konzmannb,
- Fulco Ludwigd,
- Yoshimitsu Masakif,
- Jacob Scheweb,
- Tobias Stackeg,
- Zachary D. Tesslerh,
- Yoshihide Wadai, and
- Dominik Wisseri,j
-
Edited by Katja Frieler, Potsdam Institute for Climate Impact Research, Potsdam, Germany, and accepted by the Editorial Board August 5, 2013 (received for review January 30, 2013)
Significance
Humans alter the water cycle by
constructing dams and through water withdrawals. Climate change is
expected to additionally
affect water supply and demand. Here,
model analyses of climate change and direct human impacts on the
terrestrial water cycle
are presented. The results indicate that
the impact of man-made reservoirs and water withdrawals on the long-term
global terrestrial
water balance is small. However, in some
river basins, impacts of human interventions are significant. In parts
of Asia and
the United States, the effects of human
interventions exceed the impacts expected for moderate levels of global
warming. This
study also identifies areas where
irrigation water is currently scarce, and where increases in irrigation
water scarcity are
projected.
Abstract
Humans directly change the
dynamics of the water cycle through dams constructed for water storage,
and through water withdrawals
for industrial, agricultural, or domestic
purposes. Climate change is expected to additionally affect water supply
and demand.
Here, analyses of climate change and
direct human impacts on the terrestrial water cycle are presented and
compared using
a multimodel approach. Seven global
hydrological models have been forced with multiple climate projections,
and with and without
taking into account impacts of human
interventions such as dams and water withdrawals on the hydrological
cycle. Model results
are analyzed for different levels of
global warming, allowing for analyses in line with temperature targets
for climate change
mitigation. The results indicate that
direct human impacts on the water cycle in some regions, e.g., parts of
Asia and in
the western United States, are of the same
order of magnitude, or even exceed impacts to be expected for moderate
levels of
global warming (+2 K). Despite some spread
in model projections, irrigation water consumption is generally
projected to increase
with higher global mean temperatures.
Irrigation water scarcity is particularly large in parts of southern and
eastern Asia,
and is expected to become even larger in
the future.
Terrestrial water
fluxes are affected by both climate and direct human interventions,
e.g., dam operations and water withdrawals.
Climate change is expected to alter the water
cycle and will subsequently impact water availability and demand.
Several hydrologic
modeling studies have focused on climate
change impacts on discharge in large river basins or global terrestrial
areas under
naturalized conditions using a single
hydrologic model forced with multiple climate projections (1, 2). Recently, hydrological projections from eight global hydrological models (GHMs) were compared (3).
In many areas, there was a large spread in projected runoff changes
within the climate–hydrology modeling chain. However,
at high latitudes there was a clear increase
in runoff, whereas some midlatitude regions showed a robust signal of
reduced
runoff. The study also concluded that the
choice of GHM adds to the uncertainty for hydrological change caused by
the choice
of atmosphere–ocean general circulation
models (hereafter called GCMs) (3). Expected runoff increases in the north and decreases in parts of the middle latitudes have been found also when analyzing
runoff from 23 GCMs (4).
These studies focused on the
naturalized hydrological cycle, i.e., the effects of direct human
interventions were not taken
into account. However, in many river basins
humans substantially alter the hydrological cycle by constructing dams
and through
water withdrawals. Reservoir operations alter
the timing of discharge, although mean annual discharge does not
necessarily
change much. A study with the water balance
model (WBM) showed that the impact of human disturbances, i.e., dams and
water
consumption, in some river basins is equal to
or greater than the impact of expected climate changes on annual runoff
over
the next 40 y (5). Also, rising water demands are found to outweigh global warming in defining the state of global water systems in the near
future (6). Water for irrigation is the largest water use sector, currently accounting for about 70% of global water withdrawals and
nearly 90% of consumptive water use (7). A recent synthesis of simulations from seven GHMs found that irrigation water consumption currently amounts to 1,250 km3⋅y−1 (±25%) and that considerable differences among models appear in the spatiotemporal patterns of water consumption (8).
Direct comparisons of the climate
impact and human intervention modeling studies can be difficult given
that the setups are
not identical, i.e., the input forcing data
and climate models vary. Also, because of the uncertainty of
model-specific results,
a multimodel approach is preferable in impact
modeling studies (3, 9).
This approach is similar to assessments performed within the climate
community. Here, multimodel results on current and
future water availability and consumption at
the global scale from the Water Model Intercomparison Project (WaterMIP)
within
the European Union Water and Global Change
(EU WATCH) project (9, 10), and Inter-Sectoral Impact Model Intercomparison Project (ISI-MIP) (11) are presented. (Information on how to get access to WaterMIP and ISI-MIP simulation results can be found at www.eu-watch.org and www.isi-mip.org,
respectively.) Results from these two projects are synthesized to
obtain a large ensemble of impact model results. The integration
of results from the different projects is
achieved by extracting impacts for time periods of global mean
temperature (GMT)
increases of 2 and 3 K from the simulations,
largely following the method of Tang and Lettenmaier (4). The advantage of this approach is that it allows presenting results in a way that is in line with temperature targets used
in climate mitigation discussions.
Other studies have focused on future water scarcity using results from WaterMIP and ISI-MIP, but have analyzed changes of
naturalized runoff only (3, 12).
We here aim to fill this knowledge gap by comparing the different
impacts from climate change and direct human impacts
and analyzing their interplay. The models
included take into account water withdrawals and consumption in
different sectors;
for more information, see Models and Data and Supporting Information, SI Models and Data. The objectives of this study are to (i) assess the relative contribution of anthropogenic impacts and climate change to river basin scale water fluxes, and (ii)
identify areas where climate change can be expected to cause
substantial changes in water consumption and water scarcity,
focusing on water for irrigation. The effects
of future changes in irrigated areas or irrigation practices are not
taken into
account, and only dams that currently exist
are included in the analyses. In this paper, simulations considering
man-made
reservoirs, water withdrawals, and water
consumption are referred to as human impact simulations, whereas the
simulations
without these disturbances are referred to as
naturalized simulations. The results are mainly presented in a way
intended
to give an overview of impacts at larger
spatial scales (river basin and country levels). However, some
finer-scale results
are included to reveal effects that can be
concealed at coarse spatial scales.
Results
Human Impacts Versus Climate Change.
Anthropogenic water consumption results in mean annual runoff decreases of 5% or more in many river basins during the control
period (1971–2000) (Fig. 1A and Supporting Information, River Basin Information and Results).
The effect is especially noticeable in heavily irrigated regions at
middle latitudes across Asia, and in the western part
of the United States. In some river
basins in the Middle East, central Asia, and the Indian subcontinent,
the median ensemble
runoff decrease is more than 15% as a
result of water consumption within the river basin. In several other
Asian river basins,
and in the Colorado, Nile, Orange,
Murray–Darling River basins, the ensemble median decrease in runoff
resulting from anthropogenic
water consumption is between 5% and
15%.
Fig. 1.
Comparison
of human impact and climate change effects on runoff at the river basin
level. Basin averaged runoff values are
calculated based on simulated
discharge at the outlet of the river basins, and the median ensemble
results are shown. (A) Control period (1971–2000) human impact simulations compared with control period naturalized simulations. (B) Basin averaged naturalized runoff for 2 K GMT increase, compared with control period naturalized simulations. (C) Basin averaged human impact runoff for 2 K GMT, compared with control period naturalized simulations.
Water consumption always
results in runoff decreases, whereas the climate change signal can be in
both directions. Climate
change affects naturalized runoff in
river basins in all parts of the world. Projected runoff decreases are
especially noticeable
in the Mediterranean area and in the
Middle East, but also in Central and South America and parts of
Australia (Fig. 1B). Runoff is projected to increase at northern latitudes, corresponding to areas with large projected increases in precipitation
(13). Runoff increases are also projected in parts of the Arabian Peninsula, the Horn of Africa, and the Indian subcontinent
(Fig. 1B).
The pattern of the total impacts, i.e., runoff changes caused by both 2 K GMT increase and human impacts (Fig. 1C), is dominated by the impacts of climate change alone (Fig. 1B).
However, noticeable differences exist in southwestern United States and
central Asia. To highlight the relationship between
the human impacts and climate change
effects, differences between the absolute values of the individual
impacts are presented
(Fig. 2). This comparison shows that, in several river basins, current water consumption affects annual averaged runoff more than
climate change (2 K) is expected to impact naturalized runoff. Fig. 2A shows the river basins in which the climate signal mitigates the human impact signal to some extent or even exceeds it, e.g.,
in the Nile River basin. Fig. 2B shows the river basins in which the impact of climate change adds to the human impact signal. The combined effect is hence
enhancement, e.g., in the Colorado and the Indus River basin.
Fig. 2.
(A) The difference between the absolute values in Fig. 1 A and B in basins where the human impact and climate signals are opposite, i.e., naturalized runoff increases. (B) The differences between the absolute values in Fig. 1 A and B
in basins where both the climate signal and human impact signal are
negative, i.e., runoff decreases. The red and yellow
colors indicate that the control
period human impacts are larger than future climate effects on
naturalized runoff.
Despite the locally significant decreases in runoff, anthropogenic water consumption amounts to only 1.3% of median global
terrestrial runoff (Fig. 3A).
Among the world’s large river basins, and according to the model
ensemble included in this study, the Indus River basin
is the most affected by human impacts
at the annual level. According to the median ensemble result, as much as
47% of current
runoff is consumed within the Indus
River basin (Fig. 3F). Fig. 3
also shows that the results across the model ensemble for the human
impact simulations are significantly different at the
river basin level. The interquartile
range for the Indus River basin is from 29% to 62%, and the individual
model results
vary between 18% and 79%. Large
intermodel variations are also found in the Huang He River basin (Fig. 3G),
where the simulated anthropogenic water consumption varies between 7%
and 51% of current naturalized runoff. Moreover,
for most of the river basins presented,
the impact of a 3 K GMT increase is more pronounced than a 2 K GMT
increase, both
when looking at the total effect of
climate change and human impacts and when looking at the decomposed
effects separately
(Fig. 3). In the Colorado and Mississippi River basins, and in several river basins in Asia, the human impact effect is larger than
the climate effect (Figs. 2 and 3). In the Mediterranean area, both the climate and human impact signals are negative, but the climate signal dominates (Fig. 2B).
Fig. 3.
Box plots of relative changes in runoff for (A) the world, (B) Colorado, (C) Mississippi, (D) Nile, (E) Euphrates-Tigris, (F) Indus, and (G)
Huang He for the control period (C) (1971–2000), 2 and 3 K GMT
increases. The boxes illustrate the 25th, 50th, and 75th
percentiles of the ensemble (47
members). The whiskers represent the total sample spread, and in
addition the 5th and 95th
percentiles are marked. The human
impact results (orange bars) are compared with the naturalized
simulations during the same
time period, e.g., 2 K human
impacts are compared with 2 K naturalized simulations. All climate and
combined effects (blue
and green bars) are compared with
the control period naturalized simulations.
Irrigation Water Consumption and Scarcity.
The number of water use sectors included in the results presented so far varies between the different GHMs (Models and Data). However, all GHMs include the agriculture sector, i.e., water used for irrigation, which is the largest water consumer
globally (7). Here, an index called the cumulative abstraction-to-demand (CAD) ratio (14)
is used as a measure of irrigation water scarcity. The higher this
number is, the closer the crops are to having their water
requirements fulfilled. Thus, a
decrease in CAD represents an increase in water scarcity. The highest
potential irrigation
water consumption numbers (water
consumed given water is freely available) during the control period
(1971–2000) are found
in the Indian subcontinent (Fig. 4A). Although the CAD ratio is low in the Indian subcontinent (Fig. 4B), actual water consumption (water consumed taking water availability into account) in the area is still considerable, which
is reflected in the human impact results for the Indus River basin (Figs. 1A and 3).
Fig. 4.
Irrigation water consumption and cumulative abstraction-to-demand (CAD) ratio at the grid cell level. (A) Ensemble median potential irrigation water consumption, control period (1971–2000). Light gray color represents areas where
there is no, or very little, irrigation. (B) Ensemble median CAD, control period. (C) Differences in CAD between the control period and the 2 K GMT increase period. Negative numbers mean the CAD ratio decreases.
The CAD ratio is projected to decrease with increasing GMT in most areas where irrigation exists today (Fig. 4C),
meaning an increase in irrigation water scarcity. The CAD ratio is
projected to increase in only a few scattered areas,
e.g., western India. This increase in
the CAD ratio can be linked to increased water availability in this area
(Fig. 1C). Fig. 4 reveals some areas impacted by direct human interventions that are not revealed in Fig. 1,
because subbasin variations can be concealed when presenting basin
averaged results. For example, in parts of the Mississippi
River basin, water consumption is
considerable, whereas the effect at the basin total level is small (Figs. 1–3). A decrease in the CAD ratio is projected in the United States, southwestern Europe, Pakistan, India, and China (Fig. 4C). Some statistics on the impact of 2 and 3 K of global warming on irrigation water in these areas, in addition to the global
total numbers, are presented in Fig. 5. The global median potential irrigation water consumption for the entire ensemble (47 members) is 1,171 km3⋅y−1 in the control period (Fig. 5A). The interquartile range for the same time period ranges from 940 to 1,284 km3⋅y−1. The corresponding number for the subensemble, i.e., for those models simulating both potential and actual water consumption
(29 of the 47 members; Models and Data), is 1,174 km3⋅y−1 (942–1,292 km3⋅y−1). These numbers are close to the 1,250 km3⋅y−1 (±25%) reported previously (8), and represent about 1% of mean annual terrestrial precipitation in the forcing datasets used here, and between 1% and 2%
of simulated annual terrestrial runoff.
Fig. 5.
Ensemble statistics on irrigation water consumption for the control period (C) (1971–2000), 2 and 3 K GMT increases for (A) the world, (B) United States, (C) southwest Europe (here comprising Portugal, Spain, and France, (D) Pakistan, (E) India, and (F)
China. The upper panels show annual potential and actual irrigation
water consumption. The lower panels show CAD, i.e.,
the relationship between the
actual and potential irrigation water consumption. The boxes illustrate
the 25th, 50th, and 75th
percentiles of the ensemble. The
whiskers represent the total sample spread, and in addition the 5th and
95th percentiles
are marked.
Substantial differences exist in the ensemble estimates of the amount of potential irrigation water consumed, i.e., when water
demands are always met (Fig. 5). However, potential irrigation water consumption will increase with increasing GMT, both globally and regionally (Fig. 5). Irrigation water consumed when water availability is taken into account is more similar across the ensemble, despite the
differences in human impact parameterizations (Models and Data). Global actual irrigation water consumption increases slightly with increasing GMT (Fig. 5A). The projected changes in actual irrigation water consumption are less apparent than the projected changes in potential
irrigation water consumption (Fig. 5). The spread in irrigation water consumption numbers for a given time period reflects the spread in human impacts seen for
the river basins presented in Fig. 3.
More importantly, there is a general agreement that the CAD ratio will
decrease in the areas in question, and more so the
more GMT increases. The global CAD
ratio varies from 0.4 to 0.7 across the simulations, decreasing to
0.35–0.68 at 3 K GMT
increase. The corresponding median
number decreases from 0.58 to 0.52. The smallest change in the CAD ratio
is found in India.
Here, increased water availability (Fig. 1) results in almost constant water scarcity, despite a slight increase in potential irrigation water consumption (Fig. 4). Among the areas presented in Fig. 5,
the relative decrease in the CAD ratio is most pronounced in
southwestern Europe. Here, the control period median CAD ratio
is simulated at 0.69, whereas the
median result at 3 K GMT is 0.5. Actual irrigation water consumption
does not change much
with increasing GMT, indicating that
the decrease in the CAD ratio for the areas considered is mainly caused
by an increase
in water demands.
Discussion
The climate effects on naturalized runoff presented here are broadly consistent with results presented elsewhere (3, 4, 12).
In large parts of the world, the additional impact on runoff caused by
anthropogenic water consumption does not contribute
much to the total changes. However, this
study emphasizes the importance of taking anthropogenic water
consumption into account
in areas where direct human interventions
are large, and highlights areas where water consumption leads to
substantial changes
in land surface water fluxes. It has
previously been indicated that it is unlikely that irrigation has a
significant global-scale
impact on the Earth’s climate (15), but regional predictions within global climate models can be improved by taking into account local-scale processes (15).
Surface water evaporation from
man-made reservoirs and reservoir operations causing seasonal regime
shifts across multiyears
can cause slight changes in annual runoff
numbers. However, reservoirs influence the shape of the hydrograph
profoundly in
many areas of the world and seasonally
impact discharge much more than the reduction caused by water
consumption (16, 17).
Seasonal changes in discharge caused by storing and releasing of water
in reservoirs are not presented in this study, which
focuses on annual runoff numbers. Also,
because only annual results are presented, it is not revealed whether
water scarcity
is constant over the time period
considered, or whether interannual or intraannual variations exist. The
reservoir storage
capacity within a river basin indirectly
impacts annual runoff numbers through its ability to accommodate
seasonal variations
in flow volume and hence to satisfy
irrigation water requirements. This effect has not been specifically
studied here, but
it has previously been indicated that
nearly one-half of the irrigation water extracted globally originates
from reservoirs
built for irrigation purposes (16).
The model ensemble indicates that irrigation water scarcity is expected to increase with increasing GMT. About 40% of total
agricultural production relies on irrigation (18). In light of this, the increase in water scarcity and potential decline in food production could affect people worldwide
through food price changes on the global market (19).
In areas with a projected increase in irrigation water scarcity, and
hence possible decreases in food productivity, adaptation
measures need to be addressed. To increase
food production, better water management and improved irrigation
practices (reduced
losses) have been suggested (8). Irrigation area expansion in regions with sufficient freshwater is also projected to increase food production (20). These issues must all be discussed in light of other water demands, including environmental flow requirements (8).
The areas for which irrigation water consumption and water scarcity are presented in Figs. 4 and 5 do not overlap directly with the river basins presented in Figs. 1–3. However, Figs. 4 and 5 still indicate that, if more water was available for use, the anthropogenic impacts on river basin runoff seen in Figs. 1–3 would have been even larger. The range in estimates in Fig. 3 is a result of both differences in the baseline runoff (naturalized simulations) and amount of water consumed. Parameterization
differences among GHMs that influence naturalized simulation results (9) will subsequently influence the human impact simulations. Reservoir operations and water withdrawal parameterizations further
influence the results and contribute to the rather large differences (Figs. 3 and 5).
The largest relative runoff decreases for the human impact simulations
in the Colorado River basin, for example, originate
from the hydrologic model simulating the
lowest naturalized runoff and among the highest water consumption
numbers within
the river basin. In other areas, e.g., in
the Indus and Huang He River basins, the differences are also influenced
by whether
or not multicropping is taken into account
in the hydrologic model.
It should be noted that none of
the models considers water transportation between river basins, e.g.,
water transported from
the Colorado River basin to California,
and groundwater extractions are poorly represented in most models.
Hence, the actual
irrigation water consumption numbers might
be somewhat underestimated. However, three of the GHMs assume that
anthropogenic
water demands are always met (Models and Data).
Furthermore, not all models take into account water consumption in
sectors other than agriculture, although the impact
may be small because those sectors
currently account for only a small fraction of the total. In addition,
irrigation water
withdrawals and consumption depend on the
irrigation map used (21).
These differences in human impact parameterization clearly contribute
to the spread in runoff changes and water consumption
numbers, in addition to naturalized
simulation differences. In addition, both GCMs and GHMs contribute
substantially to the
spread in future projections (3, 12).
Only climate change effects on
water demands and consumption are accounted for in this study, whereas
other variables, such
as irrigated area and irrigation
efficiencies are kept constant at the year 2000 level. Also, the
indirect effect of rising
CO2 concentrations on runoff and irrigation water consumption through its direct effect on evaporative demand is not considered.
Increasing CO2 can lead to lower irrigation water demands (20, 22). However, nutrient limitations may influence crop growth. The combined effect on crop growth, irrigation water demands,
and resulting food production is still somewhat uncertain (22). The positive trend in potential irrigation water consumption presented here is more profound than for specialized crop
models (20). Possible reasons for this lie in the different representation of agricultural land and agrohydrological processes in the
models (20).
These and other impacts on the hydrological cycle should be addressed
in future hydrological model developments and multimodel
studies. Note also that bias correction
has been applied to the GCM data (23, 24). The assumptions and implications of bias correction on forcing data used in hydrological simulations are thoroughly discussed
in the study by Ehret et al. (25). Bias correction can impact present-day simulated runoff numbers strongly, but the impact on projected relative water flux
changes, which is the focus in this paper, are much smaller (23, 26).
Conclusions
Based on a
large ensemble of simulations using eight GCMs and seven GHMs, this
study provides a comprehensive assessment of
the effects of climate change and direct
anthropogenic disturbances on the terrestrial water cycle. Despite
considerable spread
in the individual results, a number of
robust conclusions can be drawn at the regional and global scale. The
results indicate
that the impacts of man-made reservoirs,
water withdrawals, and water consumption on the long-term global
terrestrial water
balance are small. However, impacts of
anthropogenic interventions are significant in several large river
basins. In particular,
in irrigation-rich areas in Asia and in
the western United States, the effect of current anthropogenic
interventions on mean
annual runoff is stronger than the
projected changes for a 2 or 3 K increase in GMT. Climate change tends
to increase potential
irrigation water consumption on currently
irrigated lands with further detrimental effects in regions with
significant irrigation.
The climate change signal on runoff can be
positive or negative, and hence has the potential to alleviate or
aggravate irrigation
water scarcity. Globally, the relationship
between actual and potential irrigation water consumption is expected
to decrease,
indicating an increase in irrigation water
scarcity.
Models and Data
Seven GHMs are included in this study. The nature and magnitude of human disturbances at which direct anthropogenic impacts
like dams, water withdrawals, and water consumption are included in the models vary (Table 1 and Supporting Information, SI Models and Data). All models were forced with climate data from a total of eight GCMs included in the Coupled Model Intercomparison Project
3 (CMIP3) and CMIP5 archives (Table 2). CMIP3 data were prepared for the hydrological model simulations within the WATCH project (3, 23), and the CMIP5 data were prepared for ISI-MIP (24).
Included in the analyses presented here are results when using forcing
data from the A2 emission scenario (CMIP3 models)
and RCP8.5 (CMIP5 models). Thirty-year
periods of GMTs at 2 and 3 K above preindustrial level are extracted
from the GCMs
(Table 2). The control period (1971–2000) is assumed to be 0.4 K above preindustrial level for all GCMs.
View this table:
Table 1.
Hydrologic models
View this table:
Table 2.
First year of 30-y periods for each GCM and mean GMT increases above preindustrial level
All hydrological models are run at a daily time step at a spatial resolution of 0.5° latitude by longitude, and runoff is
routed through the DDM30 river network (38). Simulation results are submitted for the period 1971–2099. Not all GHMs are run using input data from all GCMs (Table 2).
Simulated discharge at the basin outlets are used when calculating
basin averaged, or world total, runoff numbers. In this
paper, potential water consumption
represents water consumed given water is freely available. All models
included in the study
simulate this quantity. Four of the
models—H08, the Lund-Potsdam-Jena managed land dynamic global vegetation
(LPJmL), the
PCRaster global water balance model
(PCR-GLOBWB), and the variable infiltration capacity macroscale
hydrologic model (VIC)—also
simulate actual water consumption, which
is defined as water consumed when water availability is taken into
account. The CAD
ratio (14) is used as a measure of irrigation water scarcity (Supporting Information, Glossary). Both actual and potential irrigation water consumption are calculated at a daily temporal resolution, and hence subannual
variations are imbedded in the final CAD numbers.
Annual runoff and water
consumption numbers are calculated for each GCM–GHM combination
independently, creating an ensemble
of up to 47 annual time series for the
period 1971–2099. Differences between simulations are thereafter
calculated for each
time period of interest (Table 2)
for each ensemble member. Finally, median numbers and other statistic
measures are calculated. All results are treated equally,
and no attempt to give weights to GCMs or
GHMs based on performance has been made.
Acknowledgments
We acknowledge the World Climate Research Programme’s Working Group on Coupled Modeling, which is responsible for CMIP, and
we thank the climate modeling groups (Table 2)
for making available their model output. We appreciate the reviewers’
constructive and valuable comments. This study was
conducted in the framework of the ISI-MIP
project and the EU WATCH Integrated Project (Contract 036946), in
collaboration
with the Global Water System Project. The
ISI-MIP Fast Track project was funded by the German Federal Ministry of
Education
and Research (Bundesministerium für
Bildung und Forschung) with project funding reference number OlLS1201A.
Y.M. and N.H.
were supported by the Environment Research
and Technology Development Fund (S-10) of the Ministry of the
Environment, Japan.
Footnotes
- 1To whom correspondence should be addressed. E-mail: ingjerd.haddeland@nve.no.
-
Author contributions: I.H., J.H., F.L., and J.S. designed research; I.H., J.H., H.B., S.E., M.F., N.H., M.K., Y.M., T.S., Z.D.T., Y.W., and D.W. performed research; I.H. and J.H. analyzed data; and I.H., J.H., H.B., S.E., M.F., N.H., M.K., F.L., Y.M., J.S., T.S., Z.D.T., Y.W., and D.W. wrote the paper.
-
The authors declare no conflict of interest.
-
This article is a PNAS Direct Submission. K.F. is a guest editor invited by the Editorial Board.
-
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1222475110/-/DCSupplemental.
References
Constraints and potentials of future irrigation water availability on agricultural production under climate change
- Joshua Elliotta,b,c,1,
- Delphine Deryngd,
- Christoph Müllere,
- Katja Frielere,
- Markus Konzmanne,
- Dieter Gertene,
- Michael Glotterf,
- Martina Flörkeg,
- Yoshihide Wadah,
- Neil Besta,
- Stephanie Eisnerg,
- Balázs M. Feketei,
- Christian Folberthj,
- Ian Fostera,b,
- Simon N. Goslingk,
- Ingjerd Haddelandl,
- Nikolay Khabarovm,
- Fulco Ludwign,
- Yoshimitsu Masakio,
- Stefan Olinp,
- Cynthia Rosenzweigc,q,
- Alex C. Ruanec,q,
- Yusuke Satohr,
- Erwin Schmids,
- Tobias Stacket,
- Qiuhong Tangu, and
- Dominik Wisserv
-
Edited by Hans Joachim Schellnhuber, Potsdam Institute for Climate Impact Research, Potsdam, Germany, and accepted by the Editorial Board October 4, 2013 (received for review January 31, 2013)
Significance
Freshwater availability is
relevant to almost all socioeconomic and environmental impacts of
climate and demographic change
and their implications for sustainability.
We compare ensembles of water supply and demand projections driven by
ensemble
output from five global climate models.
Our results suggest reasons for concern. Direct climate impacts to
maize, soybean,
wheat, and rice involve losses of
400–2,600 Pcal (8–43% of present-day total). Freshwater limitations in
some heavily irrigated
regions could necessitate reversion of
20–60 Mha of cropland from irrigated to rainfed management, and a
further loss of 600–2,900
Pcal. Freshwater abundance in other
regions could help ameliorate these losses, but substantial investment
in infrastructure
would be required.
Abstract
We compare ensembles of water
supply and demand projections from 10 global hydrological models and six
global gridded crop
models. These are produced as part of the
Inter-Sectoral Impacts Model Intercomparison Project, with coordination
from the
Agricultural Model Intercomparison and
Improvement Project, and driven by outputs of general circulation models
run under
representative concentration pathway 8.5
as part of the Fifth Coupled Model Intercomparison Project. Models
project that direct
climate impacts to maize, soybean, wheat,
and rice involve losses of 400–1,400 Pcal (8–24% of present-day total)
when CO2 fertilization effects are accounted for or
1,400–2,600 Pcal (24–43%) otherwise. Freshwater limitations in some
irrigated
regions (western United States; China; and
West, South, and Central Asia) could necessitate the reversion of 20–60
Mha of
cropland from irrigated to rainfed
management by end-of-century, and a further loss of 600–2,900 Pcal of
food production.
In other regions (northern/eastern United
States, parts of South America, much of Europe, and South East Asia)
surplus water
supply could in principle support a net
increase in irrigation, although substantial investments in irrigation
infrastructure
would be required.
A lack of
available water for agricultural production, energy projects, other
forms of anthropogenic water consumption, and
ecological use is already a major issue in
many parts of the world and is expected to grow all of the more severe
with increasing
population, higher food (especially meat)
demand, increasing temperatures, and changing precipitation patterns.
Although population
growth is generally expected to slow in the
coming decades, median forecasts typically assume that the world
population will
grow close to another 50% above the recent
milestone of 7 billion people (1). Compounding population growth are major changes to diet as rapid economic growth in much of the developing world leads
to increased wealth and demand for more processed food and animal proteins in consumer diets (2, 3).
At the same time that demand for food and animal feed is increasing at a
historic pace, countries are also increasingly
turning to agricultural commodities as a
solution to high fuel prices, energy security, and growing carbon
dioxide (CO2) emissions. Population growth adds further
stress by taking land out of agriculture for urban development. For
example, between
1982 and 2007, about 9.3 Mha of US
agricultural land were converted for development (about 1 ha every 2
min) (4).
As the availability of land for agricultural uses continues to stagnate
or even decline, focus has shifted to increased
land-use intensification and improved
management to increase yields on existing lands to meet demand
challenges and moderate
some fraction of the negative impact of
climate change (5⇓–7).
Irrigation is of paramount
importance to increasing productivity on existing agricultural lands,
and projected per-hectare
irrigation consumption is thus an important
output of global gridded crop models (GGCMs). Irrigation is also by far
the largest
component of anthropogenic demand for fresh
water and as such constitutes an essential part of the global
hydrological cycle
and thus of global hydrological model (GHM)
simulations [Haddeland et al. (8),
in this issue of PNAS]. Projected potential irrigation water
consumption by crops and managed grasses (henceforth “PIrrUse”)
is thus a rare overlap among typical GHM and
GGCM outputs. The coordinated multisector, multimodel ensembles created
in the
Inter-Sectoral Impact Model Intercomparison
Project (ISI-MIP) hence allow for not only comparison among distinct
models within
their respective sectors, but also for a
direct comparison between GHMs and GGCMs. Several studies have evaluated
the potential
impacts of future climate change on
irrigation water requirements (9, 10) and the extent to which irrigation may aid adaptation to adverse climatic change effects (5, 6). However, these studies were constrained to a single GHM or GGCM only.
The objectives of the present analysis are to (i) compare projections of PIrrUse between GHMs and GGCMs—with and without the effects on plants of increasing atmospheric CO2 ([CO2])—and (ii) estimate an upper bound for the future availability of renewable fresh water for irrigation using combined projections of
water supply from 10 GHMs (11⇓⇓⇓⇓⇓⇓⇓⇓–20) and irrigation water demand (IWD) from both GHMs and 6 GGCMs (11, 21⇓⇓⇓⇓–26) run as part of the Agricultural Model Intercomparison and Improvement Project [AgMIP (27)] and ISI-MIP [see SI Appendix, Tables S1–S5 for a summary of participating models and institutional contacts; see also Schewe et al. (28) in this issue of PNAS for a description of the GHMs and simulations and Rosenzweig et al. (29) also in this issue of PNAS for a detailed description of the GGCMs and simulations] to (iii) evaluate the potential impacts of (limited) irrigation water availability on future crop productivity and (iv)
characterize the uncertainty in projections of global potential for
irrigation-based adaptation by analyzing a consistent
cross-sectoral ensemble of 5 GCMs × 10 GHMs ×
6 GGCMs. We identify geographic regions in which a combination of
decreased
water availability and/or increased demand
may reduce water available for irrigation and thus further impact
agricultural
production beyond what is otherwise expected
from climate change, as well as regions with potential for climate
change adaptation
via intensified irrigation.
Results and Discussion
Irrigation Water Consumption in GHMs and GGCMs.
Global PIrrUse on cropland currently equipped for irrigation (30) is projected to evolve in the future with climate change (Fig. 1).
We find notable differences between projections of PIrrUse obtained
from GGCMs and GHMs that could have a material effect
on our assessment of irrigation’s
potential contribution to future yield growth and climate adaptation.
Without the effects
of increasing [CO2], GGCMs generally estimate flat or increasing consumption for PIrrUse on present irrigated area, but the trend is far less
than the strong positive trend seen in GHMs. When the effects of increasing [CO2]
are included in GGCMs, these models project a decrease in global
irrigation consumption on presently irrigated area from
8% to 15% by end of century, similar to
results found for an ensemble of GCMs by Konzmann et al. (10) based on a single model (LPJmL; highlighted in Fig. 1
for the present scenarios). With the exception of LPJmL
(Lund-Potsdam-Jena Managed Land Dynamic Global Vegetation and Water
Balance Model), which is a GHM that
includes detailed dynamic representations of plant and crop processes,
the hydrological
models did not consider the effects of
increasing [CO2] on plants. As all models are driven by the
same climate scenario data, this conflicting behavior between model
types must
stem from different representations of
agricultural land and agrohydrological processes in GHMs and GGCMs.
Fig. 1.
Comparison
of fractional change relative to the model specific average 1980–2010
baseline of projections of total global PIrrUse
in RCP 8.5 from all GCM × GHM
combination and all GCM × GGCM combination with and without the effects
of increasing [CO2]. Results from LPJmL (mean over all GCMs) with (gray dots) and without (black dots) increasing [CO2]
are shown explicitly. LPJmL is unique in that it falls into both GHM
and GGCM categories and is unique among the GHMs in
that it provides estimates for
PIrrUse both with and without the effects of increasing [CO2].
Each GHM and GGCM uses an individual mix of explicitly represented land use types. Projections of global total PIrrUse for
individual GGCMs combine results for crops not explicitly represented (SI Appendix, Fig. S1).
All GGCMs considered here simulate dynamic phenology, which accelerates
growing seasons in response to warmer climates
if there is no adjustment in management
(i.e., static sowing dates and varieties). The shortening of the period
for which
irrigation water is needed can decrease
projected consumption. Dynamic phenology is implemented in some GHMs
(e.g., LPJmL
and H08) and indeed substantial
differences in the representations of agricultural land and plant types
explain part of the
broad range of trends in GHM
projections of PIrrUse (SI Appendix, Fig. S2). See Haddeland et al. (8) in this issue of PNAS for a more detailed description of GHM representations of agricultural land and irrigation.
Reduced PirrUse from shortened cropping cycles in GGCMs is compounded by the effects of increasing [CO2]
on water use efficiency. These two mechanisms partially counteract or
even reverse increasing potential evapotranspiration
and temporal and spatial declines in
precipitation. The latter effects are the dominant drivers in irrigation
water consumption
projections of models with static
cropping period assumptions.
GHM and GGCM projections of
irrigation water consumption are both the results of simplified
representations of the complexity
of existent irrigation systems. GGCMs
here represent only single-cycle cropping systems with simple
parameterizations of irrigation
events (SI Appendix, Tables S3 and S4),
whereas regions with irrigation agriculture often cultivate multiple
cropping cycles within a year, especially at low latitudes
where no seasons are threatened by
frost. Similarly, farmers are likely to adapt to the acceleration of
maturation in single-cycle
systems by using slower maturing
varieties. This effect, along with other adaptation strategies, was
excluded in the GGCM
model setup here for most model runs,
as it complicates the analysis and attribution of climate change
impacts. GHMs on the
other hand generally ignore the effects
of increasing [CO2] on crop water use efficiency, and those with static cropping seasons likely overestimate the increase in irrigation water
consumption, especially in regions with strong seasonality in temperature (31).
The differences in
crop-specific irrigation water consumption as simulated by the GGCMs
highlight the importance of a more
complex representation of agricultural
dynamics and crop types. For example, in some GHMs that include
representations of
a limited set of crop types (e.g.,
LPJmL), crops not explicitly represented are assumed to behave like
perennial grasses with
regard to transpiration and irrigation
consumption. Given extreme differences in the projected trend of PIrrUse
for grasses
and most annual crops (SI Appendix, Fig. S1, especially cotton and sugarcane), approximating row crops with perennial grasses can lead to substantive differences in
the overall global trend of irrigation.
Water Withdrawals and Availability.
We analyze the balance of irrigation water supply and demand at the level of food production units [FPUs, composites of river
basins and economic regions following Cai and Rosegrant (32) with modifications by Kummu et al. (33); SI Appendix, Fig. S3]. We estimate potential irrigation water withdrawal or demand (PIrrWW) from PIrrUse based on average current irrigation project
efficiencies from Rost et al. (34) and assume that freshwater is freely distributable within FPUs without substantial transportation costs. These large-scale
assumptions average significant spatial variability in infrastructure availability (35) and water policy (36)
at the local level which may substantially reduce the amount of water
available (especially for new irrigation projects)
in practice. For these reasons we
consider the resulting estimates of water availability for irrigation as
upper bounds in
most FPUs. We account for
environmental flow requirements and the limits from seasonal
distribution by assuming an upper availability
of 40% of total annual blue water
supply (SI Appendix, Figs. S4 and S5)
and subtract water consumption for other sectors as projected by The
WaterGAP model (Water – A Global Assessment and Prognosis)
[SI Appendix, Fig. S5 and Flörke et al. (37)] from the available water, assuming that irrigation water always has the lowest priority of all water consumers (which is
almost always the case).
Irrigation Potential and Constraints.
If PIrrWW in a currently irrigated area is projected to be greater than or equal to the projected available renewable water,
the agricultural production in that FPU is irrigation constrained (denoted by red in Fig. 2). If projected PIrrWW is less than the projected available renewable water, the FPU has an irrigation adaptation potential
equal to the difference (green in Fig. 2). As the major uncertainty of these FPU-balances lies in the different assessment of IWD in GHMs and GGCMs, we consider two
distinct scenarios for this input: (i) the median of all GCM × GHM combinations (IWDhydro; set represented by gray bars in Fig. 1) and (ii) the median of all GCM × GGCM combinations (IWDcrop; set represented by yellow bars in Fig. 1). Fig. 2
summarizes the spatial patterns of water availability/deficiency for
these two scenarios at the FPU level. In general, ensemble
elements within the IWDhydro scenario
show higher baseline irrigation demand in most FPUs, less water
available for the expansion
of irrigation, and more FPUs requiring
contraction of irrigated areas with especially notable differences
across the western
United States, Mexico, and much of
Asia. Even though estimates of total projected irrigation consumption
differ substantially
in an absolute sense between the crop
and water models, the spatial patterns of consumption are similar (SI Appendix, Figs. S6 and S7).
Fig. 2.
Median potential end-of-century renewable water abundance/deficiency in average cubic kilometers per year under RCP 8.5 for
(Left) all GCM × GHM combinations (IWDhydro scenario) for both supply and demand and (Right)
using all GCM × GGCM combinations for irrigation demands (IWDcrop
scenario). Positive values indicate areas with irrigation
adaptation potential and negative
values indicate irrigation constrained areas. Dark green FPUs are
saturated at 50 km3/y.
Agricultural Potential with Irrigation and Climate Adaptation.
We
used the GGCM simulations to derive the possible future yield increase
due to conversion of rainfed cropland to irrigated
cropland in FPUs with irrigation
adaptation potential, and, similarly, the possible future yield decrease
due to conversion
of irrigated cropland to rainfed in
FPUs that are projected to be irrigation constrained. The magnitude of
these effects is
determined by the level of water
limitations in rainfed agriculture (sustained only by green water, i.e.,
on-field precipitation
and soil moisture). Consequently,
semiarid regions where crops are currently cultivated under rainfed
conditions typically
show the greatest yield increase under
irrigation (Fig. 3). It is apparent by comparison with Fig. 2
that many regions with the largest potential for yield increases from
increased irrigation are also those most likely to
have binding constraints on water
availability. For maximum consistency with the assumptions of the GHMs
and GGCMs used to
construct the two scenarios of
irrigation water availability/deficiency in Fig. 2, we combine irrigation scenario IWDhydro with the climate impacts and per-hectare irrigation-based yield improvements without
the effects of increasing [CO2] and scenario IWDcrop with the production factors with increasing [CO2].
These choices also lead to scenarios that better span the space of
possible future trajectories of climate impacts and
irrigation-based adaptation, as the
more optimistic/pessimistic water availability scenario
(IWDcrop/IWDhydro) is combined
with the more optimistic/pessimistic
climate impact scenario (with/without the projected beneficial effects
of increasing
[CO2]). Irrigation-based yield improvement factors for scenarios without the effects of increasing [CO2] are very similar to those in Fig. 3.
Fig. 3.
Median potential per hectare increase in maize (Upper Left), wheat (Upper Right), soybean (Lower Right), and rice (Lower Left) yields at the end-of-century from irrigation applied on what are currently rainfed areas for scenarios with the effects
of increasing atmospheric CO2 concentrations included. Maps show median values across all 30 GCM × GGCM combinations in the ensemble for RCP 8.5.
When assuming maximum conversion of rainfed cropland to irrigated cropland in FPUs with irrigation adaptation potential and
reduced irrigation water use in irrigation constrained FPUs (Fig. 2), total caloric production of maize, soybean, wheat, and rice is changed regionally (Fig. 4) according to the projected yield increases under irrigation in Fig. 3. The two scenarios (IWDhydro and IWDcrop) are similar, although differences in the western breadbasket of the United States
(most notably the Missouri River Basin) and throughout much of China are significant.
Fig. 4.
Potential
change in total production of maize, soybean, wheat, and rice at
end-of-century given maximal use of available water
for increased/decreased
irrigation use on what are currently rainfed/irrigated areas in total
calories. (Left) Median of 156 GCM × GHM × GGCM combinations for scenarios constructed using GHM estimates of present-day irrigation demand.
(Right) Median of 202 GCM × GHM × GGCM combinations for scenarios constructed using GGCM estimates of present-day irrigation demand.
Global Adaptation Potential and Uncertainties.
Aggregated
globally, expansion of irrigation agriculture has the potential to
increase production on current cropland. However,
model projections indicate that even
under the most optimistic assumptions about freshwater distribution and
transportation
within FPUs, the beneficial effect
would be exhausted by detrimental climate change effects on crop yields
by 2070 at the
latest, for irrigation scenario IWDcrop
and crop yields estimated with the inclusion of the effects of
increasing [CO2] (Fig. 5). By 2090, 57% of the median 730-Pcal reduction due to climate change with effects of increasing [CO2]
could be ameliorated by the net expansion of irrigation according to
the more optimistic irrigation scenario (IWDcrop).
Under the more pessimistic irrigation
scenario (IWDhydro), the limitations on irrigation water supply
availability further
constrain the potential ameliorating
effect of expanded irrigation to only 12% of the 1,840-Pcal reduction in
2090 due to
climate change without effects of
increasing [CO2], highlighting the need to improve
agricultural productivity by other means. This general mechanism is
valid for all GCM
× (GGCM or GHM) combinations, although
there is considerable variation among the projections of individual
ensemble members
(Fig. 5).
Fig. 5.
Comparison
of the total annual global calories of maize, soybean, wheat, and rice
for RCP 8.5 as projected by four sets of
ensemble simulations. The first
two sets assume no change in irrigated areas and consist of (i) 30 GCM × GGCM combinations with CO2 effects and (ii) 22 GCM × GGCM combinations without CO2 effects. The second two sets consist of (iii) 202 GCM × GHM × GGCM combinations with CO2 effects and a global net expansion in irrigated areas according to the IWDcrop scenario, and (iv) 156 GCM × GHM × GGCM combinations without CO2 effects and a global net expansion in irrigated areas according to the IWDhydro scenario.
Our analysis is subject to
considerable uncertainties which we address in part here. Agricultural
PIrrUse and corresponding
increases in productivity have been
simulated by the GGCMs only for irrigation management with a 100%
saturation threshold
for applications (i.e., once an
irrigation event is triggered, water is applied until soil moisture is
optimal; SI Appendix, Tables S3 and S4). Because the efficiency of irrigation water use (yield per unit water) declines at higher irrigation levels (38),
water sharing and deficit irrigation could have an overall beneficial
effect in constrained regions. Another source of
uncertainty relates to our assumptions
regarding fossil groundwater availability. Our results indicate that
many regions with
high shares of irrigated agriculture
are likely to be constrained by future freshwater availability. Because
we are concerned
with the long-term sustainable supply
of freshwater, we assume no water supply from fossil groundwater. This
is consistent
in some areas with the observed
depletion of (fossil) groundwater reserves (e.g., ref. 39), but disregards the time it will take to fully deplete these resources and the possibility that aquifers may expand across
FPUs and thus contribute to a better distribution of irrigation water in space and time.
Our assumption of 40%
freshwater availability is a valid threshold for maximum runoff
extraction at global scale, but may
be high or low in specific river
basins, for example, where irrigation infrastructure is prohibitively
expensive, those in
which periods of inundation are needed
for the functioning of riparian ecosystems, or those where flushing of
solid waste
and sediment is essential for stream
flow, water quality control, or denitrification. Regions with irrigation
constraints
may need to explore options to increase
irrigation project efficiency, which can easily double the irrigation
water supply
(34). This need for improved irrigation efficiencies is also generally true if irrigation is to play a role in reducing detrimental
climate change impacts on agricultural productivity (Fig. 5).
The effectiveness of CO2 fertilization is a source of major uncertainty, with respect to not only crop productivity [Deryng et al. (31) in this issue on PNAS] but also IWD (10). It may be the only mechanism that can alleviate some climate change impacts on agricultural irrigation water consumption
and crop yields (Fig. 5),
which otherwise decline rapidly with increasing temperatures. There are
additional socioeconomic issues associated with
irrigation consumption that we do not
address here. Whereas it may be technically possible to increase yields
by a relatively
small 5–10% per year in the eastern
United States and across much of Europe through irrigation, for example,
it may not be
economical to do so in practice due to
the cost of irrigation relative to the potential increase in production.
Additional
socioeconomic issues such as
transboundary disputes on appropriate river discharge rates will
continue to be a problem in
many arid regions.
Conclusions
We demonstrate
in a unique and broad model intercomparison across two different but
closely interrelated impact sectors that
a conversion of currently rainfed cropland
to irrigated cropland (to the extent possible given actually available
water resources)
would be insufficient to compensate
detrimental climate change impacts on current agricultural land. The
main drivers of this
effect are projected water limitations,
mainly in regions with already large fractions of irrigated agriculture,
and the detrimental
effects of climate change on agricultural
productivity. Both those regions that are projected to suffer water
limitations
and those that are projected to have
potential to expand irrigation could benefit from reduced water losses
in conveyance
and application and also from better-tuned
deficit irrigation to increase overall efficiency of irrigation water
use. Depending
on local conditions, increases in
irrigation capacity and efficiency need to be complemented by efforts to
increase water
use efficiency and soil conservation in
rainfed systems as well, which have a demonstrated capacity to boost
crop yields without
further exploiting freshwater resources in
rivers and aquifers (40).
Further efforts to increase productivity, including other means of
intensification, water saving, and land-use/land-cover
change are needed to close what is
projected to be a growing gap between agricultural production on current
cropland under
climate change and increasing demand for
agricultural commodities. Effective climate mitigation must also be
among the foremost
measures to maintain current productivity
on rainfed and irrigated land.
Uncertainties in these
projections that result from our crop and hydrology models are generally
somewhat higher than those
that result from the five climate models
that we use to drive the impact models, but the ensemble overwhelmingly
supports
the general conclusions. Nevertheless,
impact model differences need to be better understood especially with
respect to their
implications for manageability of water
consumption and climate change impacts.
Materials and Methods
Throughout this analysis we used downscaled, bias-corrected outputs of five GCMs from the Fifth Coupled Model Intercomparison
Project [CMIP5 (41)] summarized in the SI Appendix, Table S2. See Hempel et al. (42) for a discussion of the bias correction approach and Hagemann et al. (43) for a discussion of the impact of using bias corrected climate model output with GHMs. For simplicity we have considered
only a single representative concentration pathway [RCP 8.5 (44)] throughout this analysis.
Water Availability.
To
calculate water availability (blue water potentially available for
irrigation) for each of 309 FPUs, we use simulated runoff
provided by each GHM at grid cell
level. Thus, we only consider the renewable surface water, including
subsurface runoff,
assuming that no fossil groundwater
is available. Note that due to lateral water transport along river
networks, the blue
water available within an FPU may
stem from adjacent FPUs that are (partly) located in the same river
basin. To take this
factor into account, we distributed
the overall runoff within river basins according to the average
discharge rates (taken
from the GHMs) and then aggregated
for each FPU. In addition, we assumed that only up to 40% of the thus
computed renewable
water is available for human use, so
as to account for environmental flow requirements in rivers and to stay
below thresholds
of water stress detrimental to
ecosystems and human society [following Gerten et al. (45)].
We assumed that a part of the renewable water resource is consumed for
nonagricultural purposes before, and irrespective
of, the crop IWD. Note that instead
of water withdrawal we consider water consumption, i.e., the amount of
water that is actually
lost from the system (whereas a part
of the withdrawn water remains available for downstream users due to
return flows to
the rivers).
WaterGAP Estimates for Domestic and Industrial Water Use.
We estimated spatially distributed present and future total water withdrawals for the four nonagricultural water use sectors:
domestic, manufacturing, thermoelectricity, and livestock (37).
We calculated country-wide estimates of future water use (water
withdrawals and consumption) in the manufacturing and domestic
sectors based on socioeconomic
projections following the Shared Socio-Economic Pathway 2
middle-of-the-road scenario [SSP
Database (46)] (47).
To determine the amount of cooling water withdrawn for thermal
electricity production, we multiplied for each power station
its annual thermal electricity
production by its water use intensity. Future projections of thermal
electricity production
were calculated with the Integrated
Model to Assess the Global Environment (IMAGE) model (48). Input data on location, type, and size of power stations were based on the World Electric Power Plants Data Set (49).
The water use intensity is impacted by the cooling system and the
source of fuel of the power station. We distinguished
four types of fuels (biomass and
waste; nuclear; natural gas; and oil, coal, and petroleum) with three
types of cooling systems
[tower cooling, once-through
cooling, and ponds (total nonagricultural water withdrawals summarized
in SI Appendix, Fig. S2)] (50).
Maximum Agricultural Potential with Irrigation.
To
understand the implications of changed irrigation water use for the
balance of water consumption and freshwater supply,
we translate estimates of PIrrUse
into the total PIrrWW based on the current irrigation project
efficiencies from Rost et
al. (34),
in which countries are estimated to have a total irrigation efficiency
(conveyance plus application) ranging from 0.294
to 0.855. We define maximum
agricultural potential with irrigation in an FPU to be total production
assuming that all water
available for irrigation is used.
For this analysis we consider 16 of the most important global crop types
(including grass/pasture).
Because of the extreme diversity of
global agriculture, however, it is not possible to include all crops
that are important
for irrigation in all regions. In
total, the 16 crops simulated by at least one GGCM account for 85.5% of
the global irrigated
areas recorded in MIRCA2000 (monthly
irrigated and rainfed crop areas around the year 2000) (SI Appendix, Fig. S1 and Table S5).
We consider expansion/contraction only in those agriculture lands used
for the four main staple food and feed crops in
the world: maize, wheat, soybean,
and rice. For all other crops, we assumed that irrigated areas remain
fixed at present-day
levels. If an FPU is deemed
irrigation constrained in a given year (for a given element in the GCM ×
GHM × GGCM ensemble)
we assume that a fraction of the
cropland that is equipped for irrigation must go without, producing
yields according to the
rainfed estimate for that area
instead. Thus, to reduce the irrigation demand by an amount T cubic
meters in an FPU where
the average water demand for
irrigated areas in the given year is D cubic meters per hectare, the
amount of land irrigated
must be reduced by (T/D) hectare.
Given an average irrigated yield of YI tonnes per hectare in the given FPU and an average rainfed yield of YR tonnes per hectare, the loss of production is thus 
.
We do not address here the possibility of imperfect or deficit
irrigation (i.e., that all areas equipped for irrigation
receive something less than 100% of
the demanded water, rather than some receiving zero). If more than a
single crop is irrigated
in a given FPU, we assume that the
same economic and cultural considerations that affect the present-day
distribution of irrigation
will control how changes in
irrigated area are distributed; i.e., the fraction of area irrigated for
each crop will remain
fixed at historical levels. If, on
the other hand, an FPU is deemed to have some irrigation adaptation
potential, then we
define the agricultural potential
with irrigation to be the total production in that FPU with some
fraction of the rainfed
areas converted for irrigation. Here
again we assume that increased irrigation water is distributed evenly
according to the
distribution of present-day areas
equipped for irrigation. We have not allowed for any land-cover change
(e.g., crop switching
or an increase in the total
harvested area for a given crop in a region) in this analysis.
.
We do not address here the possibility of imperfect or deficit
irrigation (i.e., that all areas equipped for irrigation
receive something less than 100% of
the demanded water, rather than some receiving zero). If more than a
single crop is irrigated
in a given FPU, we assume that the
same economic and cultural considerations that affect the present-day
distribution of irrigation
will control how changes in
irrigated area are distributed; i.e., the fraction of area irrigated for
each crop will remain
fixed at historical levels. If, on
the other hand, an FPU is deemed to have some irrigation adaptation
potential, then we
define the agricultural potential
with irrigation to be the total production in that FPU with some
fraction of the rainfed
areas converted for irrigation. Here
again we assume that increased irrigation water is distributed evenly
according to the
distribution of present-day areas
equipped for irrigation. We have not allowed for any land-cover change
(e.g., crop switching
or an increase in the total
harvested area for a given crop in a region) in this analysis.
Acknowledgments
We acknowledge the World Climate Research Programme’s Working Group on Coupled Modeling, which is responsible for CMIP, and
we thank the climate modeling groups (SI Appendix, Table S2)
for making their outputs available. For CMIP, the US Department of
Energy’s Program for Climate Model Diagnosis and Intercomparison
provides coordinating support in
partnership with the Global Organization for Earth System Science
Portals. The ISI-MIP Fast
Track project was funded by the German
Federal Ministry of Education and Research (BMBF) with Project Funding
Reference 01LS1201A.
This work was also supported in part by
the National Science Foundation (NSF) under Grants SBE-0951576 and
GEO-1215910. The
research leading to these results has
received funding from the European Union’s Seventh Framework Programme
FP7/2007-2013
under Grant Agreement 266992. Computing
was provided by a number of sources, including the University of Chicago
Computing
Cooperative, the University of Chicago
Research Computing Center, and through the National Institutes of Health
with resources
provided by the Computation Institute and
the Biological Sciences Division of the University of Chicago and
Argonne National
Laboratory, under Grant S10 RR029030-01.
Part of the computing was facilitated using the Swift parallel scripting
language,
supported in part by NSF Grant
OCI-1148443. S.N.G. was supported by a Science, Technology, and Society
Priority Group grant
from the University of Nottingham. Y.M.
was supported by the Environment Research and Technology Development
Fund (S-10) of
the Ministry of the Environment, Japan.
Q.T. is supported by the 973 Program of China (2012CB955403). S.O.
acknowledges support
by the Formas Strong Research Environment
“Land Use Today and Tomorrow.” This work has been conducted under the
framework
of ISI-MIP and in partnership with the
AgMIP community. K.F. was supported by Federal Ministry for the
Environment Grant 11
II 093 Global A SIDS and LDCs.
Footnotes
- 1To whom correspondence should be addressed. E-mail: jelliott@ci.uchicago.edu.
-
Author contributions: J.E., K.F., D.G., C.R., and A.C.R. designed research; J.E., D.D., C.M., M.K., D.G., M.G., M.F., Y.W., N.B., S.E., B.M.F., C.F., S.N.G., I.H., N.K., F.L., Y.M., S.O., Y.S., E.S., T.S., Q.T., and D.W. performed research; J.E., D.D., C.M., D.G., M.F., Y.W., S.E., C.F., S.N.G., I.H., N.K., F.L., Y.M., S.O., Y.S., E.S., T.S., and Q.T. contributed new analytic tools; J.E., K.F., and M.K. analyzed data; and J.E., D.D., C.M., and I.F. wrote the paper.
-
The authors declare no conflict of interest.
-
This article is a PNAS Direct Submission.
-
Data deposition: The data reported in this paper associated with the Inter-Sectoral Impact Model Intercomparison Project is hosted at http://esg.pik-potsdam.de.
-
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1222474110/-/DCSupplemental.
References
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
First look at changes in flood hazard in the Inter-Sectoral Impact Model Intercomparison Project ensemble
- Rutger Dankersa,1,
- Nigel W. Arnellb,
- Douglas B. Clarkc,
- Pete D. Falloona,
- Balázs M. Feketed,
- Simon N. Goslinge,
- Jens Heinkef,g,
- Hyungjun Kimh,
- Yoshimitsu Masakii,
- Yusuke Satohj,
- Tobias Stackek,
- Yoshihide Wadal, and
- Dominik Wisserm,n
-
Edited by Hans Joachim Schellnhuber, Potsdam Institute for Climate Impact Research, Potsdam, Germany, and accepted by the Editorial Board August 31, 2013 (received for review January 31, 2013)
Abstract
Climate change due to anthropogenic greenhouse gas emissions is expected to increase the frequency and intensity of precipitation events, which is likely to affect the probability of flooding into the future. In this paper we use river flow simulations from nine global hydrology and land surface models to explore uncertainties in the potential impacts of climate change on flood hazard at global scale. As an indicator of flood hazard we looked at changes in the 30-y return level of 5-d average peak flows under representative concentration pathway RCP8.5 at the end of this century. Not everywhere does climate change result in an increase in flood hazard: decreases in the magnitude and frequency of the 30-y return level of river flow occur at roughly one-third (20–45%) of the global land grid points, particularly in areas where the hydrograph is dominated by the snowmelt flood peak in spring. In most model experiments, however, an increase in flooding frequency was found in more than half of the grid points. The current 30-y flood peak is projected to occur in more than 1 in 5 y across 5–30% of land grid points. The large-scale patterns of change are remarkably consistent among impact models and even the driving climate models, but at local scale and in individual river basins there can be disagreement even on the sign of change, indicating large modeling uncertainty which needs to be taken into account in local adaptation studies.Climate change due to anthropogenic greenhouse gas emissions is expected to change the distribution and variability of precipitation (1, 2). As the water holding capacity of the atmosphere is increasing with temperature, global warming is expected to increase the intensity of precipitation events (3⇓–5). Such an increase in precipitation intensity is likely to have an impact on the frequency of (pluvial and fluvial) flooding.Although an increase in flooding under climate change is often speculated on, few studies have actually provided projections of fluvial flooding at a continental or global scale (for examples see refs. 6⇓⇓–9). Most studies that have appeared to date focus on individual river basins or countries (e.g., refs. 10⇓⇓–13), and the use of different models and scenarios makes it impossible to generalize the results, to compare regional sensitivities, or to make global-scale risk assessments. Assessing the impact of climate change on fluvial flooding is furthermore hampered by biases and uncertainties in both climate and impact models.The InterSectoral Impact Model Intercomparison Project (ISI-MIP, ref. 14) provides the opportunity to explore these uncertainties in projections of changes in flood hazard at a global scale. ISI-MIP offers a framework for comparing multiple climate impact models within and across different sectors, based on consistent climate and (where appropriate) socioeconomic scenarios, providing a quantitative estimate of impacts and uncertainties. The fast-track phase of the project involves 34 global impact models across 5 sectors. The aim of this paper is to look specifically at the relative importance of climate and impact model uncertainty in projected changes of flood characteristics, as well as the consistency in these projections. We use results from nine global hydrology and land surface models (hereafter referred to as impact models, IMs) that provided simulations of daily river discharge at a global 0.5° grid (Table 1). Each IM was driven by five global climate models (GCMs) that were selected to represent the range of global mean temperature change and relative precipitation changes in the Coupled Model Intercomparison Project Phase 5 (CMIP5) simulations (see ref. 14 for further details). In this paper we only include simulations using representative concentration pathway RCP8.5, giving us an ensemble of 45 experiments, each consisting of a historical and future period.View this table:
Table 1.
Overview of models used in this paperAs an indicator of flood hazard we estimated the 30-y return level of river flow (Q30) at each grid cell for two 30-y periods (1971–2000 and 2070–2099). The Q30 is a moderately extreme discharge level that will be exceeded only very infrequently. Whereas the probability of exceedance (Pe) in any given year is 1/30, in any given 10-y period it amounts to almost one-third (0.29). In flood risk management it is therefore common to take even higher flow levels (such as the 100-y return level) as a threshold for protection measures or planning regulations.Results
The large-scale pattern of regions with projected increases and decreases in flood hazard is remarkably robust across the IMs and even the GCMs. Fig. 1 (Upper) shows the number of experiments that agree on an increase or decrease in the magnitude of Q30 by more than 10%. The magnitude of change in each GCM/IM combination is shown in Figs. S1–S5). Broadly speaking, increases in flood hazard are projected consistently in central and eastern Siberia and parts of Southeast Asia, including India. Likewise, the Q30 is generally projected to decrease in northern and eastern Europe, and parts of northwestern North America. The relatively large agreement between GCMs may be the result of consistent changes in flood-generating precipitation (2), as well as similar responses to change in the IMs. It cannot be ruled out that the bias correction applied to the climate simulations also has an influence, but this needs further investigation.Fig. 1.
(Upper) Number of experiments (out of 45 in total) showing an increase (Left) or decrease (Right) in the magnitude of Q30 of more than 10% in 2070–2099 under RCP8.5, compared with 1971–2000. (Lower Left) Average change in the magnitude of Q30 across all experiments. (Lower Left) Ratio of GCM variance to IM variance. GCM variance was computed as the variance of the change in Q30 across all GCMs for each individual IM, and then averaged over the nine IMs; IM variance was computed as the variance of the change in Q30 across all IMs for each individual GCM, and then averaged over the nine GCMs. In dark green (purple) areas GCM (IM) variance predominates.The pattern described above resembles to some extent the findings of earlier studies (6, 9) that were based on a much smaller number of models. Both these studies noted that especially at northern higher latitudes, decreases in flood hazard can occur even in areas where the intensity and/or frequency of precipitation events is projected to increase, and attributed this to a decrease in snow accumulation in winter under warmer climate conditions. Some high-latitude areas in Fig. 1, however, show a consistent increase in Q30 magnitude, tentatively reflecting an increase in snow accumulation over winter. Clearly, in rivers where the annual hydrograph is dominated by snowmelt any changes in the magnitude of the peak flows depend on a fine balance between warmer and possibly shorter winters, and increases or decreases in the total amount of precipitation during the winter season. Similar patterns were also found for northern Europe by refs. 7 and 8.Although the global pattern of changes in flood hazard shows a relatively large degree of consistency across the experiments, at the scale of individual river basins there can be important differences, and projections can disagree even on the sign of change. This can be seen for example in the River Nile, where IM simulations driven by the Geophysical Fluid Dynamics Laboratory Earth System Model (GFDL-ESM2M) show a relatively robust increase in Q30 magnitude, whereas the Hadley Centre Global Environment Model version 2 - Earth System (HadGEM2-ES) simulations generally show a decrease. This highlights the fact that at regional and local scale the impact of climate change on fluvial flood hazards can be highly uncertain.This uncertainty has various causes, including climate and impact modeling uncertainty. An estimate of GCM and IM uncertainty can be obtained by looking at the variance across both types of models (Fig. 1, Lower). Especially (but not exclusively) in areas with snowmelt, IM uncertainty can predominate over GCM uncertainty, and outside the tropics there are few areas where GCM uncertainty is considerably larger than the uncertainty brought about by the different IMs.Changes in the magnitude of the Q30 translate into an increase or decrease of the probability of occurrence of the historical 30-y discharge level. Fig. 2 summarizes the extent that the return period of the historical Q30 is expected to become longer or shorter, indicating a decrease or increase in the probability of flooding, respectively. The corresponding plots for each GCM/IM combination can be found in Figs. S6–S10). The results suggest that under these climate scenarios the historical Q30 will occur more frequently in large parts of the world: whereas at roughly one-third of the land grid points of the IMs the future return period is projected to lengthen to more than 40 y, at around 40–60% it is estimated to be less than 20 y and at 20–45% it is less than 10 y, suggesting that here the historical Q30 discharge level is 3 times more likely to occur by the end of this century. At a smaller fraction of land grid points (between 5% and 30%) the future return period is even estimated to be less than 5 y, which would mean that the current Q30 flow level can be expected to be exceeded every few years. The fraction of land where this occurs depends to some extent on the GCM/IM combination that is used, but it is worth noting that at the global scale the range in impact between IMs is of similar magnitude to the range between GCMs (Fig. 2).Fig. 2.
Fraction of land grid points where the estimated future return period of the historical Q30 is more than 40 y (red colors), less than 20 y (light green), less than 10 y (cyan), and less than 5 y (blue), grouped by driving GCM (horizontal axis). Results from individual impact models (denoted by their first three letters) are indicated with different symbols. Note that the exact number of land points can be different for each GCM/IM combination.Discussion
The results presented in the previous section suggest that under conditions of climate change flood hazard is not increasing uniformly across the globe. Decreases in the magnitude of Q30 occur particularly (but not exclusively) in areas where the hydrograph is dominated by snowmelt in spring. Nevertheless, in most model experiments an increase in flood hazard was found in more than half of the global land grid points. At a much smaller fraction (between 5% and 30%) the current 30-y flood peak is expected to occur more than once every 5 y in the future. This includes areas that are currently sparsely populated (e.g., eastern Siberia), but also some densely populated regions such as Southeast Asia. It should also be emphasized that these projections are for a high-end climate change scenario (RCP8.5), which means that a comparison with any observed trends (refs. 15 and 16) is not yet feasible.The large-scale pattern of changes in flood hazard presented above resembles, to some extent, those found in earlier studies (in particular refs. 6 and 9) that were based on a much smaller number of experiments. At the same time our analysis demonstrates that both climate and impact model uncertainty contribute to the spread in the projected changes, especially at the regional scale. A large contribution of IM uncertainty has also been found in analyses of the ISI-MIP hydrological simulations (e.g., refs. 17 and 18) Further work is necessary to better understand and explain some of these uncertainties, and tentatively relate them to model performance under present-day conditions. Note that IM uncertainty seems to be large especially in regions that are dominated by snowmelt, albeit not everywhere. Ref. 19 found that the scheme used for simulating snow is an important cause for differences in model performance between land surface and hydrological models, and potentially this also affects projections of future changes in peak flows in these areas.Also note that not all sources of uncertainty are taken into account in our analysis; for example, we have only looked at one scenario of future greenhouse gas concentrations. The subset of five GCMs used in ISI-MIP may not sample the full uncertainty in projections of extreme precipitation in the CMIP5 ensemble. Estimates of the Pe of extreme flow levels are inherently uncertain as well, especially for return periods beyond the length of the data (here 30 y). For a more extreme discharge level, ref. 8 found that natural variability in the climate simulations can yield differences in the estimated Pe even without a change in climate. Here this effect might be smaller because we use a less extreme flow level as indicator of flood hazard, but its exact importance may only be determined by including more ensemble members of the same climate model simulations, where available. We also have not explored to what extent the bias correction method that was applied to the GCM simulations (Materials and Methods) has influenced the results, but this is likely to add further uncertainty to the projections (20, 21). Finally, no attempt has been made here to weigh the models according to their skill in simulating present-day conditions.The analysis presented in this paper explores uncertainties in the potential impacts of climate change on flood hazard, and is not meant to provide specific guidance on adaptation to climate change in an individual river basin. The models used in this paper all have a global scope and may not provide an accurate description of the climatological and hydrological system at a given location. Vulnerability is also not taken into account in the current analysis (for examples of this see, e.g., refs. 22⇓–24). Actual changes in flood hazard and indeed flood risk may depend on socioeconomic factors such as land use changes, flow control structures, flood defenses, and changes in exposure, as much as climate change, and although these human factors may dominate the response in individual river basins, they could not be taken into account in a global-scale study like this. However, the results may have a clear implication for local adaptation: studies that are based on only one, or a limited set of model simulations, may underestimate the uncertainty in the climate change signal, and the corresponding change in flood hazard. It should also be noted that large uncertainty even about the direction of change does not equate to no change. It may be worth approaching this uncertainty from a risk management perspective: if all of the model simulations can be regarded as possible or even plausible scenarios of changes in flood hazard due to climate change, those model solutions that appear less likely but nevertheless carry high risk may not simply be ignored. In those regions where even the sign of changes is uncertain, adaptation plans may therefore need to be flexible to changes in both directions (25).Materials and Methods
The ISI-MIP Setup.
We used natural river flow simulations from nine global hydrology and land surface models (here referred to as IMs; Table 1). With the exception of PCRaster Global Water Balance Model (PCR-GLOBWB) and Water Balance Model (WBM), these models participated previously in the Water Model Intercomparison Project (WaterMIP), and more information on the model characteristics and general hydrological performance can be found in ref. 19. A more thorough evaluation of how well these IMs are able to reproduce observed flood characteristics and trends would be desirable, but is difficult to achieve due to a lack of suitable river flow data in many regions of the world, and the simple fact that many river systems are heavily influenced by human interference such as water extraction, reservoir construction, and structural engineering works. However, the models used here can be considered state-of-the-art in global hydrological simulations.Each IM was driven by five GCMs that were selected to represent the range of global mean temperature change and relative precipitation changes in the CMIP5 simulations (26); see ref. 14 for further details. Each GCM in turn was driven by up to four scenarios of future levels of atmospheric greenhouse gas concentrations [representative concentration pathways (RCPs); ref. 27]. In this paper we only include simulations using scenario RCP8.5, assuming that higher levels of global warming will result in stronger hydrological changes (although locally and regionally this may not always be the case, e.g., ref. 7). We thus have an ensemble of 45 experiments, each consisting of a historical and future period.The output of each GCM was bias corrected to ensure statistical agreement with the observational data set of ref. 28 over the period 1960–1999, and downscaled to the same 0.5 × 0.5° grid. The bias correction method that was used preserves the absolute trends in globally averaged temperature and relative trends in land-averaged precipitation (29), and is a modification of the transfer function approach that has been used in WaterMIP. Ref. 30 found that applying bias correction to the output from a regional climate model can improve the realism of subsequent hydrological simulations and result in a better reproduction of observed flood statistics. However, good model performance in the present day does not necessarily imply plausible projections into the future, and it has been argued that bias correction hides rather than reduces the uncertainty of the predictions (20). An implicit assumption in bias correction is that biases in the climate simulations remain the same into the future, which is probably not very realistic. Bias correction may therefore add further uncertainty to climate impact projections (e.g., ref. 21), but this is not explored further in this paper.Analysis of Flood Hazards.
For each experiment (that is, each GCM/IM combination), the time series of simulated daily river discharge at each land grid cell was smoothed to 5-daily running averages, which were thought to be more appropriate than daily values considering the typically generalized routing parameters used in global models. For each year, the annual maximum 5-daily flow was determined for both the historical and the future period. For each grid cell we thus have two distributions of 30 annual peak flows. To estimate the Q30 a generalized extreme value distribution (GEV) (31, 32) was fitted separately to these two sets of peak flows using a maximum likelihood approach. The goodness of fit was tested by calculating the Kolmogorov–Smirnov and Anderson–Darling statistics. Grid points with annual maxima close to 0 m3⋅s−1 were excluded from the analysis as these generally resulted in poor fits. This occurs mostly in desert areas, but not in all IMs. The total number of land points that was included in the analysis is therefore slightly different in each experiment. Finally, a likelihood ratio test (31) was performed to test whether the shape parameter of the GEV is significantly different from zero, or if a Gumbel distribution should be used instead. Note that the Q30 (technically the 31-y return level) can, in principle, also be estimated directly from a 30-y time series, but a GEV-based estimate is more robust as the shape of the distribution is based on all peak flows during this period.Changes in Q30 may be expressed as a change in magnitude between the historical and future period, suggesting higher or lower flood levels. Alternatively, the future return period (based on the future GEV) of the historical Q30 can be calculated, suggesting a change in probability and thus frequency. Although estimates of return levels can be uncertain, especially beyond the period of the data, it is important to keep in mind that any changes in the Q30 are caused by changes in the underlying set of peak flows between the historical and future period that determines the shape of the GEV.Acknowledgments
The constructive comments of two anonymous reviewers helped to improve the manuscript. R.D. thanks K. Williams, J. Davie, and R. Kahana for technical assistance. We acknowledge the World Climate Research Programme’s Working Group on Coupled Modelling, which is responsible for CMIP, and we thank the climate modeling groups (listed in Table 1) for producing and making available their model output. For CMIP the US Department of Energy’s Program for Climate Model Diagnosis and Intercomparison provides coordinating support and led development of software infrastructure in partnership with the Global Organization for Earth System Science Portals. The ISI-MIP Fast Track project was funded with Project Grant 01LS1201A by the German Federal Ministry of Education and Research. R.D. and P.F. were supported by the Joint UK Department of Energy and Climate Change/Department for Environment, Food and Rural Affairs Met Office Hadley Centre Climate Programme (GA01101). Y.M. was supported by the Environment Research and Technology Development Fund (S-10) of the Ministry of the Environment, Japan. H.K. and Y.S. were jointly supported by the Japan Society for the Promotion of Science “KAKENHI” (23226012) and the Japanese Ministry of Education, Culture, Sports, Science and Technology “SOUSEI” Program. This work has been conducted under the framework of ISI-MIP.Footnotes
- 1To whom correspondence should be addressed. E-mail: rutger.dankers@metoffice.gov.uk.
-
Author contributions: R.D. designed research; R.D., D.B.C., B.M.F., S.N.G., J.H., H.K., Y.M., Y.S., T.S., Y.W., and D.W. performed research; R.D., N.W.A., and P.D.F. analyzed data; and R.D. wrote the paper.
-
The authors declare no conflict of interest.
-
This article is a PNAS Direct Submission.
-
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1302078110/-/DCSupplemental.
References
HydrobiologiaThe International Journal of Aquatic Sciences
© Springer Science+Business Media Dordrecht 2013
10.1007/s10750-013-1780-6
Primary Research Paper
River of the dammed: longitudinal changes in fish assemblages in response to dams
Jonathan A. Freedman1, 3 , Benjamin D. Lorson2, Richard B. Taylor2, Robert F. Carline1 and Jay R. StaufferJr.2
(1)
Pennsylvania Cooperative Fish and Wildlife Research Unit, The Pennsylvania State University, University Park, PA 16802, USA
(2)
School of Forest Resources, The Pennsylvania State University, University Park, PA 16802, USA
(3)
Department of Biology, Stetson University, 421 N. Woodland Blvd., Unit 8264, DeLand, FL 32723, USA
Received: 20 July 2013Revised: 1 December 2013Accepted: 3 December 2013Published online: 13 December 2013
Handling editor: Katya E. Kovalenko
Abstract
Although dams are a common feature
on rivers throughout the world, their effects on diversity,
composition, and structure of fish assemblages are often unclear. We
used electrified benthic trawls and stable isotope analysis of δ13C and δ15N
to determine the complex relationships between taxonomic diversity and
food web structure of fish assemblages among sites in the free-flowing
and impounded reaches of the Allegheny River, Pennsylvania, USA. We
found higher gamma and beta fish diversity in the free-flowing section,
where Brillouin diversity increased in a downstream direction; however,
in the impounded section, we found decreasing diversity downstream.
Analysis of similarity and non-metric multi-dimensional scaling revealed
longitudinal differences in Bray–Curtis similarity between assemblages
from impounded and those from free-flowing sites. Finally, using stable
isotope analysis, we showed that fishes in the free-flowing section
derived nutrients primarily from benthic sources while fishes in the
impounded section had a stronger reliance on pelagic nutrients. Our
findings reveal that dams can reduce fish taxonomic diversity, driven
primarily by decreases in lotic taxa, while shifting resource use from
benthic toward pelagic nutrients. A multi-faceted approach to assess the
cumulative effects of dams on aquatic communities is, therefore,
recommended.
Keywords
Stable isotope analysis
Diversity partitioning
Community ecology
Food web
Nutrient dynamics
Impoundment
Introduction
The majority of large-river systems throughout the world are affected by dams (Nilsson et al., 2005)
for purposes that include flood control, hydroelectric power
generation, and facilitation of navigation or recreation. Irrespective
of their purpose, the presence of dams alters the natural flow of rivers
(Graf, 1999, 2006; Poff & Hart, 2002).
Nutrient and sediment dynamics are affected, as detritus and sediment
accumulate behind dams, thus becoming unavailable downstream (Kondolf, 1997; Vorosmarty et al., 2003; Graf, 2006). By altering flow, dams decrease the natural heterogeneity of rivers (Kondolf, 1997; Poff et al., 1997, 2007),
as pool and lentic habitats predominate, and the only proxy for riffle
habitats is usually immediately downstream of dams where turbulence and
oxygen content of the water can be relatively high (Ward & Stanford,
1983).
Altered flow regimes, and the transformation from lotic riffle-pool-run
sequences to lentic habitats, also leads to subsequent changes in
biotic assemblages (Power et al., 1996; Poff et al., 1997; Bunn & Arthington, 2002; Miranda et al., 2008).
Species adapted to fast-flowing water are especially susceptible to
such changes, while a variety of trophic shifts may occur with the
arrival and dominance of lentic species at multiple trophic levels (Poff
et al., 1997; Bunn & Arthington, 2002; Lytle & Poff, 2004).
In impounded reaches, aquatic
vegetation and periphyton can be negatively affected by higher turbidity
and sedimentation rates, and the subsequent reductions in light
penetration and changes in substrate composition (Rivier & Seguier, 1985; Poff et al., 1997).
Fish species in lithophilic reproductive guilds, such as many darters
(Percidae: Etheostomatini), require rocky and gravel habitats, in
addition to well-oxygenated, flowing water, in which to spawn and care
for their eggs (Page, 1983; Simon, 1998).
The loss of these habitats due to increased turbidity and sedimentation
can render such habitats unsuitable for reproduction even if adults are
able to survive (Berkman & Rabeni, 1987).
Sedimentation-induced changes to aquatic invertebrate assemblages can
also affect fish foraging behavior and efficiency (Harvey, 1986; Berkman & Rabeni, 1987; Milner & Piorkowski, 2004).
Assessing direct effects of these types of disturbance on fish
assemblages can be quite challenging, particularly in large-river
systems; determining more subtle indirect effects and ecological shifts
mediated by dams present a greater challenge still.
Longitudinal patterns along stream
river gradients have been described using theoretical models such as
the nutrient spiraling concept (Webster & Patten, 1979), river continuum concept (Vannote et al., 1980), process domains concept (Montgomery, 1999), and flood pulse concept (Thorp & Delong, 1994);
however, these models generally assume uninterrupted continua and do
not account for disruptions to water and nutrient flow caused by dams.
The serial discontinuity concept (Ward & Stanford, 1983)
showed how dams can not only create lentic conditions above the dam,
but below the dam can effectively “reset” environmental conditions to
states reflecting lower order streams. Paradoxically, dams can thus
provide refugia for lotic species in impounded rivers, with higher flow
and oxygenated water immediately below dams (Freedman et al., 2009a; Argent & Kimmel, 2011).
Although longitudinal patterns in
relative fish abundance and diversity along a river continuum can thus
be affected by the presence of dams (Ward & Stanford, 1983; Miranda et al., 2008).
However, most studies have focused on smaller rivers, or on large
bodied fish taxa or those that can be sampled in near shore habitats.
Furthermore, while there are other studies that separately examine the
effects of dams on either taxonomic diversity or nutrient dynamics, the
complex relationships between these factors is not well understood. Our
objectives were, therefore, to use a novel sampling gear (electrified
benthic trawl; Freedman et al., 2009b)
in conjunction with stable isotope analysis to examine the effects of
dams on benthic fish assemblages and food webs by sampling dam-impacted
and free-flowing reaches of the Allegheny River, Pennsylvania. The
Allegheny River is important because it is the most northeast extension
of the rich Ohio River (and thus also of the Mississippi River
watershed); its diverse fauna was derived from the rich
Teays/Mississippi valley via the developing Ohio River and from glacial
meltwaters of that formed the Great Lakes (Hocutt et al., 1986).
We hypothesized that fish diversity would be lower in the impounded
section, with shifts from lotic to lentic species dominating the
community. As habitat would be less diverse (contiguous deep pools in
the impounded section; riffle-pool-run sequences in the free-flowing
section), we also expected to see higher biotic homogenization in the
impounded section (Olden et al., 2004; Poff et al., 2007).
Furthermore, because our impounded sites were located downstream from
our free-flowing sites, we expected that differences due to river
distance between sites [measured in river kilometers (rkm)] between
free-flowing and impounded sites would be greater than within these
categories. Regular interruptions in nutrient and water flow caused by
navigation dams would also be expected to increase homogeneity and
disrupt any longitudinal patterns in diversity. We used diversity
partitioning to determine the relative contribution of α
(within sample) and beta (among sample) diversity to the overall
(gamma) diversity of the river. Finally, we examined how dams affect
nutrient flow and food webs using stable isotope analysis; with
greater mean depth and habitat homogenization, we expected that fishes
at impounded sites would derive fewer nutrients from benthic sources
relative to those at free-flowing sites.
Materials and methods
Study area and sampling
The Allegheny River has a total length of 523 km and a watershed of approximately 30,000 km2, and is comprised of three main sections (Fig. 1).
From its headwaters in Pennsylvania, the upper section of Allegheny
River flows into New York State before reentering Pennsylvania, and is
unregulated above a hydroelectric and flood-control dam that forms the
Kinzua Reservoir at River Kilometer 325. Below the Kinzua Dam, the
middle section of the river is free-flowing for 211 km. The lower
section’s 113 km are regulated by a series of eight navigation
lock-and-dam systems until its confluence with the Monongahela River in
Pittsburgh forms the Ohio River. Glacial alluvial gravel and rocks
comprise the dominant substrate in the Allegheny River. Commercial
gravel dredging has occurred throughout most of the nine navigation
pools on the Allegheny River (Freedman et al., 2013), but only at one site above the navigation pool influence. Annual mean discharge of the Allegheny River is 189 m3 s−1 at our uppermost site near Kinzua Dam (USGS Gauging Station 03012550; River Kilometer 316), 623 m3 s−1 at Parker, PA, located 20 km upstream of the navigation limit (USGS Gauging Station 03031500; River Kilometer 133), and 920 m3 s−1
at Lock & Dam 4 at Natrona, PA (USGS Gauging Station 03049500;
River Kilometer 39). While subject to some point- and non-point source
stressors such as sewage discharges and agriculture, 139 km of the
middle section of the Allegheny River is designated as a National Wild
and Scenic River, and is yet relatively pristine.
Fig. 1
Map of Allegheny River watershed (shaded),
showing the lower section impounded by multiple navigation
lock-and-dams, and the free-flowing middle section. For reference, the
upper free-flowing section above the Kinzua dam is also denoted
We used electrified benthic trawls (Freedman et al., 2009b) to sample benthic fish assemblages at 66 sites in the middle and lower sections of the Allegheny River (Fig. 1).
We sampled 26 sites in the lower, impounded river, with three or four
sites in each navigation pool from 2 to 9; these included sites located
above and below each dam, with one or two sites located in the middle of
the pool (Fig. 2).
We sampled 40 sites in the middle, free-flowing section of the river,
from below Kinzua dam to just above the upper navigation limit in
navigation pool 9. Three to eight (mean ± SD; 4.44 ± 1.63) 2-min timed
trawls were conducted at each site depending on the width of the river.
All fishes were identified to species in the field when possible;
representative samples were retained and photo vouchers were taken for
laboratory verification.
Fig. 2
Fish diversity (upper panel) and mean depth (lower panel) of sites along a longitudinal gradient in the Allegheny River. Open circles and dotted lines represent observed Brillouin diversity, while solid circles and lines were calculated using a three-site moving average. The navigation limit (dashed line) forms the break between the impounded lower section and free-flowing middle section, while the Kinzua dam (dashed line) is the upstream limit of the middle section of river. Navigation lock-and-dam structures are denoted by solid triangles, and mean depths by open triangles
Stable isotope sampling and preparation
Stable isotopes can be used to
provide information about both realized trophic scenopoetic
(environmental conditions) and bionomic (interactions with other
organisms) niche dimensions (Newsome et al., 2007).
In other words, we used them not only to determine what an organism was
eating, but also whether it was deriving nutrients from benthic or
pelagic sources (e.g., Post, 2002b; Vander Zanden & Vadeboncoeur, 2002; Vander Zanden et al., 2005; Newsome et al., 2007). Stable carbon isotope signatures (δ13C)
vary across both primary producers and in response to differences in
environmental variables. For instance, periphyton and phytoplankton
differ in δ13C signatures, as do producers from deep or shallow water (Vander Zanden & Rasmussen, 1999; Post, 2002b). Variation in primary producer δ13C in streams and rivers is largely driven by CO2 availability: in shallow or turbulent water, the boundary layer effect increases the availability of “fresh” CO2, while in slow or deep water CO2 is less available and is, therefore, “recycled” by primary producers (Peterson & Fry, 1987; Finlay et al., 1999; Trudeau & Rasmussen, 2003). The heavier stable isotope of nitrogen (15N) is conserved in organic tissues, and passes to higher consumers via bioaccumulation. Thus, δ15N
is enriched at a relatively constant rate (2–5‰, mean 3.4‰) across
trophic levels, and, therefore, serves to estimate trophic position
within a food-web (Vander Zanden & Rasmussen, 1999; Vander Zanden & Rasmussen, 2001; Post, 2002a; Vanderklift & Ponsard, 2003).
While stable isotope analysis has been used to gain insights into
biotic changes that occur as a response to anthropomorphic stress, such
research has tended to focus on point- and non-point-source additions to
aquatic environments (Costanzo et al., 2001; Vadeboncoeur et al., 2003; Gray et al., 2004; Grey et al., 2004; Anderson & Cabana, 2005; Vander Zanden et al., 2005)
rather than physical alterations to the environment. The ability of
stable isotope analysis to differentiate food sources and detect trophic
positions thus makes it a powerful tool for examining the effects of
dams on riverine fish assemblages.
Two sites were sampled above the
influence of navigation dams (free-flowing sites), and three sites were
sampled in the upper navigation pools of the Allegheny River (Pools
7–8) during late summer, 2007. Adult fishes were collected using a
combination of Missouri- and PSU-benthic trawls (Herzog et al., 2005; Freedman et al., 2009b).
Fish samples were immediately frozen until processed in the laboratory.
All fishes were identified to species, with the exception of shiners (Notropis
spp.) which were not identified to species prior to stable isotope
sampling, and were, therefore, grouped together and analyzed as shiner
spp. Several individuals of each fish taxon (range 2–48 individuals per
site) to compensate for inherent inter-individual variability, and of
different size-classes where relevant, were sampled for stable isotope
analysis. White muscle tissue was used if sufficient material could be
obtained for fish samples as previous studies have shown it to be less
variable than other tissues, with a moderate stable isotope turnover
rate on the order of weeks to months (Hobson, 1999);
smaller fishes were eviscerated and decapitated. To compensate for
inherent differences among sites, samples were pooled by taxon for both
the two free-flowing and the three dam-impacted sites.
All samples were rinsed with
deionized water, placed into a clean glass vial, and dried in a drying
oven at 60°C for 24–48 h. Dried samples were homogenized to a fine
powder using mortar-and-pestle, or using a glass stirring rod within the
vial. Samples were weighed into 0.2 mg (± 10%) aliquots, placed into
5 mm × 3.5 mm tin capsules, and analyzed for δ13C and δ15N
using either a Thermo-Finnigan Delta Plus or Delta XP isotope-ratio
mass spectrometer interfaced with a Carlo Erba NC2500 Elemental Analyzer
via the Conflo II or Conflo III at the Stable Isotopes in Nature
Laboratory at the University of New Brunswick, Canada.
From each sample, the ratios of 14N to 15N and of 12C to 13C were determined, and used to calculate δ15N and δ13C using the formula:
δX=[(Rsample/Rstandard)−1]×1,000,
where X refers to the rare, heavy isotope, and R is the ratio of the heavy isotope (15N, 13C) to the light isotope (14N, 12C)
in the sample and in a standard. The standard for nitrogen is
atmospheric nitrogen (AIR), and for carbon is carbon dioxide derived
from calcium carbonate in the Pee Dee Bee formation of South Carolina
(PDB). As lipids are rich in carbon relative to tissues, variable
tissue-lipid contents among samples can increase overall variability of
samples; we, therefore, used a lipid correction factor to standardize
across samples (Eq. 3, Table 1 from Post et al., 2007). For isotopic standards, standard deviations were 0.15‰ for δ13C and 0.24‰ for δ15N, for elemental standards standard deviations ranged from 0.13 to 0.15‰ for δ13C and 0.14 to 0.25‰ for δ15N, and for biologic standards, the standard deviations ranged from 0.11 to 0.14‰ for δ13C and from 0.12 to 0.14‰ for δ15N. Replicate fish tissue samples varied by an average of 0.22‰ (SD 0.24‰) for δ13C and 0.19‰ (SD 0.18‰) for δ15N.
Statistical analysis
We calculated both observed
site-to-site differences, and used three-site moving averages to
visualize longitudinal trends, in Brillouin diversity of fish
assemblages along the river (Fig. 2);
however, all analyses were conducted on the observed data. We performed
both non-metric multidimensional scaling (nMDS) and analysis of
similarity (ANOSIM) based on a Bray–Curtis dissimilarity matrix of fish
assemblages to examine differences among sites, using Primer 5.2.2
(Primer-E Ltd., Plymouth, UK). River sections (free-flowing middle and
impounded lower) were used as factors.
To quantify the effects of dams in structuring diversity, we examined the relative contributions of alpha (α, within sample) and beta (β, among sample) diversity to the gamma (γ, total) diversity of the Allegheny River (sensu Crist et al., 2003). We performed complete randomization of 10,000 iterations using additive partitioning (Partition 3.0; Veech & Crist, 2009) wherein
γ(total diversity)=α1(within site)+β1(among site)+β2(among section)
to
test for the presence of patterns across these hierarchical levels in
the Allegheny River. We tested the null hypothesis that observed fish
species richness at each hierarchical level was not significantly
different from a random distribution of these fish species among samples
at each of these levels.
We used circular statistics (Schmidt et al., 2007)
to assess differences between fish stable isotope signatures by
assessing directional changes from free-flowing to impounded sites using
the software package Oriana 3.0 (Kovach, 2009).
In circular statistics, the stable isotope data are transformed into
linear vectors for each fish species, with an origin that is
standardized as 0.0. δ13C is plotted on the X-axis, with 13C-depletion
(indicative of pelagic carbon sources) to the left (270°), and
13C-enrichment (benthic carbon sources) to the right (90°). δ15N is plotted on the Y-axis, with 15N-enrichment (higher trophic level) at 0° and 15N-depletion
(lower trophic level) at 180°. We defined the origin as being the
free-flowing site, while the other end of the vector represents the
impounded sites. The length of each vector represents the magnitude of
change of stable isotopic signatures for that species, while the angle
of the vector represents the directionality of that change. We used
Rayleigh’s Test for Circular Uniformity to test whether the distribution
of vectors was random or uniform. Alpha levels of 0.05 were used to
assess significance for all analyses.
Results
Fish distribution, abundance, and diversity
Diversity in the free-flowing
middle section generally increased from the Kinzua dam until the dam
influence near 116 rkm and was variable in the dam-impacted lower
section, but generally declined downstream (Fig. 2).
We caught more fishes in the free-flowing section (46.6 fish per trawl)
than in the impounded section (18.8 fish per trawl), as well as higher
taxonomic richness with 44 taxa in the free-flowing section compared to
34 taxa in the impounded section (Table 1).
Mean diversity was lower in the impounded section (mean Brillouin
diversity 1.04 ± 0.34 SD) than in the free-flowing section (1.44 ± 0.35;
t test, df = 64, t-stat = −4.59, P < 0.00002125; Fig. 2).
Mean depth of sites in the impounded section was 4.7 m (range
0.3–14.9 m) and in the free-flowing section was 2.2 m (range 0.3–10.0 m;
Fig. 2).
Table 1
Fish species
captured and relative abundance (number per trawl sample) in the
impounded lower section and free-flowing middle section of the Allegheny
River
Scientific name
|
Common name
|
Catch-per-sample
| |
|---|---|---|---|
Impounded
|
Free-flowing
| ||
Petromyzontidae
| |||
Ichthyomyzon bdellium
|
Ohio Lamprey
|
0
|
0.02
|
Petromyzontid sp.
|
Lamprey Larvae
|
0
|
0.01
|
Cyprinidae
| |||
Campostoma anomalum
|
Central Stoneroller
|
0
|
0.01
|
Cyprinus carpio
|
Common Carp
|
0.01
|
0
|
Erimystax dissimilis
|
Streamline Chub
|
0.57
|
3.11
|
Exoglossum laurae
|
Tonguetied Minnow
|
0
|
0.02
|
Hybopsis amblops
|
Bigeye Chub
|
0
|
0.15
|
Luxilus chrysocephalus
|
Striped Shiner
|
0
|
0.02
|
Nocomis micropogon
|
River Chub
|
0
|
0.07
|
Notropis atherinoides
|
Emerald Shiner
|
0
|
0.03
|
Notropis photogenis
|
Silver Shiner
|
0
|
0.16
|
Notropis volucellus
|
Mimic Shiner
|
0.13
|
7.50
|
Notropis spp.
|
Shiner Species
|
0.03
|
0
|
Pimephales notatus
|
Bluntnose Minnow
|
0
|
0.38
|
Catosomidae
| |||
Catostomus commersonii
|
White Sucker
|
0
|
0.01
|
Hypentelium nigricans
|
Northern Hogsucker
|
0
|
0.03
|
Moxostoma anisurum
|
Silver Redhorse
|
0.02
|
0
|
Moxostoma duquesnei
|
Black Redhorse
|
0
|
0.01
|
Moxostoma erythrurum
|
Golden Redhorse
|
0.01
|
0
|
Moxostoma spp.
|
Redhorse Species
|
0.02
|
0
|
Ictaluridae
| |||
Ictalurus punctatus
|
Channel Catfish
|
0.28
|
0.02
|
Noturus eleutherus
|
Mountain Madtom
|
0
|
0.05
|
Noturus flavus
|
Stonecat
|
0.03
|
0.01
|
Noturus stigmosus
|
Northern Madtom
|
0.01
|
0.01
|
Pylodictis olivaris
|
Flathead Catfish
|
0.03
|
0.02
|
Percopsidae
| |||
Percopsis omiscomaycus
|
Trout-Perch
|
0.06
|
0.11
|
Atherinopsidae
| |||
Labidesthes sicculus
|
Brook Silverside
|
0
|
0.01
|
Centrarchidae
| |||
Ambloplites rupestris
|
Rock Bass
|
0.01
|
0.07
|
Lepomis cyanellus
|
Green Sunfish
|
0
|
0.01
|
Lepomis macrochirus
|
Bluegill
|
0
|
0.05
|
Micropterus dolomieu
|
Smallmouth Bass
|
0.37
|
1.63
|
Micropterus punctulatus
|
Spotted Bass
|
0.01
|
0
|
Micropterus salmoides
|
Largemouth Bass
|
0.02
|
0
|
Micropterus sp.
|
Black Bass Species
|
0.02
|
0
|
Percidae
| |||
Etheostoma blennioides
|
Greenside Darter
|
0.61
|
4.65
|
Etheostoma caeruleum
|
Rainbow Darter
|
0.68
|
6.87
|
Etheostoma camurum
|
Bluebreast Darter
|
1.01
|
1.17
|
Etheostoma flabellare
|
Fantail Darter
|
0.09
|
0.34
|
Etheostoma maculatum
|
Spotted Darter
|
0.01
|
0.21
|
Etheostoma nigrum
|
Johnny Darter
|
3.36
|
0.17
|
Etheostoma tippecanoe
|
Tippecanoe Darter
|
0.80
|
0.39
|
Etheostoma variatum
|
Variegate Darter
|
0.02
|
1.17
|
Etheostoma zonale
|
Banded Darter
|
0.42
|
5.19
|
Percina caprodes
|
Logperch
|
2.44
|
1.32
|
Percina copelandi
|
Channel Darter
|
6.41
|
1.52
|
Percina evides
|
Gilt Darter
|
0.72
|
3.05
|
Percina macrocephala
|
Longhead Darter
|
0.30
|
1.44
|
Percina maculata
|
Blackside Darter
|
0
|
1.51
|
Percina (hybrid)
|
Darter hybrid
|
0
|
0.01
|
Perca flavescens
|
Yellow Perch
|
0.02
|
0
|
Sander vitreus
|
Walleye
|
0.04
|
0.02
|
Sander sp.
|
Walleye or Sauger
|
0.07
|
0.04
|
Sciaenidae
| |||
Aplodinotus grunniens
|
Freshwater Drum
|
0.19
|
0
|
Cottidae
| |||
Cottus bairdi
|
Mottled Sculpin
|
0
|
4.01
|
Mean Number of Fish per Trawl
|
18.8
|
46.6
| |
We caught a total of 10, 038 fishes comprising 54 taxa: 53 species and 1 hybrid (Table 1).
Fishes from the family Percidae (primarily darters) comprised 90.5% of
the total catch in the impounded section, while catch from the
free-flowing section comprised 62.4% percids and 24.6% cyprinids (minnow
family). However, percids were more abundant in the free-flowing
section, with a catch rate of percids (29.1 per trawl) almost double
that in the impounded section (17.0 per trawl; Table 1).
The most prevalent percids in the impounded section were tolerant
species such as Channel Darter (34.2% of total catch), Johnny Darter
(17.9%), and Logperch (13.0%). In the free-flowing section, the most
prevalent percids were species with more lotic requirements such as
Rainbow Darter (14.7% of total catch), Banded Darter (11.1%), and
Greenside Darter (10.0%; Table 1). Overall, the most prevalent species in the free-flowing section was Mimic Shiner (16.1% of total catch; Table 1),
while Mottled Sculpin (8.6%) were also prevalent, particularly in the
upper reaches of the section. The free-flowing site with the lowest
diversity (0.33) was located at river km 296.1. This site was the
deepest in the free-flowing section (9.0 m deep) and we captured just 22
fishes: 19 Trout-Perch and three Mottled Sculpin. This site was also
characterized by sandy substratum, which was also noted at the other
three sites in the impounded section where Trout-Perch were collected.
Twenty taxa (19 species and one
hybrid) were found only in the free-flowing section, while 10 taxa were
found only in the impounded section (Table 1). These contributed to a section beta diversity (β
2) of 14.5 which represented 26.9% of
gamma species richness but was not significantly different than
expected using diversity partitioning (P > 0.05; 10,000 iterations; Table 2; Fig. 3). The mean numbers of species that were not shared among sites (β
1) were higher than expected from the 10,000 randomizations (72.3% of gamma diversity versus 56.7%; P < 0.001). The mean numbers of species shared among sites (α
1) were lower than expected (27.7% versus 43.3% expected; P < 0.001; Table 2; Fig. 3).
Table 2
Additive partitioning results for fish communities among sites in impounded and free-flowing sections of the Allegheny River
Spatial scale
|
Diversity component
|
Observed mean diversity
|
Expected mean diversity
|
Contribution to gamma diversity (%)
|
|---|---|---|---|---|
River
|
γ
|
54
| ||
Section
|
α
2
|
39.5
|
39.5
|
73.1
|
β
2
|
14.5
|
14.5
|
26.9
| |
Site
|
α
1
|
10.94
|
17.11
|
27.7
|
β
1
|
28.56
|
22.39
|
72.3
|
Fig. 3
Diversity
partitioning results (10,000 iterations) for the Allegheny River showing
species richness between the free-flowing middle and impounded lower
sections (β
2), among sites (β
1), and within sites (α
1). The symbol plus indicates that observed diversity was greater than expected, while minus indicates that observed diversity was lower than expected
There was differentiation
between fish assemblages from the free-flowing and impounded sections of
the Allegheny River along Axis 1 of the nMDS, with all fish assemblages
from impounded sites having values of <0, while only one
free-flowing site value of <0 along this axis (nMDS, Stress 0.17;
Fig. 4). ANOSIM also revealed significant differences in site similarity between free-flowing and impounded sites (Global R:
0.62, significance level 0.1). Fish assemblages in the impounded
section were equally dissimilar between 0 and 59 rkm (navigation pools
2–5) and 59–115 rkm (navigation pools 6–9) subsections. Fish assemblages
from the free-flowing section were progressively more dissimilar from
impounded section assemblages with the increasing distance upstream
(Fig. 4).
Fish assemblages from sites located below dams in both the 0–59 and
59–116 rkm subsections were not more similar to free-flowing sites than
that of sites located mid-pool or above dams.
Fig. 4
nMDS plot of Bray–Curtis similarity among sites in the Allegheny River. Downward triangles represent impounded lower section sites, while upward triangles represent free-flowing middle section sites. Symbol shading represents subsections defined by rkm
Food webs and nutrient dynamics
Small fishes from free-flowing sites were less depleted in 13C, consistent with reliance on benthic-derived nutrients (Fig. 5). Fishes from impounded sites had 13C depleted δ13C
signatures, consistent with increased reliance on pelagic-derived
nutrients rather than benthic-derived nutrients. Only Mottled Sculpin
from free-flowing sites had δ13C signatures more negative than −24.00‰.
Fig. 5
Bivariate plots of δ13C and δ15N for fish assemblages in the free-flowing middle section (upper plot) and impounded lower section (lower plot) of the Allegheny River. Symbols indicate mean stable isotope values (±s.e.) of individual species while light gray crosses represent stable isotope values for individual fish
There was a directional shift in δ13C
from free-flowing to impounded sites. Circular statistics revealed that
fishes from the lower section (Rayleigh’s Test, Z = 11.437, P = 0.00000143; Fig. 6)
shifted to increased reliance on pelagic-derived nutrients at impounded
sites. There was no significant effect of dam influence on the trophic
position of fishes. Fantail Darter (from mean δ15N 13.30 ± 0.38 SD at free-flowing sites to 14.61 ± 1.41 at undredged sites) and Johnny Darter (from δ15N
10.99 ± 0.25 to 12.93 ± 1.00) appeared to be exceptions as they both
increased mean trophic position from free-flowing to undredged sites.
Fig. 6
Circular plots of δ13C (horizontal axis) and δ15N (vertical axis). Enriched δ13C indicative of benthic-derived nutrients is to the right and depleted δ13C representing pelagic-derived nutrients is to the left. Higher and lower δ15N values are indicative of higher and lower trophic positions and are oriented to the top and bottom of the plot, respectively. Individual arrows represent mean δ13C and δ15N values of individual species: vector direction indicates shifts in δ13C and δ15N
between sites in the free-flowing middle section and impounded lower
section, while the length of the vector indicates the magnitude of the
difference. The solid line is the overall mean, and the line at the circumference is the 95% confidence interval
Discussion
Effects of dams on fish distribution, abundance, and diversity
We detected significant
differences between fish communities in the impounded lower section and
free-flowing middle sections of the Allegheny River. These were largely
influenced by higher fish abundance and taxonomic diversity at sites in
the free-flowing section. Mottled Sculpins were captured at 23 of the 24
uppermost sites; they were among the most abundant taxa in the
free-flowing section, but were not captured below river km 210. This may
be due to thermal limitations and habitat preference, as this species
is generally associated with cool, clear, flowing water (Scott &
Crossman, 1973; Trautman, 1981).
Lotic-adapted darters such as Greenside, Rainbow, Banded, and Gilt
darters were more abundant in the free-flowing section than in the
impounded section. Darter species that were more abundant in the lower
section included Johnny Darter, Channel Darter, and Logperch, all of
which are better adapted to slower-moving and lentic conditions than
most darters (Page, 1983).
Streamline Chub and Mimic Shiners were also more prevalent in the
free-flowing section although they have different habitat preferences.
Both species are found in streams and rivers, but while Mimic Shiners
are more tolerant of both silt and lotic conditions than Streamline
Chub, neither species thrives in high-silt environments that
characterizes much of the impounded section (Trautman, 1981).
Fish assemblages from below dams
were dissimilar to above-dam sites. Contrary to our expectations,
however, below-dam sites were not more similar to free-flowing sites.
Species that are adapted to lotic conditions may find refugia below dams
within rivers where turbulence is greatest (Freedman et al., 2009a; Argent & Kimmel, 2011).
For instance, lotic fish species listed as threatened by the state of
Pennsylvania were found at higher abundances in dam tailwaters on the
Ohio River (Freedman et al., 2009a), and a similar trend was noted for dams in the Allegheny River (Argent & Kimmel, 2011).
Despite providing refugia for lotic species, fish assemblages from
habitats immediately below dams were equally dissimilar to free-flowing
sites as were other dam-impacted sites. This indicates a fundamental
impact of dams on these fish assemblages.
This may be driven, in part, by
the lower diversity in the impounded section than the free-flowing
section. Additive partitioning revealed higher heterogeneity in fish
assemblages between impounded and free-flowing sites than expected.
There was no evidence of increased homogeneity per se
among the impounded sites relative to free-flowing sites. At the same
time, despite generally lower diversity at lower river km in the
impounded section, there was no pattern of longitudinal changes in this
section apparent in the MDS analysis. The free-flowing section, however,
showed a downstream pattern in increased diversity that was also
apparent in MDS analysis. Fish assemblages in the free-flowing section
are thus generally consistent with the river continuum concept (Vannote
et al., 1980)
in that there were increases in taxonomic diversity and mixed
assemblages of lotic and lentic species at downstream sites. The
disruption of this pattern, with generally lower taxonomic diversity and
a sharp drop in lotic species, and a lack of longitudinal changes in
the dam-impacted section are consistent with the serial discontinuity
concept (Ward & Stanford, 1983).
While the locks in navigation
lock-and-dam systems provide access between pools, the dams inhibit fish
movement. For instance, river darter, Percina shumardi,
has been captured to the base of the second lock-and-dam of the Ohio
River (DaShields lock-and-dam in the Montgomery Pool; Freedman et al., 2009a).
Extensive sampling has not collected this species upstream of this dam
in the Ohio River or in the Allegheny or Monongahela rivers (Freedman et
al., 2009a, b; Stauffer et al., 2010; Argent & Kimmel, 2011),
suggesting that it is recolonizing the Pennsylvania section of the Ohio
River from downstream refugia rather than simply having been missed in
prior surveys. The range of this species may expand upstream into the
Allegheny and Monongahela rivers, but this dispersal will likely be
slowed by the presence of navigation dams. The extirpation from the Ohio
River of anadromous species such as Lake Sturgeon, Acipenser fulvescens, can be at least partially explained by the presence of dams (Pearson & Pearson, 1989). The presence of locks may help to mediate this issue, as juvenile paddlefish, Polyodon spathula,
stocked in the Ohio River were confirmed to have passed through locks
in both upstream and downstream directions (Barry et al., 2007).
The use of navigation lock chambers by fishes can also be confirmed by
lock chamber rotenone surveys on The Ohio River in which almost 3 × 106
fishes comprising 116 fish taxa were collected in 377 sampling events:
an average of almost 8,000 fishes per collection (Thomas et al., 2005).
Effects of dams on food webs and nutrient dynamics
Stable isotope analysis revealed
a shift toward increased reliance on pelagic-derived nutrients by
fishes at impounded sites relative to the free-flowing sites above the
navigation dam influence. This shift is consistent with shifts from
allochthonous to autochthonous and benthic to pelagic nutrients from
low-order streams to high-order rivers predicted and observed in other
studies (e.g., Vannote et al., 1980; Finlay, 2001),
and also with patterns of increased depth caused by the downstream
presence of dams at these sites. These results are also consistent with
shifts from benthic-driven primary production (e.g., periphyton) to
pelagic production (e.g., phytoplankton) as a result of cultural
eutrophication (Vadeboncoeur et al., 2003; Chandra et al., 2005; Vander Zanden et al., 2005).
Such shifts are generally considered to be the result of eutrophication
increasing concentration and productivity of pelagic primary producers,
thus starving benthic producers of both nutrients for growth and
sunlight for photosynthesis (Vadeboncoeur et al., 2003; Chandra et al., 2005). Anthropogenically increased depth and turbidity may have similar effects (Freedman et al., 2013).
As average depth increases from headwater streams to high-order rivers,
relatively less light reaches the river floor, from near 100% of
non-refracted light in small clear streams to zero in turbid and deep
water, thus decreasing benthic production. Dams increased the mean depth
from <3 m in pools in the free-flowing Allegheny River to a constant
minimum of 4–5 m (or more) in dam-impacted reaches, and can, therefore,
decrease benthic production without any influences from eutrophication
or other increases in relative turbidity. Since the free-flowing reach
of the river includes runs and riffles as well as pools, while the
homogenous impounded reaches effectively consist only of long pools
possibly with minimal lotic habitat immediately downstream of dams, a
loss of some benthic nutrient pathways becomes even more likely.
Reliance on benthic-derived nutrients at free-flowing sites appears to be high while at dam-impacted sites δ13C
signatures suggest a trend toward increasing reliance on
pelagic-derived nutrients. While a full range of benthic and pelagic
nutrients seem to be available, fishes are relying more on pelagic
nutrients; this is consistent with the theory that reliance on pelagic
nutrients (and decreased reliance on benthic nutrients) would increase
with the increased depth due to the navigation dams. Pelagic production
is also lower in lower order rivers, with zooplankton diversity and
biomass consequently increasing downstream (e.g., Vannote et al., 1980; Ward & Stanford, 1983).
The relative reliance on pelagic nutrient sources in the free-flowing
section may, therefore, be a combination of higher availability of
benthic nutrients and lower availability of pelagic nutrients.
Rafinesque (1820)
referred to the Allegheny River as being “almost perfectly clear,”
while our Secchi depths ranged from 142 to 145 cm downstream of an
active dredging operation and 157–198 cm in other impounded areas of the
river (JAF, unpublished data), thus offering supporting evidence that
this is no longer the case. We found Secchi depths in the range of
330 cm, however, in a pool above the dam influence, so there does appear
to be a negative effect of dams on water clarity. Although other
studies have found that dams can decrease turbidity through retention of
fine sediments (e.g., Kondolf, 1997),
the Allegheny River system may differ due to the size of the dams
(relatively small compared to dams constructed for hydroelectric power
generation, flood-prevention, and similar purposes) and locks, both of
which may allow for the passage and resuspension of fine sediments. Land
use does not differ greatly between the upper navigation pools and
lower free-flowing section, consisting primarily of forested land with
some residential properties. The approximately 214 km of the Allegheny
River between the Kinzua Dam and the end of the navigation dam influence
near East Brady PA, likely experience full light penetration except
during times of high discharge due to shallow depths and lower
turbidity. Despite the upstream presence of the Kinzua dam, free-flowing
sites likely represent similar reference states to the historical
condition, and with similar fish assemblages and food-web structure.
According to the river continuum
concept, nutrient sources shift from allochthonous inputs to
autochthonous primary production along the longitudinal river gradient
(Vannote et al., 1980).
These gradients can be reset by dams, creating higher-flow downstream
conditions that mimic higher order streams. Conversely, deeper and
slower flowing conditions above dams can be more similar to lower order
rivers. Plankton communities are sparse in higher order streams and
rivers relative to lower order rivers and impoundments. In the Allegheny
River, therefore, nutrient inputs in the free-flowing section would
comprise primarily benthic sources, possibly with increased
allochthonous inputs. In the dam-impacted section, however, increased
phytoplankton production due to environmental conditions combined with
lower benthic production would lead to increased reliance on such
pelagic producers. The Allegheny River also has a long history of
extractive gravel dredging. Dredged portions can exceed 20 m depth, with
no light penetration deeper than approximately 10 m. We focused our
sampling on undredged areas <7 m deep; however, in another study, we
found that dredged areas accumulate terrestrial detritus, and to alter
nutrient and sediment flow (Freedman et al., 2013).
Conclusions
We found significant differences
among fish community compositions at sites in impounded and free-flowing
sections of the Allegheny River. Furthermore, the shift from
communities characterized by lotic-adapted species and those intolerant
of silt, to those dominated by generalist and tolerant large-river
species was very abrupt. In particular, the longitudinal gradient in
fish community similarity and downstream trend toward increasing
taxonomic diversity was disrupted in the impounded section, where we
found decreasing downstream diversity but no concurrent trend in
similarity. These findings were consistent with the stable isotope
results, which showed shifts away from the benthic production that
characterized the free-flowing section toward increasing reliance on
pelagic-derived nutrients in the impounded section. These shifts were
likely due to a decrease in benthic production due to increased depth,
turbidity, and siltation (Freedman et al., 2013),
and would also be consistent with a decline of lotic and intolerant
species. By using an electrified benthic trawl, we were able to sample
small benthic fishes that are difficult to sample using traditional
methods. However, although we did also capture some non-benthic taxa,
our sampling method was biased toward the capture of benthic rather than
pelagic or littoral fishes. Since benthic fishes may be particularly
impacted by habitat alterations due to increased depth (Freedman et al.,
2013), our findings cannot necessarily be extrapolated to the entire fish community.
While dams can provide economic
benefits, it is necessary to understand the effects that they can have
on individual fish species, populations, and communities. While dam
removal can restore habitats, and subsequently invertebrate and fish
populations (Maloney et al., 2008), many factors need to be considered prior to restoration (Poff & Hart, 2002). Stable isotope analysis of δ13C and δ15N
is an appropriate tool for assessing differences in fish assemblages
between sites with varying degrees of influence from dams, and should be
considered for before-after-control-impact (BACI) study designs. It is,
therefore, important for managers and policy makers to consider not
only the direct effects of habitat alterations on taxonomic diversity,
but also indirect effects on ecosystem functioning. Furthermore,
alterations in water flow, prey availability, and migration due to dams
can even effect changes in fish ecomorphology and functional morphology
in certain species (Curry et al., 2004; Palkovacs et al., 2007; Langerhans, 2008; Freedman, 2010; Haas et al., 2010),
further confounding these issues. Dams influence riverine fish in many
ways; a complete understanding of ecological processes is, therefore,
necessary for informed conservation and management decisions.
Acknowledgments
We
thank A. Anderson, V. Cavener, D. Cooper, H. Goldstein, A. Henning, R.
Lorson, R. Lorson, T. Stecko, K. Taylor, T. Vasilopoulos, and R. Yoder
for their invaluable field and laboratory assistance. The Stable
Isotopes in Nature Laboratory at the University of New Brunswick
performed the stable isotope analysis. An earlier draft of this
manuscript was improved by addressing comments and suggestions from two
anonymous reviewers. This research received funding and support from
State Wildlife Grant T-42 administered by the Pennsylvania Fish and Boat
Commission; from the Pennsylvania Department of Conservation and
Natural Resources through Wild Resources Conservation Program Grants
WRCP-06171 and WRCP-07269; and from the United States Geological Survey,
Pennsylvania Cooperative Fish and Wildlife Research Unit.
References
Anderson, C. & G. Cabana, 2005. δ15N
in riverine food webs: effects of N inputs from agricultural
watersheds. Canadian Journal of Fisheries and Aquatic Sciences 62:
333–340.CrossRef
Argent,
D. G. & W. G. Kimmel, 2011. Influence of navigational lock and dam
structures on adjacent fish communities in a major river system. River
Research and Applications 27(10): 1325–1333.CrossRef
Barry,
P. M., R. F. Carline, D. G. Argent & W. G. Kimmel, 2007. Movement
and habitat use of stocked juvenile paddlefish in the Ohio River System,
Pennsylvania. North American Journal of Fisheries Management 27:
1316–1325.CrossRef
Berkman,
H. E. & C. F. Rabeni, 1987. Effect of siltation on stream fish
communities. Environmental Biology of Fishes 18(4): 285–294.CrossRef
Chandra,
S., M. J. Vander Zanden, A. C. Heyvaert, B. C. Richards, B. C. Allen
& C. R. Goldman, 2005. The effects of cultural eutrophication on the
coupling between pelagic primary producers and benthic consumers.
Limnology and Oceanography 50(5): 1368–1376.CrossRef
Curry, R. A., S. L. Currie, L. Bernatchez & R. Saint-Laurent, 2004. The rainbow smelt, Osmerus mordax, complex of Lake Utopia: threatened or misunderstood? Environmental Biology of Fishes 69(1–4): 153–166.CrossRef
Finlay,
J. C., 2001. Stable-carbon-isotope ratios of river biota: implications
for energy flow in lotic food webs. Ecology 82(4): 1052–1064.
Finlay,
J. C., M. E. Power & G. Cabana, 1999. Effects of water velocity on
algal carbon isotope ratios: implications for river food web studies.
Limnology and Oceanography 44(5): 1198–1203.CrossRef
Freedman,
J. A., 2010. Dams, Dredging, and Development: Effects of Anthropogenic
Disturbances on Fish Ecology. The Pennsylvania State University,
University Park.
Freedman, J. A., T. D. Stecko, R. W. Criswell & J. R. Stauffer Jr., 2009a. Extensions of the known ranges of Percina shumardi Girard and three species of Etheostoma (subgenus Nothonotus) in Pennsylvania. Journal of the Pennsylvania Academy of Science 83(1): 42–44.
Freedman,
J. A., T. D. Stecko, B. D. Lorson & J. R. Stauffer Jr., 2009b.
Development and efficacy of an electrified benthic trawl for sampling
large-river fish assemblages. North American Journal of Fisheries
Management 29: 1001–1005.CrossRef
Freedman,
J. A., R. F. Carline & J. R. Stauffer Jr., 2013. Gravel dredging
alters diversity and structure of riverine fish assemblages. Freshwater
Biology 58: 261–274.CrossRef
Graf,
W. L., 1999. Dam nation: a geographic census of American dams and their
large-scale hydrologic impacts. Water Resources Research 35(4): 1305.CrossRef
Graf,
W. L., 2006. Downstream hydrologic and geomorphic effects of large dams
on American rivers. Geomorphology 79(3–4): 336–360.CrossRef
Gray, M. A., R. A. Cunjak & K. R. Munkittrick, 2004. Site fidelity of slimy sculpin (Cottus cognatus): insights from stable carbon and nitrogen analysis. Canadian Journal of Fisheries and Aquatic Sciences 61: 1717–1722.CrossRef
Grey,
J., S. Waldron & R. Hutchinson, 2004. The utility of carbon and
nitrogen isotope analyses to trace contributions from fish farms to the
receiving communities of freshwater lakes: a pilot study in Esthwaite
Water, UK. Hydrobiologia 524(1): 253–262.CrossRef
Haas,
T., M. Blum & D. Heins, 2010. Morphological responses of a stream
fish to water impoundment. Biology Letters 6: 803–806.PubMedCentralPubMedCrossRef
Harvey,
B. C., 1986. Effects of suction gold dredging on fish and invertebrates
in two California USA streams. North American Journal of Fisheries
Management 6(3): 401–409.CrossRef
Herzog,
D. P., V. A. Barko, J. S. Scheibe, R. A. Hrabik & D. E. Ostendorf,
2005. Efficacy of a benthic trawl for sampling small-bodied fishes in
large river systems. North American Journal of Fisheries Management 25:
594–603.CrossRef
Hobson, K. A., 1999. Tracing origins and migration of wildlife using stable isotopes: a review. Oecologia 120(3): 314–326.CrossRef
Hocutt,
C. H., R. E. Jenkins & J. R. Stauffer Jr., 1986. Zoogeography of
the fishes of the Central Appalachians and Central Atlantic Coastal
Plain. In Hocutt, C. H. & E. O. Wiley (eds), Zoogeography of North
American Freshwater Fishes. Wiley, New York.
Kovach, W. L. 2009. Oriana—Circular statistics for Windows, version 3. Kovach Computing Services, Pentraeth, Wales, UK.
Lytle, D. A. & N. L. Poff, 2004. Adaptation to natural flow regimes. Trends in Ecology & Evolution 19(2): 94–100.CrossRef
Maloney,
K. O., H. R. Dodd, S. E. Butler & D. H. Wahl, 2008. Changes in
macroinvertebrate and fish assemblages in a medium sized river following
a breach of a low head dam. Freshwater Biology 53(5): 1055–1068.CrossRef
Milner,
A. M. & R. J. Piorkowski, 2004. Macroinvertebrate assemblages in
streams of interior Alaska following alluvial gold mining. River
Research and Applications 20: 719–731.CrossRef
Miranda,
L. E., M. D. Habrat & S. Miyazono, 2008. Longitudinal gradients
along a reservoir cascade. Transactions of the American Fisheries
Society 137: 1851–1865.CrossRef
Montgomery,
D. R., 1999. Process domains and the river continuum. JAWRA Journal of
the American Water Resources Association 35(2): 397–410.CrossRef
Newsome,
S. D., C. Martinez del Rio, S. Bearhop & D. L. Phillips, 2007. A
niche for isotopic ecology. Frontiers in Ecology and the Environment
5(8): 429–436.CrossRef
Olden,
J. D., N. LeRoy Poff, M. R. Douglas, M. E. Douglas & K. D. Fausch,
2004. Ecological and evolutionary consequences of biotic homogenization.
Trends in Ecology & Evolution 19(1): 18–24.CrossRef
Page, L. M., 1983. Handbook of Darters. TFH Publications, Neptune City, NJ.
Pearson, W. D. & B. J. Pearson, 1989. Fishes of the Ohio River. Ohio Journal of Science 89(5): 181–187.
Peterson, B. J. & B. Fry, 1987. Stable isotopes in ecosystem studies. Annual Reviews in Ecology and Systematics 18: 293–320.CrossRef
Poff,
N. L. & D. D. Hart, 2002. How dams vary and why it matters for the
emerging science of dam removal. Bioscience 52(8): 659–668.CrossRef
Poff,
N. L. R., J. D. Allan, M. B. Bain, J. R. Karr, K. L. Prestegaard, B. D.
Richter, R. E. Sparks & J. C. Stromberg, 1997. The natural flow
regime. Bioscience 47(11): 769–784.CrossRef
Poff,
N. L. R., J. D. Olden, D. M. Merritt & D. M. Pepin, 2007.
Homogenization of regional river dynamics by dams and global
biodiversity implications. Proceedings of the National Academy of
Sciences of the United States of America 104(14): 5732.PubMedCentralPubMedCrossRef
Post, D. M., 2002a. The long and short of food-chain length. Trends in Ecology & Evolution 17(6): 269–277.CrossRef
Post,
D. M., 2002b. Using stable isotopes to estimate trophic position:
models, methods, and assumptions. Ecology 83(3): 703–718.CrossRef
Rafinesque,
C. S., 1820. Ichthyologia Ohiensis, or natural history of the fishes
inhabiting the river Ohio and its tributary streams, preceded by a
physical description of the Ohio and its branches. Printed for the
author by W. G. Hunt, Lexington KY.
Rivier,
B. & J. Seguier, 1985. Physical and biological effects of gravel
extraction in river beds. In Alabaster, J. S. (ed.), Habitat
Modification and Freshwater Fisheries. U.N. Food and Agriculture
Organization, Rome: 131–146.
Scott, W. B. & E. J. Crossman, 1973. Freshwater Fishes of Canada, 3rd ed. Galt House Publications Ltd., Oakville, ON.
Simon,
T. P., 1998. Assessment of Balon’s reproductive guilds with application
to Midwestern North American Freshwater Fishes. In Simon, T. P. (ed.),
Assessing the Sustainability and Biological Integrity of Water Resources
Using Fish Communities. CRC Press, New York: 97–122.
Stauffer,
Jr., J. R., R. Taylor, R. Cleaver Yoder, B. D. Lorson, K. M. Taylor
& B. Fost, 2010. Larval fish survey of the navigational channels of
the Ohio River system. US Army Corps of Engineers, Pittsburgh, PA: 391
pp.
Thomas,
J. A., E. B. Emery & F. H. McCormick, 2005. Detection of temporal
trends in Ohio River fish assemblages based on lockchamber surveys
(1957-2001). In Rinne, J. N., H. R. Hughes & R. Calamusso (eds),
Historical Changes in Large River Fish Assemblages of the Americas.
American Fisheries Society, Bethesda, MD: 431–450.
Thorp,
J. H. & M. D. Delong, 1994. The riverine productivity model: an
heuristic view of carbon sources and organic processing in large river
ecosystems. Oikos 70(2): 305–308.CrossRef
Trautman, M. B., 1981. The Fishes of Ohio. Ohio State University Press, Columbus.
Trudeau,
V. & J. B. Rasmussen, 2003. The effect of water velocity on stable
carbon and nitrogen isotope signatures of periphyton. Limnology and
Oceanography 48(6): 2194–2199.CrossRef
Vadeboncoeur,
Y., E. Jeppesen, M. J. Vander Zanden, H. H. Schierup, K. Christoffersen
& D. M. Lodge, 2003. From Greenland to green lakes: cultural
eutrophication and the loss of benthic pathways in lakes. Limnology and
Oceanography 48(4): 1408–1418.CrossRef
Vander
Zanden, M. J. & J. B. Rasmussen, 1999. Primary consumer delta C-13
and delta N-15 and the trophic position of aquatic consumers. Ecology
80(4): 1395–1404.CrossRef
Vander
Zanden, M. J. & J. B. Rasmussen, 2001. Variation in delta N-15 and
delta C-13 trophic fractionation: implications for aquatic food web
studies. Limnology and Oceanography 46(8): 2061–2066.CrossRef
Vander
Zanden, M. J. & Y. Vadeboncoeur, 2002. Fishes as integrators of
benthic and pelagic food webs in lakes. Ecology 83(8): 2152–2161.CrossRef
Vander
Zanden, M. J., Y. Vadeboncoeur, M. W. Diebel & E. Jeppesen, 2005.
Primary consumer stable nitrogen isotones as indicators of nutrient
source. Environmental Science & Technology 39(19): 7509–7515.CrossRef
Vannote,
R. L., G. W. Minshall, K. W. Cummins, J. R. Sedell & C. E. Cushing,
1980. The river continuum concept. Canadian Journal of Fisheries and
Aquatic Sciences 37: 130–137.CrossRef
Veech, J. A. & T. O. Crist, 2009. PARTITION: Software for Hierarchical Partitioning of Species Diversity, Version 3.0. http://www.users.muohio.edu/cristto/partition.htm.
Vorosmarty,
C. J., M. Meybeck, B. Fekete, K. Sharma, P. Green & J. P. M.
Syvitski, 2003. Anthropogenic sediment retention: major global impact
from registered river impoundments. Global and Planetary Change 39(1–2):
169–190.CrossRef
Ward,
J. V. & J. A. Stanford, 1983. The serial discontinuity concept of
river ecosystems. In Fontaine, T. D. & S. M. Bartell (eds), Dynamics
of Lotic Ecosytems. Ann Arbor Science Publishers, Ann Arbor, MI: 29–42.
Webster,
J. R. & B. C. Patten, 1979. Effects of watershed perturbation on
stream potassium and calcium dynamics. Ecological Monographs 49: 51–72.CrossRef
Environmental Management
© Springer Science+Business Media New York 2013
10.1007/s00267-013-0212-8The Role of Published Information in Reviewing Conservation Objectives for Natura 2000 Protected Areas in the European Union
(1)
European Topic Centre on Biological Diversity, Museum National d’Histoire Naturelle, 57 Rue Cuvier, 75231 Paris, France
Received: 8 February 2013Accepted: 25 November 2013Published online: 7 December 2013
Abstract
Protected areas are designated to
protect species and other features known to be present at the time of
designation, but over time the information about the presence of
protected species may change and this should call for a continued review
of conservation objectives. Published scientific literature is one of
the possible information sources that would trigger a review of
conservation objectives. We studied how published data on new
discoveries of protected animal species were taken into account by the
nature conservation authorities in updating species lists of Natura 2000
sites in the European Union, which are the basis for conservation
planning at the site-level. Over the period studied (2000–2011) only
40 % of published new protected species records were recognized by the
authorities. The two main reasons for this seem to be a reliance on
other sources of information by authorities and the difficulty in
finding relevant information in scientific papers. The latter is because
published faunistic information is very fragmented among different
journals, and often insufficient in details. We recommend better
cooperation between authors, publishers, and nature conservation
authorities in terms of information presentation, publishing policy, and
a regular review of published information.
Keywords
Protected species
New records
Scientific knowledge
Management objectives
Published information
Natura 2000
Introduction
One of the key steps in systematic
conservation planning is the identification of conservation objectives
for protected areas (Margules and Pressey 2000; Louette et al. 2011).
For each protected area, conservation objectives are defined to
conserve targeted species, habitats, and other features, often linked to
legislation or other obligations such as the European Union (EU)
Habitats Directive or the Ramsar Convention on wetlands. However, over
time some species may be lost from a site, others may be discovered by
surveys or monitoring activities, either newly arrived or just not
previously recorded (Gaston et al. 2008).
Therefore management plans of protected areas in most countries include
monitoring and periodic review of conservation objectives, usually at a
frequency of at least once every 10 years (Kruk et al. 2010).
The 1992 EU Habitats Directive (a
legal instrument applying to all Member States) established the Natura
2000 network of protected areas targeted at habitats and species listed
in annexes to the directive together with sites designated under an
earlier directive on wild birds. The network now includes more than
26,000 sites and is one of the largest networks in the World (Sundseth
and Creed 2008; Evans 2012).
It is regarded as largely complete for terrestrial and freshwater
habitats and species but incomplete for marine features. Natura 2000
network plays a key role in addressing the 2020 target of halting
biodiversity loss in the European Union (European Commission 2011a).
The Habitats Directive includes a
system for periodic review of the conservation objectives of each site
(European Environment Agency 2012a). Each site in the network is described using an agreed format known as a Standard Data Form (European Commission 2011b).
Together with the details such as name, area, and coordinates, the form
notes each of the habitat types and species listed on Annexes I and II
of the Habitats Directive and Annex I of the Birds Directive that occur
in the Natura 2000. The form is electronic and the forms for the
>26,000 sites are stored as a database managed by the European
Environment Agency. Site management plans must take into account the
ecological requirements of the species listed on the Standard Data Forms
(European Commission 2012a)
and the European Commission demands that the forms are revised
regularly to provide up to date information on the sites. An annual
update has been suggested although to date the frequency has varied
greatly between Member States (European Environment Agency 2012a).
The date that a targeted species or habitat is first noted on the form
for a given site can be considered as the legal starting point for
planning and implementing conservation actions under the EU directives
to maintain or restore the feature at favorable conservation status.
Information on changes in species
and habitats present on a site can be collected and made available to
the national or, in some countries, regional authorities who are
responsible for revising the Standard Data Forms in several ways, and
Table 1
summarizes this process in three consecutive steps: data collection,
data communication, and data recognition. It is most likely that the
largest proportion of new information arrives from national Natura 2000
monitoring schemes designed and funded by national or regional nature
conservation authorities (e.g., Goverse et al. 2006; Mehtala and Vuorisalo 2007; Henry et al. 2008).
In this case, there should be an automatic flow of information from
field to Standard Data Forms. Data on species, and to lesser extent
habitats, are also available from other sources including scientific
projects, activities of non-governmental organizations (NGO), and
citizen science programs (Schmeller et al. 2009).
It is not clear how such information is communicated to the authorities
or how such information is dealt with by the authorities.
Table 1
Logical framework of
projected data-flow on how information on new discoveries of species in
existing Natura 2000 sites reach and are dealt with by the national
nature conservation authorities
Data-flow stages
|
Our assumptions
|
|---|---|
How collected?
| |
Authority funded study, monitoring of protected areas
|
Clear
target is to observe changes in species composition and trends. The
results would most likely be reflected in Standard Data Forms through
technical reporting
|
Ecological studies where species inventories are not a primary target
|
Even
if inventories of flora and fauna are not a subject of the study,
still, the fact that a species of interest was studied in Natura 2000
site could attract attention by the authorities. Result reflection in
Standard Data Forms is uncertain
|
Miscellaneous scientific projects and activities by individual scientists
|
Some
scientists are targeted on faunistics or floristics (particularly of
less studied taxa) and systematics. Result reflection in Standard Data
Forms is uncertain
|
NGO activity and citizen science
|
Potentially
a great source of information, especially in countries where this is
developed. Result reflection in Standard Data Forms is uncertain
|
How communicated to national authorities?
| |
Oral or written: informal
|
Probably
a rare occasion as authorities usually need written record, but this
cannot be excluded. This happens in the Member States where
administration traditions are not very formal and there is a good
communication between authorities and scientists
|
Written: technical report
|
Mostly
brings information from scientific inventory activities that were
initiated with the aim to monitor changes in Natura 2000 sites
|
*Written: published
|
Theoretically any study can be reflected there and information given may overlap with all other ways of communication
|
Online: database
|
Although
in an increasing number of European countries online databases report
new species findings, it is uncertain how and whether this information
reaches responsible authorities
|
How treated by the authorities?
| |
*Recognized and accepted
|
Best outcome, but still it is not possible to track which communication type has contributed to this
|
*Recognized but disregarded
|
High
uncertainty! Not possible to detect the fact whether it was recognized
but disregarded for a variety of reasons or simply not accepted, again
due to a number of possible reasons
|
*Not recognized
|
High
uncertainty! Probably authorities are not paying enough attention about
data that are available apart studies organized by themselves
|
In this paper we focus on data
published in scientific journals, the media that we assume forms the
majority of non-official data available to the public, including
responsible authorities. The advantages of this form of communication
are that (1) information becomes public, (2) it is usually
peer-reviewed; giving the results increased credibility, and (3) it is
likely to provide sufficient details (e.g., when, where, how many
individuals, etc.) allowing a critical examination before inclusion into
the relevant Standard Data Form(s). Published information is
appreciated by nature conservation authorities while this is not often
the case with informal ways of communication (e.g., Postiglione 2006).
Although the designation of protected areas for biodiversity has a long history (European Environment Agency 2012b),
research is still ongoing to help select the most appropriate protected
areas for species and habitats in terms of quantity, quality, and
connectivity (e.g., Parrish et al. 2003; Schabetsberger et al. 2004; Ioja et al. 2010; Jantke et al. 2011).
Only relatively recently has the need for setting conservation
objectives and implementing management actions in designated protected
areas been emphasized or re-emphasized, at least in connection with
Natura 2000 (e.g., Ostermann 1998; Louette et al. 2011).
The conservation objectives of protected areas are dependent on the
species and habitats present, so any changes in protected species
composition should be taken into account. Therefore, in this study we
firstly aimed to assess the number and availability of new species
records from existing Natura 2000 sites that are published in scientific
journals. Secondly, we checked whether such information is taken into
account by nature conservation authorities and recorded in the Standard
Data Forms and what potential factors may affect recognition of a new
species record. Finally, we investigated the relationships between the
events of protected area designation, new protected species record,
publishing of this record, and the start of conservation action for the
species in a particular protected area. Analyses of these important
links are virtually absent from the scientific literature, among related
subjects we found only studies examining factors that affected the
timing of species listings under the USA Endangered Species Act (Tear et
al. 1995; Ando 1999), but this is of greater relevance to any future review of annexes of European Union nature directives (e.g., Cardoso 2012; Maes et al. 2013).
Methods
General Assumptions
We searched for information on
four important events in the process of updating lists of protected
species in existing Natura 2000 sites: date of site designation, date of
discovery of a previously not listed protected species, date of
publication of this record, and date of entry of a new record in the
Standard Data Form of the site (Fig. 1).
We modeled possible flows of these events that would enable us to
assess how important was the role of publications in including newly
recorded species in Standard Data Forms of existing Natura 2000 sites.
Further interpretation of sequences of events and the time periods
between the events is given in Table 2.
Fig. 1
Possible sequences
of the important events for updating species lists for individual
Natura 2000 sites: four modeled scenarios. Event legends: A site designation, B discovery of new species in the site, C date of publication of this new record, D entering the record in the SDF
Value
|
Name
|
Assumptions
|
|---|---|---|
… → B
|
Species already in SDF before reported discovered by the study
|
Another,
older dataset probably used by authorities, or the ‘new’ discovery of
species comes from a new locality within existing protected areaa
|
B → C
|
Time between discovery and publication
|
Less
time indicates author’s responsibility, quality of manuscript (as most
journals claim to be peer-reviewed), and the journal’s ability to fit an
important message as early as possible in its publication schedule
|
C → D
|
Time between publication and entering the species record in SDF
|
Less
time indicates a good ability of nature conservation authorities to
review scientific news unless the fact of species discovery is not
reported in parallel from another source, which is also a good outcome
|
B → D
|
Time between discovery and entering the species record in SDF
|
Pure conservation value, sooner it is, the better
|
B → …
|
Species not reported in SDF
|
If
species record does not reach SDF, this is a useful reference for the
European Commission to ask reasons why this has not yet happened
|
This study was based on a
comprehensive literature review and by analyzing historical editions of
Natura 2000 databases submitted by European Union Member States to the
European Commission. The following steps were implemented and described
in following chapters: (1) finding new species records for existing
Natura 2000 sites in scientific literature, (2) linking these records to
an existing Natura 2000 site, (3) examining if and when this record was
entered in the Standard Data Forms.
New Species Records from Existing Natura 2000 Sites
Literature review was performed
through a combination of key-word searches using Google Scholar and
Scopus search engines and a systematic review of the contents of the
journals which publish relevant faunistic reviews, as we restricted this
study to animals (Appendix). The majority of these journals were
published by a national academic or non-governmental organization, but
exceptionally some interesting records were also found in international
journals. Our primary aim was to detect any potentially first (new)
records of Habitats Directive Annex II animal species in European Union
countries. In some cases, however, it was not possible to determine
whether the species record is actually ‘new’ or ‘first’, as the aim of
the paper was to describe the current fauna of the site without direct
reference to the previous inventories. Such publications were at first
included in this study, but rejected at later stages of data filtering
when, for example, it proved that Standard Data Forms contained a record
of a species earlier than it was reported found by the respective
publication (last option; Fig. 1).
Useful information for our
purpose was typically found in original papers of the following
categories: (1) papers reporting newsworthy findings of a species in new
geographical locations, (2) inventory results of a geographical unit,
(3) ecological research of particular species or a group of species, and
(4) literature reviews. The last category, however, was used only in
relatively few cases when reviews provided recent information collected
over a short period of time.
European and national Atlases on
the distribution of species of different taxa were not considered as in
most cases we could not obtain precise locations of observations or
dates of records. In addition, the majority of published European and
national Red Lists do not show the exact locations of threatened
species, as well as lack exact dating, and therefore could not be used.
We did not consider available online databases whose information to some
degree overlaps with that of paper publications. Some European Union
countries have well-functioning online reporting systems for species
recording, including those of Annex II of the Habitats Directive (for
example, in Sweden and the United Kingdom), which are largely based on
citizen science. We acknowledge this data source as very important
(Table 1), particularly in a view to the future, possibly even taking over the role of the published information (see “Discussion” section).
We focused on species discovery
reports that were published in a period that would correspond to the
post-designation period of most Natura 2000 sites in different groups of
European Member States. For the 15 ‘old’ Member States we examined a
publication period between 2000 and 2011, for the 12 ‘new’ Member States
who joined the European Union in or after 2004 we focused on period
between 2004 and 2011, with the exception of Bulgaria and Romania, who
joined only in 2007 (2007–2011). We systematically searched 60 journals
(Appendix). A few references came from other journals whose contents we
could not access in full and from conference proceedings. Our work
cannot be considered as a complete review of the publications reporting
new species in existing Natura 2000 sites in a given period, but we
believe it is sufficiently large to show trends in information use.
The date of the new record in
most cases was attributed to a month/year precision. Exceptionally in
some studies authors provided only the dates of the study period, but
not the exact date for a record of a particular species of interest. In
such cases we used the completion date. As regards the date of
publishing, in most cases it was also possible to record exact
month/year based on information provided in journals’ web-pages.
The Standard Data Forms have a field ‘Documentation’ where all site-specific references can be recorded (European Commission 2011b).
However, this data field is an ‘optional’ category, and such
information has not systematically recorded by all Member States. A
quick examination showed that 9 (of 27) Member States had no or very few
records in this field. Another eight states had records completed for
approximately 50 % of sites and the remaining ten had records for all or
nearly all sites. But even where records were present, they were mostly
internal references (legal acts) related to site establishment or only
references to a few species suggesting that these entries were filled at
the time of site designation and have never been updated.
During our search for new
records in existing Natura 2000 sites we came across situations when
papers reported the complete opposite—the absence of species from the
Natura 2000 site where it was previously reported (e.g., de Bruyne 2002; Romano et al. 2007; Strugariu et al. 2008).
We did not analyze such records and possible nature conservation
authority responses any further as we believe that applying the
precautionary principle, there should be more research than one negative
observation before removing the species from the Standard Data Form as
species may be re-discovered after some time (e.g., van Dijk 2006).
Instead, the appropriateness of current management should be assessed
to avoid local species extinctions, as, for example, is described by
Konvicka et al. (2008).
Linking Records of Species to an Existing Natura 2000 Sites
The information in publication
on the new species record was used to link with an existing Natura 2000
site. We used the following cues: Natura 2000 site code or name, a name
of a national protected area (often matching with the name of Natura
2000 designation), geographical coordinates (enabling us to use ESRI
ArcGIS 10.0 to establish a link with a Natura 2000 site), a detailed map
of a study area (enabling us to judge with a high confidence whether
the location falls into a Natura 2000 site), or kilometers and direction
from a local village/town, or a point on a small scale map that fell
within borders of existing Natura 2000 site. In total, our web-based
search yielded 182 relevant publications reporting at least one
potentially new record of an Annex II species in a Natura 2000 site, and
414 such records in total.
Information from the Standard Data Form
Analysis of historical Natura
2000 databases of the European Environment Agency archive provided us
with the date for when the site was designated and the date when the
species was first entered in the Standard Data Form of a particular
site. This analysis also indicated which published records were not
recognized by the authorities and were absent from Standard Data Forms
by December 2011. Dates were obtained with a minimum precision of
month-year.
Analysis of Natura 2000
databases, however, reduced the sample size as 71 species records
appeared to occur prior to site designation, suggesting that Standard
Data Forms were not complete at the time of designation. Other 113
species records, as reported by publications, appeared after the
appearance of species record in the Standard Data Forms. Thus the final
sample size comprised 190 genuinely new records for the existing Natura
2000 sites from 101 scientific papers covering 19 (out of 27) European
Union Member States and 61 species.
Given that publications were the
starting point of this study, we do not know the number of new species
records in Natura 2000 sites that were not published and thus not
detected by our literature search (i.e., many might pass event C, as in
Fig. 1).
In order to acquire at least some data about this, we examined how many
species have been added to existing Natura 2000 sites in the period
2009–2011 and cross-checked which additions might have been a result of
relevant publications detected in this study.
Statistical Methods
All statistics were performed using R software (R Development Core Team 2012).
We used survival analysis to study the ‘lifetimes’ and distributions of
published new records of protected species in Natura 2000 sites until
they are taken into account by nature conservation authorities. This was
the reverse of the usual application of this method, for example in
medical sciences, where a terminating event (hazard) of observation
means ‘death’ while in our case it is positive, i.e., recognition of a
new species record. Our dataset was right-censored as many of the
published papers may be still recognized (i.e., entered in the Standard
Data Forms) after the closing date in this study, i.e., December 2011.
We used the non-parametric Kaplan–Meier estimator of survival
probability to observe patterns of new species record acceptance
probability by authorities over time since species discovery and
publication (Fig. 3).
Chi square tests of the R function ‘survdiff’ were used to analyze
differences in lifetime distributions by different factors: publication
type, language of publication and whether a particular records belongs
to an ‘old’ or ‘new’ (joined in or after 2004) European Union Member
State.
Results
Only 40 % of a total of 190
published records of protected species in existing Natura 2000 sites
(our sample size) had been entered in Standard Data Forms by December
2011. Reported new species were discovered from a few months up to
16 years after designation of Natura 2000 site (Fig. 2, events A–B). On average, new species records were published within less than 2 years after recording (Fig. 2,
events B–C). Forty-three percent of new records were actually first
entered in Standard Data Form and only later published (on average
16.2 months later; Fig. 2,
events C–D negative values). The remaining 57 % records were first
published and then entered Standard Data Forms on average 21.7 months
later (Fig. 2, events C–D positive values).
Fig. 2
Distribution of time intervals between events with mean values and Standard Errors. Legend
A–B Time from site designation to discovery of new species. B–C Time from discovery of new species to date of publication. B–D Time from discovery of new species to entry in the SDF. C–D Time from date of publication to entry in the SDF. Negative values for C–D indicate cases when event D happened before C
Of 41 publications reporting more
than 1 new species record, 39 % had partial recognition of new records,
i.e., some species from the same publication were entered in Standard
Data Forms, some not. This indicates that authorities were somehow
selective in using published information or, alternatively, different
source of information were used that reported only the part of the
records.
The probability of the new species
record being entered in Standard Data Forms showed a steady increase
over time since species discovery and since publication of the new
record (Fig. 3). The most prominent increase of probability happened soon after discovery, but in most cases before publication (Figs. 2, 3a); this suggests the presence of informal communication between researchers and authorities. Figure 3b
shows that after approximately 80 months (or 6.6 years) after
publication the probability of recognition (i.e., entering in SDF) does
not increase. This means that approximately 80 % of the so far
non-recognized records still have a chance of being entered in SDFs, as
the margin of 80 months since publication of the record has not yet
passed. Probabilities of entering a record in a SDF were marginally
higher in new European Union Member States than ‘old’ Member States (χ
2 = 3.8; d.f. = 1; P = 0.05), but did not differ between papers published in English or other language (χ
2 = 1.2; d.f. = 1; P = 0.27) and between types of published papers (χ
2 = 1.5; d.f. = 3; P = 0.68).
Fig. 3
Cumulative
probability (Kaplan–Meier) and 95 % confidence intervals of entering the
newly discovered species into Standard Data Forms since time of
discovery (a) and since publishing the record of discovery (b)
In the 3-year period (2009–2011),
355 new species records were added to the Standard Data Forms of
existing Natura 2000 sites in 16 Member States of the European Union.
Only 14 of those records were possibly triggered by publications found
in this study.
Discussion
While the availability of biodiversity data is considered to be a major issue in conservation (Bisby 2000; Yesson et al. 2007),
our results showed that the current impact of available scientific
literature on updating management objectives for Natura 2000 sites is
low. As often indicated by authors, one of the most important
motivations for publishing faunistic news is to improve conservation of
species. Then a situation where 60 % of published records are not
recognized by the authorities is not satisfactory. At the same time we
see that the process of updating of the Standard Data Forms is still
ongoing as most Member States are apparently aware of the need for such
updates, but they are clearly not initiated by scientific papers whose
contribution accounts for less than 4 % of all new entries in Standard
Data Forms. Therefore it is essential to understand the reasons for this
situation, firstly by identifying the data sources that ‘compete’ with
published papers in bringing information to the attention of nature
conservation authorities and secondly by analyzing what does not work
with existing scientific media as there is such a low rate of using
published information in practical conservation.
It seems that the main
contributors of new faunistic data to Standard Data Forms are the Natura
2000 monitoring programmes which are run in most countries. Although
there is a variation among Member States, data collected are reported
directly to nature conservation authorities without ‘formalizing’ this
information in the form of scientific publications and thus without
making this information widely accessible. Even if such reports are
available on the web, or upon request from agencies, the search for them
could be cumbersome requiring personal contacts and possibly knowledge
of the local language (Amano and Sutherland 2013).
A portion of the new data may be later published by individual authors
participating in such monitoring programmes (some authors indicated the
source of funding in their publications), and this could be an
explanation why so often (43 %) new species records were first entered
in Standard Data Forms and only later published.
Likewise, much faunistic data have been gathered by numerous projects funded by the European Union LIFE programme (http://ec.europa.eu/environment/life/project/Projects/index.cfm).
Even though species inventories were not a major component of many
projects, an element of species inventory was probably present in many
of them and brought many new observations as LIFE projects focused
mainly on existing Natura 2000 sites. But even if each project has its
own web-page where the main achievements are described, as well as the
European Commission’s online database of all past and present LIFE
projects is freely accessible, the available reports rarely provide
necessary details of new species records and their locations.
The other potential holders of
information about the distribution of protected species are web-sites
with online databases which are usually run by national nature
conservation agencies often with contributions by citizen scientists.
Participation of the public ensures large numbers of incoming records,
in some cases with million records per year, and scientific papers will
never be able to compete with them in quantity of observations
(Schmeller et al. 2009).
Other advantages are that an online database reflects the findings
almost immediately (while publishing a new species record takes time: on
average almost 2 years as shown in this study) and also that writing up
the paper could be a challenge for some potential data providers, while
inserting a new observation in a database only takes a few mouse
clicks.
Such web-sites and public databases of species records are well-developed, for example, in Sweden (Swedish Species Gateway; http://www.slu.se/en/collaborative-centres-and-projects/artdatabanken/) and United Kingdom (National Biodiversity Network; http://www.nbn.org.uk/).
In these countries, in the field of biological recording, citizen
scientists have largely replaced professional scientists (Silvertown 2009).
This may be the reason why we could not find any relevant papers for
this study from Sweden and the United Kingdom. However, the rate at
which national nature conservation authorities are using new information
from public databases remains unknown. However, in many other
countries, for example in Eastern Europe, nature conservation is a
relatively new addition to the political agenda and public involvement
in data collection is still not actively practiced by the various levels
of government or such involvement is not well-developed (e.g., Eben 2006; Schmeller et al. 2009). Here published information should still have a role in decision-making but this could change in future.
Our study showed that there is
still a significant proportion of new species records that were missed
or possibly not recognized by authorities, suggesting that scientific
publications do contain some information that authorities cannot obtain
except by checking journals. Even though 190 relevant records found by
our study compares poorly with more than 300,000 species records in the
European Natura 2000 database of December 2011, in theory, each single
new record should be validated by the relevant authorities.
If we assume that authorities are
aware of the publication about the discovery of new protected species in
existing Natura 2000 sites, we could think of two reasons why this
information is not taken into account in updating Standard Data Forms
and site conservation objectives. First, a minority of European Union
Member States are not adding any new features on Standard Data Forms in
principle as they consider the set of species at the site designation as
final conservation objective and new species is seen as an additional
management burden at the sites. In our opinion, this approach cannot be
justified, as most often species are missing from the ‘initial lists’
due to lack of information at the time of site designation (see also
Lozano et al. 2013),
but not due to temporal changes in species presence at the site.
Secondly, the absence of recognition could be explained by the fact that
many ‘first records’ are often observations of species in extreme
locations with respect to their distribution range, often accidental
(e.g., in mammals), and one such observation does not immediately oblige
authorities to recognize it as a subject to define conservation
objectives for a site.
But apparently the main reason of
not-using published information is because it was not searched for or a
search did not yield the expected results. To search biological
information in the web, in spite of today’s powerful search engines,
including Google Scholar available to practically everyone, requires
some time and experience and such persons may not be available. It is
unlikely that administrative staff in ministries would search for
literature themselves; this is rather a job for agencies or contracted
scientists. It is also known that the recent economic crisis forced
environment ministries and agencies to significantly reduce staff that
could otherwise be directed to better monitoring of published
information.
Even if someone is willing to
search for newly published information, the following difficulties are
likely to be encountered. Using different keyword combinations (e.g.,
involving species and country names, etc.) may produce only partial
lists of potentially relevant papers. But from these papers, it is
possible to identify journals which could be searched systematically as
it is likely that they were publishing other similar papers. Still, even
a list of journals produced in such a way may not be complete. It may
prove useful to also search for European scientific societies (for
example, entomological societies) and then look for the ‘publications’
section of their web-pages. Almost in each country it is not sufficient
to look only in their home journals because many, if not most,
researchers also publish their results in journals abroad.
During our search for information
for this paper we observed that, except for some sensational new
records, very few international peer-reviewed journals publish
descriptive faunistic reviews but focus exclusively on analytical and
experimental studies. Journals with low Impact Factors, or with no
Impact Factor reported, and most often published by national or regional
scientific and conservation societies and natural history museums
(Appendix), proved most important. But many such journals mostly publish
faunistic observations from less studied exotic regions of the world
than the territory of the European Union. The ratio of home-related
articles is often at best 1:10. This is particularly true for North-West
Europe which is well surveyed and it is less likely that new species
will be found in new locations. Meanwhile, there are a number of
journals with relevant content in Eastern and Mediterranean Europe. Even
the relatively few papers from the North-West Europe did not much
contribute to knowledge about protected species in Natura 2000 sites.
For example, it was hard to find any published observations on protected
bat species from most of Finland, however, extreme observations above
the Arctic Circle were still published (Siivonen and Wermundsen 2008).
Also, searches should not be restricted to faunistic papers as we came
across a number of analytical ecology studies (e.g., habitat selection,
food analysis, behavior) on Habitats Directive Annex II species in
Natura 2000 sites (suggesting that there are abundant populations making
such study possible) but 70 % of such populations were not reported in
Standard Data Forms (e.g., reported by Scalici and Gilbertini 2007; Tartally and Varga 2008; Zografou et al. 2009).
In conclusion, the potentially useful information is much dispersed and
a search might take a considerable amount of time, particularly when
undertaken for the first time.
Further difficulties can be
encountered after an interesting title and journal has been identified.
Very often abstracts, usually available in all cases, are not enough to
validate a new species record and to check if it is linked to an
existing Natura 2000 site. Most journals where we found relevant papers
followed a policy of free-access, and full-length papers can be accessed
if the respective web-page is functioning correctly. However, many
interesting journals with higher Impact Factors require payment to
access or an institutional subscription, often not available to staff in
conservation agencies and ministries. Some journals of scientific
societies first require a membership to the society before one can
access journal contents. Some journals have not yet made their older
volumes available electronically. Eventually, if a reader has
successfully acquired a promising full-length paper, s/he may soon be
disappointed by its quality. Sometimes authors are not very cooperative
in presenting data in a complete and easily understandable way or simply
do not understand how to allow the best use of their data for practical
conservation. In our study in some papers we struggled to find even
such basic elements as the date (period) and place of observation; this
problem has been previously noted by authors of review papers (e.g.,
Boldogh 2006).
However, keeping high publishing standards is also a responsibility of
editors and reviewers as almost all of journals we studied claim to be
peer-reviewed.
On a more positive note, our
results showed that when published records of new species were
recognized by nature conservation authorities and entered in the
Standard Data Form, the mean time interval between the species discovery
and recognition (entry in Standard Data Form) was only around 2 years
(Fig. 2
B–D). An even more remarkable finding was that very often the entry in
Standard Data Forms occurred before publication, which indicates the
informal communication between researchers and authorities. This seems
to be a case particularly in relatively small European Union countries,
for example in Hungary and Latvia, where ‘everyone knows everyone’ and
new information quickly finds its way from the field to responsible
authorities (Zita Zsembery, Ilona Mendzina, pers. comm.). Also, many
extreme cases of delayed recognition probably have explanations, for
example, the snail Vertigo moulinsiana was collected in a Latvian site in 1997 and incorrectly identified as a very similar species Vertigo geyeri.
After a number of years, the collection was revisited and the
identification error was corrected and the record was published 7 years
after discovery in 2004 (Pilate 2004).
Even if the Natura 2000 network is now considered to be close to completion (European Commission 2012
b),
there will always be a need to update site records that are essential
in reviewing conservation objectives at the site-level. It seems that
the currently insufficient use of published papers in updating Natura
2000 Standard Data Forms is due to the difficulty of finding relevant
information rather than due to ignorance from the nature conservation
authorities. It is also clear that published papers are currently not
the most important source in updating information about protected areas.
Scientific journals may lose this role completely in the near future
unless they become better organized in terms of providing freely
available, exact, and concise information about new records of protected
species in the Natura 2000 network. We call for better cooperation
between field scientists/authors, publishers, and nature conservation
authorities as any new information on presence of species at the
site-level can be very useful for conservation even in such a relatively
well-studied region as Europe. To authors reporting new species records
we recommend to indicate clearly (1) precise coordinates and (2) Natura
2000 site details (code and official name can be found in:
http://natura2000.eea.europa.eu/#), date(s), numbers, as well as any
other information, e.g., habitat requirements, that could be useful for
future management planning to fulfil conservation objectives.
It could be worthwhile to examine
experience at a global scale in communicating new faunistic information.
For example an existing journal ‘Check List: Journal of Species Lists
and Distribution’ (http://www.checklist.org.br/)
is a bimonthly peer-review online journal, devoted to publishing lists
of species and notes on the geographic distribution of any taxon as such
reports have traditionally been neglected in other journals. As this
journal focuses mainly, but not exclusively, on the Americas, a similar
European journal (or electronic platform) would be most welcome. This
could help to prevent the current fragmentation of information and feed
not only the Natura 2000 process with new data but also support
red-listing and reviewing the lists of protected species.
We also recommend that national
authorities responsible for nature conservation should regularly check
the recent scientific literature for data completeness, even if they
have other sources of information available. Published literature is
probably the most cost-effective mean of obtaining information on new
species records.
Acknowledgments
Data
collection and analysis was supported within a framework of the
European Topic Centre on Biological Diversity of the European
Environment Agency. We thank Mora Aronsson, Marita Arvela, Michael
Hosek, Ilona Mendzina, Dominique Richard, Carlos Romao, Frank Vassen,
and Zita Zsembery for useful comments on earlier drafts of this paper.
Bruno Opermanis helped with R. We thank the three referees for valuable
suggestions on manuscript improvement. The views expressed are those of
the authors and should not be taken as views of the European Topic
Centre on Biological Diversity.
Appendix
List of journals which were systematically searched for this study (publishing authority in brackets)
Acta Chiropterologica (Museum and Institute of Zoology, Polish Academy of Sciences)
Acta Entomologica Slovenica (Slovenian Entomological Society)
Acta Entomologica Musei Nationalis Pragae (National Museum Prague)
Acta Herpetologica (Firenze University)
Acta Ichthyologica et Pisatoria (West Pomeranean University of Technology)
Acta Societatis Zoologicae Bohemicae (Czech Zoological Society)
Acta Theriologica (Polish Academy of Sciences)
Acta Universitatis Latviensis, Biology (University of Latvia)
Acta Zoologica Bulgarica (Institute of Biodiversity and Ecosystem Research)
Acta Zoologica Academiae
Scientarum Hungaricae (Hungarian Natural History Museum/Biological
section of the Hungarian Academy of Sciences)
Acta Zoologica Lituanica (Institute of Ecology/Nature Research Centre)
Annales Zoologica Cracoviensia (Institute of Systematics and Evolution of Animals, Polish Academy of Sciences)
Annales Zoologici (Museum and Institute of Zoology, Polish Academy of Sciences)
Beitrage zur Entomofaunistik (Austrian Socitety for Entomofaunistics)
Biharean Biologist (University of Oradea)
Biologia Bratislava (Slovak Academy of Sciences)
Bonn Zoological Bulletin (Zoological Research Museum Alexander Koenig)
Bulletin de la Société Herpétologique de France (Herpetological Society of France)
Bulletin of the Irish Biogeographical Society (Irish Biogeographical Society)
Cynthia (Catalan Butterfly Monitoring Scheme)
Entomologia Hellenica (Hellenic Entomological Society)
Entomologica Fennica (Entomological Society of Finland)
Entomologische Berichten (Netherlands Entomological Society)
Entomologisk Tidskrift (Swedish Entomological Society)
European Journal of Entomology (Czech Entomological Society)
Faunistische Abhandlungen (Museum of Zoology, Dresden)
Fragmenta Faunistica (Museum and Institute of Zoology, Polish Academy of Sciences)
Folia Malacologica (Association of Polish Malacologists)
Folia Zoologica (Institute of Vertebrate Biology (Brno))
Graelsia (National Museum of Natural Sciences, Madrid)
Gredleriana (South Tyrol Nature Museum)
Herpetozoa (Austrian Herpetological Society)
Hystrix (Italian Theriological Association)
Insect Conservation and Diversity (Royal Entomological Society)
Italian Journal of Zoology (Italian Society of Zoologists)
Journal of Insect Conservation (Springer)
Journal of Conchology (The Conchological Society of Great Britain and Ireland)
Latvijas entomologs (Latvian Entomological Society)
Lynx (Natural History Museum, Prague)
Mammalia (De Gruyter)
Malacologica Bohemoslovaca (Institute of Zoology, Slovak Academy of Sciences)
Mitteilungen der Deutschen Malakozoologischen Gesellschaft (German Malacozoological Society)
Mollusca (Museum of Zoology, Dresden)
Natura Sloveniae (Ljubljana Biotechnical Faculty/National Institute of Biology)
Natura Somogyensis (Somogy County Museums)
Nederlandse Faunistische Mededelingen (Naturalis Biodiversity Center)
New and Rare for Lithuania Insect Species Records and Descriptions (Lithuanian Entomological Society)
North-Western Journal of Zoology (University of Oradea)
Oryx (Flora and Fauna International)
Phegea (Flemish Entomological Society)
Salamandra (German Society for Herpetology)
Silva Gambreta (Sumava National Park)
Small Carnivore Conservation Newsletter (IUCN, Species Survival Commission)
Tentacle (IUCN/SSC Specialist Group)
Tiscia (Tisza Research Committee, University of Szeged)
Travaux du Museum National d’Histoire Naturelle (‘Grigore Antipa’ National Museum of Natural History, Bucharest)
Vertebrate Zoology/Zoologische Abhandlungen (Senckenberg Nature Research Society)
Vespertilio (Czech Society for Bat Conservation)
Wissenschaftliche Mitteilungen aus dem Niederosterreichischen Landesmuseum (State Museum of Lower Austria)
Zoologische Mededelingen (Netherlands Centre for Biodiversity)
References
Amano
T, Sutherland W (2013) Four barriers to the global understanding of
biodiversity conservation: wealth, language, geographical location and
security. Proc R Soc Lond B 280:20122649. doi:10.1098/rspb.2012.2649
CrossRef
Ando
AW (1999) Waiting to be protected under the Endangered Species Act: the
political economy of regulatory delay. J Law Econ 42:29–60CrossRef
Bisby FA (2000) The quite revolution: biodiversity informatics and the internet. Science 289:2309–2312CrossRef
Boldogh G (2006) The bat fauna of the Aggtelek National Park and its surroundings (Hungary). Vespertilio 9–10:33–56
Cardoso P (2012) Habitats directive species lists: urgent need of revision. Insect Conserv Divers 5:169–174CrossRef
de Bruyne R (2002) De nauwe korfslak Vertigo angustior in Nederland. Nederlandse Faunistische Mededelingen 16:11–20
Eben
M (2006) Public participation during site selection for Natura 2000 in
Germany: the Bavarian case., Environmental Science and Engineering.
Stakeholder dialogues in natural resource management. Part IIISpringer,
Berlin, pp 261–278
European Commission (2011a) The EU biodiversity strategy to 2020. European Commission, Luxembourg
European
Commission (2011b) Commission implementing decision of 11 July 2011
concerning a site information format for Natura 2000 sites (notified
under document C (2011) 4892). Off J Eur Union L 198:39–70
European Commission (2012a) Commission note on the setting conservation objectives for Natura 2000 sites. https://circabc.europa.eu/faces/jsp/extension/wai/navigation/container.jsp. Accessed 11 Dec 2012
European Commission (2012b) Environment: EU celebrates 20 years of LIFE and nature protection. Press Release 23 Oct 2012. http://europa.eu/rapid/press-release_IP-12-1134_en.htm. Accessed: 5 Feb 2013
European Environment Agency (2012a) Reference portal for Natura 2000. http://bd.eionet.europa.eu/activities/Natura_2000/reference_portal. Accessed 12 Dec 2012
European Environment Agency (2012b) Protected areas in Europe—an overview. EEA Report No5/2012, Copenhagen
Evans D (2012) Building the European Union’s Natura 2000 network. Nat Conserv 1:11–26CrossRef
Gaston
KJ, Jackson SF, Nagy A, Cantu-Salazar L, Johnson M (2008) Protected
areas in Europe. Principle and practice. Ann NY Acad Sci 1134:97–119CrossRef
Goverse
E, Smit GFJ, Zuijderwijk A, van der Muij T (2006) The national
amphibian monitoring program in the Netherlands and Natura 2000.
Herpetolo Bonn II:39–42
Henry
P-Y, Lengyel S, Nowicki P, Julliard R, Clobert J, Celik T (2008)
Integrating ongoing biodiversity monitoring: potential benefits and
methods. Biodivers Conserv 17:3357–3382CrossRef
Ioja
CI, Patroescu M, Rozylowicz L, Popescu VD, Verghelet M, Zotta MI,
Felciuc M (2010) The efficacy of Romania’s protected areas network in
conserving biodiversity. Biol Conserv 143:2468–2476CrossRef
Jantke
K, Schleupner C, Schneider UA (2011) Gap analysis of European wetland
species: priority regions for expanding the Natura 2000 network.
Biodivers Conserv 20:581–605CrossRef
Konvicka
M, Benes J, Cizek O, Kopecek F, Konvicka O, Vitaz L (2008) How too much
care kills species: grassland reserves, agri-environmental schemes and
extinction of Colias myrmidone from its former stronghold. J Insect Conserv 12:519–525CrossRef
Kruk
RW, De Blust G., Van Apeldoorn RC; Bouwma IM; Sier ARJ (2010)
NATURA2000: Information and communication on the designation and
management of Natura2000 sites: summary main report 2: organizing the
management in 27 EU Member States. Rapporten van het Instituut voor
Natuur-en Bosonderzoek, 2010 (28). INBO, Brussel, 33 pp
Louette
G, Adriaens D, Adriaens P, Anselin A, Devos K, Sannen K, Van Landyt W,
Paelinckx D, Hoffmann P (2011) Bridging the gap between the Natura 2000
regional conservation status and local conservation objectives. J Nat
Conserv 19:224–235CrossRef
Lozano
FD, Atkins KZ, Moreno Saiz JC, Sims AE, Dixon K (2013) The nature of
threat category changes in three Mediterranean biodiversity hotspots.
Biol Conserv 157:21–30CrossRef
Maes
D, Collins, S, Munguira ML, Sasic M, Settele J, Van Swaay C, Verovnik
R, Warren M, Wiemers M, Wynhoff I (2013) Not the right time to amend the
Annexes of the European Habitats Directive. Conserv Lett. doi:10.1111/conl.12030
Margules CR, Pressey RL (2000) Systematic conservation planning. Nature 405:243–253CrossRef
Mehtala
J, Vuorisalo T (2007) Conservation policy and the EU Habitats
Directive: favourable conservation status as a measure of conservation
success. Eur Environ 17:363–375CrossRef
Ostermann OP (1998) The need for management of nature conservation sites designated under Natura 2000. J Appl Ecol 35:968–973CrossRef
Parrish
JD, Braun DP, Unnasch RS (2003) Are we conserving what we say we are?
Measuring ecological integrity within protected areas. Bioscience
53:851–860CrossRef
Pilate D (2004) Vertigo moulinsiana in Latvia. Acta Univ Latviensis Biol 676:127–129
Postiglione A (2006) Impact of Natura 2000 sites on environmental licensing. Italian Report. EUFSE, Helsinki 15–16 Sep 2006. www.eufje/uploads. Accessed 5 Feb 2013
R Development Core Team (2012). http://www.R-project.org. Accessed 5 Feb 2013
Romano
A, Montinaro G, Mattocia M, Sbordoni V (2007) Amphibians of the Aurunci
Mountains (Latium, Central Italy), Checklist and conservation
guidelines. Acta Herpetol 2:17–25
Scalici M, Gilbertini G (2007) Feeding habits of the crayfish Austropotamobius pallipes in a brook in Latium (central Italy). Ital J Zool 74:157–168CrossRef
Schabetsberger
R, Jehle R, Maletzky A, Pesta J, Sztatecsny M (2004) Delineation of
terrestrial reserves for amphibians: post-breeding migrations of Italian
crested newts (Triturus c. carnifex) at high altitude. Biol Conserv 117:95–104CrossRef
Schmeller
DS, Hernry P-Y, Julliard R, Gruber B, Clobert J, Dziock F, Lengyel S,
Nowicki P, Deri E, Budrys E, Kull T, Tali K, Bauch B, Settele J, Van
Swaay C, Kobler A, Babij V, Papastergiadou E, Henle K (2009) Advantages
of volunteer-based biodiversity monitoring in Europe. Conserv Biol
23:307–316CrossRef
Siivonen Y, Wermundsen T (2008) Distribution and foraging of bats in northern Finland: Myotis daubentonii occurs north of the Arctic Circle. Vespertilio 12:41–48
Silvertown J (2009) A new dawn for citizen science. Trends Ecol Evol 24:467–471CrossRef
Strugariu
A, Sos T, Gherghel I, Ghira I, Sahlean TC, Puscasu I,
Hutuleac-Volosciuc MV (2008) Distribution and current status of the
herpetofauna from the northern Macin mountains area (Tulcea county,
Romania). Analele Stiintifice ale Universitati AI.I.Cuza Iasi s. Biol
Anim 54:191–206
Sundseth
K, Creed P (2008) Natura 2000: protecting Europe’s biodiversity. Office
for Official Publications of the European Communities, Luxembourg
Tartally A, Varga Z (2008) Host and use of Maculinea teleius in the Carpathian basin. Acta Zool Acad Sci Hung 54:257–268
Tear
TH, Scott JM, Hayward PH, Griffiths B (1995) Recovery plans and the
Endangered Species Act: are criticisms supported by data? Conserv Biol
9:182–195CrossRef
van Dijk G (2006) Dytiscus latissimus after 38 years rediscovered in the Netherlands in 2005. Nederlandse Faunistiche Mededelingen 24:1–6
Yesson
C, Brewer PW, Sutton T, Caithness N, Pahwa JS, Burgess M, Gray WA,
White RJ, Jones AC, Bisby FA, Culham A (2007) How global is the global
biodiversity information facility? PLoS One 2(11):e1124CrossRef
Zografou
K, Sfenthourakis S, Pullin A, Kati V (2009) On the surrogate value of
red-listed butterflies for butterflies and grasshoppers: a case study in
Grammos site of Natura 2000, Greece. J Insect Conserv 13:505–514CrossRef
Environmental Management
© Springer Science+Business Media New York 2014
10.1007/s00267-013-0222-6Riverine Threat Indices to Assess Watershed Condition and Identify Primary Management Capacity of Agriculture Natural Resource Management Agencies
Jeffrey D. Fore1, 5 , Scott P. Sowa2, David L. Galat1, Gust M. Annis3, David D. Diamond3 and Charles Rewa4
(1)
Department of Fisheries and Wildlife Sciences, University of Missouri, Columbia, MO, USA
(2)
The Nature Conservancy, Michigan Field Office, Lansing, MI 48906, USA
(3)
Missouri Resource Assessment Partnership, School of Natural Resources, University of Missouri, Columbia, MO, USA
(4)
NRCS Resource Assessment Division, Beltsville, MD, USA
(5)
Present address: The Nature Conservancy, West Tennessee Program Office, Jackson, TN, USA
Received: 1 September 2012Accepted: 13 December 2013Published online: 4 January 2014
Abstract
Managers can improve conservation
of lotic systems over large geographies if they have tools to assess
total watershed conditions for individual stream segments and can
identify segments where conservation practices are most likely to be
successful (i.e., primary management capacity). The goal of this
research was to develop a suite of threat indices to help agriculture
resource management agencies select and prioritize watersheds across
Missouri River basin in which to implement agriculture conservation
practices. We quantified watershed percentages or densities of 17 threat
metrics that represent major sources of ecological stress to stream
communities into five threat indices: agriculture, urban, point-source
pollution, infrastructure, and all non-agriculture threats. We
identified stream segments where agriculture management agencies had
primary management capacity. Agriculture watershed condition differed by
ecoregion and considerable local variation was observed among stream
segments in ecoregions of high agriculture threats. Stream segments with
high non-agriculture threats were most concentrated near urban areas,
but showed high local variability. 60 % of stream segments in the basin
were classified as under U.S. Department of Agriculture’s Natural
Resources Conservation Service (NRCS) primary management capacity and
most segments were in regions of high agricultural threats. NRCS primary
management capacity was locally variable which highlights the
importance of assessing total watershed condition for multiple threats.
Our threat indices can be used by agriculture resource management
agencies to prioritize conservation actions and investments based on:
(a) relative severity of all threats, (b) relative severity of
agricultural threats, and (c) and degree of primary management capacity.
Electronic supplementary material
The online version of this article (doi:10.1007/s00267-013-0222-6) contains supplementary material, which is available to authorized users.
Keywords
Agricultural conservation
Threat index
Management capacity
Watershed condition
Threat assessment
Missouri River basin
Introduction
Restoring natural resources is a
process of implementing conservation practices at the correct places to
achieve a desired set of conditions (Palmer et al. 2005).
Decisions on where to focus conservation practices are complicated in
stream ecosystems because sources of environmental stress (hereafter,
threats) can be distributed anywhere within a watershed and may be far
removed from the site of interest, thus highlighting the importance of
considering total watershed condition (Wang et al. 1997).
Managers are increasingly faced with conservation planning over large
spatial extents (e.g., states or large river basins) and need tools to
help prioritize and select streams on which to focus conservation
efforts and resources. Biological assessments of ecological condition
are one such tool, but are incomplete over large spatial extents (often
<1 % of stream miles in a basin is represented; Sowa et al. 2007).
Thus, management agencies have difficulty in selecting and prioritizing
watersheds since ecological condition of unsampled streams is largely
unknown. Managers can instead use existing geospatial datasets to
conduct watershed scale threat assessments to identify overall watershed
condition, identify potential sources of stress in known ecologically
degraded streams, and determine if an agency’s conservation practices
are suitable to address the threats in a watershed (i.e., an agency has
primary management capacity).
Threat assessments are typically
conducted by developing a multi-metric threat index that uses geospatial
data to specify the location and quantify the extent and magnitude of
human threats in a watershed by summarizing watershed condition with a
single score (Danz et al. 2007; Mattson and Angermeier 2007).
Threat indices are advantageous over landuse and individual threat
metric maps (e.g., locations of point-source discharges) because indices
represent overall watershed condition and relativize watershed
condition estimates to the most threatened watershed. Threat indices can
be quantified and mapped at a stream segment (length of stream between
two confluences) resolution over large spatial extents to include sites
lacking a direct assessment of ecological condition. Since most threat
indices are made up of multiple threat metrics that represent an array
of human disturbances, they can sometimes be used to infer the likely
source of environmental stress.
When dealing with watersheds where
ecological degradation is known, managers should use information from
threat assessments to guide conservation practice implementation.
Following the conceptual example in Fig. 1, low Index of Biotic Integrity (IBI; Karr 1981)
scores can identify ecologically degraded streams and individual IBI
metrics (e.g., proportion of lithophilous spawning fishes) that
represent functional community traits and may identify the stressor
(Leonard and Orth 1986).
Stressors can be identified using functional traits of fish communities
(e.g., lithophilous spawners) because they can be linked to specific
physical drivers or processes that are altered by human threats (Poff 1997; Sutherland et al. 2002).
Conservation planning is then improved because managers have identified
the likely threats causing degradation and are able to make informed
decisions regarding the appropriate conservation strategies to address
the appropriate threats (Fig. 1).
Identifying the likely cause of impairment is particularly important
from a logistical standpoint because most resource management agencies
have the ability and capacity to address only a select suite of threats.
Fig. 1
Conceptual diagram
illustrating potential decision pathways and outcomes of conducting
ecological condition and threat assessments. Solid arrows
represent the alternative decision paths which resource managers could
follow when conducting each assessment independent of the other. Dotted arrows and borders represent decision pathways and potential assessment outcomes when ecological condition and threat assessments are coupled. Italic font
represents intermediate outcomes of the decision path (e.g., lithophils
were identified as limiting biotic integrity in the ecological
condition assessment)
Although currently developed
threat indices are useful for conservation planning because ecological
condition can be inferred from threat information, they are not designed
to identify primary management capacity. Agencies who use estimates of
ecological degradation to focus their conservation programs may be less
successful in improving ecological condition if they work in watersheds
where they do not have primary management capacity (because the agency’s
conservation practices are not designed to remediate the threats
causing degradation). The primary management capacity of the US
Department of Agriculture’s Natural Resource Conservation Service (NRCS)
lies in working with producers on mostly privately owned agriculture
lands. The agency’s conservation practices principally address
environmental stresses caused by agriculture activities, not urban or
industrial activities. In the absence of interagency coordination,
conservation practice effectiveness would be greatest in watersheds
where the most prevalent threats are within the agency’s primary
management capacity and threats outside its capacity were minimal.
Coordination among multiple resource management agencies can be
facilitated by knowing the prevalence of threats both within and outside
each agency’s primary management capacity, thus increasing conservation
practice effectiveness. We argue that resource management agencies,
like NRCS, would benefit from having stream segment-scale threat indices
to assess the relative watershed contribution of various threats (e.g.,
agriculture vs. urban) for use in strategically allocating resources
and more effectively coordinating with other management agencies across a
large geography and multiple spatial scales.
To that end, our goal was to
develop a suite of threat indices and provide a framework for NRCS and
other resource management agencies to select and prioritize watersheds
to implement agricultural conservation practices for each of the
450,000+ stream segments in the Missouri River basin (MORB). Our
objectives were to: (1) quantify the watershed percentages or densities
of 17 threat metrics that represent major sources of ecological stress
to stream communities across the basin, (2) conduct a threat assessment
to assess total watershed condition for each stream segment using five
threat indices developed from the threat metrics: agricultural, urban,
point-source pollution, infrastructure, and all non-agricultural
threats, and (3) identify stream segments where NRCS has primary
management capacity (i.e., the threats in a watershed can best be
addressed through agricultural conservation practices applied with NRCS
assistance). A conceptual example is provided to demonstrate how
decisions regarding conservation can be influenced using threat and
ecological condition assessments. In addition, although management
capacity was not identified for agencies that address non-agricultural
threats, the threat indices or scoring criteria used to delineate
management capacity herein could be formulated to represent specific
agencies as needed.
Methods
Study Area
The MORB is well suited for
developing threat indices because of its large geographic area,
considerable variation in watershed and stream conditions, and extensive
landscape modification (Galat et al. 2005; Revenga et al. 1998). The MORB drains about 1,371,017 km2 of the United States and 25,100 km2 of Canada (Fig. 2) (Galat et al. 2005).
Restoring conditions of MORB altered riverine habitats presents
significant challenges to resource managers due to, among other things,
the size of the basin and the diversity and spatial distribution of
existing threats. Dominant land use and land cover within the basin
includes 25 % cropland, 48 % grassland/pasture, 10 % forest, 11 % shrub,
3 % urban, 2 % wetland, and 1 % open water (Homer et al. 2004).
Agriculture threats (row-crop and grazing) are most prevalent across
MORB but considerable spatial heterogeneity exists among agriculture and
non-agriculture threats (e.g., point-source pollution, urbanization,
and mining activities) making the prioritization of agriculture lands to
be enrolled in conservation practices a significant challenge.
Geographic Framework
The base stream layer was acquired from the work done for the Missouri River Basin Aquatic Gap Project (Annis et al. 2009). These stream networks represent a modified version of the 1:100,000 National Hydrography Dataset (NHD; http://nhd.usgs.gov).
The primary modification of the NHD was the repair of gross
underrepresentation of stream density in portions of the basin
corresponding to select 1:100,000 scale topographic maps. The resulting
stream networks were also processed to remove loops and braids within
the network that caused problems with geoprocessing tasks of quantifying
threat prevalence throughout the MORB. We used 30-m digital elevation
models from the NHDPlus (http://nhd.usgs.gov)
and ArcHydro Tools (ArcGIS 9.3, ESRI, Redlands, CA, USA) to create
corresponding local catchment polygons (i.e., the land immediately
draining a stream segment) for each of the 464,118 individual stream
segments in the resulting MORB stream network. The resulting stream
segments and catchment polygons were used as the spatial framework for
quantifying and mapping the individual threat metrics and multimetric
threat indices for this project.
Rationale and General Approach to Threat Index Development
Seventeen threats metrics were
used to develop five multimetric indices representing major categories
of sources of ecological stress and to identify NRCS management
capacity: agricultural, urban, point-source pollution, infrastructural
(those occurring directly in stream channels), and all non-agriculture
threats (Table 1).
The threat metrics used in the indices were chosen because data were
publicly available, reasonably consistent in coverage across the MORB,
and represent the major threats to aquatic systems. The agricultural
threat index represents the major agricultural threat metrics to aquatic
systems (Table 1). Row-crop agriculture and grazing affect sedimentation regimes (Waters 1995) while channelization directly modifies channel structure, physical habitat (Frothingham et al. 2001), and hydrology (Rhoads et al. 2003). The threat metrics in the urban index (Table 1) represent hydrologic alterations from impervious surfaces (Roy et al. 2005), pollution from densely populated areas (Hatt et al. 2004; Young and Thackston 1999), and potential increases in sedimentation due to construction from increasing population density (Wolman and Schick 1967). The point-source pollution index represents pollution sources that have potential direct effects on aquatic biota (Table 1). The infrastructure index represents threat metrics occurring in a stream channel that can be readily mapped in a GIS (Table 1). Road and rail stream crossings affect physical stream habitats (Bouska et al. 2010) and dams cause alterations in physical habitat (Kondolf 1997; Ligon et al. 1995) and hydrologic regime (Poff et al. 2007).
The non-agriculture index collectively represents all threats from the
urban, point-source pollution, and infrastructure indices (Table 1).
Table 1
Threat metrics and their data sources used to calculate five threat indices within the MORB
Threat dataset (measurement unit)
|
Modified
|
Threat index
|
Data sources
|
Source date
| ||||
|---|---|---|---|---|---|---|---|---|
AG
|
UR
|
PSP
|
IN
|
NAG
| ||||
Row-crop agriculture (% of watershed)
|
No
|
X
|
U.S.G.S.—2001 NLCD
|
2006
| ||||
Canada National Land and Water Information Service
|
2007
| |||||||
Estimated grazing (% of watershed)
|
Yes
|
X
|
U.S. Department of Agriculture—2006 Agriculture Census
|
2006
| ||||
U.S.G.S.—2001 NLCD
|
2006
| |||||||
Channelized streams (km/km2)
|
Yes
|
X
|
U.S.G.S.—24 k NHD
|
Varies
| ||||
U.S.G.S. and EPA—100 k NHD
|
2006
| |||||||
U.S.G.S. Wetland Mapper Team—National Wetlands Inventory
|
2006
| |||||||
Impervious surface ( % of watershed)
|
Yes
|
X
|
X
|
U.S.G.S.—2001 NLCD
|
2006
| |||
Canada National Land and Water Information Service
|
2007
| |||||||
Population density 2000 (#/km2)
|
Yes
|
X
|
X
|
U.S. Census Bureau—2000 Block Data
|
2000
| |||
Statistics Canada
|
2007
| |||||||
Population change 1990–2000 (#/km2)
|
Yes
|
X
|
X
|
U.S. Census Bureau—1990 Block Data
|
1990
| |||
U.S. Census Bureau—2000 Block Data
|
2000
| |||||||
Statistics Canada
|
2007
| |||||||
Coal mines (#/km2)
|
Yes
|
X
|
X
|
EPA—Better Assessment Science Integrating Point and Non-point Sources
|
2001
| |||
Canada National Pollutant Release Inventory Data
|
2008
| |||||||
University of Nebraska—Lincoln
|
1996
| |||||||
Iowa Department of Natural Resources
|
2003
| |||||||
Lead mines (#/km2)
|
No
|
X
|
X
|
EPA—Better Assessment Science Integrating Point and Non-point Sources
|
2001
| |||
Other mines (#/km2)
|
Yes
|
X
|
X
|
U.S.G.S.
|
2005
| |||
Canada National Pollutant Release Inventory Data
|
2008
| |||||||
CERCLIS sites (#/km2)
|
Yes
|
X
|
X
|
EPA—Envirofacts
|
2007
| |||
Toxic release inventory sites (#/km2)
|
Yes
|
X
|
X
|
EPA—Envirofacts
|
2007
| |||
RCRA sites (#/km2)
|
Yes
|
X
|
X
|
EPA—Envirofacts
|
2007
| |||
NPDES sites (#/km2)
|
Yes
|
X
|
EPA—Envirofacts
|
2006, 2008
| ||||
Landfills (#/km2)
|
Yes
|
X
|
X
|
EPA—Better Assessment Science Integrating Point & Non-point Sources
|
2001
| |||
Missouri Department of Natural Resources
|
2006
| |||||||
Canada National Pollutant Release Inventory Data
|
2008
| |||||||
Dams (#/km2)
|
No
|
X
|
X
|
National Inventory of Dams. U.S. Army Corps of Engineers
|
1996
| |||
Canadian National Topographic Database
|
Unknown
| |||||||
Road stream crossing (#/km2)
|
Yes
|
X
|
X
|
Census Bureau—TIGER
|
1999
| |||
Statistics Canada
|
2008
| |||||||
Missouri Resource Assessment Partnership—Streams
|
2009
| |||||||
Rail stream crossing (#/km2)
|
Yes
|
X
|
X
|
Census Bureau—TIGER Database
|
1999
| |||
Statistics Canada
|
2008
| |||||||
Missouri Resource Assessment Partnership—Streams
|
2009
| |||||||
Each threat metric and index
represents the potential “risk” of environmental stress as a function of
its prevalence. Several authors (Table 2)
have used weighting schemes that relate threat metrics to expected or
empirically derived biological responses so that ecological degradation
can be inferred from threat metric or index scores. For example, Mattson
and Angermeier (2007)
used professional judgment to weight threat severity by the degree to
which each threat was perceived to affect the five components of
biological integrity (Karr et al. 1986). Esselman et al. (2010)
used multivariate correlations with threat metrics to biological
endpoints to weight threat severity and indices across the conterminous
U.S. However, we assumed risk to be equal across all threats (i.e., all
threat metrics were assumed to have the same potential risk) and
positively associated with threat prevalence because our intent was to
represent management capacity and not ecological condition. In a perfect
world, weighting threat metrics could be done objectively because a
dose/response relationship would exist for every (or most) threat
metrics as they relate to multiple biological endpoints. We acknowledge
that a vast literature exists on species/habitat relationships but there
is generally incomplete or no knowledge regarding thresholds of
biological change or degradation related to prevalence of threat metrics
such as those used in this study. There are, however, two notable
exceptions that when agricultural land use generally exceeds 50 %
watershed area (Wang et al. 1997) and urban land use exceeds 10 % (Snyder et al. 2003; Wang et al. 2001; Wang et al. 2000)
watershed area, there are often significant biological changes. Our
threat metrics or indices were not weighted because we believe that
weighting should be done objectively for all metrics and not a select
few. Of the 17 threat metrics we used, there was evidence to suggest
that only two metrics (row-crop agriculture and urban lands) could be
objectively weighted to biological degradation. Lastly, research has
demonstrated that subjectively weighted threats indices do not provide
additional information over unweighted indices (Paukert et al. 2011)
and we believe use of a subjective weighting system would decrease the
ability of our indices to identify management capacity since weighting
would be inconsistent and inaccurate across all threat metrics.
Table 2
Comparison of published threat indices, their components, and weighting procedures
Index component
|
Mattson and Angermeier (2007)
|
Danz et al. (2007)
|
Sowa et al. (2007)
|
Annis et al. (2009)
|
Esselman et al. (2010)
|
Fore et al. (this study)
|
|---|---|---|---|---|---|---|
Spatial extent for development
|
Upper Tennessee River basin (55,400 km2)
|
U.S. Great Lakes basin (765,000 km2)
|
Missouri, USA (180,533 km2)
|
EPA Region 7 (IA, KS, MO, NE; 739,769 km2)
|
Conterminous USA (8.08 × 106 km2)
|
MORB (1.39 × 106 km2)
|
Spatial grain
|
Subwatersheds of 8-digit HUC
|
0.3–17,000 km2
|
Aquatic ecological system types ~250–1,500 km2
|
Stream segment 1:100,000 NHD
|
Stream segment 1:100,000 NHD
|
Stream segment 1:100,000 NHD
|
Feature representation
|
Polygon
|
Polygon
|
Polygon
|
Polygon and line
|
Polygon and line
|
Polygon and line
|
Specificity to ecosystem
|
Relative to Upper Tennessee River basin
|
Relative to US Great Lakes basin
|
Relative to state of Missouri
|
Relative to EPA Region 7
|
Relative to conterminous USA
|
Relative to Bailey’s (1983) divisions
|
Assessment units fixed or continuous
|
Fixed
|
Fixed
|
Fixed
|
Continuous—upstream and local watersheds
|
Continuous—upstream and local watersheds
|
Continuous—upstream and local watersheds
|
No. of threats
|
12
|
86
|
11
|
36
|
17
|
17
|
Spatial scale of threat quantification
|
Total contributing area
|
Total contributing area
|
Total contributing area
|
Total and local contributing area
|
Total and local contributing area
|
Total and local contributing area
|
“Dose” quantification
|
Dose
represented for individual threats as product of rank-based frequency
and severity scores; ranks based on thresholds in literature and
quartile scores
Cumulative threats represent as sum of individual threats
|
Multivariate representation of threat categories via principal component axes
Cumulative threats represented by summing axis scores for each threat category
|
Dose for individual threats represented as discrete rank scores; based on thresholds in literature and quartile scores
Cumulative threats represented as numeric combination of individual threats
|
Dose
is standardized and non-discrete for individual and cumulative threats;
also standardized to stream size and across measurement units
|
Multivariate representation of threats via principal component axes; standardized to stream size
|
Dose
is standardized and non-discrete for individual and cumulative threats;
also standardized by ecoregion and across measurement units
|
Threat density weighting
|
Used literature values to identify thresholds and quartile scores
|
Weighting not used
|
Used literature values to identify thresholds and quartile scores
|
Used non-discrete ranking procedure
|
Weighting not used
|
Used non-discrete ranking procedure
|
Threat severity weighting
|
Used expert judgment, as perceived to affect to biological integrity
|
Weighting not used
|
Weighting not used
|
Used distance weighting for some threats; weights subjectively assigned
|
Used biological weighting via multivariate correlation to threats
|
Weighting not used
|
Empirical construction or validation
|
Not empirically constructed. Validated by Paukert et al. (2011); found density and severity weighting produced similar results to no weighting.
|
Indices related to fish and bird ecological condition metrics
|
Not empirically constructed or validated
|
Showed positive relationship with macroinvertebrate index of biological integrity—an overall decline in ecological condition
|
Empirically constructed using biological samples to weight threat severity
|
Not empirically constructed or validated
|
Modified Threat Metric Data
Four threat metrics required
creation or modification from their original form. Grazing and stream
channelization threats were not appropriately represented in existing
data sources and were modified. Impervious surfaces were overestimated
in the NLCD and population change information needed to be quantified.
Refer to online supplemental information for details of data
modification.
Quantifying Threat Prevalence
The number of threats evaluated
in an index affects its comprehensiveness and ability to identify all
potential sources of stress. Threats were quantified to assess their
prevalence and were recorded as unit density, usually as proportion of
watershed (e.g., proportion of row-crop) or number of units per
watershed area (e.g., number of discharges per square kilometer), for an
assessment region. The most precise prevalence estimates are those
represented by the actual threat density or proportion of watershed
value (e.g., 25 % of watershed area) and least precise are estimates
that categorize prevalence (e.g., 0–25 % = 1 etc.). The spatial grain at
which threat prevalence is quantified affects the ability of threat
indices to inform decisions regarding placement of conservation
practices. As spatial grain increases (e.g., from local contributing
areas to 8-digit Hydrologic Unit Codes), the representation of threat
prevalence becomes more generalized and the ability to identify
fine-scale spatial patterns is reduced or eliminated.
Individual threat metric
prevalence was quantified within each of the local catchment polygons so
that the threat assessment can be summarized regionally while retaining
the resolution to inform localized planning. Then we used customized
Arc Macro Language (ESRI, Redlands, CA, USA) programs to sum all these
values for each individual stream segment’s entire watershed (i.e., the
local catchment and all upstream catchments that a segment drains). We
then divided these summed values by the overall watershed area to
quantify the prevalence, per unit area or as proportion of watershed, of
each threat metric within the watershed of each segment.
Threat Metric and Index Calculations
We normalized and calculated scores for each threat metric
so that comparisons could be made among threats recorded in different
measurement units (e.g., to compare proportion of watershed vs. point
densities) . Each threat index was normalized to a common scale so that direct comparisons could be made among indices and ecoregions.
Threat metrics were removed from our dataset, to reduce redundancy in our representation of threats (Stoddard et al. 2008),
if they were significantly correlated with a threat metric that could
appropriately represent the removed metric (e.g., road density can be
represented as an impervious surface, but not by cropland). Metrics were
removed if they were correlated with at least one variable and their
Pearson correlation coefficient was >0.55 and P < 0.05 (corrected for multiple comparisons using Bonferroni adjustments).
Threat metric scores (Ts ) were calculated as
Tsi,j,k=[Tri,j,kmax(Tri,j,k)]×100,
where Tr
i,j,k
is the ranked value of threat prevalence (as total contributing area) for every i
stream segment (i.e., the stream segment with the lowest threat
prevalence received a rank of one and the stream segment with the
highest density received the highest rank) in the j ecoregion for the k
threat metric. Ties in threat prevalence were given the same rank.
Since nearly all the metrics were point densities (e.g., #/km2)
they were often very skewed, especially in large watersheds. Ranking
improved their normalization and each segment’s ranked threat prevalence
score (Tr, numerator in the above equation) was divided by maximum
ranked value for its corresponding ecoregion [max(Tr
i,j
), denominator in the above equation] and
multiplied by 100 (range 0–100). Ranking was used because metrics
represented as unit densities (e.g., #/unit area) were generally skewed
and this method of transformation improved normalization. Ranking had
little effect on the distribution of metrics recorded as proportion of
watershed. Ts values of 100 represent the highest threat prevalence.
Threat index scores (TI) were
calculated and normalized by summing the corresponding threat metric
scores (Ts) for each index (Table 1)
TIi,j=[∑nk=1Tsi,j,kmax (∑nk=1Tsi,j,k)]×100,
where Ts
i,j.k
is the threat metric score for every i stream segment in j ecoregion for threat metrics scores k through n.
The Ts scores were summed for each index (numerator in above equation)
and divided by the maximum summed value of Ts within each ecoregion.
Index scores of 100 represent stream segments with highest potential
stress within each ecoregion. Final threat index scores were then
incorporated into a seamless stream layer database and mapped in ArcGIS
(ESRI, Redlands, CA, USA).
Agriculture Conservation Program Primary Management Capacity Matrix
A matrix was developed to
determine the degree of NRCS management capacity for stream segments
based on watershed condition. Since conservation programs target a
limited suite of threats, implementing agricultural conservation
practices in watersheds where NRCS has primary management capacity
should increase conservation practice effectiveness. We did not
distinguish between private and public lands, and it should be noted
that public lands are generally not within NRCS management capacity;
management capacity of public lands is under the agency responsible for
managing those lands. The matrix was used to determine the relative
degree of NRCS management capacity by assessing the potential stress
from agricultural threats, relative to potential non-agricultural stress
for each segment. For each stream segment, its agriculture and
non-agriculture threat index scores were given a quartile score (i.e.,
index score 0–25 = 1, 25–50 = 2, etc.; Table 3).
(Different scoring criteria would be acceptable to formulate this
matrix if quartile scores were deemed too coarse a resolution; this is
intended to be an illustrative example.) The upper half of the matrix
was populated by dividing the agriculture (y) and non-agriculture (z) quartile scores for each X
yz
. Matrix scores were then transposed to the corresponding X
yz
on the lower half of the matrix and
given negative values. Positive scores indicate stream segments where
NRCS is most likely to have primary management capacity and the more
positive the score, the more likely NRCS is to have greater management
capacity (Table 3).
Matrix scores ≥2 were considered to represent primary NRCS management
capacity. (The threshold of ≥2 is presented here as an illustration and
could be altered to suit an agency’s needs.) We used a scoring matrix
because the interpretation of agricultural/non-agriculture ratios was
unintuitive when non-agricultural threats were greater than agricultural
threats. For example, a 75:25 agriculture/non-agriculture ratio equals
3, but a 25:75 ratio equals 0.33 where scoring matrix values would be 3
and −3, respectively.
Table 3
Scoring matrix
used to identify stream segments where NRCS had primary management
capacity (see text for explanation) in the MORB
Non-agriculture quartile scores
|
Agriculture threat quartile scores
| |||
|---|---|---|---|---|
1 (0–25)
|
2 (25–50)
|
3 (50–75)
|
4 (75–100)
| |
1 (0–25)
|
1
|
2
|
3
|
4
|
2 (25–50)
|
−2
|
1
|
1.5
|
2
|
3 (50–75)
|
−3
|
−1.5
|
1
|
1.3
|
4 (75–100)
|
−4
|
−2
|
−1.3
|
1
|
Illustrative Example of Coupling Threat and Ecological Condition Assessments
We obtained fish IBI scores from sites in MORB (Fig. 3) to identify ecologically degraded streams and to illustrate how the conceptual process outlined in Fig. 1
could be utilized by resource managers. This comparison was not
intended to validate the threat indices nor was it an attempt to
establish empirical relationships with threat index scores and IBI
scores. This illustration is meant to demonstrate how information from
threat indices can be used by managers to infer likely causes of
ecological degradation when adequate ecological condition assessments
have been performed.
We haphazardly selected four
streams that spanned the overall range of IBI scores. The IBI scores
were computed for the US Environmental Protection Agency’s (EPA)
Regional Environmental Monitoring and Assessment Program in EPA Region 7
(M. Combes, unpublished data). The IBI was applicable to streams across
the entire EPA Region 7 and contained the metrics (all metrics were
evaluated as “number of”): native species, native families, native
individuals, sensitive individuals, tolerant individuals, benthic
species, native sunfish species, minnow species, long-lived species,
introduced species, trophic strategies, native carnivore species, native
omnivore and herbivore species, and reproductive strategies. The
individual metrics for the IBI were evaluated to identify potential
stressors causing ecological degradation. Threat index scores for each
site were then compared relative to the overall IBI score and the
individual IBI metrics to illustrate how information from threat indices
and ecological condition assessments can inform conservation decisions.
Results
Percent impervious surface was significantly and highly correlated with developed open (r = 0.59), low (r = 0.96), medium (r = 0.91), and high (r = 0.66) urban land-use variables from the NLCD as well as road density (r = 0.85);
therefore, these metrics were not included in the threat indices. As a
result, percent impervious surface in a contributing area was used to
represent correlated urban land use and road density threats.
Patterns of potential stress were
evident at regional scales, but we identified localized patterns of
potential stress that showed considerable spatial heterogeneity.
Agriculture threats are most prominent across the MORB and on average
stream segments have higher potential agriculture stress (Table 4). Mean scores for all threat indices, thus potential stress, significantly varied among the five ecoregions of the MORB (Table 4). Visual examination of mapped output for the agriculture threat index (Fig. 4)
illustrates that regional patterns in threat stress exist, e.g., high
agriculture threats in the east-central portion of the basin. Although
potential agricultural stress appears consistent within ecoregions,
there was considerable spatial heterogeneity in potential agriculture
stress at localized scales (i.e., within regions of high agricultural
stress, there are many watersheds with low to moderate agricultural
stress; see inset in Fig. 4). Similar patterns exist when examining non-agriculture threat stress across the basin (Fig. 5).
Non-agriculture threats primarily occur in or near urban areas, but
were sometimes common and high outside of urban areas (e.g., within
areas of high agricultural stress). These patterns were not evident on
landcover maps because they do not account for watershed condition (see
inset in Fig. 5).
Table 4
Mean and standard error of threat index values for the MORB calculated within Bailey’s (1983) divisions
Threat index
|
MORB
|
Division
| ||||
|---|---|---|---|---|---|---|
Hot Continental
|
Prairie
|
Temperate Desert
|
Temperate Steppe
|
Temperate Steppe Regime Mountains
| ||
Agriculture
|
43.24 (0.03)
|
32.30a (0.15)
|
48.81b (0.05)
|
33.29c (0.14)
|
46.76d (0.03)
|
26.75e (0.08)
|
Urbanization
|
28.57 (0.02)
|
35.47f (0.15)
|
34.07g (0.06)
|
27.04h (0.11)
|
26.57h (0.03)
|
25.65i (0.07)
|
Infrastructure
|
16.55 (0.02)
|
18.90k (0.11)
|
22.27l (0.05)
|
18.56k (0.12)
|
14.73m (0.03)
|
13.10n (0.06)
|
Point-source pollution
|
17.26 (0.01)
|
23.52p (0.08)
|
17.52q (0.02)
|
26.58r (0.05)
|
16.00s (0.01)
|
17.23t (0.02)
|
Non-agriculture
|
22.94 (0.01)
|
28.07u (0.09)
|
25.83v (0.03)
|
31.16w (0.08)
|
20.94x (0.01)
|
22.20y (0.04)
|
Fig. 4
Map of the
agriculture threat index scores (target threats) for every stream
segment within the US portion of the MORB. Threat index scores were
calculated using threat prevalence information quantified for every
stream segment’s upstream watershed area. Threat index scores were
calculated separately for each division classification (see Fig. 2). Maximum threat scores are relative to the most threatened stream segment in each division (Color figure online)
Fig. 5
Map of the
non-agriculture threat index scores (non-target threats) for every
stream segment within the US portion of the MORB. Threat index scores
were calculated using threat prevalence information quantified for every
stream segment’s upstream watershed area. Threat index scores were
calculated separately for each division classification (see Fig. 2). Maximum threat scores are relative to the most threatened stream segment in each division (Color figure online)
Matrix scores ≥2 were considered
representative of segments where NRCS had primary management capacity.
Based on this criterion, NRCS had primary management capacity for 60 %
of stream segments in MORB. NRCS had primary management capacity in 55 %
of Prairie Division, 76 % of Temperate Steppe Division, 24 % of Hot
Continental Division, 14 % of Temperate Desert Division, and 29 % of
Temperate Steppe Regime Mountain stream segments. Regional patterns in
NRCS primary management capacity were evident across MORB and generally
followed the patterns of potential agriculture stress (Fig. 6).
Fig. 6
Map of NRCS primary
management capacity for every stream segment within the US portion of
the MORB. Streams with management capacity scores ≥2 (see text and
Table 2) were considered to be under NRCS management capacity (Color figure online)
The four sites evaluated for fish
biotic integrity ranged from poor (site 1 = 14) to excellent (site
4 = 96) ecological condition and two sites were intermediate (Table 5). Sites 1, 2, and 3 had low benthic metric scores (Table 6) (fishes that feed and reproduce in the benthos and are sensitive to sedimentation; Barbour et al. 1999). Sites 1 and 3 had the highest agricultural threat index scores (Table 5).
Site 2 had a relatively high urban threat index score and moderate
point-source pollution and infrastructure threat index scores (Table 5). All threat indices in site 4 had low scores (Table 5).
Table 5
Fish IBI and threat index scores from four stream sites in MORB
Site
|
IBI score
|
Index
| ||||
|---|---|---|---|---|---|---|
Agriculture
|
Point-source pollution
|
Urbanization
|
Infrastructure
|
Non-agriculture
| ||
1
|
14
|
78.76
|
24.66
|
42.97
|
27.47
|
33.96
|
2
|
37
|
14.04
|
46.72
|
93.19
|
55.14
|
67.23
|
3
|
63
|
41.15
|
15.22
|
19.09
|
7.75
|
16.46
|
4
|
96
|
6.33
|
18.44
|
13.10
|
25.93
|
20.90
|
Table 6
Fish IBI and metric scores for four streams in the MORB
Site
|
IBI score
|
Number of IBI metric scores
| |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
NAT
|
NAF
|
IND
|
SENS
|
TOL
|
BNTH
|
SUN
|
MIN
|
LOL
|
INT
|
TRO
|
NAC
|
NOH
|
REP
| ||
1
|
14
|
0.94
|
2.41
|
1.52
|
0
|
0
|
0
|
0
|
1.9
|
0
|
10
|
0
|
0
|
0
|
2.91
|
2
|
37
|
2.05
|
4.65
|
3.75
|
0
|
0.32
|
0
|
4.55
|
1.19
|
3.47
|
10
|
2.07
|
10
|
10
|
0
|
3
|
63
|
7.06
|
5.75
|
6.73
|
0
|
4.26
|
1.51
|
8.32
|
8.74
|
7.85
|
10
|
7.65
|
10
|
4.46
|
6
|
4
|
96
|
10
|
10
|
10
|
10
|
7.53
|
10
|
10
|
10
|
10
|
10
|
10
|
10
|
10
|
10
|
Discussion
Threat indices were used to
identify regional and local patterns of multiple agriculture and
non-agriculture threats for every stream segment in MORB. The threat
patterns were similar to those of landcover maps; however, unlike
landcover maps our threat indices represent watershed condition for
multiple threats. Within highly impacted agriculture regions, there was
considerable variation in watershed condition among stream segments.
This highlights the importance of cautiously using landcover map
information to make resource conservation decisions.
Coupling ecological condition and
threat assessments allows resource managers to identify ecologically
degraded sites, the threats most likely causing degradation, and helps
provide information needed to select appropriate conservation practices.
The spatial resolution (stream segments) of our threat indices allows
them to be coupled with field-based ecological data and potential
sources of stress can be evaluated for essentially any biological stream
sample. However, users should recognize that the indices cannot inform
farm-scale planning efforts because our maps and resulting threat index
scores represent total watershed condition.
The IBI scores and their
underlying metrics provide an informative example of how threat index
information can be used by managers to infer likely stresses and their
sources in ecologically degraded streams. This example serves as an
illustration of how managers can use threat indices and biological
information to quickly and efficiently reduce uncertainty regarding the
threats most likely causing ecological degradation. Our focus was to
develop threat indices that management agencies can use as a coarse
filter to guide their management actions based on watershed condition
and not ecological degradation. Furthermore, the IBI example does not
need to be comprehensive (i.e., include a larger sample of sites)
because we were not attempting to validate the usefulness of our indices
at identifying ecological degradation. Using Fig. 1
and our IBI data, sites 1, 2, and 3 had overall IBI scores that
indicated ecological degradation and the benthic IBI metric was low in
each stream (relative to the least disturbed site 4). This suggested
that sedimentation was an ecological stress (Barbour et al. 1999).
The high agricultural threat index scores from sites 1 and 3 suggested
that potential sedimentation stresses most likely originated from
agriculture threats and possibly from urbanization threats in site 1.
This indicates that agricultural conservation practices administered by
agencies like NRCS would be the most appropriate for stream restoration.
Threat index scores for site 2 suggested that sedimentation stress
originated from urban threats and that point-source pollution and
infrastructure threats may contribute additional stresses (Table 5).
In this instance, conservation practices or policies administered by
state or federal water quality authorities (e.g., USEPA and local
municipalities) would be most appropriate. Finally, site 4 had the
highest ecological condition and correspondingly low threat index
scores, suggesting a need for proactive (i.e., preventing further
degradation) rather than restorative conservation practices.
For threat indices to be widely
used by multiple resource agencies, indices need to be applicable to
multiple ecological indicators. Most published threat indices (Table 2)
are limited in their use outside of the taxa or ecological indicators
they were developed for because the indices account for threat severity
by weighting the relative influence of threat metrics to an ecological
indicator (e.g., Esselman et al. 2010; Annis et al. 2009). Unfortunately, different taxa have been shown to respond differentially to the same source of stress (Berkman et al. 1986),
therefore severity weights for threat metrics are likely applicable
only to the ecological indicator being evaluated. We were unable to
objectively weight our threat indices to ecological indicators because
dose/response relationships are not quantified between many of the
threat metrics we used (e.g., number of mines) and potential ecological
indicators. Therefore, we decided that if threat metric weighting could
not be done objectively for all threats that it was inappropriate to
weight a limited suite of threat metrics (e.g., agriculture landcover
and impervious surfaces) and misrepresent the true thresholds of
severity. In a comparative analysis, Paukert et al. (2011)
found that weighting threat indices produced nearly identical scores
relative to scores from an unweighted index. This suggests that
weighting threat indices is unlikely to increase biological realism,
especially when weighting schemes involve subjectivity. Instead of
accounting for severity, we argue that resource managers would be better
off by establishing empirical relationships between threat indices and
an ecological indicator to account for threat severity. Our contention
is that because relationships between a threat index and ecological
indicator are likely to vary by region and taxon (Frimpong and
Angermeier 2010)
that threat index scores should not be altered depending on the
ecological indicator being evaluated. Instead, managers can alter their
interpretation of threat index scores by establishing empirical
relationships between threat indices and ecological indicators (e.g., a
threat index score of 50 in one region and a score of 65 in another
region may represent the equivalent degree of degradation). Doing so
allows threat indices to be easily computed, avoids making assumptions
about threat impacts to ecological indicators across different regions,
and increases an index’s applicability to resource managers. Improving
the consistency of how relationships between threat metrics and
ecological indicators are defined, better mapping and reporting of
threat metric data are significant challenges to future development of
threat indices that appropriately account for severity.
In most watersheds, multiple threats affect ecological condition (Diana et al. 2006; Zorn and Wiley 2006),
and it is likely that addressing conservation concerns for an area will
involve multiple agencies who have distinct management authorities. We
estimated that agriculture conservation programs such as those
administered by NRCS would have primary management capacity for a
majority (60 %) of the stream segments in MORB. However, some lands
within the MORB are under public ownership and management (e.g.,
national forests and grazing lands therein are managed by the US Forest
Service), and may contain agricultural threats. Public lands are areas
where NRCS would not have primary management capacity, as threats on
those lands would be addressed by the managing agency. Given the
existing regional patterns, there is considerable heterogeneity in NRCS
primary management capacity across the landscape as non-agriculture
threats are often prevalent enough outside of urban areas. Three of the
five MORB ecoregions had NRCS primary management capacity in only
14–29 % of stream segments, indicating that relative to agriculture,
non-agriculture threats are the predominate threat. Effective
conservation in those ecoregions may require collaboration among
multiple resource agencies. Although management capacity was not
identified for non-agriculture conservation programs, the threat indices
or scoring matrix could be reformulated to meet desired needs. For
example, our point-source pollution index could be viewed as best
addressed by the U.S. Environmental Protection Agency because they
permit and regulate point-sources of pollution (US Environmental
Protection Agency 2001).
Developing comprehensive management programs (among multiple management
agencies) requires agencies to examine the relative contribution of
stress from the threats within their management capacity to those of
another agency so collaboration can be successful.
Conducting stream conservation
efforts at large geographic areas involves identifying and prioritizing
conservation areas (Groves et al. 2002).
Threat indices can be used as tools to explicitly identify potential
sources of stress on the landscape, identify agencies that have
management capacity over assessment units, and establish relationships
with ecological indicators to determine potentially degraded systems.
The threat indices and primary management capacity scoring systems we
developed could be used in a winnowing process to identify and select
subsets of stream segments for stream conservation. The regional
patterns we observed in potential stresses could be used as a
“first-cut” in a winnowing process to select broad geographic areas to
focus conservation program implementation. In the MORB example, the
Prairie and Temperate Steppe Divisions tend to be the most
agriculturally threatened and could be the focus of NRCS conservation
programs. Next, threat indices can be used to identify individual stream
segments within regions where NRCS has primary management capacity, and
presumably a greater chance of achieving conservation success. Where
ecological condition is known, resource managers should use threat
indices to establish relationships among ecological indicators because
ecological indicators respond differently to threats (Danz et al. 2007; Frimpong and Angermeier 2010).
Once relationships among indicators are successfully established,
resource managers can then interpret their threat index scores in an
ecologically meaningful manner and select the appropriate conservation
programs to implement.
Acknowledgments
We
thank the United States Department of Agriculture’s Natural Resource
Conservation Service under the Conservation Effects Assessment Project
for funding this project. Michael Morey provided technical assistance
and Aaron Garringer conducted much of the GIS work. Matt Combes of
Missouri Department of Conservation provided fish IBI data. This
manuscript was improved by helpful reviews from Charles Rabeni and two
anonymous reviewers.
Electronic supplementary material
Below is the link to the electronic supplementary material.
References
Annis
GM, Sowa SP, Diamond DD, Combes MD, Doisy KE, Garringer AJ, Hanberry P
(2009) Developing synoptic human threat indices for assessing the
ecological integrity of freshwater ecosystems in EPA region 7. Final
report, Environmental Protection Agency (submitted)
Bailey RG (1983) Delineation of ecosystem regions. Environ Manag 7:365–373CrossRef
Barbour
MT, Gerritsen J, Snyder BD, Stribling JB (1999) Rapid bioassessment
protocols for use in streams and wadeable rivers: periphyton, benthic
macroinvertebrates and fish, 2nd edn. EPA 841-B-99-002, U.S.
Environmental Protection Agency, Office of Water, Washington, D.C.
Berkman
HE, Rabeni CF, Boyle TP (1986) Biomonitors of stream quality in
agricultural areas: fish versus invertebrates. Environ Manag 10:413–419CrossRef
Bouska
WW, Keane T, Paukert CP (2010) The effects of road crossings on Prairie
Stream habitat and function. J Freshw Ecol 25:499–506CrossRef
Danz
N, Niemi G, Regal R, Hollenhorst T, Johnson L, Hanowski J, Axler R,
Ciborowski J, Hrabik T, Brady V, Kelly J, Morrice J, Brazner J, Howe R,
Johnston C, Host G (2007) Integrated measures of anthropogenic stress in
the U.S. Great Lakes Basin. Environ Manag 39:631–647CrossRef
Diana
M, Allan JD, Infante D (2006) The influence of physical habitat and
land use on stream fish assemblages in southeastern Michigan. In: Hughes
RM, Wang L, Seelbach PW (eds) Landscape influences on stream habitats
and biological assemblages. American Fisheries Society, Symposium 48,
Bethesda, MD, 359–374
Esselman
PC, Infante DM, Wang L, Wu D, Cooper A, Taylor WW (2010) An initial
assessment of relative landscape disturbance levels for river fish
habitats of the conterminous United States. National Fish Habitat Action
Plan
Frimpong
EA, Angermeier PL (2010) Comparative utility of selected frameworks for
regionalizing fish-based bioassessments across the United States. Trans
Am Fish Soc 139:1872–1895CrossRef
Frothingham
KM, Rhoads BL, Herricks EE (2001) Stream geomorphology and fish
community structure in channelized and meandering reaches of an
agricultural stream. In: Dorava JM, Montgomery DR, Palcsak BB,
Fitzpatrick FA (eds) Geomorphic processes and riverine habitat. American
Geophysical Union, Washington D.C., pp 105–117CrossRef
Galat
DL, Berry CR, Peters EJ, White RG (2005) Missouri River Basin. In:
Benke AC, Cushing CE (eds) Rivers of North America. Elsevier, Oxford, pp
427–480
Groves
CR, Jensen DB, Valutis LL, Redford KH, Shaffer ML, Scott JM,
Baumgartner JV, Higgins JV, Beck MW, Anderson MG (2002) Planning for
biodiversity conservation: putting conservation science into practice.
BioScience 52:499–512CrossRef
Hatt
BE, Fletcher TD, Walsh CJ, Taylor SL (2004) The influence of urban
density and drainage infrastructure on the concentrations and loads of
pollutants in small streams. Environ Manag 34:112–124CrossRef
Homer
C, Huang C, Yang L, Wylie B, Coan M (2004) Development of a 2001
national landcover database for the United States. Photogram Eng Remote
Sens 70:829–840CrossRef
Karr JR (1981) Assessment of biotic integrity using fish communities. Fisheries 6:21–27CrossRef
Karr
JR, Fausch KD, Angermeier PL, Yant PR, Schlosser IJ (1986) Assessing
biological integrity in running waters: a method and its rationale.
Special Publication 5, Illinois Natural History Survey, Urbana, IL
Kondolf GM (1997) Hungry water: effects of dams and gravel mining on river channels. Environ Manag 21:533–551CrossRef
Leonard
PM, Orth DJ (1986) Application and testing of an index of biotic
integrity in small, coolwater streams. Trans Am Fish Soc 115:401–414CrossRef
Ligon FK, Dietrich WE, Trush WJ (1995) Downstream ecological effects of dams. BioScience 45:183–192CrossRef
Mattson
KM, Angermeier PL (2007) Integrating human impacts and ecological
integrity into a risk-based protocol for conservation planning. Environ
Manag 39(1):125–138CrossRef
Palmer
MA, Bernhardt ES, Allan JD, Lake PS, Alexander G, Brooks S, Carr J,
Clayton S, Dahm CN, Shah JF, Galat DL, Loss SG, Goodwin P, Hart DD,
Hassett B, Jenkinson R, Kondolf GM, Lave R, Meyer JL, O’Donnell TK,
Pagano L, Sudduth E (2005) Standards for ecologically successful river
restoration. J Appl Ecol 42:208–217CrossRef
Paukert
CP, Pitts KL, Whittier JB, Olden JD (2011) Development and assessment
of a landscape-scale ecological threat index for the Lower Colorado
River Basin. Ecol Indic 11:304–310CrossRef
Poff
NL (1997) Landscape filters and species traits: towards mechanistic
understanding and prediction in stream ecology. J N Am Benthol Soc
16:391CrossRef
Poff
NL, Olden JD, Merritt DM, Pepin DM (2007) Homogenization of regional
river dynamics by dams and global biodiversity implications. Proc Nat
Acad Sci 104:5732–5737CrossRef
Revenga
C, Murray S, Abramovitz J, Hammond A (1998) Watersheds of the world:
ecological value and vulnerability. Worldwatch Institute, Washington,
D.C.
Rhoads
BL, Schwartz JS, Porter S (2003) Stream geomorphology, bank vegetation,
and three-dimensional habitat hydraulics for fish in midwestern
agricultural streams. Water Resour Res 39:21–23
Roy
AH, Freeman MC, Freeman BJ, Wenger SJ, Ensign WE, Meyer JL (2005)
Investigating hydrologic alteration as a mechanism of fish assemblage
shifts in urbanizing streams. J N Am Benthol Soc 24:656–678CrossRef
Snyder
CD, Young JA, Villella R, Lemarié DP (2003) Influences of upland and
riparian land use patterns on stream biotic integrity. Landscape Ecol
18(7):647–664CrossRef
Sowa
SP, Annis G, Morey ME, Diamond DD (2007) A GAP analysis and
comprehensive conservation strategy for riverine ecosystems of Missouri.
Ecol Monogr 77:301–334CrossRef
Stoddard
JL, Herlihy AT, Peck DV, Hughes RM, Whittier TR, Tarquinio E (2008) A
process for creating multimetric indices for large-scale aquatic
surveys. J N Am Benthol Soc 27:878–891CrossRef
Sutherland
AB, Meyer JL, Gardiner EP (2002) Effects of land cover on sediment
regime and fish assemblage structure in four southern Appalachian
streams. Freshw Biol 47:1791–1805CrossRef
US
Environmental Protection Agency OoW (2001) Protecting the Nation’s
waters through effective NPDES permits: a strategic plan FY 2001 and
beyond. EPA-833-R-01-001, United States Environmental Protection Agency,
Office of Water, Washington, D.C.
Wang
L, Lyons J, Kanehl P, Gatti R (1997) Influences of watershed land use
on habitat quality and biotic integrity in Wisconsin streams. Fisheries
22:6–12CrossRef
Wang
L, Lyons J, Kanehl P, Bannerman R, Emmons E (2000) Watershed
urbanization and changes in fish communities in southeastern Wisconsin
streams. J Am Water Resour Assoc 36(5):1173–1189CrossRef
Wang
L, Lyons J, Kanehl P, Bannerman R (2001) Impacts of urbanization on
stream habitat and fish across multiple spatial scales. Environ Manag
28(2):255–266CrossRef
Waters
TF (1995) Sediment in streams: sources, biological effects, and
control. American Fisheries Society Monograph 7, Bethesda, MD
Wolman
MG, Schick AP (1967) Effects of construction on fluvial sediment, urban
and suburban areas of Maryland. Water Resour Res 3:451–464CrossRef
Young KD, Thackston EL (1999) Housing density and bacterial loading in urban streams. J Environ Eng 125:1177–1180CrossRef
Zorn
TG, Wiley MJ (2006) Influence of landscape characteristics on local
habitat and fish biomass in streams of Michigan’s lower peninsula. In:
Hughes RM, Wang L, Seelbach PW (eds) Landscape influences on stream
habitats and biological assemblages. American Fisheries Society,
Symposium 48, Bethesda, MD, 375–393
Environment Systems and DecisionsFormerly The Environmentalist
© Springer Science+Business Media New York 2013
10.1007/s10669-013-9438-5Cost-benefit and systems analysis of passively ventilated solar greenhouses for food production in arid and semi-arid regions
(1)
Centre for Energy Studies, Indian Institute of Technology, Delhi, 110016, Hauz Khas, India
(2)
Department
of Renewable Energy Engineering, Maharana Pratap University of
Agriculture and Technology, Udaipur, 313001, Rajasthan, India
Published online: 2 March 2013
Abstract
In this paper, economic
feasibility of two vegetable crops (i.e., cucumber and tomato)
cultivated in a naturally ventilated greenhouse, and the net present
worth, cost-benefit ratio, payback period, and internal rate of return
for these crops on year-round cultivation are presented. The
cost-benefit ratio demonstrated that growing cucumbers and tomatoes can
be economically viable in this climatic region. The present experimental
study was conducted in the composite climatic condition of Udaipur
(24°35′N, 73°42′E), India. The study area is defined as arid and
semi-arid region of Indian climatic conditions. Droughts are a recurring
phenomenon in arid and semi-arid regions creating a situation that
affects not only agricultural productivity but also people’s health. In
particular, the western part of the state is a desert, and its
socioeconomic status influences nutrient purchasing power. A poor diet
can lead to a vitamin and mineral deficiency. The state of Rajasthan has
good agricultural potential; interventions using protected cultivation
practices can increase the production and productivity of vegetable
crops. However, the current adoption rate of such practices in the state
remains very slow, even after a promotional scheme offered by the state
government. The government and policymakers should consider offering
demonstrations of practices at a larger level. Farmers of the state are
marginal and economically poor, requiring more financial assistance. Low
cost technologies would be suitable for these farmers.
Keywords
Off-season cultivation
Greenhouse
Economic feasibility
Crop cultivation
1 Introduction
Open-field farming is not a new
approach. It began with the appearance of civilization. This farming
approach allows use of both good and poorer fertile soil, and it
encourages cooperation among the farmers. On the other hand, it is
difficult to handle pest, and animals and weeds that spread in such
open-field conditions. Ultimately, the approach reduces net
productivity. The open farming succeeds newer and modern agricultural
technology by successfully meeting a growing demand for food by the
world’s population. Drastic increases in the yield of primary crops such
as rice and wheat in Indian climatic conditions have been reported.
This increase in food production is mainly due to advances in scientific
input, including the development of new crop varieties, the use of
pesticides and fertilizers, and the construction of large irrigation
systems. In the present context, consumers are very much aware of their
nutritive diets and prefer organic farming produces. Organic farming is
simply with the process of growing of fruits and vegetables without
using any type of pesticides and chemical fertilizer.
In the present context, high
pressure on current agricultural production systems at global level and
facing huge challenges and changes with ongoing farming practices,
declining soil fertility and crop yields, poor market access,
constrained to assess land, and high inflation are constraints in the
industry (Nelson et al. 2010; Yamano et al. 2011).
With these, both poverty level and household food insecurity are rising
across developing and underdeveloped countries (Charles et al. 2010; Kristjanson et al. 2010).
Some of the raw vegetables are used as salad in many countries. Less
than 200 g of vegetables per person per day is common, and this low
amount, often in conjunction with poverty and poor medical services, is
associated with unacceptable levels of mortality and malnutrition in
preschool children and other vulnerable groups. An increase in the
availability, affordability, and consumption of nutrient-dense
vegetables and fruits may be a way to reduce malnutrition substantially
(Keatinge et al. 2011).
The inhabitant of many developing
countries’ is considered to be malnourished due to low agricultural
production. There is universal agreement that in order to increase
productivity, the best course is to reduce postharvest losses and create
better market linkage by boosting small and marginal farmers.
Off-season cultivation in partially controlled or naturally ventilated
greenhouses offers one to agricultural production possibility and
enables farmers to cultivate vegetables and high nutrition crops
year-round. Protected cultivation minimizes seasonal fluctuations in
crop yields.
Modern and easily adaptable good
protected cultivation practices minimize contamination from pesticide
residue, bacteria, viruses, or helminths which are hostile to humans. In
the case of open-field farming pests, control is very difficult and
over-spraying may poison the farmer and his family unless appropriate
safety measures are taken. India accounts for one-third of all the
world’s pesticide poisoning cases (Indira Devi 2007).
There is no such problem with greenhouse crop production, and
nutritional crops can easily grow without any dangerous chemicals.
Offering highly nutrient-dense vegetables to school children is one way
to overcome malnutrition (Yang et al. 2007).
The application of plastic in the agricultural sector is growing and
has helped farmers to increase crop production, improve food quality,
and reduce the ecological footprint of their activity. Not only do
plastics allow for vegetables and fruits to be grown in every season,
but these products are usually of better quality than those grown in an
open field (Martin 2012).
Singh et al. (2006)
conducted an exhaustive study on the nutritional status of rural
population in desert area of Rajasthan state, and it was found that diet
of people resides in the state was grossly deficient in green leafy
vegetables, fats, pulses, and legumes and other vegetables. The frequent
occurrence of drought is accountable for it and affects the
agricultural productivity. This might be responsible for higher under
nutrition not only in children but in adults also.
Agriculture is not a profession
that is voluntarily chosen by many due to the drudgery one encounters
while working in the field. Consequently, migration from rural to urban
areas has become a common feature in our society. It is true that in the
foreseeable future, agriculture will continue to be a major occupation
among people. This implies that agriculture should become more
profitable, and the associated drudgery should be minimized (Kothari and
Panwar 2004; Panwar et al. 2009; Black et al. 2011).
The world population continues to rise. While the demand for growing
more food persists, nutritional security must also be ensured for a
healthy nation. Land and other resources being inelastic, farmers are
required to produce more nutritious food from the available resources
(Mukhopadhyay et al. 2011).
The need to ensure that people have nutritious diets to the masses
suggests that the production of vegetables and fruits should be
adequately increased. Considering the currently low productivity levels
of different crops, the targets of vegetable and fruit production appear
to be difficultly until better technologies are utilized for food
security (Kothari et al. 2006). Tiwari and Joshi (2012) explain the meaning of food security according to the Food and Agricultural Organization (1996)
as “a situation that exists when all people, at all times, have
physical, social and economic access to sufficient, safe and nutritious
food that meets their dietary needs and food preferences for an active
and healthy life.”
The greenhouse is a closed and
isolated structure in which agricultural operations such as sowing,
weeding, and irrigation, etc., can be performed. Such structures
eliminate the extensive migration of pests into greenhouses and create
more favorable environments that are essential for plant growth and
productivity (Hanafi and Papasolomontos 1999; Al-Helal and Alhamdan 2009).
Solar radiation is a source of energy for photosynthesis; therefore, a
greenhouse is covered with transparent material that it allows for
transmitting visible light (Lozano et al. 1996).
Most greenhouses are constructed with polyethylene covers because they
are easy to handle. However, a general problem with polyethylene is its
short shelf life, especially in harsh weather conditions such as high
temperatures, high solar intensity, and dust (Alhamdan and Al-Helal 2009). The estimated area under greenhouse crop cultivation across the world is presented in Table 1.
Plastic film greenhouses and large plastic film tunnels (ha)
|
Glasshouses (ha)
| |
|---|---|---|
Western Europe
|
140,000
|
29,000
|
Eastern Europe
|
25,000
|
1,800
|
Africa
|
27,000
|
600
|
Middle East
|
28,000
|
13,000
|
North America
|
9,800
|
1,350
|
Central/South America
|
12,500
|
0
|
Asia/Oceania
|
450,000
|
2,500
|
Considerable amount of diesel fuel
consumption and fertilizer usage, mainly nitrogen, is important in
energy management; much of this can be saved by using greenhouse grower.
Using direct and local market improves profitability for growers while
reducing the amount of energy used during transportation of products
(Mohammadi and Omid 2010). Canakci and Akinci (2006)
analyzed energy use pattern in greenhouse for vegetable. They found
that the operational energy and energy source requirements in greenhouse
vegetable production were varying from 23,883.5 to 28,034.7 and
45,763.3 to 49,978.8 MJ/1,000 m2, respectively. The energy
ratio of four major greenhouse vegetable crops—such as tomato, pepper,
cucumber, and eggplant—was found about 0.32, 0.19, 0.31, 0.23,
respectively. The net return was 595.6–2,775.3 $/1,000 m2. The researcher concluded that crop yields increased with increasing total energy input.
As far as the state of Rajasthan
is concerned, unavailability of regular electrical power, a low water
table and small arable land holdings make the greenhouse seems a viable
option to achieve better agricultural productivity. The state also
frequently faces drought (Soni et al. 1980).
Small greenhouse used for off-season cultivation can also improve the
net profitability of marginal farmer. Therefore, a techno-economic
feasibility study of naturally ventilated greenhouse has been carried
out to articulate the basic production information and cost of
production information under arid and semi-arid climatic conditions for
establishing a greenhouse vegetable enterprise.
2 Materials and methods
Bansal and Minke (1995)
divided India into six climatic zones; hot and dry, warm and moderate,
cold and cloudy, cold and sunny, and composite as illustrated in Fig. 1.
Udaipur (24°35′N, 73°42′E) has been chosen for this study primarily
because of its composite climate, with temperatures ranging from above
40 °C in summer to below 4 °C in winter. This type of climate is
predominant in the central part of India (Bansal and Bhattacharya 2009).
The naturally ventilated
greenhouse is comprised of a galvanized tubular structure in an
aerodynamic shape. The dimensions of the constructed greenhouse are
shown in Fig. 2.
Low-density ultra violet stabilized polythene of 200 micron thickness
was used for the surface of the greenhouse. A misting system is provided
with 80 misting nozzle connected to 16 LDPE pipes and monoblock pump
(Agritech Equipment & Services Private Limited, New Delhi, India).
This system was used during extensively hot summer conditions, which
generally occur in the month of May. A gravity-fed drip irrigation
system was provided to fulfil the water requirement inside the
greenhouse. The 15 sowing beds prepared for growing crops have widths of
75 cm, and the distance between each beds was kept at 30 cm. Seeds were
sown on both sides of each bed; about 1,500 seeds were sown. The
greenhouse cost [as per the rate of 16.34 US(INR52.00 US −1 as on Oct. 4, 2012)] was per square meter. Hence, the total cost of construction is 9,153.84 US$.
Fig. 2
Front view of naturally ventilated greenhouse
2.1 Crop production
Two high-yield crops, that is,
tomato and cucumber were selected to grow inside the greenhouse. Recent
research has revealed that eating more tomatoes and tomato products can
make people healthier and decrease their risk of conditions such as
cancer, osteoporosis, and cardiovascular disease (Burton-Freeman and
Reimers 2011). Cucumber is in common use throughout the world (Bao-Zhong et al. 2006),
and like watermelons it has 95 % water; cucumbers keep the body
hydrated and help regulate the body’s temperature. They also help the
body flush out toxins (Zimmer et al. 2012).
Both tomatoes and cucumbers are used in salad, and many times are eaten
raw. The nutritive values of both tomato and cucumber are presented in
Table 2. Variety of greenhouse crops and its selling price are presented in Table 3.
Table 2
Nutritive value of tomato and cucumber (100 g)
Principle
|
Nutrient value
| |
|---|---|---|
Tomato
|
Cucumber
| |
Energy (kcal)
|
18
|
15
|
Carbohydrates (g)
|
3.9
|
3.63
|
Protein (g)
|
0.9
|
0.65
|
Total fat (g)
|
0.2
|
0.11
|
Cholesterol (mg)
|
0
|
0
|
Dietary fiber (g)
|
1.2
|
0.5
|
Vitamins
| ||
Folates (μg)
|
15
|
7
|
Niacin (mg)
|
0.594
|
0.098
|
Pyridoxine (mg)
|
0.080
|
0.040
|
Thiamin (mg)
|
0.037
|
0.027
|
Vitamin A (IU)
|
833
|
105
|
Vitamin C (mg)
|
13
|
2.8
|
Vitamin E (mg)
|
0.54
|
0.03
|
Vitamin K (μg)
|
7.9
|
16.4
|
Electrolytes
| ||
Sodium (mg)
|
5
|
2
|
Potassium (mg)
|
237
|
147
|
Minerals
| ||
Calcium (mg)
|
10
|
16
|
Iron (mg)
|
0.3
|
0.28
|
Magnesium (mg)
|
11
|
13
|
Manganese (mg)
|
0.15
|
0.079
|
Phosphorus (mg)
|
24
|
24
|
Zinc (mg)
|
0.17
|
0.20
|
Table 3
Crop and variety grown in greenhouse conditions
Crop
|
Variety
|
Selling price US$ per kg
|
|---|---|---|
Cucumber
|
Hilton
|
00.57
|
Tomato
|
NUN-7712
|
00.38
|
2.2 Economic indicators
To assess the economic viability
of greenhouse, four different economic indicators namely net present
worth (NPW), internal rate of return (IRR), benefit cost ratio (B/C
ratio), and payback period have been used (Kothari et al. 2001).
2.2.1 Net present worth
The present values of the
future returns can be calculated through the use of discounting.
Discounting is essentially a technique by which future benefits and cost
streams can be reduced to their present worth. The most straightforward
discounted cash flow measure of project worth is the net present worth
(NPW). The net present worth may be computed by subtracting the total
discounted present worth of the cost stream from that of the benefit
stream. To obtain the incremental net benefit, gross cost is subtracted
from gross benefit or the investment cost from the net benefit.
The net present worth can be computed as follows:
NPW=∑t=1t=nBt−Ct(1+i)t
where B
t
= benefit in each year (US$); C
t
= cost in each year (US$); t = 1,2,…, n; i = discount rate (%).
(1)
2.2.2 Cost-benefit ratio
This is the ratio obtained
when the present worth of the benefit stream is divided by the present
worth of the cost stream. The cost-benefit ratio is a formal selection
criterion of acceptability of project, and it should be one or greater
(Kandpal and Grag 2003).
The ratio is computed by taking the present worth of the gross benefit
less associated cost and then comparing it with the present worth of the
project cost. The corresponding cost is the value of goods and services
over, and above those included in project costs needed to make the
immediate products or services of the project available for use or sale.
Project economic cost is the sum of installation costs, operation and
maintenance cost, and replacement costs.
Mathematically cost-benefit ratio can be computed as follow:
Cost - benefit ratio=∑t=nt=1Bt(1+i)t∑t=nt=1Ct(1+i)t.
(2)
2.2.3 Internal rate of return
The internal rate of return is
a very useful measure of greenhouse project worth. It is the rate of
return on capital outstanding per period while it is invested in the
project. It is the maximum interest that a greenhouse project could pay
for the resources used if the project is to recover its investment and
operating costs and still breakeven. The internal rate of return can be
found out by systematic procedure of trial and error to find that
discount rate which will make the net present worth of the incremental
net benefit stream equal to zero.
Internal rate of return is the discount rate, i such that
∑t=1t=nBt−Ct(1+i)t=0.
(3)
2.2.4 Payback period
The payback period is the
length of time from the installation of the greenhouse until the net
value of the incremental production stream reaches the total amount of
the capital investment. It shows the length of time between cumulative
net cash outflow recovered in the form of yearly net cash inflows.
The following assumptions were made to assess the economic feasibility of naturally ventilated greenhouse:
1.
The life of greenhouse structure is 20 years.
2.
The life of greenhouse cover is 5 years.
3.
Discount rate is 10 %.
4.
Two crops of cucumber and two crops of tomato can be grown in a year greenhouse.
3 Results and discussion
The freshly harvested greenhouse vegetable product illustrated in Fig. 3
was launched in local market. As market price of a product does not
remain constant, the average price was selected to assess its economic
feasibility. The income and expenditure of each crop grown inside the
greenhouse are presented in Table 4.
The survival rate of cucumber and tomato was found about 95 and 96 %,
respectively. The annual income from cucumber and tomato is about
6,125.86 and 4,320.00 US,respectively.Thesurchargeforthesecropswas1,175.00 US , which included labor, NPK, and CaNO3. The economic indicator used to assess the economic feasibility of the greenhouse is presented in Table 5. The net present worth for cucumber and tomato crop was found to be about 28,314.59 and 15,993.92 US$, respectively (Tables 6, 7).
The cost-benefit ratio of cucumber (2.17) was higher compared to that
of tomato (1.77). As the cost-benefit ratio is greater than one for
these crops, hence such crops seem to be economically viable. As far as
the payback period is concerned, it was about 5 years and 3 months for
cucumber and about 6 year and 11 months for tomato as presented in bold
text (Tables 8, 9).
Therefore, despite high production, tomato’s payback period was higher
than that of cucumber. The internal rate of return for cucumber and
tomato crop is about 35 and 20 %, respectively.
Fig. 3
Vegetable grown inside the greenhouse. a Tomato (NUN-7712). b Cucumber (Hilton)
Table 4
Total income and expenditure for different crops under greenhouse
Particular
|
Crops
| |
|---|---|---|
Cucumber
|
Tomato
| |
Number of sown survival plants
|
1,500
|
1,500
|
Number of survival plants
|
1,425 (95 %)
|
1,440 (96 %)
|
Yield per plant (kg)
|
2.5
|
2.6
|
Annual production (kg)
|
10,687.5
|
11,232
|
Annual income (US$)
|
6,165.86
|
4,320.00
|
Surcharges (US$); labor 865.38 + NPK (150) + CaNO3 (159.62)
|
1,175.00
|
1,175.00
|
Seed cost per year (US$)
|
492.30
|
86.53
|
Cost of cultivation (US$) = surcharges + cost of seeds
|
1,667.30
|
1,261.53
|
Initial investment (US$)
|
9,153.84
|
9,153.84
|
Cost of plastic cover to be replace every 5 years (US$
|
865.38
|
865.38
|
Table 5
Economic indicator of selected crops
Crop
|
NPW (US$.)
|
B–C ratio
|
Payback period
|
IRR (%)
|
|---|---|---|---|---|
Cucumber
|
28,341.60
|
2.17
|
5 years 3 months
|
35
|
Tomato
|
15,993.94
|
1.77
|
6 years 11 months
|
20
|
Table 6
Cash flow of cucumber crop (US$)
Years
|
Cash outflow
|
Present worth of cash outflow
|
Cash inflow
|
Present worth of cash inflow
|
Net Present Worth (NPW)
|
|---|---|---|---|---|---|
0
|
9,153.85
|
9,153.85
|
0.00
|
0.00
|
−9,153.85
|
1
|
1,667.31
|
1,515.73
|
6,165.87
|
5,605.33
|
4,089.60
|
2
|
1,667.31
|
1,377.94
|
6,165.87
|
5,095.76
|
3,717.82
|
3
|
1,667.31
|
1,252.67
|
6,165.87
|
4,632.51
|
3,379.83
|
4
|
1,667.31
|
1,138.79
|
6,165.87
|
4,211.37
|
3,072.58
|
5
|
2,532.69
|
1,572.60
|
6,165.87
|
3,828.52
|
2,255.91
|
6
|
1,667.31
|
941.15
|
6,165.87
|
3,480.47
|
2,539.32
|
7
|
1,667.31
|
855.59
|
6,165.87
|
3,164.06
|
2,308.47
|
8
|
1,667.31
|
777.81
|
6,165.87
|
2,876.42
|
2,098.61
|
9
|
1,667.31
|
707.10
|
6,165.87
|
2,614.93
|
1,907.83
|
10
|
2,532.69
|
976.46
|
6,165.87
|
2,377.21
|
1,400.75
|
11
|
1,667.31
|
584.38
|
6,165.87
|
2,161.10
|
1,576.72
|
12
|
1,667.31
|
531.26
|
6,165.87
|
1,964.63
|
1,433.38
|
13
|
1,667.31
|
482.96
|
6,165.87
|
1,786.03
|
1,303.07
|
14
|
1,667.31
|
439.05
|
6,165.87
|
1,623.67
|
1,184.61
|
15
|
2,532.69
|
606.31
|
6,165.87
|
1,476.06
|
869.75
|
16
|
1,667.31
|
362.85
|
6,165.87
|
1,341.87
|
979.02
|
17
|
1,667.31
|
329.87
|
6,165.87
|
1,219.88
|
890.02
|
18
|
1,667.31
|
299.88
|
6,165.87
|
1,108.99
|
809.11
|
19
|
1,667.31
|
272.62
|
6,165.87
|
1,008.17
|
735.55
|
20
|
0.00
|
0.00
|
6,165.87
|
916.52
|
916.52
|
Total
|
24,178.89
|
52,493.49
|
28,314.60
|
Table 7
Cash flow of tomato crop (US$)
Years
|
Cash outflow
|
Present worth of cash outflow
|
Cash inflow
|
Present worth of cash inflow
|
Net present worth (NPW)
|
|---|---|---|---|---|---|
0.00
|
9,153.85
|
9,153.85
|
0.00
|
0.00
|
−9,153.85
|
1.00
|
1,261.54
|
1,146.85
|
4,320.00
|
3,927.27
|
2,780.42
|
2.00
|
1,261.54
|
1,042.59
|
4,320.00
|
3,570.25
|
2,527.65
|
3.00
|
1,261.54
|
947.81
|
4,320.00
|
3,245.68
|
2,297.87
|
4.00
|
1,261.54
|
861.65
|
4,320.00
|
2,950.62
|
2,088.97
|
5.00
|
2,126.92
|
1,320.65
|
4,320.00
|
2,682.38
|
1,361.73
|
6.00
|
1,261.54
|
712.11
|
4,320.00
|
2,438.53
|
1,726.42
|
7.00
|
1,261.54
|
647.37
|
4,320.00
|
2,216.84
|
1,569.47
|
8.00
|
1,261.54
|
588.52
|
4,320.00
|
2,015.31
|
1,426.79
|
9.00
|
1,261.54
|
535.02
|
4,320.00
|
1,832.10
|
1,297.09
|
10.00
|
2,126.92
|
820.02
|
4,320.00
|
1,665.55
|
845.53
|
11.00
|
1,261.54
|
442.16
|
4,320.00
|
1,514.13
|
1,071.97
|
12.00
|
1,261.54
|
401.97
|
4,320.00
|
1,376.49
|
974.52
|
13.00
|
1,261.54
|
365.42
|
4,320.00
|
1,251.35
|
885.93
|
14.00
|
1,261.54
|
332.20
|
4,320.00
|
1,137.59
|
805.39
|
15.00
|
2,126.92
|
509.17
|
4,320.00
|
1,034.17
|
525.01
|
16.00
|
1,261.54
|
274.55
|
4,320.00
|
940.16
|
665.61
|
17.00
|
1,261.54
|
249.59
|
4,320.00
|
854.69
|
605.10
|
18.00
|
1,261.54
|
226.90
|
4,320.00
|
776.99
|
550.09
|
19.00
|
1,261.54
|
206.27
|
4,320.00
|
706.35
|
500.08
|
20.00
|
4,320.00
|
642.14
|
642.14
| ||
Total
|
20,784.66
|
36,778.60
|
15,993.94
|
Table 8
Payback period of cucumber crop (US$)
Years
|
Present worth of cash out flow in 20 years
|
Cash inflow
|
Present worth of cash inflow
|
Cumulative cash inflow
|
|---|---|---|---|---|
1
|
24,178.89
| |||
2
|
6,165.87
|
5,605.33
|
5,605.33
| |
3
|
6,165.87
|
5,095.76
|
10,701.09
| |
4
|
6,165.87
|
4,632.51
|
15,333.59
| |
5
|
6,165.87
|
4,211.37
|
19,544.96
| |
6
|
6,165.87
|
3,828.52
|
23,373.48
| |
7
|
6,165.87
|
3,480.47
|
26,853.95
| |
8
|
6,165.87
|
3,164.06
|
30,018.02
|
Table 9
Payback period of tomato crop (US$)
Years
|
Present worth of cash out flow in 20 years
|
Cash inflow
|
Present worth of cash inflow
|
Cumulative cash inflow
|
|---|---|---|---|---|
20,784.65
| ||||
1
|
4,320.00
|
3,927.27
|
3,927.27
| |
2
|
4,320.00
|
3,570.25
|
7,497.52
| |
3
|
4,320.00
|
3,245.68
|
10,743.20
| |
4
|
4,320.00
|
2,950.62
|
13,693.82
| |
5
|
4,320.00
|
2,682.38
|
16,376.20
| |
6
|
4,320.00
|
2,438.53
|
18,814.73
| |
7
|
4,320.00
|
2,216.84
|
21,031.57
| |
8
|
4,320.00
|
2,015.31
|
23,046.88
| |
9
|
4,320.00
|
1,832.10
|
24,878.98
|
4 Conclusion
A number of motives exist for
promoting greenhouse cultivation in the state, including the fact that
it provides a solution for food security, reduces the growing period,
produces high yields of highly nutritive vegetables, and allows for
year-round cultivation. Given the high water table and limited canal
irrigation facilities, farmers mainly grow rain-fed crops. Off-season
cultivation of vegetable in controlled environments increases farmer’s
income by approximately 25–40 % through vegetable sold even in the local
market. The income is generated during the off-season period, when
other such opportunities are not available. Women in the state tend to
be responsible for marketing vegetables, managing income, which is
mostly utilized for children’s; in this way, they also develop social
relations and gain self-confidence.
However, promoting greenhouse
cultivation faces many challenges in from a socio-economic point of
view. Not many people are aware about protected cultivation, and
experience in greenhouse management is limited. Hence, more technical
and practical training interventions are required to popularize
greenhouse use. In addition, marketing fresh product grown in a
greenhouse could be challenging; if farmers sell such product at a local
market, their net profit will be reduced marginally. Hence, the state
government should take care to develop a market for such high-value
product.
The following conclusions were drawn from the present study:
-
The survival rate of cucumber and tomato crop was shown to be found reasonably successful.
-
Yields per plant in cucumber and tomato crops of approximately 2.5 and 2.6 kg have been recorded.
-
This technology is suitable for cultivating cucumber and tomato crops in Rajasthan’s climatic conditions.
-
Economic results indicate that greenhouse cultivation increases the income of marginal farmers, and it is suitable even where irregular electrical supply is a major problem.
-
Cash crops such as medicinal crops can also be cultivated in greenhouses.
Acknowledgments
The
author (N. L. Panwar) gratefully acknowledges Maharana Pratap
University of Agriculture and Technology, Udaipur (Rajasthan), India,
and Indian Institute of Technology, Delhi, for sponsorship under the
quality improvement program of the Government of India. The financial
support extended by Indian Council of Agricultural Research, Govt. of
India is gratefully acknowledged.
References
Alhamdan
AM, Al-Helal IM (2009) Mechanical deterioration of polyethylene
greenhouses covering under arid conditions. J Mater Process Technol
209:63–69CrossRef
Al-Helal
IM, Alhamdan AM (2009) Effect of arid environment on radiative
properties of greenhouse polyethylene cover. Sol Energy 83:790–798CrossRef
Bansal
NK, Bhattacharya A (2009) Parametric equations for energy and load
estimations for buildings in India. Appl Therm Eng 29:3710–3715CrossRef
Bansal NK, Minke G (1995) Climatic zones and rural housing in India. Forschungszentrum Juelich, Germany ISBN: 3-89336-162-6
Bao-Zhong
Y, Sun J, Kang Y, Nishiyama S (2006) Response of cucumber to drip
irrigation water under a rainshelter. Agric Water Manag 81(1–2):145–158
Black
MJ, Whittaker C, Hosseini SA, Diaz-Chavez R, Woods J, Murphy RJ (2011)
Life cycle assessment and sustainability methodologies for assessing
industrial crops, processes and end products. Ind Crops Prod
34:1332–1339CrossRef
Burton-Freeman B, Reimers K (2011) Tomato consumption and health: emerging benefits. Am J Lifestyle Med 5(2):182–191CrossRef
Canakci M, Akinci I (2006) Energy use pattern analyses of greenhouse vegetable production. Energy 31:1243–1256CrossRef
Charles
H, Godfray J, Beddington JR, Crute IR, Haddad L, Lawrence D, Muir JF,
Pretty J, Robinson S, Thomas SM, Toulmin C (2010) Food security: the
challenge of feeding: billion people. Science 327(5967):812–818CrossRef
Food and Agriculture Organization (1996) Declaration on world food security. World Food Summit, FAO, Rome
Giacomelli
G, Castilla N, Vanhenten E, Mears D, Sase S (2008) Innovation in
greenhouse engineering. Acta Hortic 801:75–88 (adopted from Cabeltitz
CV. Integrated Greenhouse system for mild climates. Springer Heidelberg
Dordrecht 2011. ISBN: 9783642145810)
Hanafi
A, Papasolomontos A (1999) Integrated production and protection under
protected cultivation in the Mediterranean region. Biotechnol Adv
17:183–203CrossRef
Indira
Devi P (2007) Pesticide use in the rice bowl of Kerala: health costs
and policy options. SANDEE working paper No. 20–07, p 40. http://www.sandeeonline.org/uploads/documents/publication/764_PUB_workingpaper_20_indira.pdf. Retrieved 19 July 2012
Kandpal TC, Grag HG (2003) Financial evaluation of renewable energy technology. Macmillan India Ltd., Delhi
Keatinge
JDH, Yang R-Y, Hughes Jd’A, Easdown WJ, Holmer R (2011) The importance
of vegetables in ensuring both food and nutritional security in
attainment of the millennium development goals. Food Sec 3:491–501CrossRef
Kothari S, Panwar NL (2004) Evaluation of Greenhouse for cultivation of Babachi. Indian Farm 54(6):16–18
Kothari
S, Rathore NS, Panwar NL (2001) Techno economics of greenhouse for
cultivation of aswagandha under composite climate of Udaipur. Agric Eng
Today 25(3–4):36–40
Kothari
S, Kaushik SC, Mathur AN (2006) Greenhouse Science and Technology.
Himanshu Publication, Udaipur, India. ISBN. 8179061256
Kristjanson P, Mango N, Krishna A, Radeny M, Johnson N (2010) Understanding poverty dynamics in Kenya. J Int Dev 22(7):978–996CrossRef
Lozano G, Gonzalez D, Santos E (1996) Growing lettuce in greenhouses clad with polychromatic films. Plasticulture 110:15–22
Martin A (2012) The plastic portal available on http://www.plasticseurope.org/use-of-plastics/agriculture.aspx. Retrieved 19 July 2012
Mohammadi
A, Omid M (2010) Economical analysis and relation between energy inputs
and yield of greenhouse cucumber production in Iran. Appl Energy
87:191–196CrossRef
Mukhopadhyay
SK, Chattopadhyay A, Chakraborty I, Bhattacharya I (2011) Crops that
feed the world. 5. Sweetpotato. Sweetpotatoes for income and food
security. Food Secur 3(3):283–305CrossRef
Nelson
GC, Rosegrant MW, Palazzo A, Gray I, Ingersoll C, Robertson R, Tokgoz
S, Zhu T, Sulser TB, Ringler C, Msangi S, You L (2010) Food security,
farming, and climate change to 2050: scenarios, results, policy options.
International Food Policy Research Institute, Washington
Panwar
NL, Kothari S, Rathore NS (2009) Protected cultivation of medicinal
plant in composite climate of Indian State Rajasthan. Am Eurasian J
Agric Environ Sci 5(5):633–637CrossRef
Singh
MB, Lakshminarayana J, Fotedar R (2006). Studies on the nutritional
status of rural population in desert area of rajasthan. In: Proceeding
of national symposium, held at Jabalpur during October 19–20, 2006
Soni AL, Singh RN, Gupta BD (1980) Nutritional disorders in rural Rajasthan. Indian J Pediat 47:199–202CrossRef
Tiwari PC, Joshi B (2012) Natural and socio-economic factors affecting food security in the Himalayas. Food Sec 4:195–207CrossRef
Yamano
T, Otsuka K, Place F (2011) Emerging development of agriculture in East
Africa: markets, soil and innovation. Springer, DordrechtCrossRef
Yang
R-Y, Hanson PM, Lumpkin TA (2007) Better health through
horticulture—AVRDC’s approach to improved nutrition of the poor. Acta
Horticulturae 744:71–77
Zimmer AZ, Leonardi B, Miron D, Schapoval E, Oliverira JR, Gosmann G (2012) Antioxidant and anti-inflammatory properties of Capsicum baccatum: from traditional use to scientific approach. J Ethnopharmacol 139(1):228–233CrossRef
Humans
alter the water cycle by constructing dams and through water
withdrawals. Climate change is expected to additionally affect water
supply and demand. Here, model analyses of climate change and direct
human impacts on the terrestrial water cycle are presented. The results
indicate that the impact of man-made reservoirs and water withdrawals on
the long-term global terrestrial water balance is small. However, in
some river basins, impacts of human interventions are significant. In
parts of Asia and the United States, the effects of human interventions
exceed the impacts expected for moderate levels of global warming. This
study also identifies areas where irrigation water is currently scarce,
and where increases in irrigation water scarcity are projected.
Humans directly change the dynamics of the water cycle through dams constructed for water storage, and through water withdrawals for industrial, agricultural, or domestic purposes. Climate change is expected to additionally affect water supply and demand. Here, analyses of climate change and direct human impacts on the terrestrial water cycle are presented and compared using a multimodel approach. Seven global hydrological models have been forced with multiple climate projections, and with and without taking into account impacts of human interventions such as dams and water withdrawals on the hydrological cycle. Model results are analyzed for different levels of global warming, allowing for analyses in line with temperature targets for climate change mitigation. The results indicate that direct human impacts on the water cycle in some regions, e.g., parts of Asia and in the western United States, are of the same order of magnitude, or even exceed impacts to be expected for moderate levels of global warming (+2 K). Despite some spread in model projections, irrigation water consumption is generally projected to increase with higher global mean temperatures. Irrigation water scarcity is particularly large in parts of southern and eastern Asia, and is expected to become even larger in the future.
- See more at: http://wwfscience.org/resources/wwf-literature-digest/2605#sthash.kwfXnW7K.dpuf
Humans directly change the dynamics of the water cycle through dams constructed for water storage, and through water withdrawals for industrial, agricultural, or domestic purposes. Climate change is expected to additionally affect water supply and demand. Here, analyses of climate change and direct human impacts on the terrestrial water cycle are presented and compared using a multimodel approach. Seven global hydrological models have been forced with multiple climate projections, and with and without taking into account impacts of human interventions such as dams and water withdrawals on the hydrological cycle. Model results are analyzed for different levels of global warming, allowing for analyses in line with temperature targets for climate change mitigation. The results indicate that direct human impacts on the water cycle in some regions, e.g., parts of Asia and in the western United States, are of the same order of magnitude, or even exceed impacts to be expected for moderate levels of global warming (+2 K). Despite some spread in model projections, irrigation water consumption is generally projected to increase with higher global mean temperatures. Irrigation water scarcity is particularly large in parts of southern and eastern Asia, and is expected to become even larger in the future.
- See more at: http://wwfscience.org/resources/wwf-literature-digest/2605#sthash.kwfXnW7K.dpuf
Global Water Resources Affected by Human Interventions and Climate Change
Citation:
PNAS 111: 3251–3256
Publication Year:
2014
Humans
alter the water cycle by constructing dams and through water
withdrawals. Climate change is expected to additionally affect water
supply and demand. Here, model analyses of climate change and direct
human impacts on the terrestrial water cycle are presented. The results
indicate that the impact of man-made reservoirs and water withdrawals on
the long-term global terrestrial water balance is small. However, in
some river basins, impacts of human interventions are significant. In
parts of Asia and the United States, the effects of human interventions
exceed the impacts expected for moderate levels of global warming. This
study also identifies areas where irrigation water is currently scarce,
and where increases in irrigation water scarcity are projected.
Humans directly change the dynamics of the water cycle through dams constructed for water storage, and through water withdrawals for industrial, agricultural, or domestic purposes. Climate change is expected to additionally affect water supply and demand. Here, analyses of climate change and direct human impacts on the terrestrial water cycle are presented and compared using a multimodel approach. Seven global hydrological models have been forced with multiple climate projections, and with and without taking into account impacts of human interventions such as dams and water withdrawals on the hydrological cycle. Model results are analyzed for different levels of global warming, allowing for analyses in line with temperature targets for climate change mitigation. The results indicate that direct human impacts on the water cycle in some regions, e.g., parts of Asia and in the western United States, are of the same order of magnitude, or even exceed impacts to be expected for moderate levels of global warming (+2 K). Despite some spread in model projections, irrigation water consumption is generally projected to increase with higher global mean temperatures. Irrigation water scarcity is particularly large in parts of southern and eastern Asia, and is expected to become even larger in the future.
Humans directly change the dynamics of the water cycle through dams constructed for water storage, and through water withdrawals for industrial, agricultural, or domestic purposes. Climate change is expected to additionally affect water supply and demand. Here, analyses of climate change and direct human impacts on the terrestrial water cycle are presented and compared using a multimodel approach. Seven global hydrological models have been forced with multiple climate projections, and with and without taking into account impacts of human interventions such as dams and water withdrawals on the hydrological cycle. Model results are analyzed for different levels of global warming, allowing for analyses in line with temperature targets for climate change mitigation. The results indicate that direct human impacts on the water cycle in some regions, e.g., parts of Asia and in the western United States, are of the same order of magnitude, or even exceed impacts to be expected for moderate levels of global warming (+2 K). Despite some spread in model projections, irrigation water consumption is generally projected to increase with higher global mean temperatures. Irrigation water scarcity is particularly large in parts of southern and eastern Asia, and is expected to become even larger in the future.
Публикации
Няма коментари:
Публикуване на коментар