![[logo] US EPA](https://webarchive.library.unt.edu/eot2008/20081107231932im_/http://www.epa.gov/epafiles/images/logo_epaseal.gif){kind=logo}
Final Report: Regional Streamflow Sensitivity to Climate Change in an Urbanizing Environment
EPA Grant Number: R824806Title: Regional Streamflow Sensitivity to Climate Change in an Urbanizing Environment
Investigators: DeWalle, David R. , Swistock, Bryan R.
Institution: Pennsylvania State University - Main Campus
EPA Project Officer: Winner, Darrell
Project Period: October 1, 1995 through September 30, 1998 (Extended to September 30, 1999)
Project Amount: $120,000
RFA: Regional Hydrologic Vulnerability to Global Climate Change (1995)
Research Category: Ecological Indicators/Assessment/Restoration , Global Climate Change
Description:
Objective:The objective of this research was to estimate the effects of climate change on streamflow for rural as compared to urbanizing watersheds in Northeast, North-Central, Western, and Southern regions in the United States. The estimates of climate change impacts were based on a statistical analysis of flow and climatic data for 21 rural and 39 urbanizing watersheds over the past 50 years. Because long-term changes in land use were not known, urbanization was simulated using a relation between population density data from U.S. Census records and percentage of basin area in urban land use developed in this research. Estimates of streamflow changes were conducted for high, median, and low-flow conditions during annual as well as seasonal time periods. Summary/Accomplishments (Outputs/Outcomes):
Estimates of the impacts of climate change on streamflow generally have not included the potential concurrent effects of urbanization. Because urbanization is likely to occur over the period of climate change, and urbanization is known to significantly affect streamflow, the interaction of climate change and urbanization could greatly affect future hydrologic outcomes. Recent improvements in global circulation models and downscaling procedures now are providing estimates of potential climate change for various seasons of the year in smaller geographic areas, making predictions of high- and low-flow rates in streams at different times of the year and for different regions of the country possible. Consequently, we also have considered variations in climate change impacts on streamflow with: (1) season of year, (2) high flows versus low flows, and (3) regions in the United States. We used the average response of 21 rural and 39 urbanizing watersheds in the United States.
Urbanization Impacts. The most important finding from this research is that urbanization could completely offset streamflow decreases or greatly enhance streamflow increases caused by climate change in the future. Urbanization was found to increase average annual streamflow in rough proportion to the percentage changes in population density on a watershed. Greater amounts of streamflow occur in urbanizing areas due to runoff from impervious surfaces and reduced evaporation caused by reductions in evaporation by vegetation. A relation developed between population density and the fraction of watershed area that was urbanized showed that 100 percent urbanization was roughly equivalent to a population density of 5-6 persons per acre. This level of urbanization was estimated to cause a 103 percent increase in average annual flow for the 39 urbanizing basins.
The impact of flow increases due to urbanization would depend on the magnitude and type of climate changes. In general, streamflow is decreased by increasing air temperatures and increased by rising precipitation amounts. The net effect of a combined temperature and precipitation change usually is dominated by the precipitation change, with some notable regional exceptions described later. An extreme climate change, on an annual basis, would be a 4 C increase in air temperature and a 20 percent decrease in precipitation. For the 21 rural watersheds studied, mean annual flows would be reduced by about 40?50 percent if air temperature increased by 4 C and precipitation were reduced by 20 percent. Clearly, a 103 percent increase in flow due to urbanization would offset the 40?50 percent decrease in flow due to climate change to produce a 53?63 percent net increase in flow.
In cases where precipitation was increased, rather than decreased, by 20 percent and air temperature was increased by 4 C, the effect of urbanization would be additive with the effect of climate change. Using the average results for the 21 rural basins, mean annual flow was estimated to increase by about 15 percent if air temperatures were raised by 4 C and precipitation increased by 20 percent. A 103 percent increase in flow due to urbanization combined with a 15 percent increase in flow due to climate change would greatly enhance the flow changes in this case, and produce a combined 118 percent increase in flow.
The potential dominating influence of urbanization adds to the complexity of estimating future effects of climate change on streamflow. Combined changes in temperature and precipitation over a watershed will dictate the effects of climate change on flow in the future. The impact of urbanization on flow will depend on the changes in population density. Watersheds in rapidly developing regions, such as many basins in the Eastern United States, may experience large increases in population density. Basins in more remote and mountainous areas, typically found in the Western United States, may undergo relatively little population growth. This study suggests that one of the most important regional differences in streamflow response to climate changes may be due to differences in the extent of urbanization occurring among regions over time. Furthermore, the study clearly shows that when estimating the long-term changes in streamflow, the combined effects of climate change and urbanization should be considered.
Seasonal Interactions. Another important finding from this research is that temperature increases will have a greatly different impact on streamflow depending upon season of the year. Temperature increases in summer have a negative effect on streamflow, while temperature increases in winter have a very small negative or slightly positive effect on streamflow. Temperature increases during warmer parts of the year, like summer, increase evaporation and reduce flows. In winter, warmer periods produce greater snowmelt rates when a snow pack is present or cause precipitation to occur as rain instead of snow. Greater snowmelt rates and rain rather than snowfall both cause streamflow to increase in the winter season. It is important to realize that none of the basins studied, rural or urbanizing, were located in mountainous regions where large seasonal snow packs developed and streamflow is dominated by snowmelt runoff. Nevertheless, streamflow rates in winter responded positively to temperature increases on these basins, similar to the response predicted for mountainous basins with significant snowpacks. The timing of temperature increases with season of the year, as predicted by global circulation models, will greatly influence the predicted streamflow response on basins in this study.
Seasonal effects were less on urbanizing than rural watersheds during the summer season. Fewer urbanizing than rural watersheds showed significant effects of temperature on flow during the summer season. Loss of vegetation and soil water storage capacity caused by development and compaction on urbanizing watersheds reduces the sensitivity of streamflow to temperatures. Thus, even the seasonal sensitivity of streamflow to temperature appears to be affected by urbanization.
High Versus Low Flows. Streamflow sensitivity to temperature was most pronounced during low-flow periods. Low flows typically occur with relatively dry soil conditions when differences in evaporation caused by temperature variations have a relatively great effect on streamflow. At times of high flow, watershed surfaces are wetter and flow in streams is less sensitive to temperature-induced evaporation variations. Consequently, lower flows were much more affected by higher temperature than high flows. Because low flows commonly occur during summer and fall, temperature effects on low flows also were greater at that time.
Sensitivity of streamflow to precipitation also varied between high- and low-flow rates. Higher flow rates in streams were much more sensitive to amounts of precipitation received in each season than lower flow rates. Consequently, variations in precipitation amounts would change high flows more than low flows in all seasons.
Overall, climate change affected both low-flow and high-flow rates. Precipitation increases caused greater high flows in streams, while temperature increases caused bigger reductions in low flows. Because low flows and high flows commonly occur in summer and winter seasons, respectively, the greatest impacts on flow classes also would commonly occur in these seasons.
Regional Impacts. The most important regional impact identified in
this research is the previously described differences in urbanization among
regions in the United States. Watersheds in regions with major increases in
population density will have major differences in response to climate change
from basins in the same region without urbanization, as previously described.
We also found that streamflow on watersheds in the North-Central region
increased and decreased with varying precipitation more than in other regions.
In contrast, streamflow on watersheds in the Western region was least sensitive
to precipitation amounts, but most sensitive to temperature variations. Basins
studied in the Northeast and South regions had a similar response to
precipitation changes. Other factors being equal, basins in the North-Central
region would be expected to show the largest flow responses with precipitation
changes. Basins in the West would show the greatest streamflow decreases with
increases in air temperature. The major findings from this research are
summarized in the following table.
Summary of Factors Affecting Streamflow Response
to Climate
Change in the United States
| Urbanization | Streamflow greater on urbanizing than rural basins regardless of climate change |
| Temperature increases cause greater flow changes on rural than urbanizing basins | |
| Seasons of the Year | Temperature increases reduce streamflow in summer, but cause small increases in winter flows |
| High Versus Low Flows | Temperature increases have greater effect on low flows in summer than other seasons |
| Precipitation increases exponentially increase higher flows more than lower flows | |
| Regions | Temperature increases cause greater declines in flow on basins in the West |
| Urbanizing basins in any region will experience increased flows | |
| Precipitation increases cause greater flow increases on basins in the North-Central region |
Journal Articles on this Report: 2 Displayed | Download in RIS Format
| Other project views: | All 9 publications | 2 publications in selected types | All 2 journal articles |
| Type | Citation | ||
|---|---|---|---|
|
|
DeWalle DR, Swistock BR, Johnson TE, McGuire KJ. Potential effects of climate change and urbanization on mean annual streamflow in the United States. Water Resources Research 2000;36(9):2655-2664. |
R824806 (Final) |
not available |
|
|
DeWalle DR, Swistock BR, Johnson TE. Streamflow sensitivity to climate change on urbanizing and rural basins in the United States: seasonal, flow class and regional effects. Journal of the American Water Resources Association. |
R824806 (Final) |
not available |
hydrology, regression analysis, watersheds, precipitation, groundwater, air temperature, urbanization, climate change. , Water, Air, Geographic Area, Scientific Discipline, RFA, Water & Watershed, climate change, Ecological Risk Assessment, EPA Region, Hydrology, Watersheds, precipitation patterns, alternative urbanization scenarios, drinking water supplies, hydrologic models, watershed, climatic models, land use, Region 3, climate models, climate variability, land and water resources, region 4, Region 10, regional hydrologic vulnerability, urban growth, water resources, urbanizing watersheds, ecosystem models, streamflow sensitivity
Progress and Final Reports:
1998 Progress Report
Original Abstract