Climate and agricultural land use change impacts on streamflow in the upper midwestern United States

Abstract

Increased streamflow and its associated impacts on water quality have frequently been linked to changes in land use and land cover (LULC) such as tile drainage, cultivation of prairies, and increased adoption of soybeans (Glycine max) in modern day cropping systems. This study evaluated the relative importance of changes in precipitation and LULC on streamflow in 29 Hydrologic Unit Code 008 watersheds in the upper midwestern United States. The evaluation was done by statistically testing the changes in slope and intercept of the relationships between ln(annual streamflow) versus annual precipitation for the periods prior to 1975 (prechange period) and after 1976 (postchange period). A significant shift either in slope or intercept of these relationships was assumed to be an indication of LULC changes whereas a lack of significant shift suggested a single relationship driven by precipitation. All 29 watersheds showed no statistical difference in slope or intercept of the relationships between the two periods. However, a simpler model that kept the slope constant for the two periods showed a slight upward shift in the intercept value for 10 watersheds in the postchange period. A comparison of 5 year moving averages also revealed that the increased streamflows in the postchange period are mainly due to an increase in precipitation. Minimal or the lack of LULC change impact on streamflow results from comparable evapotranspiration in the two time periods. We also show how incorrect assumptions in previously published studies minimized precipitation change impacts and heightened the LULC change impacts on streamflows.

Reply to comment by Belmont et al. on “Climate and agricultural land use change impacts on streamflow in the upper midwestern United States”

“…With regard to the criticism on the robustness of our methodology, we would like to point out that Gupta et al [2015] used a statistical technique to test whether or not prechange and postchange streamflow versus precipitation relationships are the same or different. Our conclusion was that the shape of the relationship has not changed drastically for most watersheds and higher flows in postchange period are mainly due to higher precipitation (5 year moving average plots).…”

“…Using the 5 year moving averages, Gupta et al [2015] showed that increased flows in the postchange period are mainly clustered at higher precipitation levels with the general relationship between streamflow versus precipitation nearly the same. Also, these authors showed that annual ET has remained the same [Gupta et al, 2015, Tables S3 and S4] (Sen's slope was not significant at P 5 0.05) for all agricultural watersheds for the period of streamflow records. Gupta et al [2015] further explored the underlying reason for a constant ET between the prechange and postchange periods by comparing the ET of present and previous crops; lakes, ponds, depressions, and pan evaporation; as well as differences in consumptive use by hay before and after LULC changes.…”

“…Schilling et al [2010], Schottler et al [2014], and Foufoula-Georgiou et al [2015] have argued the adoption of soybean as one of the major reasons for increased streamflow. In their paper, Gupta et al [2015] went beyond the question of only looking at statistical analysis in making the case that the change in cropping system impact from past to present is small to likely absent. Using the 5 year moving averages, Gupta et al [2015] showed that increased flows in the postchange period are mainly clustered at higher precipitation levels with the general relationship between streamflow versus precipitation nearly the same.…”

“…In their paper, Gupta et al [2015] went beyond the question of only looking at statistical analysis in making the case that the change in cropping system impact from past to present is small to likely absent. Using the 5 year moving averages, Gupta et al [2015] showed that increased flows in the postchange period are mainly clustered at higher precipitation levels with the general relationship between streamflow versus precipitation nearly the same. Also, these authors showed that annual ET has remained the same [Gupta et al, 2015, Tables S3 and S4] (Sen's slope was not significant at P 5 0.05) for all agricultural watersheds for the period of streamflow records.…”

“…Gupta et al [2015] further explored the underlying reason for a constant ET between the prechange and postchange periods by comparing the ET of present and previous crops; lakes, ponds, depressions, and pan evaporation; as well as differences in consumptive use by hay before and after LULC changes. Based on these pieces of supporting evidence, Gupta et al [2015] postulated that the lack of cropping system change impact on streamflow may be partially due to nearly equal ET between the two periods, i.e., higher ET from row crops on an expanded area (even with some loss of wetlands) in the postchange period versus lower ET from small grain areas (even with presence of some wetlands and native prairies) in the prechange period. Using the trends in pan evaporation data, they also suggested lower vapor pressure gradient between plants and atmosphere from higher precipitation as another possible reason for near similar cropping system effects on ET between the prechange and postchange periods.…”