| Abstrakt: |
The 2007 IPCC report documents increasing air temperature and precipitation, particularly over the last 30 to 50 years. The question arises as to whether changes in historical climate may affect the performance of the curve number (CN) algorithm, which is widely used to estimate runoff by the engineering community. A study was conducted to determine the effects of climate period (period of similar precipitation accumulation) on CN algorithm components using historical data available at the USDA-ARS North Appalachian Experimental Watershed near Coshocton, Ohio. The data came from a small experimental watershed (WS130, 0.66 ha) that has been in the same land management practice (hay production) for ~74 years beginning in 1937 (monitored ~89% of time), the watershed size from which the original CN methodology was developed. Changes in precipitation mass curve slopes were used to identify seven climate periods of precipitation. Trends were evaluated in event-based CN component variables including Q (event runoff), P (event causal precipitation), Q/P, Ia (initial abstraction due to infiltration, interception, etc., in the CN method), Pe (effective precipitation, P-Ia), Ia/P, CNe (event-based CN), and m (m = Ia/S, where S = retention parameter). Due to wide variability of the data, trends in median CN components were used to identify possible trends in the data. There was a weak but notable increase in precipitation since 1937, primarily due to increasing monthly precipitation trends in each month from August through December and increases in extreme precipitation. There was no trend of CN across climate periods using the assumption of Ia = 0.2S. Medians of Q, P, Q/P, Ia Pe, and Ia/P did not show statistically significant trends with climate period using the event-based approach, but nearly all showed positive correlation. The historical climate trend increased parameter m (0.0045 m year-1 for Pe > 25 mm) and event CNe (0.29 CN year-1 for P > 25 mm). Significant trends were detected even if the data were not separated into periods of similar precipitation accumulation. Larger CNe may be due to increasing m and Ia (Ia not increasing significantly). Pe consistently showed nonsignificant negative correlation, suggesting that Ia tended to increase more than P, resulting in smaller Pe and thus supporting the observed increasing trend in m and CNe. Precipitation likely has less impact on the CN methodology than air temperature through the evapotranspiration process. However, it is difficult to separate their individual effects (the present study focused on precipitation). The results suggest that if current climate trends for precipitation and temperature continue, an occasional re-evaluation of the effects of these trends on CN model components may be necessary. The current literature recommendation is to change the CN domain by changing m from 0.2 to 0.05; however, m appears to be trending upward toward 0.2 in the recent runoff record. If climate trends increase or decrease, then m may be similarly trending, leading to uncertainty in the proper domain for the CN methodology over time. [ABSTRACT FROM AUTHOR] |
