Effects of Land Use on Runoff Curve Number

Calculating direct event run off from rainfall by the Curve Number method relies on the coefficient Curve Number" (or "CN"), for which handbooks are the customary source. However, with the availability of large data sets and electronic data processing, local CNs can be calculated using local rainfall-runoff data. In this study a large data set was used to develop a better understanding of the land use variable as it applies to the Curve Number method of modeling storm runoff. Data-derived CNs for 15 distinct land uses were calculated using event rainfall-runoff data from 177 small watersheds in the United States. Many of the land uses were agricultural in nature, and all were as defined in The National Engineering Handbook-Section 4, Hydrology (NEH-4). Curve Numbers for each land use on each watershed were calculated using the Asymptotic method and evaluated at 3 scales; local, regional and national. Significant differences (5% level) were found between the CNs of almost all the different land uses tested. Differences in land use CNs at the local level can be attributed solely to a change in hydrological response of a watershed due to the land use because all other input variables of the Curve Number model (soil type, watershed condition and cover density) remain constant. Differences in land use Curve Numbers at the regional and national scale may be due to other factors in addition to land use, such as climate. Significant differences in CN were also found on grazed and ungrazed paired watersheds, and on watersheds that had undergone land use conversions. The general magnitudes and rank order of the average land use CNs were in rough accord with expected handbook values. Meadow land use almost always produced the lowest CN at both the local and regional level while Forest land use produced the lowest overall average Curve Number at the national level, but also displayed the highest CN variability. When comparable. Curve Numbers from desert brush land use was always higher than from all other land uses. No significant differences could be determined between Curve Numbers for pasture and range at the regional scale or between row crops and small grain land use at any scale. Where comparable, pastures usually had higher CNs than meadows.