A graphic method of studying the frequency of recurrence of runoff peaks has been applied to the short‐period records from a number of small agricultural areas. The study shows that for periods as short as two to six years, a consistent pattern of runoff behavior is evident for the data available. The analysis of the data, by this method, for three watersheds in the common three year rotation of corn, grain, and hay, indicates a marked increase in magnitude of runoff peaks where tillage, considerably off the contour, is practiced, compared with contour strip cropping. Where diversion terraces are used in conjunction with the rotation contour cropping, a decrease in the number and magnitude of peak flows is indicated. This is due largely to the channel storage in the terrace which spreads out and reduces the peaks and delays the runoffFor pastured areas, the data show that the maintenance of good turf on the areas is essential, particularly in winter. The addition of pasture contour furrows definitely changed the runoff pattern and resulted in a marked reduction in runoff for smaller, low intensity rains. For the larger rains, the reduction in runoff by the furrows decreases as the depth or intensity of the rainfall increasesIn general, it appears that short‐period records of runoff can be classified on the basis of frequency of recurrence and serve as a guide in making recommendations as to runoff expectancy under similar land managements and conditions. This method of determining the frequency of any runoff magnitude directly is considered superior to other methods where runoff values are first related to the rainfall that produced them and the determination of runoff frequency is then attempted from rainfall records that are available for longer periods. The advantage of the present method is obvious when it is remembered that rainfall intensity and amount are but two elements in the production of runoff, whereas this method actually measures the distribution of runoff frequencies and takes into account the effect of other watershed characteristics such as length and degree of slope; drainage pattern; size and shape of area; cover; soil type, depth, and condition; and tillage practices. The consistency that is apparent in the plotting of the data indicates that short records may be employed to characterize runoff behavior, and while data for longer periods unquestionably are needed, the direct use of such records should not be overlooked and at present may offer the best method of obtaining usable information.
Read full abstract