Premium
A method of derivation of nonconstant watershed response functions
Author(s) -
Snyder W. M.,
Mills W. C.,
Stephens J. C.
Publication year - 1970
Publication title -
water resources research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.863
H-Index - 217
eISSN - 1944-7973
pISSN - 0043-1397
DOI - 10.1029/wr006i001p00261
Subject(s) - hydrograph , convolution (computer science) , storm , variable (mathematics) , watershed , mathematics , stage (stratigraphy) , surface runoff , function (biology) , state variable , streamflow , hydrology (agriculture) , meteorology , environmental science , mathematical analysis , geology , computer science , geotechnical engineering , geography , drainage basin , physics , ecology , paleontology , cartography , machine learning , evolutionary biology , artificial neural network , biology , thermodynamics
A practical numerical method has been developed for storm hydrograph analysis that allows variation of unit response functions during multiple periods of runoff producing rainfall. Previous methods of analytical derivation of unit responses from multiple period storms require superpositional linearity. This requirement is eliminated by substituting two stages of convolution for the conventional single stage process of storm hydrograph generation. The first stage convolves a fixed area‐characteristic function with a variable state function. The state function is determined by streamflow and differs for each increment of effective rainfall. First stage convolution thus produces a different unit response for each increment. The second stage convolves the variable unit responses and the effective rain increments. Validity of the technique is demonstrated by applying the analytical model to 10 complex storms. The conceptual components of area‐characteristic, area‐state, and variable response functions resulting from the analyses are seen to be hydrologically rational.