Stochastic Models in Hydrology 'Source Title' Hydrologic Modeling of Small Watersheds
Bibliographic details
Summary
The material presented in this chapter was divided into three major areas. the first part was a discussion of the statistical properties of hydrologic time series. This included event time and discrete tie, first and second order distribution properties, joint distribution properties, logn term properties wuch as the Hurst effect and the variance function, and different forms of asymmetry. The next part of the chapter was a discussion of available stochastic models. These included short memory processes wuch as and autoregressive, integrated, moving average. Long memory models such as fast fractional Gaussian noise, filtered fractional noise, broken line, and some forms fo the autregressive moving average processed were then described. Following a comparison of some of the short and long memeory models The generation of daily data by models such as the short noise process was described. Finally the disaggregation process and space time rainfall models were presented. The last part of the chapter was concerned with the subject of model selection and parameter estimation. Topics covered included incomplete data sets; properties of paramete estimates such as bias, minimum variance, and consistency; discussion of random number generators; and several different method of parameter estimation. The estimation method described included the method of moments, method of least squares, method of maximum likelihood, and the methods of Bayesian statistics.