Exploring the Variability in Suspended Sediment Yield using BASINS-HSPF and Probabilistic Modeling: Implications for Land Use Planning

Abstract

The purpose of this study is to examine the use of Monte Carlo simulation method together with BASINS-HSPF, a deterministic water quality simulation model to investigate the variability of sediment yield from various land cover types in the East Fork Little Miami River watershed, Ohio, UISA. The study has two main objectives: first, to obtain interval estimates of suspended sediment yield from urban, agricultural and forest land; and second, to generate probability density and cumulative distribution functions for those estimates. As a result, we will know not only the range of possible values of sediment yield in the watershed but also the probability with which those values are likely to occur. BASINS (Better Assessment Science Integrating Point and Non-Point Sources), a modeling system which provides a Window-based interface for the Hydrologic Simulation Program – Fortran (HSPF), is used in the analysis to simulate streamflow and suspended sediment yield. The daily values of suspended sediment yield from various land cover types generated by BASINS-HSPF are used as input to the Monte Carlo simulation technique that generates probability density and cumulative distribution functions. The distributions are used to draw conclusions about the uncertainty in the hydrologic model predictions in the form of confidence intervals for the predicted sediment yield. The study found that the width of the confidence interval has a critical importance for the evaluation of the model results.

Keywords: BASINS, HSPF, hydrologic response units, Monte Carlo simulation, probability distributions, sediment yield, streamflow