Triple diagram models for prediction of suspended solid concentration in Lake Okeechobee, Florida

Summary: Lake Okeechobee in Florida is a major component of the greater Everglades hydrologic system and provides a number of valuable uses to society and nature, such as water supply, navigation, wildlife habitat, and fishery. Suspended solid concentration (SSC) affects directly the lake’s conditions for some of these applications. Therefore, accurate prediction of SSC can enhance the management of the water quality and the long-term protection of Lake Okeechobee. Extensive data, including wind speed, flow velocity, flow direction and SSC, have been collected. Models for predicting suspended solid concentration based on 10 different scenarios are developed using these measurements. Data are divided into two groups as training and testing for the construction of the models. SSC is predicted by the Kriging interpolation technique. Criterions of mean relative error, root mean squared error and coefficient of efficiency (CE) are used to determine the prediction errors of the developed models. In general, mean relative error is below 7% and coefficient of efficiency stays above 0.92 for the models presented. Graphs, results, and interpretations are given in detail in this paper. [Copyright &y& Elsevier]