科研之友
微信
新浪微博
Facebook
分享链接
摘要
Riparian ecosystems are threatened worldwide, necessitating conservation strategies. Numerical models tailored for specific geographic areas have been developed as management support tools. However, few models are suitable for multiple river conditions, and developing these models or evaluating their suitability has become an emerging topic. The dynamic riparian vegetation model (DRIPVEM) is a numerical model developed for steep and gravelly Japanese rivers, where it has been successfully tested. Our objective was to assess the performance of DRIPVEM in a river with a low slope and fine sediment, similar to the characteristics of continental rivers. A reach of the Hii River was selected for testing the model's ability to predict the distribution of Salix spp. (willow) and herbs, as well as herb biomass and tree age. The model was calibrated based on field investigations of a selected river section. Simulation of the studied reach was carried out for the past five decades, depending on data availability. Non-parametric tests were used to compare the simulated and observed results. The simulated and observed vegetation distribution maps agreed fairly well and the sensitivity of the model for simulation of trees, herbs and bare areas was greater than 0.6. The kappa coefficients of agreement values were 0.48 and 0.49, indicating fair agreement. Moreover, the simulated biomass and tree age agreed well with observation. We conclude that the DRIPVEM simulated the observed conditions in the Hii River well, indicating that the model is applicable to rivers characterized by low slope and fine sediment grain size.
- 出版日期2017-3
