Answers are grounded in this paper and cite their sources.
In the present study, Artificial Neural Networks (ANNs) and Genetic Programming (GP) soft-computing models were established to estimate streamflow in Tel River Basin, Odisha, India. In most of the previous time-series soft computing models/studies, the time delay or basin lag has been considered without any physical basis for streamflow estimation/forecasting. Considering these limitations, the 02 isochrones (zones) were delineated up to Kesinga Station based on its topography for incorporating the time delay phenomenon. The monsoon rainfall of the period of 1998–2011 (14 years) in each zone was estimated and delayed as per their respective time-delay zone and fed into both the models for discharge estimation. It was observed that after incorporating the time delay factor, the accuracy of streamflow estimation increased for both the models. However, the ANNs could not map the peak discharge and the model efficiency was less as compared to the GP model. The best GP model achieved an efficiency of 74%, where ANNs could reach up to only 59%. After establishing the soft computing based discharge estimation model for the basin, the flood inundation between Kesinga and Kantamal stations was studied using the hydrodynamic model for the estimated peak discharge.
No takes yet. Share an insight, caveat, or question.
Garg et al. (2019) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: