Title: Bootstrap-based artificial neural network analysis for estimation of daily sediment yield from a small agricultural watershed

Authors: Gurjeet Singh; Rabindra K. Panda

Addresses: School of Earth, Ocean and Climate Sciences, Indian Institute of Technology, Bhubaneswar, India ' School of Earth, Ocean and Climate Sciences, Indian Institute of Technology, Bhubaneswar, India

Abstract: Accurate estimation of sediment yield from watershed and its sub-watersheds is a prerequisite for effective watershed management. Reported study was undertaken in a small agricultural watershed namely Kapgari in Eastern India for estimation of daily sediment yield. On the basis of drainage network and land topography, the watershed was subdivided into three sub-watersheds. Bootstrap technique was used to develop unbiased artificial neural network (ANN) models to estimate the daily sediment yield with limited quantum of continuously monitored sediment yield data from the watershed. Bootstrap-based artificial neural network (BANN) were developed using only major weather variables such as rainfall and temperature for estimation of daily sediment yield. Results illustrate that the highest coefficient of simulation efficiency values of 0.887, 0.869, 0.904 and 0.898 for estimation of daily sediment yield from watershed and its sub-watersheds were observed by addition of one day lag rainfall and present day maximum and minimum temperature with present day rainfall.

Keywords: artificial neural networks; ANNs; bootstrap; daily sediment yield; lag rainfall; small watersheds; agricultural watersheds; sub-watersheds; watershed management; India; drainage network; land topography; simulation.

DOI: 10.1504/IJHST.2015.072634

International Journal of Hydrology Science and Technology, 2015 Vol.5 No.4, pp.333 - 348

Received: 21 Jan 2015
Accepted: 24 Jun 2015
Published online: 22 Oct 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article