A simulation of the rainfall-runoff process using artificial neural network and HEC-HMS model in forest lands

Simulation of the runoff-rainfall process in forest lands is essential for forest land management. In this research, a hydrologic modelling system (HEC-HMS) and artificial neural network (ANN) were applied to simulate the rainfall-runoff process (RRP) in forest lands of Kasilian watershed with an ar...

Full description

Saved in:
Bibliographic Details
Published in:Journal of forest science (Praha) 2021-01, Vol.67 (4), p.165-174
Main Authors: Gholami, Vahid, Khaleghi, Mohammad Reza
Format: Article
Language:English
Subjects:
Antecedent moisture
Artificial neural networks
Climatology
flood
forest lands
Forest management
Forests
Hydrologic models
Hydrology
Hydrometric stations
initial loss
kasilian watershed
Lag time
Land management
Multilayer perceptrons
Neural networks
optimization
Rainfall-runoff relationships
Runoff
Simulation
Soil conditions
Soil conservation
Soil infiltration
Soil moisture
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Simulation of the runoff-rainfall process in forest lands is essential for forest land management. In this research, a hydrologic modelling system (HEC-HMS) and artificial neural network (ANN) were applied to simulate the rainfall-runoff process (RRP) in forest lands of Kasilian watershed with an area of 68 square kilometres. The HMS model was performed using the secondary data of rainfall and discharge at the climatology and hydrometric stations, the Soil Conservation Service (SCS) for simulating a flow hydrograph, the curve number (CN) method for runoff estimation, and lag time method for flow routing. Further, a multilayer perceptron (MLP) network was used for simulating the rainfall-runoff process. HEC-HMS model was used to optimize the initial loss (IL) values in the rainfall-runoff process as an input. IL reflects the conditions of vegetation, soil infiltration, and antecedent moisture condition (AMC) in soil. Then, IL values and also incremental rainfall were applied as inputs into ANN to simulate the runoff values. The comparison of the results of simulating the RRP in two scenarios, using IL and without IL, showed that the IL parameter has a high effect in increasing the simulation performance of the rainfall-runoff process. Moreover, ANN predictions were more precise in comparison with those of the HMS model. Further, forest lands can significantly increase IL values and decrease runoff generation.
ISSN:1212-4834
1805-935X
DOI:10.17221/90/2020-JFS