|
|
|
Development of an interface in ArcGIS® for watershed runoff estimation
Dr A. Sarangi
Scientist (SS)
Water Technology Centre, IARI, New Delhi,
India
Email: asarangi@iari.res.in
Miss Jyoti Patil
M.Sc Student
Division of Agricultural Engineering, IARI, New Delhi-110012
Email: jyotipp2003@yahoo.com
Dr O.P. Singh
Head, Div of Agril. Engg., IARI
IARI
Email:
Rainfall and runoff processes are important part of watershed hydrologic responses. Accurate, fast and timely estimation of runoff from precipitation depths are essential not only for planning water conservation measures, recharging the ground water zones, reducing the sedimentation and flooding hazards downstream, but also for integrated watershed management leading to enhanced biomass production. The estimation of surface runoff within Geographic Information System (GIS) environment was attempted in this study by developing an interface in ArcGIS®. The interface was developed in visual basic for application (VBA), the built-in macro programming language of ArcGIS® that incorporated the concepts of surface runoff estimation using the Natural Resources Conservation Service (NRCS) Curve Number (CN) method and modified NRCS-CN methods. The modified NRCS-CN methods included the antecedent moisture conditions and the infiltration characteristics in detail, which were not considered in original NRCS-CN method. The developed interface was operated for Banha watersheds (1751ha) under Damodar Valley Corporation, Hazaribagh, Jharkhand, India to evaluate the performance of original and modified CN techniques. The input data to the interface were land use and hydrologic soil group information of the watershed besides the event based rainfall depths. The interface estimated the runoff depth for 63 rainfall events spanning over the period from the year 1993 to 1997 using the original NRCS-CN and modified methods. It was revealed that the NRCS-CN- Mod-I (with no initial abstraction) resulted in a better prediction of the runoff depth over the Banha watershed with coefficient of determination (R2) of 0.72, where as the original NRCS-CN resulted in a R2 of 0.64. Moreover, the developed interface needs to be operated on different watershed regions to ascertain the superiority of NRCS-CN modifications over the original CN method and confirm the results of the watershed taken up in present investigation.
|
|
|