Improving a drought early warning model for an arid region using a soil-moisture index

Vijendra K. Boken

Research output: Contribution to journalArticlepeer-review

25 Scopus citations

Abstract

This study modifies a drought early warning model for Jodhpur district of Rajasthan State in India. The model had employed only two variables derived from the daily rainfall data and estimated pearl millet yield in order to issue a drought early warning. In this study, the model is modified by including an additional variable derived from a soil-moisture index. The modified model explained up to 77.3 percent of the yield variation. When tested, the mean absolute percent error in the estimated yield was 13.7 percent in the case of the modified model as opposed to 18.5% in the case of the existing model. The soil-moisture index and other variables derived from the rainfall data could be potential candidates for developing drought early warning models for other arid regions.

Original languageEnglish (US)
Pages (from-to)402-408
Number of pages7
JournalApplied Geography
Volume29
Issue number3
DOIs
StatePublished - Jul 2009

Keywords

  • Drought
  • Pearl millet
  • Soil-moisture index
  • Yield modeling

ASJC Scopus subject areas

  • Forestry
  • Geography, Planning and Development
  • General Environmental Science
  • Tourism, Leisure and Hospitality Management

Fingerprint

Dive into the research topics of 'Improving a drought early warning model for an arid region using a soil-moisture index'. Together they form a unique fingerprint.

Cite this