Validation of local knowledge drought forecasting systems in the Limpopo River Basin in Southern Africa
Abstract
Purpose
This research paper is informed by a study to assess performance of local knowledge drought forecasts (LKDFs) in the Mzingwane catchment which is located in the Limpopo River Basin in Zimbabwe. The purpose of this paper is to validate local traditional knowledge (LTK) indicators being applied in Mzingwane catchment and verify their accuracy and reliability in drought forecasting and early warning.
Design/methodology/approach
LTK forecast data for 2012/2013 season were collected through structured questionnaires administered to 40 selected household heads and focus group discussions. Observations and key informant interviews with chiefs and the elderly (>55 years) were also used to collect additional LTK forecast data. Meteorological data on seasonal rainfall were collected from the meteorological Services Department of Zimbabwe (MSD). Two sets of comparisons were conducted namely the hind-cast comparison where the LKDF system results were evaluated against what the season turned out to be and forecast comparison where local LKDF system results were compared with downscaled meteorological forecasts.
Findings
The results showed that the majority of the LTK indicators used were accurate in forecasting weather and drought conditions when compared to the observed data of what the season turned out to be. LTK forecasts were found to be more accurate than meteorological forecast at local scale. This study has shown that the reliability of LTKs is high as demonstrated by the fact that the predicted event occurs.
Research limitations/implications
Further validation be carried out for a number of seasons, in order to standardise the LTK indicators per geographical area.
Originality/value
The research creates platform for adoption of LTKs into formal forecasting systems. The research is useful to both meteorological researchers and resource constrained communities in Mzingwane catchment.
Keywords
Acknowledgements
The authors wish to acknowledge the financial support from the Drought Early Warning and Forecasting for Africa (DEWFORA) Project for field work only. The authors would also like to thank Mr A. Ndlovu and the communities that participated in this research.
Citation
Chisadza, B., J. Tumbare, M., R. Nyabeze, W. and Nhapi, I. (2014), "Validation of local knowledge drought forecasting systems in the Limpopo River Basin in Southern Africa", Disaster Prevention and Management, Vol. 23 No. 5, pp. 551-566. https://doi.org/10.1108/DPM-02-2014-0032
Publisher
:Emerald Group Publishing Limited
Copyright © 2014, Emerald Group Publishing Limited