Authors: Desmond Manatsa Geoffrey Mukwada Emmanuel Siziba Tafadzwa Chinyanganya
Publish Date: 2010/02/19
Volume: 102, Issue: 3-4, Pages: 287-305
Abstract
This study focuses on a framework of methodologies used for analyzing the frequency and spatiotemporal characteristics of agricultural droughts in Zimbabwe from a vulnerability context By employing an empirical orthogonal analysis method the study revealed that relatively strong spatial and temporal station drought relationships prevail making the drought spatiotemporal characteristics of the country to be considered highly homogeneous Thus agricultural droughts were characterized temporarily using the Standardized Precipitation Index derived from rainfall data for the longer but sparse data period from 1901 to 2004 At the same time higher spatial density analysis was achieved from shorter but denser database for the period 1941 to 1999 The results indicated that drought is a natural climatic feature of the region and occurs from time to time in defined periods However severe and extreme droughts tend to concentrate near the end of the time series suggesting that during the earlier period of the twentieth century droughts have been smaller or less pervasive The extreme droughts appear to inherit the coincidence of both very high values of spatial extent and intensity in a single event This offers a possible explanation to why extreme droughts in Zimbabwe usually have dire consequences on agriculture and the national economy By showing that the related national drought impacts on staple maize food production can be estimated this study has demonstrated that it is possible to anticipate future drought hazard impacts and predict periods of food insecurity As far as the forecasting of agricultural droughts is concerned the recently discovered Indian Ocean dipole/zonal mode seems to perform better than the traditional El Niño–Southern Oscillation as a potential drought predictor during the twentieth centuryThe Zimbabwe Meteorological Services is greatly thanked for providing the station rainfall data Bindura and Cape Town Universities are both thanked for providing facilities and partially financing the writing of this paper The provision of both ENSO and IODZM SST data by JAMSTEC is also greatly appreciated This does not leave out the two reviewers whose recommendations made this paper to make it in this international journal
Keywords: