The meteorological and agricultural assessment techniques that have be
en developed in South Africa are discussed with reference to their str
engths and weaknesses, application and suggested future developments.
Future challenges in drought assessment as influenced by democratizati
on are also considered. We conclude that meteorological methods have s
everal inherent deficiencies which render them risky for the optimum a
llocation of drought relief. Agricultural drought assessment is sugges
ted to be more preferable. Current agricultural drought models should
nevertheless be improved to deal more successfully with the dynamic na
ture of rangeland. Evidence indicates that no absolute objective bioph
ysical criteria for the quantification of the onset and the end of a d
rought have been identified and that all current criteria have an inhe
rent subjectivity. Choosing appropriate interpolation techniques for m
apping drought extent is a cause for concern, and there is scope for f
uture research. Remote sensing techniques, integrated with crop modell
ing techniques, are options to pursue further. We finally suggest that
in a democratic South Africa, more attention should be given to a mul
ti-disciplinary approach where impacts, other than meteorological and
agricultural, on previously disadvantaged black farmers ave also asses
sed. (C) 1998 Elsevier Science Ltd. All rights reserved.