Abstract:
The objective of the investigation was to apply fuzzy logic technique to
drought risk identification in Buriram province.
The standard criteria were utilized including meteorological data, soil
characteristics, topographical data, and hydrological/hydro-geological data, which
influenced the occurrence of drought phenomena. Experts were requested to classify
the significant score of primary factors (criteria/weighting) and secondary factors
(factor/rating). All the above data and factors were then analyzed with fuzzy logic and
potential surface analysis (PSA). The areas were divided into three categories high,
moderate and low potential for drought risk values. Results from both methods were
subsequently compared with drought-area data surveyed by the Department of Land
Development in 2006, in order to compare the effectiveness of the fuzzy logic
technique with PSA.
The results of the fuzzy logic technique found that the high potential areas for
drought were approximately about 8,780.44 km2 or 85.14%. The moderate potential
areas for drought were approximately about 1,532.45 km2 or 14.86 %. And the PSA
results for drought risk identification found that the high potential areas for drought
were approximately about 1,664.93 km2 or 16.14 %. The moderate potential areas for
drought were approximately about 8,097.45 km2 or 78.52%. The low potential areas
for drought were approximately about 550.51 km2 or 5.34 % of the total area.
The drought area identified by Department of Land Development was then
compare with the output of drought risk identification area from both mentioned
procedures and the accuracy, assessed by the Kappa index method. The fuzzy logic
technique revealed the percentages of accuracy, which is 0.75. At the same time, PSA
demonstrated the percentage of accuracy, which is 0.47. The accuracy of the fuzzy
logic technique was superior to PSA.