Fuzzy-GIS development of land evaluation system for agricultural production in North West Libya
Froja, Nagib Mohammed
MetadataShow full item record
The continuing deterioration of land and water resources occurring in several regions of the world is partly as a result of the mismatch between land suitability or capability and land use. Failure to achieve a perfect match between land capability and use can be particularly problematic for agricultural production because cultivating the wrong crops on wrong soils can only result in poor yields and its associated financial and other losses. There is therefore, a pressing need for effective land evaluation through better matching of land characteristics with land use to achieve optimal utilisation of available land resources for sustainable agricultural production. As far as agriculture is concerned such an exercise will result in defining which part of an area is suitable for particular crops, based on the available land resources and other production inputs, and which parts are better left for other uses. In this study, a land evaluation system for predicting the physical suitability of land for key crops, namely Wheat, Barley and Olive in the north west of Libya was developed based on matching land use requirement for these crops with the available land resources in the area. It involved a modelling strategy based on Boolean and Fuzzy logic sets, implemented within a Geographic Information System (GIS) environment. While the Boolean method assumes that the attributes of a given soil type are known with certainty and the boundaries between soil types are clearly defined, Fuzzy logic can be used to accommodate uncertainties in the available knowledge on these attributes through the use of membership functions. The GIS-based models developed comprise four layers; namely, soil, climate, slope and erosion hazard all of which have been shown directly influence land suitability for agricultural production. This resulted in the classification of the soil into 4 suitability classes, i.e. high suitability, moderate suitability, marginal suitability and not suitable. The results show that for Barley for example 52% of the soil in the north western Libya is highly suitable using Fuzzy approach while the corresponding figure for the Boolean is 62%. The two approaches were compared on cell by cell basis using map agreement. The comparison shows that there were reasonable agreements in evaluations by the two approaches for barley, wheat and olive of 51%, 46% and 56% respectively.