عنوان مقاله [English]
Land suitability evaluation and classification is one of the most useful efforts in land resources management and utilization optimization and the base of such studies are soil maps. In conventional land suitability and soil studies, the information of one studied point (profile) has been generalized to the whole of a map unit. Thus, are the spatial variability of soil properties disregarded and only a unique management is adopted for the whole of that unit of soil map. Such approach causes the efficiency of land evaluation studies decrease, due to the impurity and variabilities in soil map units. Fuzzy approach in land suitability classification enables to interpret the gradual variations of soil spatial properties and so to increase the purity of soil map units. Therefore, applying of a unique managerial method for each unit of soil map will more be rational. Soil maps as a conventional sources of land suitability analysis information, have encountered the problems, mainly due to the generalization of a representative pedon results to all of a soil unit area with no taking consideration of temporal and spatial variabilities. Thus, the present study attempts to introduce a method in order to take in account the uncertainty of land suitability variations and consequently improve of soil units purity by using of fuzzy set theory approach.
2-Materials and Methods
The area under investigation has a size of approximately 1300 ha. It is located between 32Â°17â and 32Â° 20â N, and 51Â° 3â and 51Â° 5â 30ââ E in the province of Chaharmahal-Va-Bakhtiari, Central Iran. As well as representative pedon, another 31 pedons were also excavated and investigated in the soil map unit. Soil land suitability classes of all pedons were determined using fuzzy approach and conventional method suggested by Sys et al. (1991). Then, comparison of the results were done for both methods and also compared to the results of representative pedon. The later allows understanding if the land suitability class of representative pedon could show the reality of the soils in the map units or not.
3-Results and Discussion
In conventional (traditional) method, the unit P was distinguished as a consociation map unit with the suitability class of S2. According to the results of all pedons in this unit, only 16.1% of the pedons have S2 suitability class. This means that generalization of the results of the representative pedon to whole the unit could be mis-leading considerably. Considering fuzzy logic approach, the amount of land index for representative pedon is 53.1% which is close to the border of S2 and S3 classes of conventional method. Using fuzzy membership degrees, the suitability class of representative pedon could be S3 not S2. Using Î±-cut of 85%, the purity of unit P is 34.4 and 96.9 for S2 and S3 classes, respectively. This indicates S3 class is more reasonable to use for management purposes.
In conventional soil surveys, the results of representative pedon are used to management purposes because analyzing all pedons in a map unit is time and cost-consuming. The results suggest that generalization of the results from representative pedon to the whole land unit can be mis-leading considerably. So, finding the methods which is more accurate and closer to reality and also use representative pedon data could be more useful. Fuzzy approach is a good option to solve the problems arises from the biased results when a suitability class lies close to the border of two different classes.