Prediction of spatial distribution for some land use allometric characteristics in land use planning models with geostatistic and Geographical Information System (GIS) (Case study: Boein and Miandasht, Isfahan Province, Iran)
Although traditional census can present unbiased information about different land uses, it is spatial independent and do not present particular information about spatial distribution of studied characteristic. In this study, we used geostatistic and Geographical Information System (GIS) to estimate some different land uses allometric characteristics in Isfahan Province (Iran). Thus, samples information was surveyed considering their geographic position in the studied area. After optimizing variogram parameters, empirical variogram was prepared to investigate spatial structure of different land uses allometric characteristics. Our results confirme that spatial structure for the quantitative characteristics of different land uses has a moderate degree of spatial correlation, except for type variable that has no spatial structure. Nugget effect for variogram obtained from the quantitative characteristics of different land uses was equal to 35 to 64%. We used ordinary Kriging for preparing Kriging map and Kriging standard deviation of different land uses. Also, we used geostatistic and GIS to compare geostatistical and algebraic interpolation methods and nine different interpolation methods (Kriging, local polynomial methods, inverse distance weighted, radial basis functions, global polynomial, moving average weighted, natural neighbor, nearest neighbor and triangulation with Linear Interpolation) were investigated. Spatial distribution of different land uses quantitative characteristics were validated with ordinary Kriging and algebraic methods. Our results confirm that ordinary Kriging has more accuracy than other methods for spatial prediction of different land uses quantitative characteristics.
Key words: Geostatistic, interpolation method, land use allometric characteristics, Kriging.