TY - JOUR
T1 - Modeling of soil erodibility in the Khoor and Biabanak region, using remote sensing indices
TT - مدلسازی فرسایشپذیری خاک در منطقۀ خور و بیابانک با استفاده از شاخصهای دورسنجی
JF - deej
JO - deej
VL - 5
IS - 13
UR - http://deej.kashanu.ac.ir/article-1-465-en.html
Y1 - 2017
SP - 67
EP - 80
KW - soil erodibility
KW - K factor
KW - SEI index
KW - regression analysis
KW - Khoor and Biabanak.
N2 - Soil erosion by water and wind processes are carried out widely. So one of the best ways to estimate soil loss and land degradation is regional erodibility model. The aim of this research is codification regional erodibility model by statistical methods, the relation survey in Khoor and Biabanak between physical properties of soils and remote sensing indices. Sampling Method field was linear transect method and picked up 33 soil samples from depths of 0 to 50 cm. By transferring the samples to the laboratory, were calculated the elements of sand, clay, silt, organic matter and carbon, and then measurement erodibility values using the K-factor and SEI index. Then, was taken OLI sensor image accordance time sampling from USGS website. And were extracted the indices by enforcing functions the amount of Clay, NDVI, GOSAVI, SAVI, TSAVI, NDSI, SCI and GSAVI. Finally, were calculated using Pearson correlation between physical elements and soil erodibility with Remote sensing indices. At the end of for modeling SEI and K, through simple and multiple regression attempts to gauge the relationship between these parameters and selected best models from among with higher preference value. The results of modeling SEI represents the maximum linear correlation with the indices SAVI, NDVI and TSAVI respectively, with coefficient of 0.69, 0.66 and 0.63, and K factor linear correlation with the indices NDSI, SAVI and SCI, respectively, with a coefficient of determination 0.63 0.61 and 0.58 is at level 99%. Multiple regression analysis also showed a high correlation with group consensus SEI, K and Remote Sensing indices, respectively, with a coefficient of determination 0.517 and 0.564 and estimated error 0.0031144 and 0.0092369 at level 99%. Therefore, can estimate the amount of soil erodibility by remote sensing indices and univariate and multivariate statistical models. Which enables fast and precise estimation of SEI and K values in the Khoor and Biabanak region.
M3
ER -