Comparative analysis on spatial variability of soil moisture under different land use types in orchard

Spatial variability of soil moisture under different land use types of orchard in Weibei rainfed highland of China was investigated by means of geostatistics. Classical statistical analysis showed that the average soil moisture content of clean tillage, grass land and straw land ranged from 9.36 to 20.86, 8.78 to 23.37 and 10.84 to 20.49, respectively. In 0-40 cm soil layer, 10% < CV% < 100%, which belongs to moderate variability. Depth in 120-160 cm and 160-200 cm, CV <= 10%, which belong to weak variability. And with the increasing of depth, coefficient of variation decreased. S-W test values pointed out that the distribution of soil moisture content accorded with normal distribution. The results of variation function analysis indicated that the ranges of grass land were larger than clean tillage. Spatial autocorrelation in surface soil was weaker than that of the bottom soil. The reason was that surface soil was more easily influenced by exterior factors. Fractal dimension value decreased with the increasing of soil depth. While the average content of soil moisture showed the opposite trend. According to kriging interpolation, the high-value areas of clean tillage mainly concentrated in (13.4, 27.8), (28.0, 32.5) and (27.0, 10.4), and low-value areas were in (13.8, 6.6). However, the high-value of grass areas mainly focused on around (11.5, 11.5), and low-value were in (11.2, 11.2) and (25.5, 25.5). The change of the other areas' soil moisture were not obvious in a wide range. It has a good consistency. Therefore, in the dry season, it was advisable to choose straw mulching, and reduce grass cultivation, to achieve the goal of improving soil moisture content. (C) 2016 Elsevier B.V. All rights reserved.