Effects of water conditions and land use types on the spatial variability of soil fertility in Loess Plateau | 水分和土地类型对土壤肥力因子空间变异性的影响

chinois. 研究水分条件和土地利用类型对黄土区部分土壤肥力因子空间变异的影响。采用网格法结合GPS采集耕层（0~20 cm）土样364个，并用土壤养分状况系统研究法（ASI法）分析土壤部分肥力因子含量，从地统计学角度研究不同水分条件（旱地、水浇地）和土地利用类型（粮田、苹果园、葡萄园、蔬菜地）下，土壤有机质、氨态氮、有效P和有效K的含量特点和空间变异性。常规统计结果表明，水浇地土壤铵态氮、有效P和有效K平均含量比旱地分别高26.0%，74.5%，85.3%，有机质含量则比旱地低25%，不同土地利用类型之间各肥力因子的平均含量和变异系数表现出较大差异。地统计学结果表明，同一养分因子在不同水分条件下的空间变异特征并不完全一致，各肥力因子在粮田、葡萄园和苹果园均表现出强烈的空间相关性，而在蔬菜地表现出不同于其他土地利用类型的空间变异特征。在土壤肥力因子空间变异中，结构性因素占主导作用，人为因素尚无法达到破坏其原有空间格局的程度，尺度内结构性变异部分主要是由作物类型引起的。

anglais. The study explored the effect of water conditions and land-use types on the variability of soil fertility factors in Loess Plateau. This study analyzed the content characteristics and spatial variability of soil fertility factors (organic matter, NH+4-N, available P and available K) from 2 kinds of water conditions and 5 kinds of land-use types based on the data of 364 top-layer (0C20 cm). Soil samples were obtained by grid sampling method in Xinji village of Shaanxi Province. The results of traditional statistics showed that the average levels of NH+4-N, available P and available K were 26.0, 74.5 and 85.3 percent higher in irrigated than in non-irrigated soil, with organic matter 25 percent less than in non-irrigated. For the same soil fertility, its average levels and coefficient of variation (CV) in different land-use types were inconsistent. The results of Geostatistics showed that the spatial variability characteristics of soil fertility factors in different water conditions were not exactly the same, in which the structural variability comes from type of cultivation in study scale. All soil fertility factors were strongly spatially dependent in crop-land, grapery and apple orchard, because the rate of nugget to sill was less than 25%.The spatial variability characteristics of vegetable were totally different from other land-use types because of its special management measures. This spatial variability is fundamentally affected by structural factors, and human factor hasn’t changed the original space distribution.