EUR J SOIL SCI | 西北荒漠草原不同土地利用方式下土壤磷组分的变化
Changes in soil phosphorus fractions under different land uses in desert grasslands in northwestern China
论文类型：研究型论文
作者单位：1State Key Laboratory of Grassland Agro-ecosystems, Key Laboratory of Grassland Livestock Industry Innovation, Ministry of Agriculture and Rural Affairs, Engineering Research Center of Grassland Industry, Ministry of Education, College of Pastoral Agriculture Science and Technology, Lanzhou University, Lanzhou, China
2Laboratory of Ecology and Agriculture, Northwest Institute of Eco-Environment and Resources, Chinese Academy of Sciences, Lanzhou, China
引用文本：Wang, Z. R., Li, F. R., Liu, L. L., & Yang, K. (2022). Changes in soil phosphorus fractions under different land uses in desert grasslands in northwestern China. European Journal of Soil Science, e13255.
摘要
人类通过将原生植被转化为农业用地和非农用地，影响荒漠生态系统的结构和功能。然而，荒漠草地上不同的土地利用如何通过创造不同的土壤理化和微生物特性来影响土壤中具有不同稳定性的磷（P）组分，目前还知之甚少。为了解决这个问题，在中国西北干旱区进行了一项实地研究，选择了一个包含天然荒漠草地的区域和三个相邻的不同管理方式的荒漠草地利用转化：即28年的雨养灌木-梭梭（Haloxylon ammodendron）人工林、35年的灌溉乔木-杨树（Populus gansuensis）人工林，以及33-40年的灌溉与施肥农田。本研究收集了4种土地利用类型的土壤无机磷和有机磷（Pi和Po）（不稳定P：即树脂和NaHCO3可提取P；中度不稳定P：即NaOH-可提取P；顽固性P：即HCl-可提取P和残留P)，并利用土壤变量作为预测因子。使用增强回归树分析（boosted regression tree analysis）来评估个体预测因子对P组分变化的相对贡献。结果表明，与天然荒漠草地相比，（1）农田全磷组分（77%–418%）中树脂P增幅最大；（2）除树脂P外，其余各磷组分均在人工林中（52%–367%），不稳定Po含量增幅最大；（3）灌丛人工林NaHCO3-P和HCl-Pi含量(14%–134%)均高于天然荒漠草地，其中不稳定Pi含量增幅最大。土壤性质对不同P组分的变化具有显著的控制作用，可解释51%–93%的磷组分变化，但各P组分间的主要影响因素存在差异。研究结果表明，不同土地利用方式对荒漠草地土壤中磷的组分组成和有效性存在显著差异，土壤磷循环可通过改变土地利用方式和管理水平来调节。
2. 试验方法
2.1 试验地点
Figure 1 The study area in the western Gansu Province in northwestern China. The four studied land-use types included (a) natural desert grassland (NG), (b) shrub plantation (SP), (c) poplar tree plantation (TP), and (d) cropland (CL)
图1研究区位于中国甘肃省西部。研究的4种土地利用类型包括（a）天然荒漠草地（NG）、（b）灌木人工林（SP）、（c）杨树人工林（TP）和（d）农田（CL）。
2.3. 土壤取样
2017年八月中旬。
2.4. 土壤理化性质测定
土壤含水量，pH，有机碳，全氮，全磷，微生物量碳，碱性磷酸酶。
2.5. 土壤磷组分测定
(1) the most available P (resin-Pi);
(2) labile P (Pi and Po extracted by NaHCO3; Olsen P);
(3) moderately labile P (Fe and Al associated Pi and Po extracted by NaOH);
(4) poorly soluble P (Ca associated Pi and Po extracted by dHCl and cHCl);
(5) the most resistant and insoluble P (residual-P, which was calculated as the difference between soil total P and the sum of the first eight extracted P fractions).
(6) dHCl-Pi and cHCl-Pi into a single HCl-Pi fraction.
2.6.1. 数据分析
增强回归树（boosted regression trees，BRT）分析，以量化单个预测变量的相对重要性（RI）来解释每个P组分的方差。
3. 结果
3.1. 土壤P组分绝对和相对浓度变化
Table 1 Changes in absolute concentrations (mg kg-1) of measured soil inorganic and organic phosphorus (Pi and Po) fractions of decreasing lability (labile, moderately labile, and recalcitrant P) after conversion of natural desert grassland to rainfed shrub (Haloxylon ammodendron) plantation, irrigated poplar tree (Populus gansuensis) plantation, and irrigated and fertilized cropland
3.2. 土壤理化和微生物特性的变化
TABLE 2 Changes in the proportions (%) of measured soil inorganic and organic phosphorus (Pi and Po) fractions in the soil total P content after conversion of natural desert grassland to rainfed shrub (Haloxylon ammodendron) plantation, irrigated poplar tree (Populus gansuensis) plantation, and irrigated and fertilized cropland. Soil Pi and Po fractions were grouped into three pools of decreasing lability (labile, moderately labile, and recalcitrant P)
TABLE 3 Changes in measured soil physicochemical and biological properties as predictors after conversion of natural desert grassland to rainfed shrub (Haloxylon ammodendron) plantation, irrigated poplar tree (Populus gansuensis) plantation, and irrigated and fertilized cropland
3.3. 土壤磷组分和预测变量的关系
FIGURE 2 RDA (redundancy analysis) two-dimensional ordination diagram of the first two axes showing sites (land use types), soil P fractions, and soil predictor variables. Circles represent natural desert grasslands, rhombuses represent shrub plantations, triangles represent tree plantations, and asterisks represent croplands. Soil predictor variables are soil pH (pH), soil bulk density (SBD), soil water content (SWC), soil organic carbon (SOC), soil total nitrogen (TN), soil N:P ratio (N:P), soil C:P ratio (C:P), soil microbial biomass carbon (MBC), and soil alkaline phosphatase activity (ALP). Vector overlays present the direction (arrow) and strength (length) corresponding to the variance that can be explained by the predictor variables
Table 4 The results of linear regression analyses for quantifying the relationships between changes in concentrations of eight measured soil inorganic and organic phosphorus (Pi and Po) fractions and changes in soil predictor variables after conversion of natural desert grassland to differently managed land-use systems (shrub plantation, tree plantation, and cropland)
3.4. 预测变量在调节土壤P动态的相对重要性
FIGURE 3 The relative influences of the predictor variables on the patterns of soil total P and eight measured soil inorganic and organic phosphorus (Pi and Po) fractions. The predictor variables included soil pH (pH), soil bulk density (SBD), soil water content (SWC), soil organic carbon (SOC), soil total nitrogen (TN), soil N:P ratio (N:P), soil C:P ratio (C:P), soil microbial biomass carbon (MBC), and soil alkaline phosphatase activity (ALP). R2 denotes the proportion of the variation in each P fraction explained by the BRT models. The statistical significance of the BRT model is tested using the Monte Carlo permutation test (*p < 0.05, **p < 0.01, ***p < 0.001)
4. 讨论
4.1. 荒漠土壤P组分对不同土地利用方式的差异响应
4.2. 不同P组分对土地利用变化响应的关键预测因子
5. 结论
这项研究解决了关于三种不同的人为利用石灰性荒漠土壤对土壤P组分的去向以及干旱气候下土壤P变化的主要原因的长期（>28年）后果的关键认识差距。本研究证明，三种不同的沙漠土壤利用和管理策略导致了不同的磷获取、循环和积累模式，导致三种土地利用系统之间土壤P组分组成和有效性存在显著差异。自然荒漠草地向灌溉和施肥农田的转变导致表层土壤全磷、不稳定和中度不稳定磷含量的增加最大，其次是自然荒漠草地向灌溉和不施肥的乔木种植园的转变，然后是不灌溉和不施肥的灌木种植园。8个磷组分对土地利用变化的响应主要受土壤水分、pH、有机碳、全氮、微生物量C和碱性磷活性的调节，但各磷组分的主导影响因素存在差异。
土地利用变化对特定P组分的影响可以通过改变相应P组分变化的关键决定因素的值来调节。这些发现有几个重要的启示。土地利用变化对特定P组分的影响可以通过改变相应P组分变化的关键决定因素的值来调节。这些发现有几个重要的启示。首先，本研究的结果有助于更准确地理解人为土地利用和管理变化在调节土壤磷循环和生态系统磷有效性方面所起的关键作用。其次，本研究的结果强调了在未来的磷循环模型中纳入不同磷组分的土地利用变化驱动的响应以及不同的磷组分-土壤环境关系的重要性，以提高模型预测的可靠性。最后，也是最重要的是，本研究的研究结果为通过调整土地利用类型和管理强度来管理旱地石灰性土壤中的磷循环提供了重要的见解。
