Estimation of soil water storage using temporal stability in four land uses over 10 years on the Loess Plateau, China
Estimation of soil water storage using temporal stability in four land uses over 10 years on the Loess Plateau, China
复制标题
利用中国黄土高原四种土地利用 10 年来时间稳定性估算土壤储水量
DOI：
10.1016/j.jhydrol.2014.06.003
复制
发表时间：
2014-09-19
影响因子：
6.4
通讯作者：
Shao, Ming'an
中科院分区：
文献类型：
--
作者：
Liu, Bingxia;Shao, Ming'an
了解黄土高原土壤水分在时间上的稳定性，对于优化监测策略和预测土壤水分状况具有重要意义。本研究测试并验证了估计多年平均SWS的可行性SWS在选定的位置。利用中子探针在农田、草地、休闲地和灌丛4种土地利用类型各11个样点采集了0-1、1- 2、2 -3和3-4 m土层的土壤水样品。在2004年7月至2005年12月的校准期间，在20次采样中选择了各层的最具时间稳定性的位置和每种土地用途的中坡位置。使用2006年1月至2013年10月的验证数据集来检验平均SWS估计值保持有效的时间长度。SL和GL的SWS随植物生长而减少，SL和GL的时间变化大于FL和CL。SWS的时间稳定性很高，10年以上的秩相关性超过显著性阈值（a = 0.05）。土壤水分的时间稳定性大小顺序为CL > FL > GL > SL，SL和GL土壤水分的时间稳定性随观测时间的延长而降低，表现为各土层的平均斯皮尔曼相关系数均较低。从校准期选择的MTSL可以准确地估计不同层的平均SWS下的四种土地利用，估计误差小于10%，在八年。研究证实，在每种土地利用的中坡处的单个位置可以进行采样，以减少所需的样本数量，节省时间和劳动力，同时保持多年的预测精度高。(C)2014由Elsevier B. V.出版
An understanding of the temporal stability of soil water storage (SWS) in deep soil profiles is critical to optimize monitoring strategies and to predict the status of soil water on the Loess Plateau. This study tested and validated the feasibility of estimating mean SWS over multiple years by the SWSs at selected locations. The SWSs in 0-1, 1-2,2-3, and 3-4 m layers were collected using neutron probes at 11 sites in each of four land-use types: cropland (CL), grassland (GL), fallow land (FL), and shrubland (SL). The most time-stable locations (MTSLs) for the various layers and the location at mid-slope for each land use were selected on 20 sampling occasions during a calibration period from July 2004 to December 2005. A validation data sets from January 2006 to October 2013 was used to test the length of time the estimates of mean SWS remained valid. The SWSs in SL and GL decreased with plant growth, and the temporal variations were larger in SL and GL than in FL and CL The temporal stability of the SWSs was high for all soil layers in four land uses, with the rank correlations over the threshold of significance (a = 0.05) over 10 years. The degree of temporal stability of SWSs was ranked as CL > FL > GL > SL, and the temporal stability of SWSs in SL and GL decreased with increasing lengths of observation period, as indicated by lower mean Spearman's correlations for all soil layers. The MTSLs selected from the calibration period could accurately estimate mean SWSs for diverse layers under four land uses with estimation errors less than 10% over eight years. The study verified that a single location at mid-slope of each land use could be sampled in order to reduce the required number of samples and save time and labor while maintaining a high accuracy of prediction over multiple years. (C) 2014 Published by Elsevier B.V.