| Northwest arid region of china is the most fragile and vulnerable area in the world,which is typical of arid climate,scare precipitation,and soil moisture is under negative balance condition in this region.Water distribution heterogeneity and shortage will become more severe under humanity activities and climate changes in the further,and will restrict the sustainable development of society and economy and eco-environmental construction.However,there were knowledge gap on the regional-scale spatial-temporal variability of soil moisture and the dried soil layer(DSL)on the northwest arid area of China.Therefore,based on traditional sampling and GLDAS Noah025 datasets,we studied the spatial-temporal variability of soil moisture and the intensity of DSL using traditional statistics,geo-statistics,Manner-Kendall analysis,redundancy analysis and temporal-stability methods.And we also estimated the soil hydraulic parameters(field capacity,FC,and permanent wilting point,PWP)by state-space models.The main results were as follows:1.Spatial variability of soil moisture.The soil mass water content(SMWC)and soil relative extractable water(REW)increased as the depth increased.SMWC were7.76%,7.38%,7.67%and 8.15%in the root-zone(0-100 cm),0-10 cm,10-40 cm and40-100 cm soil layers,respectively.REW were 14.14%,-1.64%,13.13%and 14.45%in the root-zone,0-10 cm,10-40 cm and 40-100 cm soil layers,respectively.SMWC was significantly decreased in the order of cropland>grassland>forestland(p<0.05).Bare-land had the smallest SMWC,but was not significantly different with the forestland(p>0.05).REW of the cropland was significantly higher than the other land uses(p<0.05).Traditional statistics indicated that SMWC and REW were,respectively,moderate and strong variations on the northwest arid area of China.Semivariograms indicated that SMWC were moderate dependence with nugget ratios of 47.8~50.0%and large ranges of 649.0~857.0 km.REW was pure nugget semivarigram on the northwest arid region.2.The driving factors of spatial variability of soil moisture.Both large-scale climatic and small-scale terrain-related(soil properties and other local factors)factors had significant influences on the spatial variability of soil moisture.The contributions were 23.3~30.7%,14.1~26.2%and 9.2~16.9%of the soil properties,local factors and climatic factors,respectively.The interactions between soil properties,climatic factors and local factors were increased with soil depths increasing.The interaction between climatic and local factors was weak in depth.However,soil properties rather than climatic factors were the driving factors which influenced the spatial variability of soil moisture in the hyper-arid area(Xinjiang).3.Preliminary investigation of DSL.The DSL was distributed in 68.75%(55 in80)of northwest arid areas of China.The probabilities of occurrence of DSL in farmland,grassland and forestland was 34.62%,84.85%and 85.71%,respectively.The mean soil water content within DSL(DSL-SWC),initial formation depth(DSL-FD)and DSL thickness(DSL-T)were 5.99%,1.61 m and 2.74 m,respectively.The DSL of grassland had the highest development intensity,and its thickness is 3.44 m,which is significantly higher than that of cropland(1.77 m)and forestland(2.13 m).The development of DSL was severer in the southeastern Hexi Corridor and the western Xinjiang region.Redundancy analysis indicated that climatic factors,soil properties and local factors would explain 37.0%,27.1%and 33.2%variations of DSL states(p<0.01),respectively.Climatic factor was the driving factors impacting the DSL,which could explain 8.7%variations of DSL independently(p<0.05).4.GLDAS Noah025 data sets had significant relationship with the surface soil moisture(R~2=0.238,p<0.001),but had poor correlations with the soil moisture below10 cm depth.The seasonal distributions of soil moisture were obvious.The soil moisture in summer was higher than that of other seasons in Hexi Corridor and southern Xinjiang,while the wetter season was spring in northern Xinjiang.Soil moisture in depth of 10-100 cm had significant increasing tendency during 2000 to 2017(p<0.05),and the change ratios increased with increasing soil depth.The surface soil moisture was significantly increased on the inter-annual scale in the Hexi Corridor(p<0.05).5.Traditional statistics indicated that soil moisture were weak and moderate variations on the seasonal and interannual scales,respectively.The proportions of weak and moderate variation of root-zone soil moisture were 32.81%and 67.19%,respectively,on the seasonal scale,and 51.72%regions showed weak variation on the inter-annual scale.The standard deviation of relative difference increased with increased soil water content.All of climatic,soil and topography characteristics had influences on the seasonal and inter-annual dynamics of soil moisture.The temporal stability locations of soil moisture were mainly distributed in Mulei and Balikun in Xinjiang.6.The FC and PWP in 0-20 cm soil layer were 12.7±6.0%and 7.5±3.4%in the Hexi Corridor.Both of these two parameters were moderate variations with variation coefficients of 45.0~47.5%.The optimal state-space models could explain 99.9%and99.7%variations of soil moisture,which had better performance than the multiple linear regression with R~2 of 0.599~0.805.Soil texture,land use and bulk density had significant impacts on FC and PWP.This study focuses on the spatial-temporal dynamics and the driving factors of soil moisture.Soil moisture has obvious seasonal and inter-annual dynamic patterns,and both climatic and terrain-related factors have significant influences on the spatial-temporal variability of soil moisture.For the first time this study reveals DSL developments of this region,and grass has the severest development of DSL.This study collected large scale land surface characteristics and established prediction models of soil hydraulic parameters based on the state-space models.The research is beneficial to crop production,irrigation regime,water resources regulation and eco-environment construction. |
PDF Full Text Request