Spatio-temporal Pattern Of Soil Moisture Along A Slope Farmland In Red Soil Hilly Region

Red soil is an important land resource in southern China,which is mainly located in a warm and humid subtropical monsoon climate.Although the precipitation in these areas is abundant,the seasonal distribution was seriously uneven.Meanwhile,agro-ecological system in red soil region faces many environmental problems,(e.g.,floods,seasonal drought,and soil erosion)which restricted the sustainable development of regional agriculture.Therefore,the main objectives of this dissertation are:(1)to clarify the spatial distribution of soil particle composition in both horizontal and vertical direction under natural conditions;(2)to obtain the spatio-temporal variability of soil moisture and its control factors;(3)to analyze the temporal stability of soil moisture at the hillslope scale;(4)to discuss the prediction method of soil moisture on the basis of temporal stability;(5)to obtain the convenient method of predicting soil moisture at high space-time resolution by choosing two kinds of typical local vegetation types(Peanut and Citrus orchard)along an agricultural hillslope in the red soil hilly region of Jiangxi province.The results can provide a scientific data for the management of soil water and regional agriculture in red soil hilly region.The results were as follows:(1)The sand,silt and clay contents in red soil hilly region were affected by slope position,soil depth and vegetation types.In the whole slope,sand,silt and clay contents all exhibited moderate spatial heterogeneity,with the coefficients of variation changing between 17.6%and 23.2%,10.7%and 18.8%,13.5%and 17.0%,respectively.Due to the deposition of coarse particles,higher sand content was observed on bottom slope than that on top and middle slope for Peanut and Citrus orchard land(P<0.05).The clay content on bottom slop was,however,significantly lower than that on other slope positions(P<0.05).Soil sand content decreased whereas the clay content increased with increasing soil depth(P<0.05)for both vegetation types,mainly due to the movement of more clay particles with the infiltration process of soil water.The silt content did not change significantly in both horizontal and vertical directions.Sand content had stronger variation than silt and clay contents among slope positions and soil depths.In addition,the distribution of soil particles were also affected by vegetation types and cropping systems.The migration of soil sand along the slope was more significant in Peanut land than that in Citrus orchard.The migration rate of clay in the vertical direction was stronger in Citrus land than that in Peanut land.However,vegetation types and cropping systems had little effects on the distribution of soil silt content.(2)At the slope scale,soil water content increased with the enhancement of soil depth,but the increment was gradually decreased.The coefficient of soil water content changed in the medium level,varying between 8.2%and 18.8%among different soil depths(0-10,10-20,20-40,40-60,60-80 and 80-100 cm),and the spatial and temporal heterogeneity of soil moisture in the shallow layer(0-20 cm)was stronger than that in the deeper layer(40-100cm).Soil water content showed a clear seasonal distribution:spring>summer>winter>autumn.The climatic factors such as rainfall and atmospheric temperature were important factors influencing the time variation of soil water content,and the response of soil moisture to climate factors in shallow layer(0-20 cm)is more sensitive than that in deeper layer(40-100 cm).The main controlling factors for spatial heterogeneity of soil moisture included elevation,bulk density,soil particle composition,organic matter and directions of slope.(3)Vegetation types had a major impact on soil moisture in the slope,soil moisture in the shallow layer(0-20 cm)was significantly lower than that in the deep layer(40-100 cm)between the two vegetative types(P<0.05).In the same depth,the variation coefficients of soil moisture in Peanut was significantly higher than that in Citrus orchard(P<0.05).(4)There existed a strong temporal stability(TS)of soil moisture among different depths,and the spatial pattern similarity of soil moisture among the six soil layers was significant at the 0.01 level by the Spearman rank correlation coefficient.Two temporal stability indices(SDRD and Spearman’s ranking coefficients)showed that the temporal stability of soil moisture increased with the soil depth.(5)The mean value of soil moisture in the study area could be predicted with the technology of temporal stability in red soil region.By comparing the three common methods for identifying the representative measuring points,it was found that the prediction accuracy of sample points was the highest with the minimum SDRD value,which was an indirect prediction of representative measuring points.Temporal stability technology could be used to accurately predict the average values and the spatially distributed soil water storage at high time-space resolution by combining the advantages of high time low space resolution data and high space low time resolution data.Once the temporal stability location and temporal stability relationship were identified,only one auto-measured location was needed to estimate the mean and the spatial distribution accurately for the whole slope at any time.