| Soil moisture is the basis for studying surface runoff,soil erosion and other hydrological processes.The transport and redistribution of precipitation by soil directly affect the runoff production mechanism of the watershed,which determines the soil moisture distribution and hydrological processes such as surface runoff,interflow and deep infiltration,and then affects the erosion and sediment production of the watershed.Mountain areas account for 24%of the earth’s land area.The topographic,soil,bedrock and other surface structures of small mountains watershed are the most closely related to the formation of runoff.The spatial distribution and change of soil properties directly affect the soil moisture pattern and hydrological process of the watershed.The sediment yield response of small mountainous watersheds shows complex spatial and temporal distribution characteristics due to the combined effects of meteorology,soil,topography,water system,vegetation and land use conditions within the small watersheds.In the context of global climate change and frequent extreme rainfall events,the study of moisture patterns and erosion and sediment production response characteristics in mountainous areas not only helps to reveal the spatial and temporal evolution of hydrological processes in watersheds under changing environments,but also can provide theoretical support for scientific watershed management and effective prevention of soil erosion and flash flood disasters in small watersheds.Therefore,this study takes Zhangjiachong mountainous watershed in the Three Gorges Reservoir Areas as the study area,and takes a small mountainous watershed with a single parent material as the study object.Through intensive watershed soil sampling,and on the basis of investigating the soil hydrological properties of the watershed,three typical slopes were selected for long time series and high frequency observations of watershed soil moisture as well as watershed meteorology and runoff sediment,to systematically study the soil hydrological properties under the characteristics of typical mountainous watersheds.The spatial variability of soil hydrological properties in typical mountainous watersheds was systematically studied to clarify the spatial and temporal response of soil moisture in different landscape locations of the watershed and the controlling factors,to reveal the runoff production process under the combined effect of soil moisture and rainfall,and to clarify the erosion and sediment production characteristics of small watersheds driven by runoff and soil moisture.The main findings are as follows:（1）The results of geostatistical analysis showed that the nugget coefficient of all surface soil properties in the study area was less than 25%,and the spatial correlation of soil properties was strong.Soil properties formed a significant spatial pattern in the watershed,which showed that the sand content was lower in the middle and lower reaches of the watershed and higher in the northeastern part of the upper reaches;soil bulk density was lower in the middle and lower reaches and relatively higher in the upper reaches;and soil organic carbon was higher in the northeastern part of the middle and upper reaches of the watershed.Soil properties differed hydraulically between the surface layer and the layer above the relatively impermeable layer.The results of the analysis of hydrologically sensitive areas in the watershed indicate that the soil topography index,which integrates the topography and soil properties of the watershed,is very effective in describing the watershed runoff generation areas,which is important for the subsequent exploration of the spatial heterogeneity of soil properties driving the spatial and temporal variability of soil moisture,which in turn affects the watershed moisture distribution and flow production processes.（2）The results show that the Gaussian mixture model can effectively cluster the hidden states behind different soil moisture,and the hidden Markov model to express the transition patterns of different states can better explore the influencing factors behind soil moisture changes,and the combination of the hidden state time series can well express the changes of soil moisture dominant control factors on the time scale.Through the observation of rainfall events,combined with the comprehensive analysis of hidden Markov chains and soil moisture time series,it is found that the variability of soil moisture is significantly higher on the lower slope than on the upper slope,and there are obvious differences between the hidden states of soil moisture at two adjacent locations.The most important factors influencing soil moisture changes over the whole time series are rainfall and the nature of the soil itself.The introduction of the hidden state analysis method can be effectively applied to the description and prediction of soil moisture.（3）The response of soil moisture to rainfall on different slopes in the watershed showed significant spatial variability,and the soil moisture response was influenced by the interaction of rainfall characteristics,landscape characteristics,and soil characteristics.The frequency of preferential flow varies with landscape location and slope.For slope H1 in the upper part of the watershed,preferential flow frequently occurred at 30 cm and 40 cm depths.Soil moisture changes were observed at all depths during almost all rainfall events for slopes H2 and H3,but for slope H2,preferential flow frequently occurred at 30 cm depth;the preferential flow response at 30 cm on the upper slope occurred 3-4 times more frequently than at 20 cm.The proportion of preferential flow response occurring on the H3slope at the bottom of the watershed was relatively low.（4）The complexity of the soil moisture response was further well demonstrated by quantifying the relative change in volumetric soil moisture content（ΔS）,the response time of peak rainfall and peak moisture(Tp2p),and the velocity of wetting front(Vwf),which are three key indicators of the soil moisture response.ForΔS,the most important explanatory variables were soil bulk dendity,antecedent soil moisture content（AMC）,and rainfall.The maximum 30-min rainfall intensity,advance precipitation index（API）and mean rainfall intensity had the greatest influence on Tp2p.The most important explanatory variables for Vwf were antecedent soil moisture content（AMC）,mean rainfall intensity and soil bulk density.Among the topographic variables,slope was identified as an explanatory factor forΔS,and elevation was the second most important explanatory factor forΔS.In conclusion,among all factors rainfall characteristics and soil properties were the key explanatory factors forΔS,Tp2p,and Vwf.（5）Intra-annual variability in soil moisture,runoff,and sediment load explains how91%of the sediment loads were transported in 5%of the time.At the event scale,soil moisture significantly influenced hydrologic processes in the watershed;in addition,four of the 16 events produced 84.59%of the sediment loads.Correlation analysis of hydrologic elements showed that runoff and peak flow were most closely correlated with sediment loads.The highest frequency of clockwise lags in the runoff-sediment concentration hysteresis loop relationship indicates that the primary source of erosion is near the outlet of the watershed.Although the counterclockwise pattern occurs least frequently,the occurrence of the counterclockwise pattern is associated with wetter periods（i.e.,high moisture content）when the watershed is more hydrologically connected and fewer sources of sediment are available for transport.Soil moisture-runoff hysteresis analysis confirms that 68.8%of flood events occur under soil moisture content conditions and are responsible for the vast majority of runoff and sediment loads.This study addresses the critical issue of erosion and sediment production in small mountainous watersheds,especially those with a single soil-forming material,and investigates the spatial distribution of soil hydrological properties in watersheds,combining high-frequency observations of soil moisture,meteorology,runoff and sediment in watersheds.Moreover,it can provide theoretical support for scientific management of watersheds and effective prevention and control of soil erosion and prediction of flash floods in small watersheds. |
PDF Full Text Request