| In the past few decades,a series of soil and water conservation measures and eco-rehabilitation strategies have been implemented on the Loess Plateau.Consequently,these artificial measures have altered the underlying surface conditions of the region,leading to significant changes in the runoff and sediment load in the Yellow River.Clarifying the response of runoff and sediment to changing environment at different spatial and temporal scales will benefit deep perception of runoff and suspended sediment transport dynamics,and provide valuable information for soil and water conservation planning and integrated catchment management.Therefore,the thesis takes Xichuan River catchment,a typical hilly-gully region in the Loess Plateau,as the study area.Based on the meteorological and hydrological data collected from Zaoyuan hydrological station and six rainfall gauge stations distributed across the study area,this study investigated the runoff and sediment load variations at different temporal scales and quantified the relative contributions of climate variability and human activities to runoff and sediment load variations.A total of 112 events occurred during 1974-1989 and 2007-2013 were selected as the statistical sample,K-medoids and discriminant analysis were employed to classify the flood events according to runoff depths,flood duration,and peak discharge,then the runoff-sediment dynamics under different flood patterns were examined.Meanwhile,a method named“Paired flood events method”was proposed to distinguish the impacts of human activities and climate variability on runoff and sediment load change at the flood event scale.Furthermore,the field simulated rainfall was used to investigate the effects of vegetation coverage and different vegetation components on reducing surface runoff and erosion,as well as the mechanism of regulating to runoff generation and sediment yiled process at the hillslope scale.The specific results and conclusions are as follows:?1?Human activity was the main driving force of runoff and sediment load reduction at interannual scale,while the contribution of precipitation change was relatively low.Significant decreasing trends in runoff and sediment load were detected?P<0.05?,whereas no statistically significant trends were detected in the annual precipitation.The changing points of both the annual runoff and the sediment load occurred in 1996 at the 95%confidence level with no abrupt change points in the annual rainfall detected.The double mass curves analysis indicated human activity was responsible for approximately 37.4%and 54.7%of total runoff and sediment load reduction in the Xichuan River catchment,respectively,and precipitation change could only explain about 5.7%and 8.3%reduction in runoff and sediment load,respectively.?2?The temporal variation of runoff and sediment load at monthly and daily scale and its response to soil and water conservation have been analyzed.Both the monthly runoff and sediment load presented unimodal type distribution during baseline period and changing period,and the peak occurred in July.The reduction could be detected for the runoff and sediment load in different monthes of flooding season with the implement of soil and water conservation.However,the the ratio of runoff and sediment load of floodind season to the annual runoff and sediment load in the changing period was roughly consistent with the ratio in the baseline period.In comparison to the baseline period,the relationships between daily discharge and sediment delivery have been altered in the changing period.A significant decline was observed in the,sediment delivery,and suspended sediment concentration from 1974 to2012.Furthermore,the average maximum daily discharge,sediment delivery,and suspended sediment concentration decreased by 37.3%,42.6%,and 18.1%during the period of 1996-2012 relative to the values between1974 and 1995,respectively.During the period of 1974-1995,the Ln 3 distribution was the optimal distribution function of extreme daily discharge in Xichuan River basin.The extreme daily discharge of 5-year,10-year,25-year,50-year,and100-year return levels is 63.5 m3·s-1?90.7 m3·s-1?132.1 m3·s-1?168.0 m3·s-1,and 208.4 m3·s-1,respectively.In contrast,extreme daily discharge series could be well matched by Wakeby distribution function from 1996 to 2012.The corresponding extreme daily discharge of 5-year,10-year,25-year,50-year,and 100-year return levels is 25.1 m3·s-1?28.6 m3·s-1?34.5 m3·s-1?41.5 m3·s-1,and 52.1 m3·s-1,respectively.Moreover,For all return periods,the threshold of extreme daily discharge during the period of 1974-1995 was much higher than that from1996to 2012.?3?The WASASED hydrological model was calibrated and validated during the period from 1988 to 1991.The results revealed that both Nash-Sutcliffe coefficientand determination coefficient for daily discharge exceeded 0.6,which indicated that the model results met the demand of precision,and could be used to predicte the daily discharge in the Xichuan River catchment.This study simulated the runoff yield under four precipitation scenarios.Keeping the other inputs the same,the four precipitation scenarios were input to the validated SWAT model,and the daily discharge was simulated during the period of 1988 to 1991.The runoff yield increase with increasing precipitation and decrease with decreasing precipitation.When precipitation increased by 5%and 10%,the runoff yield increased by 22.8%and 35.2%,respectively.When precipitation decreased by 5%and 10%,the runoff yield decrease by 13.8%and 31.0%,respectively.?4?Explore suspended sediment dynamic and runoff-sediment transference relationships under different flood patterns.Based on K-medoids clustering and discriminant analysis method,one hundred and twelve flood events data were classified into four flood patterns.The hydrologic features of different flood patterns exhibit significant differences.Each flood pattern resulted in differing levels of the event sediment yield,the contribution of different flood patterns to the summed sediment load was 24.5%?1.48×107 t?,11.2%?6.77×106 t?,21.4%?1.29×107 t?and 42.9%?2.59×107 t?for Patterns A,B,C,and D,respectively.Of the four flood patterns,the average sediment yiled produced by Pattern A,C was significantly larger than that by Patterns B and D.Therefore,the sediment control should firstly aimed at large sediment-producing patterns such as Pattern A and Pattern C.The SSC-Q hysteresis loops were further applied to investigate the dynamic changes of the relationship between suspended sediment concentration and runoff.The majority of flood events presented counter-clockwise hysteretic loops and figure-eight loops across the entire study period.All the events in pattern A belonged to complex hysteresis loops.The main hysteretic loops type for Pattern B was counter-clockwise hysteretic loops.The complex hysteresis loops were the main type for Pattern C.For the Pattern D,the figure-eight hysteretic loop was the main type.In addition,the proportion of complex hysteresis loops decreased in the changing period compared with the baseline period,whereas the the proportion of counter-clockwise hysteretic loops increased.?5?Quantify the relative contribution of the climate variability and human activities on the runoff and sediment load changes by using the“Paired flood events method”.Based on the pre-set thresholds,a total of eight sets of paired events with approximate or identical values of rainfall amount and rainfall intensity were selected to investigate the runoff and sediment load changes in response to human activities.The analysis result indicated that the human activities contributed 20%-79%to the change in runoff,with an average contribution rate of47.0%.Similarly,the contribution to sediment load changes by human activities varied from55%to 95%,with an average value of 81.0%.?6?Both the overland flow rate and sediment yield rate show a downward trend with the increase of vegetation coverage.There was a linear functional relation between vegetation coverage and overland flow rate for both 1.5 mm·min-1 and 2.0 mm·min-1 rainfall intensities,whereas the relationships between vegetation coverage and overland flow rate could be described with exponential function.Under the two rainfall intensities,the time required for overland flow initiation gradually extended with the increase of vegetation coverage.Compared with the bare soil plot,the mean overland flow rate reduction efficiency for different vegetation coverage plots?20%,40%,60%,90%?were 24%,36%,53%,79%under the 1.5 mm·min-1 rainfall intensity,whereas reduced sediment yield by 37%,73%,78%,94%,respectively.Under the 2.0 mm·min-1 rainfall intensity,the plots with different vegetation coverage?20%,40%,60%,90%?reduced overland flow rate by 17%,28%,37%,71%,respectively,and the the mean sediment yield reductions were 27%,67%,78%,89%,respectively.The sediment reduction effect of grass is greater than the flow reduction effect under the same vegetation coverage condition.?7?For the grassland plots,the litter layer played a predominant role in reducing overland flow rate,while the vegetation roots were performed better in soil erosion control.Under the two rainfall intensities,the time required for overland flow initiation was the longest in the intact grass plot,followed by the no litter plot,while the initial time of overland flow generation in the bare soil slope was the shortest.The litter layer contributed to reducing overland flow rate by 26%and 17%under 1.5 mm·min-1 and 2.0 mm·min-1 rainfall intensities conditions,respectively.The contributions of the vegetation roots to reducing overland flow rate were 12%and 11%under 1.5 mm·min-1 and 2.0 mm·min-1 rainfall intensities conditions,respectively.Stems and leaves parts made contributions of 7%and 10%to reducing overland flow rate were 12%and 11%under 1.5 mm·min-1 and 2.0 mm·min-1 rainfall intensities conditions,respectively.Vegetation roots showed the highest sediment yield reduction with a value of 51%for both 1.5 mm·min-1 and 2.0 mm·min-1 rainfall intensities conditions,followed by litter layer,which contributed 36%and 30%to sediment yield reductions under the 1.5mm·min-1 and 2.0 mm·min-1 rainfall intensities,respectively.The stems and leaves components made the minimum contributions to reducing sediment yield,with the value of 3%and 7%under the 1.5 mm·min-1 and 2.0 mm·min-1 rainfall intensities,respectively. |
PDF Full Text Request