Research On The Ecohydrological Sensitivity Of Watershed And Its Influencing Factor Analysis

Vegetation-water relationship is an important subject in Ecohydrology.However,the understanding of possible effects of vegetation change on water at multiple spatial scales remains incomplete.There are consistent results of hydrological responses to vegetation change in small watersheds.By contrast,the impact of vegetation change on streamflow in large watersheds has been less investigated due to limitations such as the lack of high quality and sufficient precipitation and flow data,and commonly-accepted methods to quantify streamflow variation attributed to vegetation change.Meanwhile,due to lack of a suitable indicator to assess the impact of vegetation change on streamflow,it is difficult to compare the differences of ecohydrological processes between climate,vegetation,topography,soil and landscape.In this study,we selected 14 large watersheds vary along different climate zones in China as examples,to detect whether the climate,hydrological and vegetation data have statistically significant upward or downward trends on multiple temporal scales(annual,dry season,wet season)by use of non-parametric test;to quantify the effects of climate variability,vegetation change and other factors(e.g.,road construction,agriculture,dams,mining,urbanization)on seasonal and annual flows by an improved single watershed approach;to introduce ecohydrological sensitivity and compare the differences of ecohydrological sensitivity between climate gradient,dominant soil type and hydrological regime;to assess the effects of watershed heterogeneity on ecohydrological sensitivity and simulate ecohydrological sensitivity based on significantly related watershed heterogeneous indices.These findings highlight ecohydrological processes in large watersheds are complicated and also associated with topography,soil condition and landscape pattern.This requires for effective managements to alleviate negative effects on water resource caused by vegetation change under climate change.Key results are:(1)According to the trend analysis,significant increase trends were detected in maximum,minimum and mean temperature(annual,dry season and wet season)in most watersheds in response to global warming(?<0.05),whilst insignificant tendency took place in evapotranspiration and precipitation(?>0.05)during the study period.Besides,negative relationships were identified between streamflow and vegetation change in wet season.(2)According to the improved single watershed approach,vegetation change generally had significant effects on streamflow in watersheds.Differences in dominant vegetation type and types of vegetation change lead to different hydrological responses in wet season.Wet season flows increased by 4.6%,3.3%,0.5%and 5.1%,respectively in the Xiangshui,Upper Zagunao,Zagunao and Gengzhang watersheds,decreased by 6.3%in the Dongchuan watershed as a result of LAI reduction ascribed to vegetation change.On the contrary,vegetation growth resulted in 21.3mm,52.9mm,97.0mm,104.7mm,8.4mm,37.2mm and 18.8mm reductions in the Pingjiang,Tangwang River,Upper Heishui River,Heishui River,Gongbujiangda,Jingchuan and Rui River watersheds in wet season,respectively,whereas 31.9mm and 0.6mm increments in the Xinancha River and Heshuichuan watersheds in wet season,respectively.Dry season flow variations were closely related to soil conditions,soil moisture and underground water discharge.Vegetation reduction increased dry season flows in Pingjiang,Upper Zagunao,Zagunao,Gengzhang,Dongchuan and Jingchuan watersheds,whilst decreased dry season flows in the Tangwang River,Xinancha River and Rui River watersheds.In comparison,dry season flows increased in the Xiangshui,Upper Heishui River,Heishui River,Gongbujiangda and Heshuichuan watersheds in response to vegetation growth.In addition,we failed to identify relationships between vegetation growth and dry season flow reduction in study watersheds.The relative contributions of vegetation change and climate variability to streamflow variation can imply their impact strength,both offsetting and addictive effects were detected in this study.(3)Ecohydrological sensitivities in large watersheds were highly variable along climate gradient,dominant soil type and hydrological regime.Annual and dry season ecohydrological sensitivities were positively related to dryness index.Streamflow responses to vegetation change were more sensitive in water-limited watersheds than in energy-limited watersheds.Negative relationships were identified between annual and dry season ecohydrological sensitivities and permeability of dominant soil type.There was significant difference of dry season ecohydrological sensitivity between rain-dominated watersheds and rain-snow hybrid watersheds.At the same time,ecohydrological sensitivities were higher in dry season than that in wet season.(4)According to this study,ecohydrological sensitivity was determined not only by the vegetation change but also by watershed properties such as climate gradient,topography,soil condition and landscape pattern.Annual and wet season ecohydrological sensitivities were highly associated with soil properties,while dry season ecohydrological sensitivity was closely related to topography.(5)Based on watershed heterogeneous indices,Multiple Linear Regression(MLR)was employed to simulate ecohydrological sensitivity.The models grouped with the best parameters were selected to construct the prediction model.Here,the best prediction model took place in dry season,the R~2 and p value reached to 0.966 and 0.051,whilst the accuracy of wet season ecohydrological sensitivity model was lower than that of dry season and annual models whose R~2 and p value were 0.501 and 0.064.The R~2 and p value of annual ecohydrological sensitivity model were 0.902 and 0.035,respectively.Through prediction model,ecohydrological sensitivity could be simulated based on watershed heterogeneous indices.Understanding their interactive effects is important for sustainable water supply in large watersheds.