| The Loess Plateau is the main source of sediment for the Yellow River and has serious soil erosion problems.Since the large-scale vegetation restoration activities were carried out in the 1990s,the vegetation cover on the Loess Plateau has increased rapidly,and the ecological environment has improved significantly.However,the vegetation water demand has increased significantly,which inevitably interferes with the regional water cycle process,and evapotranspiration,as a key link in the water-heat balance,maintains a dynamic equilibrium with vegetation in the natural state,while the rapid increase of artificial vegetation will inevitably break this equilibrium.Therefore,accurately estimating evapotranspiration on the Loess Plateau,clarifying the impact of vegetation restoration on evapotranspiration,and further exploring the water-heat balance relationship in the watershed under large-scale vegetation restoration are of great significance for further understanding the response mechanism of water cycle to changing environments and guiding vegetation restoration work in the Loess Plateau.Six typical river basins on the Loess Plateau(Fen River,Kuye River,Wuding River,Beiluo River,Jing River and Wei River basin)was taken as the research objects.The Penman-MonteithLeuning(PML)model considering vegetation factors was used to estimate the actual evapotranspiration from 2000 to 2019.Then the effect of vegetation restoration on actual evapotranspiration was quantitatively estimated.To investigate the response mechanism of evapotranspiration water consumption to vegetation change,the estimated evapotranspiration was introduced into the water-heat balance model,and the dynamic evolution of the water-heat balance relationship in each basin was analyzed from the perspective of the relationship between aridity index and evaporation rate.And the driving factors of the change of water-heat balance relationship are revealed.The main conclusions are as follows:(1)The Mann-Kendall trend test and Kriging method were used to analyze the characteristics of water and heat status and vegetation conditions in each typical watershed.The results showed that the average annual temperature(T)in all typical river basins except Beiluo River Basin showed an increasing trend.The annual potential evapotranspiration(ET0)showed an insignificant upward trend in the Fen River Basin and Wei River Basin,but a downward trend in the other river basins.The annual rainfall(P)of all basins showed an increasing trend in different degrees,among which the increasing trend was significant in the Kuye River Basin,Wuding River Basin and Jing River Basin.The average annual leaf area index(LAI)of each typical watershed showed a significant increasing trend.The cultivated land area continued to decrease,the forest area increased as a whole,the shrub area increased as a whole except for the Beiluo River Basin,the grassland area increased in Beiluo River Basin,Jing River Basin and Wei River Basin,and decreased in other basins.(2)The parameters of the PML model were optimized,and the actual evapotranspiration(ET)of each typical basin was estimated based on meteorological and vegetation data.The reliability of the estimated results was verified,and the trend of ET was analyzed.The results showed that the maximum stomatal conductance of leaves(gsx)and soil moisture index(f)of the PML model were optimized,and the results of R2,NSE and RMSE showed that the optimization results were reasonable,so the model was applicable in the study area.The simulation results were validated using the multi-year water balance method,which further demonstrated the reliability of the simulation results.The annual ET increases in different degrees in each typical river basin,among which the Fen River basin,Kuye River Basin,Wuding River Basin and Jing River Basin showed significant increasing trends.(3)Correlation analysis was used to analyze the response relationship of evapotranspiration to different influencing factors,and PML model combined with detrending method was used to estimate the influence of vegetation restoration on evapotranspiration.The results showed that among the three influencing factors of atmospheric evapotranspiration demand(T),water supply(P)and vegetation conditions(LAI),ET was the highest correlation with LAI in all typical basins,indicating that vegetation conditions had the greatest influence on ET changes in each typical basin.Vegetation restoration resulted in a maximum increase of 37.22mm in the Wuding River Basin,followed by the Fenhe River Basin,Beiluo River Basin and Jing River Basin with an increment range of 31.64 to 34.02mm,and a smaller increase of 23.94 mm and 17.99 mm in the Kuye River Basin and Wei River Basin,respectively.(4)The aridity index(AI),evaporation rate(ER)and residual water-energy ratio(WER)of each basin were calculated for each basin,and their evolution patterns were analyzed.The results showed that AI values of the typical basins ranges from 1.68 to 2.69,belonging to semi-arid areas,among which the Wei River Basin had the lowest AI value(1.68),and the Kuye River Basin has the highest value(2.69),and the overall AI trend was decreasing.The ER of the typical basins varied from 0.88 to 0.95 and the trend was not significant,which was mainly attributed to the offsetting trends of increasing precipitation and evapotranspiration.Residual water-energy ratio(WER)was calculated for all typical river basins.The WER of each typical river basin showed an increasing trend to different degrees,among which the WER of Beiluo River and Jing River basins are larger,while the WER of Kuye River Basin and Wuding River Basin are relatively smaller,indicating that the hydrothermal conditions of Kuye River Basin and Wuding River Basin are worse.(5)The AI-ER dependent structure relationship was constructed based on Coupla function.Through the selection of edge function,Copula function is optimized to obtain the most suitable Copula function for each sequence.The Pettitt test was used to diagnose the variation of AI-ER dependent structural sequences.The results showed that the Copula function that best fits the AIER j oint distribution is Clayton Copula in the upper Fenhe River Basin,Frank Copula in the Kuye River Basin,Wuding River Basin and Jing River Basin,and Gumbel Copula in the other basins.There were no obvious variation points in AI-ER dependent structure sequences in each typical basin.However,the K value of Kuye River,Wuding River and Jinghe River basin was larger than that of other river basins although it does not exceed the threshold value.Therefore,there may be slight changes in the AI-ER relationship during the study period.(6)The direct and indirect effects of the influence factors on AI-ER relationship were analyzed by using the path analysis method.The results show that without considering the interaction among the influencing factors,the influence of each typical basin on the AI-ER relationship is P<LAI<T.The influence of P on AI-ER is mainly direct.The direct influence of LAI on AI-ER in all river basins except Beiluo River Basin was relatively small,but the indirect influence was great,which ultimately led to the greater influence of LAI on AI-ER.The direct effect of T on AI-ER and the indirect effect of P and LAI on AI-ER through T were small. |
PDF Full Text Request