| The hydrometeorological elements system under natural conditions is stable,but it has been in a significantly non-stationary state in recent years due to the impact of human activities and climate change.The research on hydrological and meteorological elements in traditional theories and models is based on the assumption of stationarity,and it has been unable to adapt to the prediction needs of non-stationary changes in hydrology and meteorology under the influence of current climate change and human activities.However,the non-stationary study of hydrometeorological elements depends on long-term series data,and a large number of watersheds,especially important water source conservation mountain areas,lack long-term hydrometeorological observation data..Therefore,the study of the spatial differentiation of non-stationarity of hydrometeorological elements and the establishment of a non-stationary prediction model of hydrological and meteorological elements with regional natural characteristics will help to carry out rapid prediction of non-stationary of large-scale hydrometeorological elements,overcome the lack of basic hydrometeorological observations,and improve the management efficiency of relevant departments such as water resources,agriculture,and meteorology.The upper reaches of the Minjiang River are located in the alpine canyons on the edge of the Qinghai-Tibet Plateau,with fragile habitats and extreme sensitivity to climate change.As an important water conservation area of the Chengdu Plain,the non-stationary nature of its hydrological and meteorological elements directly affects the water supply and flood control and drought resistance of the Chengdu Plain and even the upper Yangtze River Basin.This study takes the upper reaches of the Minjiang River as the study area,uses 7 types of meteorological elements(precipitation,maximum temperature,average temperature,minimum temperature,sunshine hours,relative humidity and wind speed)and 2 types of hydrological elements(potential evapotranspiration and actual evapotranspiration)from 1961 to 2017 in the basin.In this study,the non-stationary degree and complexity of the interannual changes of 9 types of hydrometeorological elements were calculated,and the interannual change trends of each element were analyzed.This study also explored the correlation between the non-stationarity of each element and the impact factor.And on this basis,a prediction model for the non-stationarity of hydrometeorological elements was established.The main research conclusions are as follows:(1)The non-stationary degree and complexity of hydrometeorological elements in the upper reaches of the Minjiang River have obvious spatial differences.Among them,the non-stationarity of precipitation,minimum temperature,sunshine hours,relative humidity,and potential evapotranspiration is higher in most areas of the basin.The non-stationarity of the highest temperature and the average temperature is moderately low in the entire basin.The non-stationarity of wind speed and actual evapotranspiration has obvious north-south differences in the basin,showing high south and low north.(2)The interannual change trend of hydrometeorological elements in the upper reaches of Minjiang River has obvious spatial differences.Among them,the change trend of the highest temperature,average temperature,lowest temperature and potential evapotranspiration showed a significant increase in the entire basin.The change trend of precipitation and wind speed is significantly reduced only in a small part of the southern part of the basin.The changing trends of sunshine hours,relative humidity and actual evapotranspiration have different significant changes with different geographical conditions in the basin.(3)The non-stationarity of hydrometeorological elements in the upper reaches of the Minjiang River has different significantly related influencing factors.This study screened a total of eight kinds of influencing factors in three categories:vegetation,soil and terrain:normalized vegetation index(NDVI),leaf area index(LAI),soil humidity(SH),elevation(EL),slope(SLOPE),profile curvature(PC),aspect(ASPECT),terrain humidity index(TWI).The influencing factors of precipitation and potential evapotranspiration are EL,LAI,SLOPE,PC,TWI,SH;The highest temperature influencing factors are EL,LAI,NDVI,SLOPE,TWI,SH;The influencing factors of average temperature,sunshine hours and wind speed are EL,LAI,NDVI,SLOPE,PC,TWI,SH;The influence factors of the lowest temperature and relative humidity are EL,LAI,NDVI,TWI,SH;The actual influencing factors of evapotranspiration are EL,LAI,NDVI,SLOPE,ASPECT,PC,TWI,SH.(4)In the prediction model based on the influence factors of different non-stationarity of hydrometeorological elements in the upper reaches of the Minjiang River,the prediction accuracy of the model established by the artificial neural network is higher than that of the model established by multiple linear regression analysis.In the non-stationary prediction model of hydrometeorological elements established by multiple regression analysis,the R~2 for precipitation,highest temperature,average temperature,lowest temperature,sunshine hours,relative humidity,wind speed,potential evapotranspiration and actual evapotranspiration are 0.313,0.358,0.259,0.290,0.422,0.116,0.377,0.280 and 0.439,respectively.In the non-stationary prediction model of hydrometeorological elements established by artificial neural network,the R~2 for precipitation,highest temperature,average temperature,lowest temperature,sunshine hours,relative humidity,wind speed,potential evapotranspiration and actual evapotranspiration are 0.617,0.663,0.652,0.676,0.655,0.565,0.674,0.580 and 0.721,respectively.The above researches haave important scientific support for the management of water resources in the upper reaches of the Minjiang River and the upper reaches of the Yangtze River under climate change.They also have important implications for the study of non-stationarity of hydrometeorological elements in other regions. |
PDF Full Text Request