Relationship Between Eco-hydrological Indices And Groundwater In The Erdos Plateau

How the terrestrial vegetation ecosystem is affected by the climatic gradients and groundwater? This is a basic issure in the study area of eco-hydrogeology.In this thesis,the problem is investigated for the Erdos Plateau,a region with arid and semi-arid climate,via the proposed concept of distributed eco-hydrological indices(EHIs).An EHI is refered to a quantitative index from which one of the ecological conditions,such as water supply,living state of the species,hydrological functions and ecological water consumption,etc.,for the terrestrial vegetation ecosystem.This study proposes 4 EHIs,including the aridity index,vegetation index,precipitation-runoff index and vegattion water consumption index.The aridity index(?)reflects the climatic background.The vegetation index reflects the covrage of plants such as the enhanced vegetation index(EVI)and leaf area index(LAI).The precipitation-runoff index(C_r)reflects the runoff production capacity which depends on the residual precipitation after loss by the land-surface evapotranspiration.In the arid region,C_rindicates the capacity of groundwater recharge.The vegetation water consumption index(Cw)refelects the degree of wateruse inthe vegetation zone.These EHIs are estimated for elements in remote-sensing data,with a resolution of that from 250 m to 1 km.In this study,the region along the Erdos Plateau from the Dosit River in the west to the Wuding River in the east is taken as the concerned area.The distributed data of the 4 EHIs and groundwater depth(d)data are organized for the study region,on the basis of the collected meteorological data,remote sensing data and groundwater observation data.Among them,the C_r and Cw values are estimated from the actual evapotranspiration,which is identified from the original 1-km-resolution data of MOD16-ET data and MuSyQ-ET data,via water balance analysis at the catchment scale.It is indicated by statistical analysis that the stochastic distributions of the EHIs and groundwater depth do not follow the normal Gaussion distribution patterns.Based on the 250-m-resolution data,the relationships between the vegetation index,EVI,and the aridity index and the groundwater depth are analyzed in detail.With increasing? and d,EVI shows decresing trend as a general trend.However,the variation characteristics could not be sufficantly revealed only by analysis on the impact of one factor.In this study,a double-factors(?-d)coordinates is built.The ?-dspace is divided into several gridded zones to calculate the statistical values of EVI(p=50%? p=98%)from the data points.Then contours of the specific EVI values are drawn.This map in the double-factors space reveals the joint influence of the aridity index and groundwater depth on the vegetation index.It is indicated that when d<7 m,the vegetation index is sensitive to change in groundwater depth.In semi-arid climate,EVI generally decreases with increasing d.However,when ?>6,EVI decreases with decreasingd.This trend is caused by the salinity problem in soils underlying shallow groundwater driven by the strong evaporation.Based on the 1-km-resolution data,complex relationships also exist among the other EHIs.The 250-m-resolution groundwater depth data are upscaled as the effective groundwater depth(d_e)at the 1-km-resoluton with a reasonable algorithm,in preparing for the analysis on the relationships among the EHIs and multiple controlls.With the increasing aridity index,the MOD16-based C_r data and Cwdata both show decresing trends,but the MuSyQ-based data shows low correlation with the aridity index.The leaf area index,LAI,shows significant impacts on the C_r and Cw data: with the increasing LAI,C_r decreases but Cw increases.Thus,in this suty another double-factors coordinate is built with LAI and de,to analyze the joint influence of vetegation index and groundwater on the C_r and Cw data.When LAIis low,the C_r and Cw data of p=50% and p=90% will reach to maximum or minimun at a certain range of de;whereas when LAIis high the relationships among them are not significant.It is revealed through correlation and co-variation analysis that the correlation between C_r-Cwand LAI is significantly higher than that between C_r-Cwand ?.In addition,change in groundwater depth will influence theLAI-?correlation and the C_r-?correlation.The study results provide new insights on the plant-ecological patterns and the controls in the Erdos Plateau.The impacts of vegetation index on the precipitation-runoff and groundwater recharge processes are also preliminary revealved in the statistical aspects.