Modeling Evapotranspiration With Its Components And Water Supply In Jiangxi Province's Terrestrial Ecosystems Based On Remote Sensing And Meteorological Data

The combine process of soil evaporation,interception evaporation and transpiration from vegetation are called evapotranspiration. It is the most important way to exchange water and energy between soil, vegetation and atmosphere. Accurate to estimate terrestrial ecosystems'evapotranspiration plays an important role in global climate change and water resources, environmental problems, and management of water resources,monitor drainage of agriculture and the prediction of drought for increase water use have important significant.In this paper, we take the spatio-temporal distribution pattern of jiangxi province's terrestrial ecosystems'evapotranspiration with its components and water supply as the scientific problem, and aim to validate and improve PT-FI model which based on the ecosystems mechanism process use key data sources including MODIS remote sensing data and eddy covariance flux/meteorological data from China Terrestrial Ecosystem FLUX Research Network (ChinaFLUX) Qianyanzhou site. And then we forward specific solution schemes for these problems and build a ET model(PT-WE model). With the PT-WE model and precipitation data, modeling site and regional scales evapotranspiration with its components and water supply, and assess the spatio-temporal distribution pattern of jiangxi province terrestrial ecosystem. The research work and main conclusions are as follow:(1) Based on remote sensing PT-FI model conceptual model, we summary the advatanges of estimating ET via PT-FI model and analyze potential problems lying in the estimation of ET. And then we forward specific solution schemes for these problems and verification and improvement and build a PT-WE model to estimate jiangxi province terrestrial ecosystem ET at site and regional scales.(2) With the verification and improvement to the evapotranspiration model (PT-WE model), we simulate the ET and its components of the planted coniferous forest ecosystem at Qianyanzhou(QYZ) during 2003 and 2008 by using remote sensing data from MODIS and meteorological data. The estimated ET is compared with the observation from eddy flux tower sites. The results show that the simulated annual ET is about 2.4% lower than measurement, with a decision coefficient value(R2) of 0.83 and root mean square error of 0.61mm-d-1. As to the ET components, soil evaporation is 12% of simulated ET with relative stability at seasonal and interannual scale. With obvious seasonal and interannaual variability, interception evaporation contribute 23% of the ET and with the same change tendency as precipitation. Canopy transpiration is about 65% of ET with obvious seasonal variation and interannual stability. During January to March, photosynthesis isn't to be strong, T/ET less than 30%, with the enhance of canopy transpiration, T/ET increases rapidly after April, in the end of July up to about 90%. Because of the data is easy to get at regional dimension in this model, which can lay a foundation for the further scaling up of evapotranspiration estimation in regional scales.(3) Temperature,Relative humidity (RH) and net radiation are necessary input parameter in the study of the regional ecological system. Net radiation is a important data for calculate ET, which is obtained through computing total solar radiation using Angstrom equation and Chang Jen-Hu modification formula.8-day temperature and Relative humidity (RH) and net radiation are obtained through using spatial different interpolation model (AUSPLIN) to interpolate the meteorological data in stations of China state meteorological bureau.(4) Spatial database of the jiangxi province's terrestrial ecosystems are constructed, containing the 8-day 8-day normalized differential vegetation index(NDVI) and enhance vegetation indices(EVI), which are obtained through computing surface reflectance of MODIS, and Harmonic analysis of time series (HANTS) is used to processing the abnormal data of vegetation.(5) Based on regional scale of meteorological and remote sensing data, we use PT-WE model to estimate jiangxi province's terrestrial ecosystems' evapotranspiration with its components and produce each 8-day ET data in 2001-2009. The result shows that the average of ET is about 510mm in jiangxi, ET over jiangxi decreases from the south to the northeast, with 8-day transpiration from vegetation value ranging from 70～510mm, the average of which is about 378mm about 75.8% of ET; with 8-day soil evaporation value ranging from 18-70mm, the average of which is about 64mm about 12.6% of ET; with 8-day interception evaporation value ranging from 15-100mm, the average of which is about 54mm about 11.3% of ET.(6) According to the principle of water balance and based on spatialized precipitation,ET data of jiangxi province to estimate the water supply of jiangxi province in 2001-2009.The average of total water supply is about 1044mm and ranging from 790-2100mm, for approximately 1740 billion cubic metres, The water supply over jiangxi decreases from the east to the west. The runoff coefficient is 0.67. The spatial distribution is similar to water supply.