Estimation Of The Day Evapotranspiration In Semiarid Area Using Remote Sensing

Sanhuajian of the Yellow River (from Sanmenxia to Huayuankou)is a mostly semiarid area in Middle China. The characteristic of this region is the precipitous topography, the complicated surface type, heavy the coefficient of runoff, easily arid in spring and rainstorms in summer. Monitoring the state of energy balance and moisture status of the surface by accurately estimating regional evapotranspiration, it has an important realistic meaning for studying water balance at watershed scale, forecasting rainstorm and flood, monitoring large area arid and flood, supplying the numerical weather forecast and estimating crop output of farmland.Traditional climate and hydrology methods for estimating surface evapotranspiration limited to observe and estimate at discrete point, To obtain regional evapotranspiration,there are some limitations which include low-precision interpolation, poor timeliness, vegetation canopy temperature is not easy to acquire and large-scale and high-density observation charge is costly. 1970s, with the rapid development of satellite remote sensing and computer technology, estimation of evapotranspiration by a large scale distributed becomes fast and accuracy. Multi-temporal, multi-resolution, Multi-spectral and multi-angle satellite remote sensing data can objectly image the geometry structure and the states of Water and Heat of the surface, and then obain important parameters of studying on evapotranspiration such as Turbulent Heat Fluxes of Surface Layer, the water content of soil and so on, all what make the remote sensing method superior to the regular method based on microclimate and hydrology. However, there are shortcomings about evapotranspiration estimation using satellite remote sensing: on the one hand, precision of Land Surface Temperature(LST) that is one of the most critical parameters in evapotranspiration model needs to be raised, especially leaded by non-homogeneous surface, the other hand, the operating process of retrieving surface parameters by remote sensing is complex and difficult to control. Therefore, it is necessary that simplified model of estimating day evapotranspiration established by remote sensing.In the paper, based on remote sensing data, supplemented by a little regularly meteorological data, verified by the day evapotranspiration data obtained through surveying and combined with the energy balance model, we carry on a series of study about day evapotranspiration estimation at Sanhuajian of the Yellow River where the terrain is complicated and the surface type is multiplicity. Main study contents and fruits of this study are as follows:1. The precision of land surface temperature(LST) retrieved by remote sensing is a key factor to restrict the estimation of evapotranspiration. In this paper, the factors, elevation, slope and aspect, etc, that influence on the precision of land surface temperature are considered by introducing separate and integrated method in the mountain- climatology to this study. Obtained revisory coefficient of topography by statistical analysis, corrected topographically the land surface temperature and verified the retrieved result by real measure temperature. The estimation precision of evapotranspiration when the land surface temperature by topographical revision input SEBAL is better than not considering topographical effect, the average relative errors of hydrological station decrease from 13.1% to 9.5%.2. Through the contrast and analyse of evapotranspiration between that estimated from multi-original Remote Sense Images and a single Remote Sense Images, the result is that estimation precision of evapotranspiration is inproved becaused of using multi-original Remote Sense Images. 3. The possibility relations can be concluded from the key influence factor of evapotranspiration through analyzing between the condition of water supply, difference of temperature from the surface to atmosphere, NDVI and the day evapotranspiration. The other similar research district can choose the corresponding formula to estimate the evapotranspiration after confirming the decisive factors according to the actual conditions.