ESTIMATION OF AREAL EVAPOTRANSPIRATION BY REMOTE SENSING 
AND GIS TECHNIQUES 
Chung-Hyun, AHN 
Division of Global Environmental Information, 
Svstems Engineering Research Institute 
P.O. BOX 1, Yusung-Gu, Taejon, 305-600, KOREA 
e-mail:hyun@serimail.seri.re.kr 
Commission IV 
KEY WORDS: GIS, Evapotranspiration, LANDSAT TM, Energy Balances Mapping 
ABSTRACT 
Remote Sensing data with ancillary ground-based meteorological data provides the capability of computing three of 
the four surface energy balance components(i.e. net radiation, soil heat flux and sensible heat flux) at different 
spatial and temporal scales. As a result, this enables the estimation of the remaining term, latent heat flux. One of the 
practical applications with this approach is to provide evapotranspiration maps over large areas. This results could 
estimate and reproduce areal evapotranspiration over large area as much as several hundred square kilometers. 
Moreover, some calculating simulations for the effects of the land use change on the surface heat flux has been mad 
by this method, which is able to estimate evapotranspiration under arbitrary presumed conditions. 
From the simulation of land use change, the result suggests that the land use change in study area can be produce 
the significant changes in surface heat flux. This preliminary research suggests that the future research should 
involve development of methods to account for the variability of meteorological parameters brought about by 
changes in surface conditions and improvements in the modelling of sensible heat transfer across the surface- 
atmosphere interface for partial canopy conditions using remote sensing information. 
1. Introduction 
An important application of remote sensing homogeneous area surrounding the instruments but the 
information is in the evaluation of the energy and water results are not applicable to large diverse areas. 
budgets of natural and agricultural land surfaces. One of Although in recent years some sophisticated models 
the main objectives of some recent hydrometeorological are being used in an operational mode, the amounts of 
studies has been to test the feasibility of evaluating the information required and the fine tuning of some of the 
surface energy and water balance of regional scales with geophysical parameters may make it difficult to 
models using satellite data. Because the surface energy implement in area where there is limited meteorological, 
and water budgets have important implications in plant physical, and soils information. . The only visible 
modeling a wide range of geophysical processes, means of mapping the spatial distribution of 
particular focus has been placed on obtaining reliable evapotranspiration on regional or local scales is using of 
estimates of evapotranspiration at various spatial and remote sensing data. 
temporal scales. Evapotranspiration is generally estimated by using 
The magnitude of evapotranspiration has a broad thermal infrared data acquired by satellite based sensors 
range of applications in plant physiology and irrigation and conventional ground-based meteorological data as 
practices and to regional and global scale hydrology and inputs to a one-dimensional boundary layer model. 
meteorology. For many applications, accurate values of Although remote sensing may provide information 
daily evapotranspiration are necessary from field science relating to evapotranspiration at the field scale(e.g.. 
of order 10km* to mesoscale of order 10km> The vegetation index, surface temperature, albedo), local 
evaporation of water from soil and plant surface is a meteorological conditions such as wind speed, air 
component of the surface energy balances that is of both temperature, water vapor pressure etc, cannot be 
theoretical and practical interests. Conventional ground- assessed with remote sensing. In addition, remote 
based methods for estimating evapotranspiration, such sensing data from satellite are essentially instantaneous 
as Bowen ratio, provide accurate measurements over a and may only be available once a day. This has resulted 
30 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B4. Vienna 1996 
in th 
instant 
the wh 
Re 
intens: 
popul 
impac! 
short. 
incom 
and it 
Be 
evapot 
energy 
chang 
chang 
from 
means 
latent 
tempe 
Th 
latent 
expres 
an att 
values 
residu 
GIS te 
balanc 
reclan 
2. De 
an 
Th 
compx 
storag 
neglig 
where 
is the 
unit 0