] 
  
| 
| 
so | + -41 A +: 
| | 
I i 1 
2 | i 
AN | 
40 ft Dana it + L + 
A | 
; Y 
= @ / / IN 
Oo 30 + AHA} FA; — 
z | /[ £5 EN 
/ [ A d xt 
V V // “ON | a 
3 | $ | Sn 0e Le 
^ / h - N 
0 BJ i +H - ne . d iA | + 
i F& | 
| | 
| , Los 
v] 
D = t — A — A 
0 re op si — += — A 
8-Mar 24-Mar  9.Apr  25.Apr  TMay 27 May Qjun 28Jun — Wu 300! BAug IlAug 
NOVI 
  
  
  
  
  
  
  
  
|^ aliua nda Jan — 
  
5.Mar 24 Ma! 9.Apı 25Apr  nMay 27-May Qe 28 Jun M-Jui 30-51  f$.Aug  3*Aug 
Figure 2 Observed NDVI for non-irrigated (left) and irrigated sunflower (right) using high resolution satellite for Seville 1992 for 
selected fields 
  
30 free men | 
| 
5 ER ER | 
| 
| 
20 | i i re = e ep 
- | 
> | 
a | 
> | $ 
® | 1 A s. } 
| | A 
| | 
v | 3, / m Na i 
| , / 7v ® | 18 
$/ IE { 
| g-— á i leo sud 
" + 
| : T 1 
oi + = 4 + -—— b. ond 
8Mar 24.Mar 9Ap $.Api — "May 27-Mar Quen 28Jun — Ww 3 b-Aug — 31Avg 
  
  
  
  
   
| 
| 
l^ | 
X | | 
- M -4 "i d Rien 
le NI 
\ i \ i 
NINA 
X NN à 
BOTEN 
\ TiEN VEN 
+ T N 
SN N V Y 
Mm 
NE 
Hu } 
+ 
i 
i 
1 1 | 
i +14 erm cman San aT RE ER an uma Pm Lame ge etico a han cora med 
8 Mar 24 Mar 9 Apr 25 Api "May 27 May Q X^ 28 men LIP 30 suit S'Aug 3Aug 
Figure 3 Observed WDVI for non-irrigated (left) and irrigated sunflower (right) using high resolution satellite for Seville 1992 for 
selected fields 
5 EVAPOTRANSPIRATION MAPPING AND 
SIMULATION 
5.1 Estimation of evapotranspiration 
Remote sensing observations in the thermal infra red part of the 
spectrum (TIR) provide suitable measures of surface temperature. 
Since surface temperature is a state variable resulting from 
incoming and outgoing energy fluxes, it can be used to estimate 
surface energy fluxes. The partitioning of net available energy 
from incoming short and long wave radiation is governed largely 
by available soil moisture. On wet surfaces, available energy is 
consumed by evaporation of water, leaving little energy for 
heating the surface. On dry surfaces more (or all) energy is used 
for heating, thus resulting in relatively high surface temperatures. 
Surface temperature, as observed by TIR remote sensing, can 
therefore be used to estimate surface evapotranspiration (ET). A 
more exact quantification is, however, a complex procedure since 
evapotranspiration is determined by other factors as well, such as 
surface roughness, vegetation cover, vapour pressure deficit, etc. 
A method to determine the ET rate through the energy balance is 
implemented in the Surface Energy Balance Algorithm for Land 
surfaces SEBAL (Bastiaanssen,1995). SEBAL solves the surface 
energy balance pixel-by-pixel according to micro-meteorological 
theories. Usage is made of three derived remote sensing products: 
(i) surface reflectance of visible and near infra-red radiation, r,; 
(11) surface temperature, Ty, derived from thermal infrared remote 
sensing; and (iii) the Normalized Difference Vegetation Index 
(NDVI). The energy balance for land surfaces reads: 
Q -G,-H-AE=0 (Wm?) (4) 
where Q' is net radiation, G, soil heat flux, H sensible heat flux 
and AE latent heat flux, i.c. the amount of energy, À (J kg) 
required in the liquid-to-vapour transition of E (kg m? s^). 
The surface energy balance equation can be expressed into a 
latent heat flux density: 
JE = {1-n)KI- oT» &oT; - p.o/ (4) CT, - T) p/ra) (T, - T) 
Wm? (5) 
where r, is surface reflectance, KL (W m^] global solar radiation, 
€oT,'[W m^] long wave sky emittance with & being the apparent 
emissivity of the atmosphere, o the Stefan Boltzmann constant 
and T, [K] screen height air temperature, &oT, [W m?] long 
wave surface emittance with T, [K] the surface temperature, Pa 
[kg m] air density, cy [J kg! K^] air specific heat, r,, the mean 
aerodynamic resistance to heat transport in air, p, [kg m^] soil 
density, c, [J kg! K'!] soil specific heat, rg, the soil resistance to 
heat transport. 
The surface temperature is interpreted in energy balance studies 
as being the result of partitioning of net energy between latent 
heat and sensible heat. In SEBAL the difference between surface 
and air temperature (T, - Ty) is coupled linearly to surface 
temperature, and is obtained by inversion of the equation for 
sensible heat transfer after solving it for two extreme situations: 
one where H = 0 (wet), and one where AE = 0 (dry). These 
146 Intemational Archives of Photogrammetry and Remote Sensing. Vol. XXXII, Part 7, Budapest, 1998 
extre 
To an 
Basti 
the a 
clim 
AE 
In cc 
actua 
evap 
Az) 
With 
instal 
fracti 
total 
instar 
actua 
E»: 
5.2 
simu 
The : 
km. 
used 
CGN 
deptl 
Transpiration (mm/day) 
NS 
Figu 
crop