IWRA Executive Office

abstract 248

Crop water stress detection from remote sensing using the SSEBI-2 algorithm

Author(s): a case study in Morocco
Author(s): Claire Jacobs(1), Gerbert J. Roerink(1), Ali Hammani(2)
(1) ALTERRA, Wageningen, The Netherlands (2)Institut Agronomique et Vétérinaire Hassan II, Rabat, Morocco

Keyword(s): water stress, irrigation, evapotranspiration, monitoring, remote sensing, SSEBI-2

Article:

Poster:

Session: Poster session 4

AbstractAccurate estimations of evapotranspiration (ET) give insight in

water use, crop water stress, and production levels of crops. For planning and management of water resources,

water managers and specifically irrigation engineers need to dispose of temporal and spatial values of ET.

The

use of remote sensing data has proven to be useful to estimate ET for large spatial areas. A simplified method is

provided by the SSEBI algorithm, developed to estimate surface fluxes from remote sensing images. SSEBI

provides accurate ET estimations and has the advantage that limited field data are required (Roerink et al., 2000).

So far, SSEBI is applied to separate remote sensing images, representing individual days within a season. A method

for temporal integration of the images into a constant time series of daily ET maps has been lacking. The current

paper describes the application of a new, easy to use procedure that applies such a temporal integration: the SSEBI

-2 algorithm. SSEBI-2 derives daily actual and potential ET maps from remote sensing. The approach allows a

quick temporal and spatial assessment of seasonal water consumption for large river basins or irrigation systems,

with a minimum amount of input data required.

The SSEBI-2 algorithm is applied to the Oum Er Rbia Basin in

Morocco, which includes the Tadla irrigation perimeter, one of the most productive irrigation areas in Morocco. A

number of 26 low resolution MODIS images of 2006 has been used to derive evaporative fraction maps. These

maps are subsequently integrated in equal intervals (days, decades, months), after which for each interval the net

radiation flux is calculated on the basis of standard meteo data, combined with remote sensing input. Finally actual

and potential ET on a daily basis have been derived for the Tadla irrigation scheme. For validation, surface flux

measurements are used.

A comparison of actual and potential ET reveals the occurrence of crop water stress

at specific moments in the irrigation season of 2006 in Tadla. This information can be used to adapt irrigation

strategies or propose alternative cropping calendars. In addition, when looking at a basin scale, Tadla is consuming a

large amount of water within the whole river basin, leaving little room for agricultural perimeters located downstream

in the basin.

Conclusions

Although the project is still under way, promising results have been obtained

so far, which demonstrate the usefulness of SSEBI-2 to derive constant time series of evapotranspiration for

irrigation assessments, in a way that is understandable and reproducible by non remote sensing experts.

The

use of MODIS appears highly suitable for the calculation of ET time series, due to its (i) high temporal resolution, (ii)

standard atmospheric correction and (iii) standard geometrical correction. An additional advantage is the easy access

of MODIS images (free of charge through internet) which provides the opportunity for routine processing of images.

This is promising for operational crop monitoring applications that require processing at a near-real-time

basis.

Acknowledgements

The research described in this paper is conducted in the framework of the

AquaStress project (2005-2009), an EU funded integrated project

(http://www.aquastress.net/).

References

Roerink, G.J., Z. Su and M. Menenti, 2000. S-SEBI: A simple

remote sensing algorithm to estimate the surface energy balance, Phys. Chem. Earth, Vol. 25, No. 2, pp 147-

157.

Events

The congress

Programme

Practical Infos

Crop water stress detection from remote sensing using the SSEBI-2 algorithm