Super typhoon Haiyan makes landfall
Super typhoon Haiyan
Subscribe here to receive special images

VEGETATION - 2000

Lake Maggiore - Italy, 3-6 April 2000


Interest of MIR data from VEGETATION for the monitoring of climatic phenomena impact on crops, a case study

Thierry Fourty

Herve Kerdiles, Frederic Biard
(Geosys, France)

Paper (pdf file, 244 k)

Remote sensing approaches applied to agriculture are undergoing profound changes. Actually, two kinds of approaches, requiring both different technologies or tools, are developed.

Precision farming is the most promising approach. However, mainly due to lacks in the understanding of radiative transfer processes occurring inside the crops (vegetation/soil) and atmosphere, results remain approximate.

The second approach is the monitoring of crops on a global basis in both time and space dimensions. This approach is very interesting for private companies since it doesn’t require new tools or sensors to respond to the emergence of economic markets.

To show and test an operational use of such an approach, we propose to investigate the use of data acquired with the VEGETATION sensor to monitor crops at a regional scale. This study will focus on the potential of the MIR band to show crop water stress in a region located around the Aude in south west of France, showing an important difference in rainfall between 1998 and 1999.

From an industrial point of view, crop monitoring based on remote sensing requires 2 mains stages.

1) The first stage consists in determining and delimiting homogeneous zones in terms of crop or their behavior.

Three different approaches will be investigated. The first one consists in a photo-interpretation based on color compositions. The second one consists in fitting linear relationships between bands to show a possible significance of residuals in terms of type of crops (or land cover). The last one consists in doing an unsupervised classification.

For each approach, the strengths and weaknesses of the additional use of MIR information will be evaluated.

2) The second stage consists in processing previously defined zones to detect abnormalities in crop behavior, related to the crop water status due to rainfall.

For this purpose, based on the results of the first stage, defined zones provided by the best approach will be selected. Then, possible relations between radiometric information, including or not MIR (crop water sensitive) and the measured data of cumulative rainfalls between 1998 and 1999 will be researched.

Results will be discussed with regards to industrial applications, sensibility and accuracy.