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


Lake Maggiore - Italy, 3-6 April 2000

Sub-pixel mapping of Sahelian wetlands using multi-temporal SPOT-VEGETATION images

Jan Verhoeye, Robert De Wulf
University of Gent
Faculty of Agricultural and Applied Biological Sciences
Coupure Links 653
9000 Gent, Belgium

Paper (pdf file, 169 k)

Africa supports some of the world’s largest swamps. Some of the most extensive of these occur within the Sahelian zone: the floodplains of the Senegal River, the Interior Niger Delta, the mid Niger floodplain, the Chad Basin (comprising Lake Chad, the Logone-Chari floodplains and Hadejia, Jama’are and Komadugu floodplains).

During the last 40 years the growth of the human population and the associated increased demand for irrigation water and arable land, has put increasing pressure on these wetland ecosystems. In an effort to improve the economic situation of the local populations, large-scale hydro-agricultural projects are being planned, comprising large dams and irrigation schemes, which threaten the wetlands.

These wetlands are very extensive and they constitute very dynamic eco-systems. These characteristics make the wetlands suitable objects for study using satellite images with coarse spatial but high temporal resolution, such as SPOT-VEGETATION images.

At coarse spatial resolutions pixels inevitably become mixed. Traditional classification techniques are ’hard’ in the sense that a single pixel is assigned to a single land cover class. For mixed pixels ‘soft’ classifiers can be used, which assign a pixel to several land cover classes in proportion to the area of the pixel that each class covers. The result consists of a number of fraction images. This step will be called sub-pixel classification. The next step, sub-pixel mapping, consists of assigning the land cover fractions to the sub-pixels and results in a hard classification at a higher resolution than the original SPOT-VEGETATION images.

The sub-pixel classification is based upon the hypothesis of the linear spectral unmixing model: the image spectra are the result of mixtures of surface materials, shade and clouds, and each of these components is linearly independent from the other. In fact the value of each pixel can be modelled as a linear combination of the land cover spectra present in the image and their respective fractions. Multiple linear regression techniques can be used to solve these models. This technique has been applied to the combination of a high-resolution SPOT-XS image and a time-series of low-resolution SPOT-VEGETATION images. The results show that presented method is capable of accurately estimating the sub-pixel fractions.

The key problem of sub-pixel mapping is determining the most likely locations of the fractions of each land cover class within the pixel. Assuming a spatial dependence within and between pixels can solve this. The coarse pixels are divided into smaller units and the land cover is allocated to the smaller cells within the larger pixels, in such a way that spatial dependence is maximised. This problem can be cast into a traditional linear programming format. The spatial dependence can be represented by various measures, varying from simple averages to values calculated using spatial statistics. Preliminary results indicate that the classification accuracy at 500 m resolution closely approaches the original accuracy at 1000m.