Remote Sensing

Please note this page in currently under construction

An NDVI vegetation index derived from a Landsat ETM+ image for 30th December, 2000 (registered Plan Vivo farmers’ plots are shown in blue)

Satellite imagery

The satellites used include the Landsat TM, ETM+ and MSS and Terra and Aqua.

Sensors on the satellites provide the information needed to create the images. These include the MODIS (Moderate Resolution Imaging Spectroradiometer), and the

Using the data from the satellites, indices can be derived. For example, the NDVI value can be found and images created.

In addition to the satellite images, ground truthing was used to verify results. GPS locations were also plotted onto the images to show major features such as rivers and roads.

Results

NDVA

The data from the Landsat can be used to derive the NDVA, which is a measure based on the amount of red and near infra-red to red light detected by the satellite sensor. Initial images (see above) show that the ratio of near infra-red to red light from the surrounding forested areas is much higher than for the cleared agricultural plots.

The resolution for this Landsat imagery is 30m pixels which is detailed enough to include the 1-2ha Plan Vivo plots. The image also shows areas of disturbance or deforestation, and if images taken at different times are compared can show temporal as well as spatial changes. Deforestation rates can then be derived from this which are required to calculate the baselines for forest carbon assessments. The deforestation within the Gorongosa National Park can also be tracked. From initial analysis, it is apparent that the encroachment stems mostly from around towns, and also alongside roads.

Land cover types

Different reflectivities of the vegetation can be used to identify different land cover types, for example a variety woodland types, grassland, water and agriculture. Eight land cover types were identified using this method.

Land use

Land use can be determined using remote sensing techniques. An initial study in 2004 (Spadeveccia et al 2004) shows land use maps of the area.

Future

As technnology progresses in this area, and images are more available, there is potential for increasing use of remote sensing capabilities for a project such as this one.

Specifically uses which have not yet been applied, include the SEVERI (Spinning Enhanced Visible and Infrared Imager) which is avalable on Metiosat-8, a geostationary satellite updated ever 15 minutes. This can be used to identify fires, and its resolution means it can detect hotspots > 5 ha in size.