Welcome to the training areas section of this website. This section will describe the derivation of a global network of training sites using Landsat imagery and various other data sources.

We selected 156 Landsat scenes, mostly from the Multispectral Scanner System (MSS), to identify over 9,000 pixels in the PAL data where we have high confidence that the labeled cover type occurs. An extended list of the path/rows we chose scenes from is available on the web. These training data provide the basis for carrying out a global land cover classification described in the second part of this website. They also provide data for validating other land cover classification products. The methodology and Landsat images used for deriving these training data for classification of AVHRR data at 8km resolution can also be applied to 1km AVHRR data and, in the future, MODIS data at 250m and 500m resolution.

Included in this section:

  • Introduction
  • Data
  • Methods
  • Results and Conclusions

    For further information, contact:

    Ruth DeFries
    Laboratory for Global Remote Sensing Studies
    Department of Geography
    2181 LeFrak Hall
    University of Maryland at College Park
    College Park, MD 20742-8225 USA

    email: rd63@umail.umd.edu