The classification of tropical fragmented landscapes and moist forested areas is a challenge due to the presence of a continuum of vegetation successional stages, persistent cloud cover and the presence of small patches of different land cover types.
The Upper Guinean forests of West Africa have experienced a dramatic decrease of their original extent. Logging, mining, hunting and human population growths are still placing extreme stress on this biodiversity hotspot. Therefore, forest mapping in this area is extremely important for conservation and planning issues and with respect to the emerging REDD+ (Reducing Emissions from Deforestation and Forest Degradation in Developing Countries) program. Detailed mapping is also important for national planning in many tropical countries, where local communities rely on woody vegetation as a primary source of products and energy.

Gaia Vaglio Laurin and Riccardo Valentini (IAFENT Division), in a study recently published on the International Journal of Applied Earth Observation and GeoInformation investigated the potential of combining optical and radar sensors for discriminating land cover classes for a moist tropical area which has never been classified by high resolution remote sensing before.
The study area (covering a total of 7749 km²) spans the border of Sierra Leone and Liberia and includes the recently established Gola Rainforest National Park (GRNP) and most of the Liberian Gola National Forest. To classify one such study area in West Africa researchers integrated the optical sensors Landsat Thematic Mapper (TM) and the Advanced Visible and Near Infrared Radiometer type 2 (AVNIR-2) with the Phased Arrayed L-band SAR (PALSAR) sensor.
The analysis resulted in fine and accurate vegetation mapping in a previously untested region, exploiting all available sensors synergies and highlighting the advantages of each dataset.

Read the integral version of the paper.