Optimal Resolution for Large-scale Vegetation Mapping Using Air- Borne Multispectral Data

There is an increasing interest in the use of high-resolution multispectral data (acquired by small format digital air-borne imaging sensors) for large-scale vegetation mapping. Recent studies have focused on the potential of air-borne multispectral data as a means of identifying plant communities. This study is concerned with establishing a formal relationship between spatial resolution and large-scale vegetation mapping. The study was necessitated by the fact that economic considerations take precedence over technical recommendations when selecting imagery data for large-scale vegetation mapping in Uganda. The question is whether there is any scientific justification in acquiring sub-meter imagery data for large-scale mapping of ‘low-value’ natural resources such as vegetation. Rigorous assessment of both thematic and geometric integrity of vegetation patches mapped from digital air-borne imagery data shows that a formal relationship between large-scale vegetation mapping and high-resolution imagery data can be established. The findings show that there is no scientific justification to acquire sub-meter imagery data for large-scale vegetation mapping. It is concluded that digital air-borne data acquired at spatial resolutions greater than sub-meter is optimal for cost-effective large-scale vegetation mapping. * Corresponding author.