Selecting representative high resolution sample images for land cover studies. Part 1: Methodology

This is the first of two articles which explore thr combined use of coarse and fine resolution data in land cover studies. It describes the developmen t and evaluation of an objective procedure to select a representative sampl e of tiles of high resolution images that complements a coarse coverage of an entire region of interest. The second article explores that use of the p rocedure for an accurate estimation of cover type composition at the region al scale. The Purposive Selection Algorithm (PSA) assumes that a relationsh ip exists between land cover compositions at the two spatial scales. It sel ects one tile at a time, seeking the sample which most closely resembles th e composition of the coarse resolution map. Two selection criteria were use d, fraction of cover types and contagion index. PSA was evaluated using two land cover maps for a 288 kmx165 km area in central Saskatchewan, Canada d erived from Landsat Thematic Mapper images (30 m pixels) and Advanced Very High Resolution Radiometer (AVHRR, 1000 m pixels), each divided into 64 til es. The performance of an intermediate sensor (480 m pixels) was assessed b y resampling the TM map. When using cover type composition alone, it was fo und that the procedure rapidly converges on a representative set of tiles w ith land cover composition very similar to the full coverage. The match bet ween the domain and sample cover type fractions was very close, with errors less than 0.002% once about 1/5 to 1/3 of the tiles were selected and no d iscernible bias in the selected sample. Compared to the TM whole area cover age, samples selected with AVHRR classification were as representative as t hose obtained using the TM map. The performance of samples selected by a co mbination of cover composition and contagion index responded to the charact eristics of individual tiles in terms of the selection criteria. A rigorous application of the algorithm with spatial heterogeneity measures such as t he contagion index is computationally very demanding. It is concluded that PSA provides an efficient and effective tool to select a representative sam ple for land cover studies in which both large area coverage and local deta il are desired. (C) Elsevier Science Inc, 2000.