Large datasets, such as the Global Land 1-km Advanced Very High Resolu
tion Radiometer (AVHRR) Data Set (Eidenshink and Faundeen 1994), requi
re compression methods that provide efficient storage and quick access
to portions of the data. A method of lossless compression is describe
d that provides multiresolution decompression within geographic subwin
dows of multi-spectral, global, 1-km, AVHRR images. The compression al
gorithm segments each image into blocks and compresses each block in a
hierarchical format. Users can access the data by specifying either a
geographic subwindow or the whole image and a resolution (1, 2, 4, 8,
or 16 km). The Global Land 1-km AVHRR data are presented in the Inter
rupted Goode's Homolosine map projection. These images contain masked
regions for non-land areas which comprise 80 per cent of the image. A
quadtree algorithm is used to compress the masked regions. The compres
sed region data are stored separately from the compressed land data. R
esults show that the masked regions compress to 0.143 per cent of the
bytes they occupy in the test image and the land areas are compressed
to 33.2 per cent of their original size. The entire image is compresse
d hierarchically to 6.72 per cent of the original image size, reducing
the data from 9.05 gigabytes to 623 megabytes. These results are comp
ared to the first order entropy of the residual image produced with lo
ssless Joint Photographic Experts Group predictors. Compression result
s are also given for Lempel-Ziv-Welch (LZW) and LZ77, the algorithms u
sed by UNIX compress and GZIP respectively. In addition to providing m
ultiresolution decompression of geographic subwindows of the data, the
hierarchical approach and the use of quadtrees for storing the masked
regions gives a marked improvement over these popular methods.