Outlook for future satellites and data use in forestry

Abstract

The outlook for future satellites and data use is assessed in this article by reviewing the experience to date with satellite sensors, primarily the LANDSAT multispectral scanner, which have provided data for over a decade, and by estimating the extent to which second‐generation satellite data sources will be able to overcome present limitations. The principal improvement of second‐generation data sources is in spatial resolution, but the improved spectral bands and radiometric resolution of the Thematic Mapper will also be important for some information requirements. Synthetic aperture radar data will provide additional new opportunities in the 1990s. The ability of new sensors to overcome present limitations is discussed with respect to generating forest inventories and updating inventories for changes caused by logging, burns, insect damage, and regeneration. Applications of satellite data for generating change information are likely to be more important than those for creating initial forest inventories. Even second generation satellite data will not generally provide sufficient detail for the initial inventory, although they may provide a good high‐level stratification as a first stage. It will take considerable effort in research, development, and technology transfer to realize the potential of second generation satellite sensors for forestry applications. These topics are discussed in some detail to emphasize the most important problem areas including geometric corrections; the addition of context, texture, and ancillary information to the image classification problem; and appropriate matching of technological capabilities to user needs. The future of data use will depend heavily on availability and ease of use of the data. If there is a good supply of reasonable‐cost data and the data are easy to obtain and easy to use, there is little doubt they will be used to an increasing extent to improve forest management around the world.