ZONING OF TIMBER EXTRACTION IN THE BRAZILIAN AMAZON

The state of Para (1,248,042 km(2)), in eastern Amazonia, produces 65% of Brazil's roundwood. Logging is now spreading across this state in an unplanned and unregulated fashion. Using a geographic information s ystem (GIS), we combined and analyzed spatial information on forest co ver legal land classification, log processing industries, biodiversity , and infrastructure for the entire state. We first used this GIS, in combination with economic data, to analyze the spread of logging activ ities in Para. We found that in the mid-1990s, the potential already e xisted (in economic terms) to harvest timber from 80% of Para's forest ed lands: 21% of Para's forest was accessible for the harvest of all c ommercial species, including those of low value; nn additional 30% was accessible for logging a select group of medium-value species; and a final 29% of the the state's forest was accessible for the logging of mahogany a high-value species. Although 29% of Para's lands are legall y protected from logging, protection is weak: almost three-quarters of these ''protected'' lands fall within the zone in which timber can no w be profitably harvested. We also used the GIS to develop a rationale for zoning where logging might be permitted, as well as prohibited, i n Para. First, we noted that 19% of the state contained lands without timber (12% deforested, 6.3% nonforest vegetation types, and 0.7% wate r). The land that we designate for logging, based both on economic and conservation considerations, would be approximately 32% (400,000 km(2 )) of the state. This would include areas where logging is already und erway (24%), production reserves and buffer areas (3%), and remote are as with no conservation restrictions (5%). We propose that the remaini ng area (49%, oi 611,540 km(2)) be protected from logging, at least fo r the time being. These lands include forested areas where logging is already ''officially'' prohibited (28% of Para) as well as nonprotecte d areas with a high conservation priority (21%) (i.e., areas that are especially rich in biodiversity).