An alternative approach for detecting and monitoring selectively logged forests in the Amazon

A methodology is presented for estimating the area affected by selective logging in Amazonia's terra firme forests. Areas of land cleared for temporarily storing logs in the forest (log landings) were detected within soil fraction images derived from linear mixture models of forested areas. Based on field data, a 180 m timber extraction radius was assumed for all log landings. The extraction area calculated for detected log landings provides an estimate of the area potentially affected by logging. The resultant image classifications revealed excellent agreement with field data. This methodology can be a powerful tool for the monitoring and control of logging in the Amazon.