Multiscale Landscape Pattern Monitoring Using Remote Sensing

Department of Agriculture, restoration project supported by the Collaborative Forest Landscape Restoration Program (CFLRP). The objective of the 4FRI is to restore fire-adapted conditions on 2.4 million acres of ponderosa pine (Pinus ponderosa) spanning four National Forests. Robust monitoring techniques are needed to assess the impacts of landscape-scale management and progress toward desired conditions, and to facilitate adaptive management. Remotely sensed information was identified as a major source of pre- and post-treatment data useful for evaluating changes in horizontal forest structure (or landscape pattern) at multiple spatial and temporal scales. With sponsorship from the Remote Sensing Steering Committee, the Remote Sensing Applications Center developed and tested a workflow and methods for assessing and monitoring landscape pattern using high-resolution imagery. The workflow and methods, which are applicable to other forests, consist of 1) creating a tree/nontree patch map of the analysis area from high-resolution imagery, 2) dividing the patch map into sublandscapes and computing landscape pattern indices (LPIs) for each one, 3) assessing the accuracy of the LPIs and calibrating them, and 4) summarizing and analyzing the LPIs at multiple scales to track change through time. In addition to describing the workflow and methods, this report also provides a comprehensive discussion of special considerations for implementing a landscape pattern monitoring program for other large geographic areas.