Rates of Disturbance vary by data resolution: implications for conservation schedules using the Alberta Boreal Forest as a case study

Abstract

Investigations of biophysical changes on earth caused by anthropogenic disturbance provide governments with tools to generate sustainable development policy. Canada currently experiences one of the fastest rates of boreal forest disturbance in the world. Plans to conserve the 330 000 km(2) boreal forest in the province of Alberta exist but conservation targets and schedules must be aligned with rates of forest disturbance. We explore how disturbance rate, and the accuracy with which we detect it, may affect conservation success. We performed a change detection analysis from 1992 to 2008 using Landsat and SPOT satellite image data processing. Canada's recovery strategy for boreal caribou (Rangifer tarandus caribou) states that ≤35% of a caribou range can be either burned or within 500 m of a man-made feature for caribou to recover. Our analyses show that by 2008 78% of the boreal forest was disturbed and that, if the current rate continues, 100% would be disturbed by 2028. Alberta plans to set aside 22% for conservation in a region encompassing oil sands development to balance economic, environmental, and traditional indigenous land-use goals. Contrary to the federal caribou recovery strategy, provincial conservation plans do not consider wildfire a disturbance. Based on analyses used in the provincial plan, we apply a 250 m buffer around anthropogenic footprints. Landsat image analysis indicates that the yearly addition of disturbance is 714 km(2) (0.8%). The higher resolution SPOT images show fine-scale disturbance indicating that actual disturbance was 1.28 times greater than detected by Landsat. If the SPOT image based disturbance rates continue, the 22% threshold may be exceeded within the next decade, up to 20 years earlier than indicated by Landsat-based analysis. Our results show that policies for sustainable development will likely fail if governments do not develop time frames that are grounded by accurate calculations of disturbance rates.