One of the main ways we can do to track ecosystem status is by looking at its extent change over time, as decreasing extent is associated with reduced carrying capacity, niche diversity, and resources, leading to increased competition, predation, and threats. In our recent study, we demonstrated a new method of using satellite data to estimate ecosystem extent and trend that uses all Landsat images available over the region while requiring no field data.
We used Hukaung Valley Wildlife Sanctuary as our case-study area. Here, goldmining has become an important source of income for local communities, but these mines have resulted in extensive deforestation. We aimed to quantify forest extent change in the region between 1999-2018.
By taking advantage of Google Earth Engine’s capacity to run more than 650 classification models based on Landsat imagery, we then used Generalised Additive Mixed Models (GAMMs) to calculate ecosystem extent trend through time while accounting for factors that may affect our estimate and returning results with explicit confidence intervals. We found a significant decline of forest extent at a rate of 0.274±0.078% per year within our study region.
Our methods will allow us to provide better estimates of ecosystem change and can be applied to any ecosystem globally where satellites are able to detect change. We hope the methods here will be used to inform global conservation targets, such as the UN Sustainable Development Goals, the Aichi Biodiversity Targets, and the IUCN Red List of Ecosystems etc., providing more robust data that can be used to drive policies.
Lee, C.K.F., Nicholson, E., Duncan, C. & Murray, N.J. (2020) Estimating changes and trends in ecosystem extent with dense time‐series satellite remote sensing. Conservation biology.
Calvin Lee 