Principal Investigator:
Peter Potapov

CO Investigators:
Matthew Hansen

Sponsor:
USGS

Time period:
2020
to
2021

Researchers:

Abstract:

The main objective of the proposed activity is to enhance the capacity of the Ministry of Environment (Ministerio del Ambiente, MINAM) of Peru in national satellite-based land cover monitoring. To achieve this objective, we will strengthen the capacity of MINAM in Landsat ARD time-series data processing and characterization in order to develop (a) annual forest extent and dynamic maps and sample-based area estimates that meet international standards for reporting; (b) land cover maps suitable for national land cover change reporting for GHG emissions estimation; and (c) operational near-real-time forest monitoring capability. Specific activities will include:

a. Building capacity for forest and other land cover themes monitoring using Landsat ARD. The Landsat ARD data and tools developed by the UMD team have a potential to greatly improve the efficiency and accuracy of annual land cover mapping. The Landsat ARD will allow MINAM to perform land cover mapping and change detection independently using existing hardware capacity. This objective will be achieved through development of the steps-by-step training materials and providing remote training on ARD processing and characterization system deployment and operation.

b. Application of UMD analysis-ready products for national applications. The UMD team have developed a number of global and regional analysis ready products that can greatly improve the land cover mapping and monitoring capacity of MINAM without the need to perform in-house satellite data analysis. Such products include annual forest change detection, forest cover and height maps, annual and seasonal water extent, global crop area, and near-real-time forest loss alert products. The UMD team will provide remote training on using these products for national land cover monitoring.

c. Sample analysis application for national land cover area and change reporting. Sample analysis is a recommended “good practice” method for estimation of unbiased area and uncertainty for land cover and change area reporting. The UMD have developed a set of tools for national sample analysis using (a) Landsat-based wall-to-wall maps to support stratified sampling design to improve sampling efficiency; (b) web-based tools for sample data visualization and interpretation; and (c) statistical tools for estimation of area and uncertainty from sample data. This project will improve the capacity of MINAM in sample analysis by providing novel tools and training by the UMD team.