MSc thesis topic: Toward Efficient Forest Monitoring: Optimizing Sample-Based Approaches for repeated REDD+ Reporting

Since REDD+ payments are performance-based, countries must reliably measure and report their emission reductions. This makes the development of robust and transparent forest monitoring systems a high priority. According to IPCC guidance (2003), countries should aim to produce greenhouse gas (GHG) inventories that are neither significantly over- nor underestimated, with uncertainties minimized as much as possible.

Background

Given the complexity of land use and land-use change, especially in dynamic tropical regions, systematic approaches are needed to reduce uncertainty. A key part of this involves accurate estimation of land-use changes (also known as activity data), such as deforestation. Two main methods are recommended for tracking these changes: spatially explicit land-use change maps and sample-based surveys.

Land-use change maps are helpful for visualizing trends, but they are prone to classification errors. The latter lead to substantial bias if change areas are directly estimated from map pixel counts (Sandker et al., 2021). Therefore, sample-based surveys have increasingly been used to estimate activity data for REDD+ reporting, as they can provide statistically sound estimates and allow for uncertainty analysis. Many of these sample-based assessments have been retrospective—focusing on past land-use changes. For REDD+ reporting, there is a growing need for operational monitoring systems that can provide regular, up-to-date assessments. Such systems must be efficient and capable of detecting relatively rare events, like localized deforestation.
Despite the growing use of sample-based approaches, there are still important gaps in understanding which sampling strategies are most suitable for real-time, operational monitoring in the context of REDD+.

This thesis will explore different sample-based approaches for monitoring forest and land-use change in an operational context. It will investigate how existing sample data can be reused or adapted to support ongoing monitoring and reporting for REDD+. Several sampling design adjustments will be tested using simulated reference maps, with the aim of identifying practical and efficient strategies that can improve consistency and accuracy in continuous REDD+ reporting.

Objectives

• Quantify and compare area estimation of forest change using sampling options suited for continuous and operational monitoring.
• Evaluate the efficiency of sampling options in terms of statistical and practical implications.

Requirements

• Geodata processing (geoscripting, geotools)
• Statistical skills (e.g., using R or Python) or strong motivation to learn it.

Literature

• Sandker, M.; Carrillo, O.; Leng, C.; Lee, D.; d’Annunzio, R.; Fox, J. The Importance of High–Quality Data for REDD+ Monitoring and Reporting. Forests 2021, 12, 99.
• OpenMrv Sampling design for estimation of area and map accuracy.
• Arévalo, P., Olofsson, P., & Woodcock, C. E. (2020). Continuous monitoring of land change activities and post-disturbance dynamics from Landsat time series: A test methodology for REDD+ reporting. Remote Sensing of Environment, 238, 111051.
• Tang, X., Woodcock, C. E., Olofsson, P., & Hutyra, L. R. (2021). Spatiotemporal assessment of land use/land cover change and associated carbon emissions and uptake in the Mekong River Basin. Remote Sensing of Environment, 256, 112336.
• Tsendbazar, N., Herold, M., Li, L., Tarko, A., de Bruin, S., Masiliunas, D., Lesiv, M., Fritz, S., Buchhorn, M., Smets, B., Van De Kerchove, R., & Duerauer, M. (2021). Towards operational validation of annual global land cover maps. Remote Sensing of Environment, 266, 112686.
• Melo 2023 Satellite-based global maps are rarely used in forest reference levels submitted to the UNFCCC.

Theme(s): Modelling & visualisation; Integrated Land Monitoring