ALBERT

Description: National-scale assessments of post-deforestation land-use are crucial for decreasing deforestation and forest degradation-related emissions. In this research, we assess the potential of different satellite data modalities (single-date, multi-date, multi-resolution, and an ensemble of multi-sensor images) for classifying land-use following deforestation in Ethiopia using the U-Net deep neural network architecture enhanced with attention. We performed the analysis on satellite image data retrieved across Ethiopia from freely available Landsat-8, Sentinel-2 and Planet-NICFI satellite data. The experiments aimed at an analysis of (a) single-date images from individual sensors to account for the differences in spatial resolution between image sensors in detecting land-uses, (b) ensembles of multiple images from different sensors (Planet-NICFI/Sentinel-2/Landsat-8) with different spatial resolutions, (c) the use of multi-date data to account for the contribution of temporal information in detecting land-uses, and, finally, (d) the identification of regional differences in terms of land-use following deforestation in Ethiopia. We hypothesize that choosing the right satellite imagery (sensor) type is crucial for the task. Based on a comprehensive visually interpreted reference dataset of 11 types of post-deforestation land-uses, we find that either detailed spatial patterns (single-date Planet-NICFI) or detailed temporal patterns (multi-date Sentinel-2, Landsat-8) are required for identifying land-use following deforestation, while medium-resolution single-date imagery is not sufficient to achieve high classification accuracy. We also find that adding soft-attention to the standard U-Net improved the classification accuracy, especially for small-scale land-uses. The models and products presented in this work can be used as a powerful data resource for governmental and forest monitoring agencies to design and monitor deforestation mitigation measures and data-driven land-use policy.

Description: The goals of the Paris Agreement (PA) on collectively managing climate change can only be reached if all parties to the United Nations Framework Convention on Climate Change (UNFCCC) commit to actions supporting their Nationally Determined Contributions (NDCs). Developing-economy nations play a crucial role in reaching the PA targets, particularly in the Agriculture, Forest, and Other Land Uses (AFOLU) sector. However, developing country Parties also face several constraints in tracking and communicating progress towards their climate policy targets and implementation of their NDCs. The operationalization of Biennial Transparency Report (BTR) and Enhanced Transparency Framework (ETF) under the PA will bring stricter reporting timeframes and advanced transparency for all parties. With these requirements rapidly coming into force, addressing reporting gaps is now a pressing priority. The present study analyzes the NDCs, and Biennial Update Reports (BURs) submitted by developing country Parties under the UNFCCC. In an illustrative exercise, our in-depth analysis concentrates on reporting on the AFOLU sector and identifies issues impeding a comprehensive and comparable Global Stock Take (GST): (i) issues of consistency in reporting timeframes (ii) issues in transparency of reporting on mitigation sectors and on relevant progress indicators (iii) incomparability of methodological approaches proposed and used, and (iv) the implications of limited national capacity for transparent reporting. The UNFCCC and developed country Parties now have the opportunity of providing specialized support for developing country Parties. This could include tailored guidance to address gaps in both greenhouse-gas (GHG) emissions accounting, and reporting challenges, to ensure consistent, comprehensive, and transparent reporting to reinforce capacities moving forward following the next GST.

Description: Transparent, accurate, comparable, and complete estimates of greenhouse gas emissions and removals are needed to support mitigation goals and performance assessments under the Paris Agreement. Here, we present a comparative analysis of the agriculture forestry and other land use (AFOLU) emission estimates from different datasets, including National Greenhouse Gas Inventories (NGHGIs), FAOSTAT, the BLUE, OSCAR, and Houghton (here after updated H&N2017) bookkeeping models; Emissions Database for Global Atmospheric Research (EDGAR); and the US Environmental Protection Agency (EPA). We disaggregate the fluxes for the forestry and other land use (FOLU) sector into forest land, deforestation, and other land uses (including non-forest land uses), while agricultural emissions are disaggregated according to the sources (i.e., livestock, croplands, rice cultivation, and agricultural fires). Considering different time periods (1990–1999, 2000–2010, and 2011–2018), we analyse the trend of the fluxes with a key focus on the tropical regions (i.e., Latin America, sub-Saharan Africa, and South and Southeast Asia). Three of the five data sources indicated a decline in the net emissions over the tropics over the period 1990–2018. The net FOLU emissions for the tropics varied with values of 5.47, 5.22, 4.28, 3.21, and 1.17 GtCO2 year−1 (for BLUE, OSCAR, updated H&N2017, FAOSTAT, and NGHGIs, respectively) over the recent period (2011–2018). Gross deforestation emissions over the same period were 5.87, 7.16, 5.48, 3.96, and 3.74 GtCO2 year−1 (for BLUE, OSCAR, updated H&N2017, FAOSTAT, and NGHGIs). The net forestland sink was −1.97, −3.08, −2.09, −0.53, and −3.00 GtCO2 year−1 (for BLUE, OSCAR, updated H&N2017, FAOSTAT, and NGHGIs). Continental analysis indicated that the differences between the data sources are much large in sub-Saharan Africa and South and Southeast Asia than in Latin America. Disagreements in the FOLU emission estimates are mainly explained by differences in the managed land areas and the processes considered (i.e., direct vs indirect effects of land use change, and gross vs net accounting for deforestation). Net agricultural emissions from cropland, livestock, and rice cultivation were more homogenous across the FAOSTAT, EDGAR, and EPA datasets, with all the data sources indicating an increase in the emissions over the tropics. However, there were notable differences in the emission from agricultural fires. This study highlights the importance of investing and improving data sources for key fluxes to achieve a more robust and transparent global stocktake.