Publications

In this paper, we assess the potential of spatial, temporal and spatio-temporal deep learning methods for large-scale classification of land-use following tropical deforestation using dense satellite time series on the pan-tropical scale. Continental models performed better than the pan-tropical model, while overall spatio-temporal models performed better than spatial or temporal ones.

This study quantified post-deforestation land use across the tropics for the period 1990–2000. This dataset was then combined with a pan-tropical AGB map at 30 m resolution to refine emission factor from forest conversion by matching deforestation areas with their carbon stock before and after clearing and to assess spatial dynamics by follow-up land use.

This study assesses the impact of achieving cereal self‐sufficiency by the year 2050 for 10 sub-Saharan countries on GHG emissions related to different scenarios of increasing cereal production, ranging from intensifying production to agricultural area expansion.

Greenhouse gas emissions reduction from the land use sector requires that accurate, consistent and comparable datasets are available for transparent reference and progress monitoring. Through an online survey, we investigated stakeholders’ data needs for estimating forest area and change, forest biomass and emission factors, and AFOLU GHG emissions. Our results show that current open and freely available datasets and portals are only able to fulfil stakeholder needs to a certain degree. We also identify key elements for increasing overall transparency of data sources, definitions and methodologies.

Land use change in South America, mainly deforestation, is a large source of anthropogenic CO2 emissions. This study quantifies proximate drivers of deforestation and related carbon losses in South America based on remote sensing time series in a systematic, spatially explicit manner.