Supervisor
Professor David Edwards
Brief summary
Wood production has severe, growing, and likely underestimated environmental impacts on tropical ecosystems. Two key challenges to effectively predicting both the historic and future consequences of wood production are understanding the extent and intensity of wood harvests. Within tropical environments, wood is primarily sourced by selectively logging of rainforests, often over very large spatial scales and through both legal and illegal means. Although remote sensing efforts are increasingly able to detect small-scale degradation events, we still have a poor understanding of where wood has been harvested, in what volumes, and how this has changed through time. This PhD will use cutting-edge remote-sensing techniques and integrated-modelling approaches to: Estimate at high-resolution the spatial extent and locations of selectively logged forests through time; generate the first estimation of annual logging intensities (m3/ha) across tropical forests; develop models predicting the spatial and socio-environmental drivers of historic selective logging expansion and intensity, and potentially use these to predict future areas where selective logging expansion and intensification is most likely; and explore how accurately nationally reported harvest footprints and volumes align with already existing and candidate-developed measures of forest degradation/harvests.
Project Summary
Wood production has severe, growing, and likely underestimated environmental impacts on ecosystems around the world, especially in the tropics. Two key challenges to effectively predicting both the historic and future consequences of wood production are understanding the extent and intensity of wood harvests. Within tropical environments, wood is primarily sourced by selectively logging of rainforests, often over very large spatial scales and through both legal and illegal means. Although remote sensing efforts are increasingly able to detect small-scale degradation events, we still have a poor understanding of where wood has been harvested, in what volumes, and how this has changed through time. This limits our ability to understand both the ecosystem-scale consequences of different ways of producing wood, and the recovery potential of vast areas of tropical forest estate. Meanwhile, without accurate models capable of estimating the key historic drivers of selective logging and the associated intensity of harvest, our conservation policies will continue to be reactive rather than proactive into the future.
While detecting logging using remote sensing is fundamental for developing a consistent understanding of wood production over large spatial scales, a wealth of Food and Agricultural Organization of the United Nations (FAO) and national-scale statistical data describing wood-harvesting volumes and end-uses also exists. Combining these FAO and national-scale harvesting data with geospatial, remote sensing-derived estimates of harvesting footprints and volumes offers a unique opportunity for understanding larger-scale wood production dynamics. This integrated approach could enable a richer understanding of national-scale harvesting patterns, but it is unclear how well remote sensing-derived estimations of wood production and degradation will align with nationally reported harvesting volumes. This is particularly important to understand across different contexts, since countries are likely to vary substantially in how accurately they report domestic and exported timber volumes, in their harvesting intensities, and in their share of legal or illegal logging.
What will the successful applicant do?
This PhD will use cutting-edge remote-sensing techniques and integrated-modelling approaches to:
- Estimate at high-resolution the spatial extent and locations of selectively logged forests through time.
- Generate the first estimation of annual logging intensities (m3/ha) across tropical forests.
- Develop models predicting the spatial and socio-environmental drivers of historic selective logging expansion and intensity, and potentially use these to predict future areas where selective logging expansion and intensification is most likely.
- Explore how accurately nationally reported harvest footprints and volumes align with already existing and candidate-developed measures of forest degradation/harvests.
The candidate will have experience in using a suite of remote-sensing products (e.g. GEDI, LANDSAT), and be aware of their potential limitations and uses. The successful applicant will join the Centre for Global Wood Security, which is directed by Edwards, and will be embedded within the Cambridge Conservation Initiative housed in the David Attenborough building.
References
Edwards DP, Tobias J, Sheil D, Meijaard E, Laurance WF (2014) Maintaining ecosystem function and services in logged tropical forests. Trends in Ecology and Evolution 29: 511-520. doi.org/10.1016/j.tree.2014.07.003
Hethcoat, M.G., Edwards, D.P., Carreiras, J.M.B., Bryant, R.G., França, F.M., Quegan, S., 2019. A machine learning approach to map tropical selective logging. Remote Sensing of Environment 221, 569–582. doi.org/10.1016/j.rse.2018.11.044
Vancutsem, C., Achard, F., Pekel, J.-F., Vieilledent, G., Carboni, S., Simonetti, D., Gallego, J., Aragão, L.E.O.C., Nasi, R., 2021. Long-term (1990–2019) monitoring of forest cover changes in the humid tropics. Science Advances 7, eabe1603. doi.org/10.1126/sciadv.abe1603