I have recently finished my PhD at the School of GeoSciences, which was supervised by Prof. Mathew Williams and Prof. John Grace. My current research interests are landscape and ecosystem ecology, forest ecology, biogeochemical cycles, remote sensing and GIS, ecophysiology and plant productivity. Other interests include terrestrial carbon accounting and carbon sequestration, international carbon policy (REDD+) and policy implementation.
My project title was " Carbon dynamics in African miombo woodlands: from the leaf to the landscape" and my research lies within the global change research group. The aims of my PhD were:
1. Investigate seasonal limitations to photosynthetic productivity of miombo woodland tree species:
Miombo woodlands are seasonally dry woodlands covering a large extent of southern Africa. Very little is known of their leaf-level productivity and seasonal fluctuations in photosynthesis, so that it is unclear to what extent seasonal fluctuations occur, and whether productivity is limited by stomatal limitations on leaf gas exchange or leaf trait variations between dry and wet seasons. Future climate change predictions are for increased temperatures and prolonged drought in Africa, increasing the need for a more detailed understanding of how miombo woodland productivity responds to climate variability. To shed more light on seasonal patterns of photosynthetic productivity, I took ecophysiological and hydrological measurements at the leaf scale on key tree species in two seasons, at the end of the dry season and at the end of the wet season. Leaf traits were also determined. I found that large inter-specific seasonal variations in productivity occurs between miombo woodland tree species; some species were able to maintain photosynthetic productivity between seasons while others were not. These inter-specific differences were attributed to varied access to soil water, hydraulic conductivity, and photosynthetic capacities related to varied leaf developmental stage at the time of measurement in the dry season. Furthermore, I found that productivity was not different between seasons when assessed on combined data, indicating that miombo woodlands are able to take advantage of the higher light environment during the late dry season when new leaves emerge.2. Determine how carbon stocks are distributed in a miombo woodland landscape, at what scales does variability occur, and what determines their distributions:
To increase our understanding of the distribution and spatial variation of vegetation and soil C stocks in a miombo woodland landscape we sampled along a 5 km transect, using cyclic sampling scheme to allow geostatistical analyses. Multiple variables, such as above-ground woody biomass, soil C, grass biomass, litter, and soil texture were sampled simultaneously to allow correlation analyses to be performed. To read the published results, please click here, or click the link on the left for more information.
3. Create a simple spatial model of deforestation and degradation to estimate risk of forest loss for a case study area in central Mozambique:
Biomass loss through deforestation and degradation is a spatially complex issue, caused by several proximate and underlying driving forces. In Africa, biomass loss has many drivers including charcoal production and shifting cultivation. I created a simple spatial model using the hypothesis that if forested areas are easily accessible, are cultivable, have extractable value and are unprotected (The ACEU hypothesis), the land area is at high risk of deforestation and degradation. Using GIS techniques I applied spatial data of the ACEU parameters to the case study site in central Mozambique, and created a risk map of areas more likely to have deforestation and degradation activites. The risk map was validated against a time series of biomass maps produced from radar backscatter. Using this model and the resultant risk map we can and ask fundamental questions of our understanding of drivers in the study site, and inform land managers and policy makers for targeted monitoring efforts. This study was developed in conjunction with Ecometrica, and has been implemented as an interactive on-line tool to demonstrate the outcomes of this project.
Below is a wordle of my thesis, showing you the most common words and themes of the research
This PhD was funded by a CASE studentship from the Natural Environment Research Council in conjunction with Ecometrica