Risk modelling and decision-making for UK tree planting

Project description: Climate risk modelling of national tree planting to support commitments towards Net Zero emissions.

1. Linear programming

Utilised linear programming to identify optimal locations and species for tree planting across a range of possible futures. My results show that policy of which species to plant, and where to plant, differs drastically depending on the assumed climate and economic futures.

2. Risk modelling and optimisation under uncertainty

Implemented state-of-the-art stochastic optimisation algorithms to identify optimal locations and species for tree planting under uncertainty. In particular, my algorithm identifies combinations of tree planting that minimises the Conditional Value-at-Risk (CVaR). These results show that these stochastic optimisation algorithms can successfully reduce the risk profile of UK tree planting initiatives - but not by much.

3. Policy appraisal

Extended the stochastic optimisation algorithms to identify optimal mix of carbon dioxide removal technologies in the UK in the presence of an alternative carbon removal technology. My analysis further shows that tree planting is still preferable over comparable carbon dioxide removal technologies, even though it is inherently risky in its costs of implementation.

These results are currently in review in an academic journal, please do not cite. Part of a funded research project supported by NERC, QUEX Institute, the Alan Turing Institute and BBSRC.

Full GitHub repository: GitHub Repo