University College London Featured PhD Programmes
University of Sussex Featured PhD Programmes
University of Oxford Featured PhD Programmes
University of Oxford Featured PhD Programmes
The Hong Kong Polytechnic University Featured PhD Programmes

Shaping the future of the UK’s landscapes – building a decision support system for sustainability


Project Description

Project background

The landscapes of the UK serve varied functions – e.g. supporting farmers’ yields, forestry, biodiversity and recreation. For the future, there are several major questions. What should determine the nature of our landscapes in coming decades? How can landscapes be resilient to climate change and be supported sustainably? There are major challenges to building a system that can inform us in detail about these questions but there are also major opportunities with new data infrastructure, high resolution earth observation data, and powerful analytical tools and models.

Currently, we can analyse and model land use either at national scale at coarse resolution, or at high resolution on a small scale. Fine-scale pattern is important in driving land use decisions, but a national system is critical for policy makers. The challenge here is to develop tools and algorithms that can scale between field and UK, to bridge the scale/resolution trade-off that currently limits decision support.

Research questions

This project will use a range of data sets, statistical methods and machine learning approaches to address the following questions:

1. What is the potential to simplify models and representations of landscapes? Can these simplifications be used to link models of landscapes at different scales? What extra uncertainty is introduced into the decision-making process by these simplifications?

2. How can information be transferred across scales? Methods will be investigated for using multiple datasets at different scales to inform the modelling through calibration and validation.

3. For exemplar regions across the UK, what is the current balance of agricultural yield, forest biomass accumulation, soil carbon storage and soil moisture? What are the sensitivities of these key outputs to global change (climate, CO2) and how can decisions be taken to optimise these outputs? How should the techniques developed be used to support policy decisions in practice?

Methodology

We will use an existing landscape model known as DALEC as an exemplar. DALEC is a coupled carbon-water model developed by Williams (one of the supervisors) which is currently incorporated in a model-data fusion framework CARDAMOM. The objective will be to improve CARDAMOM by developing statistical methods so that models and data can be linked across different scales. This linkage will enable better decision-making as users will be able to access modelling results at multiple resolutions within the same framework. Modelling dependencies between scales enables the transfer of information across resolutions so that insights and data at a coarse scale can inform insights at a fine scale and vice versa. This transfer of information will ultimately result in a reduction in uncertainty as the data and models are being used more efficiently. A range of spatial data from earth observation, soil maps, land use maps and climate models will be integrated with the modelling to support calibration, validation, uncertainty analysis and scenario testing.

Our aim is to understand decision making in the context of a few key factors: agricultural and forest yield, soil carbon storage, and soil moisture. These three factors link (i) farm incomes, forest industry; (ii) climate treaty obligations regarding carbon storage; and (iii) hydrology, drought risk for crops, and flood risk. While these are only a subset of ecosystem services, they present a core set of economic and ecological trade-offs as a proof of concept for this studentship. Other factors can be introduced if time allows.

This project arises from a major new UK research programme on ‘Evidence Based Decisions for UK Landscapes’, with the involvement of Research Councils, the UK government, and academia. The student will connect to this activity through attendance at relevant workshops and inter-disciplinary research. The student will also share the results of the project with the National Centre for Earth Observation (Williams is a member).

Training

A comprehensive training programme will be provided comprising both specialist scientific training and generic transferable and professional skills. The student will receive advanced training in statistical methods for modelling and simulating uncertainty through the Academy for PhD Training in Statistics and the Maxwell Institute Graduate School. Williams’ group of researchers and PhD students will provide a supportive environment for model-data fusion activities linked to global change ecology. There will be opportunities to collaborate with the statistics group in the School of Mathematics and with experts in uncertainty quantification through the Alan Turing Institute where Wilson and Dent are fellows (including the possibility of the student visiting the Institute through its “enrichment scheme” PhD visitor programme).

Funding Notes

Project funded through the E4 DTP based in Edinburgh: View Website

Find more information on the application process here: View Website

Requirements

A background in statistics, data science, ecological and/or environmental science, or related biological discipline is favoured, but transfers from physical sciences are possible. Strong quantitative skills are vital, and experience or interest in data analytics would be helpful. An interest in ecological and environmental issues would be advantageous.

References

References

Smallman, T.L. and Williams, M., 2019. Description and validation of an intermediate complexity model for ecosystem photosynthesis and evapotranspiration: ACM-GPP-ETv1. Geoscientific Model Development, 12(6), pp.2227-2253.

O’Hagan, A., 2006. Bayesian analysis of computer code outputs: A tutorial. Reliability Engineering & System Safety, 91(10-11), pp.1290-1300.

Cumming, J.A. and Goldstein, M., 2010. Bayes linear uncertainty analysis for oil reservoirs based on multiscale computer experiments. O’Hagan, West, AM (eds.) The Oxford Handbook of Applied Bayesian Analysis, pp.241-270.

How good is research at University of Edinburgh in Mathematical Sciences?
(joint submission with Heriot-Watt University)

FTE Category A staff submitted: 56.80

Research output data provided by the Research Excellence Framework (REF)

Click here to see the results for all UK universities

Email Now

Insert previous message below for editing? 
You haven’t included a message. Providing a specific message means universities will take your enquiry more seriously and helps them provide the information you need.
Why not add a message here
* required field
Send a copy to me for my own records.

Your enquiry has been emailed successfully





FindAPhD. Copyright 2005-2019
All rights reserved.