Supervisors:

Richard Essery (richard.essery@ed.ac.uk), Mat Williams (mat.williams@ed.ac.uk)

Eleanor Blyth, CEH Wallingford (emb@ceh.ac.uk)

Predictions of changes in permafrost distributions due to climate warming using outputs

from climate models differ considerably in their severity, but almost all suggest that there will

be significant permafrost degradation over the twenty-first century. This will have major

impacts on the hydrology, ecology and infrastructure of cold regions, so reduced uncertainty

in predictions of permafrost change would be valuable. Modelling discontinuous permafrost

in marginal regions, which are particularly vulnerable to warming, is particularly challenging.

In these regions, there is a delicate balance between patterns of frozen ground, snow and

vegetation. Increases in either tall vegetation, which traps more wind-blown snow, or thick

moss layers will change the ground thermal regime through increased insulation. Changes in

permafrost distributions, in turn, will change the local hydrology and so influence vegetation

distributions. These landscape processes are not captured in the current land surface

schemes that are used to provide energy and mass boundary conditions in climate models.

This project will make use of the JULES community land surface model

(http://jchmr.org/jules/), which will be the land surface scheme incorporated in the NERC

QUEST Earth System Model (QESM; http://www.quest-esm.ac.uk) and is managed by a

consortium from the Met Office, the Centre for Ecology and Hydrology and several

universities, including Edinburgh. JULES can represent the near-surface hydraulic and

thermal processes of organic and mineral soils in freezing conditions, but it does not yet have

representations of deep soils or ice bodies necessary for modelling long-term permafrost

dynamics; these will be developed and implemented by the student. The model will be run at

high spatial resolutions with representations of snow redistribution, variable solar heating on

slopes, lateral water flows and vegetation dynamics for selected landscapes. Outputs from

this model will be used to develop parametrizations of permafrost dynamics for large-scale

climate models that cannot explicitly resolve landscape processes. Hypotheses to be tested

by modelling in this project include:

• melting of permafrost will lead to changes in distributions of tundra vegetation

• changes in vegetation will lead to changes in ground insulation and so influence

permafrost distributions

• warming air temperatures and decreasing snow cover could lead to either

increasing or decreasing ground temperatures, depending on the balance between

the two

The range of techniques to be used in this project will provide superb training

opportunities for the student in climatological, hydrological and ecological modelling, model

integration and evaluation of models with observations. The supervisors have wide

experience between them in these areas. A placement at the Centre for Ecology and

Hydrology in Wallingford will provide experience of work in a large public sector research

organization. Although the modelling will make use of existing datasets, opportunities may

arise for the student to participate in field research. This project will suit a student with a good

degree (or nearing completion) in a physical or numerate subject, some experience of

computer modelling and ability to work both independently and in collaboration. Experience

of Fortran programming and knowledge of environmental or climate processes are desirable.

Background:

ACIA, 2004. "Impacts of a Warming Arctic: Arctic Climate Impact Assessment", Cambridge

University Press. Available online at <http://www.acia.uaf.edu/pages/overview.html>