The response of the Arctic regions to changing climate

Lead Research Organisation: University of Edinburgh
Department Name: Sch of Geosciences


The anthropogenic burning of fossil fuels is affecting the climate system via raised atmospheric CO2 concentrations. But are there aspects of the natural Earth system that, when forced by imposed climate change, have themselves a major impact on the carbon cycle? Such climate-carbon (C) cycle feedbacks might be positive - that is, they may cause a reduction in the current capability of natural systems to mitigate anthropogenic CO2 emissions, or may even force natural systems to become direct sources of CO2. Such positive feedbacks are a major cause for concern, and their initiation could be regarded as a climate 'tipping point'. It has been hypothesised that the Arctic land surface could be one such 'tipping point', whereby global warming is sufficient to induce soil C losses greater than any extra draw-down of C through enhanced tundra (shrub) and boreal forest growth in a warmer, CO2-enriched environment. In addition, a warmer climate will impact on the on the energy and water cycles, with less snow cover in Northern Latitudes reducing surface reflectivity and so inducing additional warming. In reality the energy, water and C cycles are all strongly linked and need to be modelled interactively to provide robust estimates of the future. Global Circulation Models (GCMs) are designed to emulate the climate system and the global carbon cycle, and in the process, provide pointers to potential 'tipping points' in the climate system. But their predictions for the Arctic region will only be as good as the hydrology, ecology and energy interactions depicted in the land surface model for that region. This project therefore has two aims - to provide much more physical realism in land surface models, and then to see how this enhancement impacts on modelled future climate. Does the Arctic region eventually enhance human-induced climate change by increasing future levels of atmospheric carbon dioxide? Modelling the land surface for the Arctic region is complicated. To get this correct, we will need to capture how the vegetation may grow and expand in a warmer environment, and how this might change soil C stocks. We also need to model how the snow interacts with vegetation - snow cover will change with climate, and will influence the energy inputs, the water cycle, the frozen ground and the vegetation distribution. For example, deeper snow will occur in areas of tall vegetation and thus vegetation structure influences not only the timing of snow melt but also the thermal regime as deeper snow actually insulates the soils. There is thus a knock-on effect on soil respiration and vegetation growth. This project will model all of these features, dynamically, such that the impact of future temperature and snowfall patterns on the Arctic ecosystems can be assessed. Extensive use will be made of existing observational datasets developed by the PIs and others over the last decade and, in particular the International Polar Year. This new knowledge of the Arctic land surface will be introduced within a pan-Arctic gridded modelling system. The local and regional behaviours will be integrated to determine net land-atmosphere CO2 fluxes. However, major future changes in land surface behaviour might have strong feedbacks on other aspects of the climate system e.g. surface temperatures and soil moisture. Hence, the last component of this project is to make coupled land-atmosphere simulations, thereby capturing all feedbacks. We will achieve this through our existing and on-going collaboration with the Hadley Centre (a world leading centre for modelling the climate system, who make predictions with their family of GCMs). This link will allow a final assessment to be made of whether the Arctic land-surface could pass an unwelcome climate 'tipping point', and thus feedback on existing warming, either locally through enhancing warming further or through the global C cycle.


10 25 50
publication icon
LUNDIN J (2017) Firn Model Intercomparison Experiment (FirnMICE) in Journal of Glaciology

publication icon
Ménard C (2014) Modelled sensitivity of the snow regime to topography, shrub fraction and shrub height in Hydrology and Earth System Sciences

publication icon
Rastetter EB (2010) Processing arctic eddy-flux data using a simple carbon-exchange model embedded in the ensemble Kalman filter. in Ecological applications : a publication of the Ecological Society of America

Description Carbon (C) allocation and turnover in arctic mosses is largely unknown, but their response to climatic change has potentially significant impacts on arctic ecosystem C budgets. Our experiments and modelling results are the first to show differences in C partitioning between arctic moss species in situ. We highlight the importance of modelling C dynamics of moss separately from vascular plants for a realistic representation of vegetation in arctic C models (Street et al. 2012).

Continuous time-series estimates of net ecosystem carbon exchange (NEE) are routinely made using a technique called eddy covariance. Process models are used for making predictions of NEE, and can be compared against data to determine their validity. Our model predicts NEE based only on leaf area, irradiance, and temperature and has been well corroborated for all the major vegetation types in the Low Arctic using chamber-based data. This is the first application of the model to eddy covariance data. The model effectively tracked with independent observations. We were able to improve the model by increasing leaf area over the study period, consistent with expected changes. However, we also found the model was improved by including diurnal changes in leaf area, which are not consistent with observations. It is likely that an effect of atmospheric humidity on photosynthesis, which is not included in the simple model, was the cause of this mismatch.

Soil organic matter is a vast store of carbon, with a critical role in the global carbon cycle. Despite its importance, the dynamics of soil organic carbon decomposition, under the impact of climate change or changing litter inputs, are poorly understood. We have shown how different assumptions about how soil C is decomposed generate models consistent with observations. However, these models react very differently to changes in litter inputs to soils, or to changes in climate. It is vital that we use the most appropriate model structure for making earth system predictions.
Exploitation Route The development and testing of models relating to mosses, soils, and canopies in Arctic environments, and their carbon-climate sensitivity has been a focal part of this research. Our model studies will underpin the development and improvement of earth system models used for climate change analyses, for instance JULES.
Sectors Environment

Description Our findings have been disseminated by publications, and have provided support for generating further research grant income from the NERC Arctic Research Programme, i.e. - CYCLOPS project. Our findings are being used to upgrade a dedicated ecosystem model, DALEC, a core model for the UK National Centre for Earth Observation.
First Year Of Impact 2012
Sector Environment
Impact Types Policy & public services

Title Factorial Snow Model 
Description A model for energy and mass balance of snow on the ground with switchable parametrizations of snow albedo, thermal conductivity, compaction, liquid water storage and interactions with the atmosphere 
Type Of Material Computer model/algorithm 
Year Produced 2015 
Provided To Others? Yes  
Impact The model is being used as the basis for collaborations with researchers in Canada, France and Switzerland