Modelling Mass Balance of Svalbard Glaciers
The Arctic climate is currently warming at a faster rate than observed elsewhere on Earth and future projections suggest this trend will continue into the 21st century. With glaciers and ice caps covering ~36 600 km2, Svalbard is one of the largest glaciated areas in the Arctic. Future climate change will significantly alter the mass balance of glaciers and ice caps across the archipelago with important consequences for sea level. We are currently developing a numerical mass balance model, which will ultimately be used to calculate the spatial and temporal variations in mass balance of the archipelago’s ice masses. It is currently being trained and validated for the glacier Midre Lovénbreen, NW Spitsbergen where we have been working for the last few years (Figure 1). The model uses an energy balance approach to determine surface melt variations (Arnold and others, 2006) and air temperature and precipitation to calculate patterns of accumulation. Recent improvements to the model include a subsurface routine to deal with the processes of conduction and melt water refreezing within the snowpack (Wright, 2005).
We have used ERA-40 reanalysis data for the years 1970 - 2000 together with surface meteorological data from Ny-Ålesund over the same period, to develop linear regression transfer functions to convert gridded ERA-40 data to surface observations (Figure 2). The corrected ERA-40 data are used to calibrate the summer and winter surface mass balance model for Midre Lovénbreen for the mean climatic conditions between 1970 and 2000. The calibrated model is then used to simulate the spatial and temporal mass balance variations from 1970-2000 (e.g. Figure 3). We validate the simulated mass balance time series using measured summer, winter and net mass balance variations (Figure 4).
Future work will involve: i) modelling past variations in Midre Lovénbreen mass balance using ERA-40 reanalysis data extending back from 1970 to 1957 and from 2000 to present using operational forecasting data; ii) assessing the effects of “degrading” the spatial resolution of the model to 50m, 100m and 200m; iii) extending the spatial domain of the model to calculate past variations in the mass balance of all of Svalbard’s glaciers and ice caps; iv) running the model into the future using downscaled GCM output.
Our work is being done in collaboration with Jack Kohler (Norsk Polar Institute, Tomsø, Norway).
- Arnold, N.S., Rees, W.G., Hodson, A.J. and Kohler, J. 2006 Topographic controls on the energy balance of a high Arctic glacier. Journal of Geophysical Research. 111 F02011, doi 10.1029/2005JF000426 (15 pp).
- Wright, A.P. 2006. The impact of meltwater refreezing on the mass balance of a High Arctic glacier. PhD Thesis, University of Bristol, UK).
Papers relating to this project
- Rye, C., Willis, I. And Arnold, N. 2007. Modelling the 21st Century mass balance of Svalbard glaciers using ERA-40 reanalysis and general circulation models. In The Dynamics and Mass Budget of Arctic Glaciers (Extended Abstracts). IASC Working Group on Arctic Glaciology Meeting. Pontresina (Switzerland). IMAU.
Figure 1. Topographic Map of Midre Lovénbreen and surrounding region (source: Wright, 2005)
Figure 2. Comparison of Ny-Ålesund and ERA-40 meteorological variables: a) temperature; b) precipitation; c) solar radiation; d) long wave radiation; e) wind speed; f) relative humidity. Data are averaged over the period 1970-2000. The ERA-40 have been corrected for systematic bias using linear regression. Temperature, solar radiation and long wave radiation have strong correlations (>0.9); Precipitation correlation was lower (~0.6). The lowest correlations were found for wind speed and humidity.
Figure 3 Modelled mean mass balance of Midre Lovénbreen between 1970 and 2000. Units are m w.e. a-1. The model shows a strongly negative mass balance at the glacier snout. At higher elevations this trend becomes less negative with positive balances at the edges of the upper basins.
Figure 4. Measured and modelled annual net mass balance time series for Midre Lovénbreen 1970-2000. The model was calibrated to fit the the mean mass balance over this time period using parameter values from the literature. The model captures the negative trend observed over time and reproduces the year-to-year variation well.