These maps show the number of days with polar bear habitat in the Barents Sea. They are based on polar bear telemetry data collected and analyzed (similar to Durner et al. 2009) by the Norwegian Polar Institute (Merkel & Aars in prep). Three seasons based on annual sea ice fluctuations and the biology of polar bears were defined to calculate available habitat.
The three seasons are:
- Winter: 1 November – 30 April
- Sea ice break-up: 1 May – 31 July
- Sea ice freeze-up: 1 August – 31 October.
Each season as well as the months of minimum (September) and maximum (April) ice cover are shown in three different time periods to illustrate the change in available polar bear habitat. The time periods are; Mid-90’s: 1992-1996 Mid-00’s: 2002-2006 Now: 2013 Additionally, the change in habitat from the mid-90’s to 2013 is also visualized in one map for each season as well as September and April. Monthly maps include one map for each time period for September and April, respectively. The colors range from white (= 0 days with habitat) to orange (= 30/31 days in September/April). Seasonal maps contain a map for each season defined above for each time period, respectively. The color palette for the sea ice break-up and freeze-up periods range from white (= 0 days with habitat) to orange (= 92 days) as these seasons span 3 months each. Consequently, the winter season, ranging over 6 months, has its maximum at 182 days (orange). Change maps are composed of the available habitat for each season as well as September and April in the year 2013 minus the available habitat for each respective time period in the mid-90’s. Habitat gained in 2013 is marked in green (color intensity increasing with increasing number of days gained) while habitat lost compared to the mid-90’s is shown in purple (color intensity increasing with increasing number of days lost). No change in available habitat is shown in white. Of note here is the substantial change and shift northwards during the last 22 years in available habitat in September (Link to change September mid-90’s to 2013) while no such extensive change is visible for the month April (Link to change April mid-90’s to 2013).
Bathymetry, land mask and sea ice concentration data were used for the analysis. The International Bathymetric Chart of the Arctic Ocean (IBCAO) Version 3.0 (Jakobsson et al. 2012) was used to acquire ocean depth values. Land polygons were obtained from the Norwegian Polar Institute (NPI) mapping section, and are generalized from official NPI Svalbard and GITBarents data. Further, sea ice concentration data were retrieved from the University of Hamburg (Kaleschke et al. 2001, Spreen et al. 2008). The original data resolution of 12.5 x 12.5 km was interpolated to 1 x 1 km for these images. Further, the data points around the land mask have been interpolated into the land to avoid a pixelated view of the coastlines. Therefore, the maps published here should only be used for visualization and not for scientific studies.
REFERENCES ASI Algorithm SSMI-SSMIS sea ice concentration data, originally computed at and provided by IFREMER, Brest, France, were obtained as 5-day median-filtered and gap-filled product for the Norwegian Polar Institute, Tromsø, Norway from the Integrated Climate Date Center (ICDC, http://icdc.zmaw,de/), University of Hamburg, Hamburg, Germany"
Durner, G. M., et al. (2009). "Predicting 21st-century polar bear habitat distribution from global climate models." Ecological Monographs 79(1): 25-58.
Jakobsson, M., et al. (2012). "The International Bathymetric Chart of the Arctic Ocean (IBCAO) Version 3.0." Geophysical Research Letters 39.
Kaleschke, L., et al. (2001). "SSM/I sea ice remote sensing for mesoscale ocean-atmosphere interaction analysis." Canadian Journal of Remote Sensing 27(5): 526-537.
Spreen, G., et al. (2008). "Sea ice remote sensing using AMSR-E 89-GHz channels." Journal of Geophysical Research-Oceans 113(C2).