1. Climatology of polar lows in the Nordic and Barents seas over 1995- 2008 based on satellite data

2. Polar lows and their general characteristics  Polar lows are short-living intense mesoscale maritime atmospheric low pressure weather systems, observed over high latitudes, both in Arctic and Antarctic, during wintertime  Short lifetime: from several hours to 2 days (average 15÷20 hours)  Small size: 100÷1000 km  High surface wind speed: > 15 m/s (some time > 30 m/s)  Typically marine phenomenon: polar lows rapidly break down over land and ice cover

3. Areas of occurrence of polar lows in the Arctic Labrador sea Arctic polar lows are significantly more intensive than Antarctic ones due to large fluxes of heat and moisture Most intensive Arctic polar lows are called “Arctic hurricanes” Nordic and Barents seas is one of the main genesis areas for polar lows in the Arctic

4. Existing polar low climatologies YearAuthors and PaperPeriodType of dataAreaMethod 2011 G. Noer, Ø. Saetra, T. Lien, Y. Gusdal (2011). A climatological study of polar lows in the Nordic Seas. Quarterly Journal of the Royal Meteorological Society, 137(660), 1762–1772 2000-2009Visible imageryNordic SeasSubjective analysis 2008 Zahn, M., and H. von Storch (2008). A longterm climatology of North Atlantic polar lows. Geophys. Res. Lett., 35, L22702, doi:10.1029/2008GL035769 1948 - 2006 Model: NCEP/NCAR re- analyses data and CLM data Sub-Arctic region of the North Atlantic Objective analysis of pressure fields using digital filter 2008 Blechschmidt, A.-M. (2008). A 2-year climatology of polar low events over the Nordic Seas from satellite remote sensing. Geophys. Res. Lett., 35, L09815, doi:10.1029/2008GL033706 2004 ‑ 2005 Combined use of thermal infrared AVHRR imagery and SSM/I derived wind speeds from HOAPS Nordic SeasSubjective analysis 2008 Thomas J. Bracegirdle*, Suzanne L. Gray. (2008). An objective climatology of the dynamical forcing of polar lows in the Nordic seas. Int. J. of Climatol., 14(28), 1903- 1919 January 2000 to April 2004 Cyclone Database developed by Hewson objectively identified from the UK Met Office global operational model Norwegian and Barents seas Objective analysis 2006 Kolstad, E.W. (2006). A new climatology of favorable conditions for reverse-shear polar lows. Tellus, 58A, 344– 354 1948 - 2005 Model: ERA-40 reanalysis data Latitudes over 60°N Objective analysis 1999 Harold, J.M., Bigg, G.R. and Turner, J. (1999). Mesocyclone activities over the north-east Atlantic. Part 1: vortex distribution and variability. Int. J. Climatol.,19, 1187–1204 October 1993- September 1995 Infrared AVHRR North-East Atlantic and Nordic Seas Eye inspection of AVHRR images 1985Wilhelmsen, K. 1985. Climatological study of gale- producing polar lows near Norway. Tellus, 37A, 451–459 1972-1977Weather mapsNorwegian and Barents seas Scrutinizing weather maps and documenting mesoscale cyclones with gale-force wind and horizontal size 100-500 km

5. Shortcomings of existing climatologies  Often polar lows are not detected on the surface weather analysis maps  Mesoscale cyclones/polar lows are under-represented in current reanalysis datasets (Condron et al., 2006): o Only up to 80 % of cyclones larger than 500 km can be detected in mean sea level (MSL) pressure o Only up to 40 % of cyclones larger than 250km can be detected in MSL pressure o Only 20 % of cyclones larger than 100 km can be detected in MSL pressure  Modal size of AVHRR-derived mesoscale cyclones/polar lows is 100-150 km (Harold et al., 1999)

6. New approach for polar low monitoring and climatology creating Approach: o Retrieval of atmospheric columnar water vapor (CWV) fields from satellite passive microwave data (e.g., SSM/I and AMSR-E) o Revealing vortex structures in these fields o Identification of these structures with polar lows o Verification of identity of each revealed vortex structure with polar low using other satellite data (e.g., AVHRR imagery and data on sea surface wind speed derived from SSM/I or radar scatterometers) o Polar low parameters (life time, size, location, moving speed) estimation and trajectory tracking 31 January 2008 11:14 UTC AMSR-E Advantages: o independence on day time o independence on clouds o regularity and high temporal resolution in polar regions o existing of 35 + -year record of continuous satellite passive microwave observations, since 1979 till now and beyond Retrieved parameters: o sea surface wind speed o atmospheric columnar water vapor o total cloud liquid water content Bobylev et al., IEEE TGRS, 2011

7. NN-based polar regional algorithms for CWV retrieval from SSM/I and AMSR-E Bobylev et al., IEEE TGRS, 2010 Based on comparison of SSM/I and AMSR-E retrievals with polar island station radiosonde data from Jan Mayen, Bjornoya and Torshavn (http://weather.uwyo.edu/) NN-configurations Validation

8. Accuracy of NN polar regional algorithms: comparison with Wentz global algorithm Accuracy of polar NN-algorithm is 40% higher than that of Wentz global operational algorithm Comparison of NN-based polar regional algorithm for SSM/I CWV retrieval with widely used Wentz global operational algorithm based on Torshavn radiosonde measurement data σ = 1.34 kg/m 2 σ = 1.90 kg/m 2 NN polar regional algorithm Kg/m 2 SSM/I 1 July 2005, 7:44 Wentz global operational algorithm CWV in polar lows might be just 2÷3 kg/m 2 higher than in surrounding areas and polar low detection thus becomes impossible if retrieval error is close to these values.

9. Polar low detection and tracking by satellite microwave radiometers Q, kg/m 2 31 January 2:10 UTC AMSR-E31 January 9:35 UTC AMSR-E31 January 14:30 UTC SSM/I 31 January 11:14 UTC AMSR-E 31 January 3:50 UTC AMSR-E 31 January 12:47 UTC SSM/I Polar low trajectory 31 January 7:58 UTC SSM/I 75N Polar low over Norwegian Sea on 30-31 January 2008 tracked by SSM/I and AMSR-E

10. Uncertainties: o number of different definitions of term “polar low” used by various researchers o specific features of detection schemes applied for catching polar lows Definition of polar lows for our climatology: o mesoscale cyclones with horizontal size ≤ 1,000 km and surface wind speed ≥ 15 m/s o such climatology might include the following types of polar lows (following classification by Turner and Bracegirdle): reverse shear systems trough systems boundary layer fronts cold lows comma clouds baroclinic-wave forward shear orographic polar lows Creating polar low climatology: definitions

11. Verification of polar low detection Polar low in Barents Sea on 19 January 2000 CWV field from SSM/I dataWind speed field from SSM/I data (RSS) AVHRR thermal infrared imagery

12. Polar low climatology creation Portal, SSM/I F13, T B for all channels and all circuits Columnar water vapor retrieval with enhanced resolution - 12.5 km NN-algorithm Detection of vortex structures in water vapor fields Verification of polar low detection Wind speed > 15 m/s: 1995-2011 – RSS, SSM/I 2002-2011 - QuickScat Cloud structure typical for mesocyclones: NOAA AVHRR Detected polar low for climatology Estimation of polar low parameters from passive microwave and other available data Polar low climatology for Nordic and Barents seas over 1995-2008 Generation of columnar water vapor fields for Nordic and Barents seas

13. Polar low climatology over the Nordic and Barents seas for 1995-2008 Spatial distribution of detected polar lows Year Number of cyclones Number of polar lows per year (637 total) Polar low size distribution Number of cyclones Cyclone diameter (km) Polar low life-time distribution Number of cyclones Cyclone life-time (hr)