Therefore, the polar low climatological data are of critical importance. There are several existing polar low climatologies created using various techniques. But, most of them are based on modeled and reanalysis data, weather maps and visible/infrared imagery. However, polar lows are often not detected at the weather maps and are under-represented in current reanalysis datasets (Condron et al., 2006). Thus, only up to 80 % of cyclones larger than 500 km can be detected in mean sea level pressure, only up to 40 % for 250-km-scale cyclones, and only 20 % for 100-km-scale cyclones. The same time modal size of AVHRR-derived mesoscale cyclones/polar lows is 100-150 km (Harold et al., 1999). Using just visible and infrared imagery does not ensure the tracking of polar lows due to insufficient temporal resolution of this imagery. Moreover, visible images can't be obtained during night and winter.
Conventional observations are too sparse, and sometimes completely unavailable to provide data for polar low studies. The most informative polar low studies include the comprehensive joint analysis of different satellite data from various instruments providing the most complete information about storm development. Among the satellite sensors, microwave radiometers have important advantages for detection and tracking of polar lows. This is independence on day and night time and clouds, and, what is extremely important, regularity and high temporal resolution in polar regions, provided by past and current satellite radiometers, such as SMMR, SSM/I, AMSR-E, SSMIS and AMSR 2. In this content authors developed an approach for detection and tracking of polar lows based on satellite passive microwave data from above listed radiometers (Bobylev et al., 2011). This approach consists of two stages. During the first stage the total atmospheric water vapor fields are calculated from passive microwave measurements using precise retrieval Neural Network Algorithms, applicable over the open water, and having high retrieval accuracies under a wide range of environmental conditions (Bobylev et al., 2010). During the second stage the vortex structures are detected in these fields, and polar lows are identified and tracked.
Based on this approach, the climatology of polar lows in the Nordic seas was created over 1995-2012, as a first step. All polar lows have been identified for this period on SSM/I and SSMIS imagery by means of analysis of atmospheric water vapor fields. Other satellite data, such as QuikSCAT SeaWinds, Metop ASCAT, Terra MODIS, NOAA AVHRR and Envisat ASAR, were used as additional information for polar low parameter retrieval and analysis.