Satelitarna charakterystyka zachmurzenia ogólnego nad Svalbardem w roku 2007 w powiązaniu z cyrkulacją atmosfery

W artykule przedstawiono rozkład czasowy i przestrzenny zachmurzenia ogólnego nad Sval-bardem w 2007 r. Wszystkie prezentowane wielkości zachmurzenia wyliczono z maski chmur, będącej jednym z produktów powstałych w wyniku przetworzenia danych satelitarnych radiometru MODIS, umieszczonego na satelitach Terra i Aqua. Analizie poddano średnie miesięczne, średnią roczną oraz średnie zachmurzenie w po-szczególnych 11 typach uproszczonej klasyfikacji Niedźwiedzia, tak dla całej powierzchni archipelagu, jak i jego poszczególnych części.

One of the fundamental problems in cloud climatology research is a lack of high spatial and temporal resolution data. Conventional, surface-based visual observations are limited to a small number of locations and represent atmospheric conditions only within a small vicinity of the stations. This is particularly true in the Arctic, which is inadequately sampled due to extreme weather condi-tions and maritime character of this area. As an alternative, satellite data can be utilized as a base for cloud climatology studies. In this paper Moderate Resolution Imaging Spectroradiometer (MODIS) observations are used as a source of cloud data for investigating the relation between total cloud cover and atmospheric circulation patterns over Svalbard. MODIS data were obtained as a Cloud Mask product – a 1 km resolution raster with four classes reflecting cloud detection confidence: 'confident clear', 'probably clear', 'uncertain clear' and 'certain cloudy'. Each class was arbitrary turned into fractional cloud cover as 0%, 33%, 66% and 100% respectively. Total number of 5607 MODIS passes over Svalbard was analyzed (about 16 a day). Area of study was divided into three regions – Spitsbergen (1) with subregions: 1a (north-western part), 1b (north-eastern part), 1c (southern part); Nordaustlandet (2); Barents Island and Edge Island (3). Mean monthly and annual cloud amount was calculated for each region as a ratio of cloudy pixels (weighted by 0%, 33%, 66% and 100%) to all pixels within given region/subregion. MODIS-derived information was then set against Niedźwiedź (2007) circulation type classification. Classification is an application of Lamb (1972) subjective classification, reduced in this study from 21 to 11 types: 5 cyclonic, 5 anticyclonic and 1 undetermined. As the results show, mean total cloud cover over Svalbard in 2007 amounted to 74%, varying from 61% in February up to 85% in August. The greatest mean monthly cloud cover (88%) was observed over Nordaustlandet in August, while the lowest (57%) over southern part of Spitsbergen in February. The cloudiest parts of Svalbard in 2007 were Nordaustlandet and Edge Island with 76% and 77% of annual mean cloud cover respectively – slightly more than Spitsbergen (73%). Spatial distribution of annual mean cloud cover Svalbard was controlled by topography and atmospheric circulation conditions. Atmospheric circulation over Svalbard in 2007 was dominated by advection from N-E-S directions and non-advective situations (center of cyclone or cyclonic trough). Average cloud cover was nearly constant throughout all circulation types, ranging from 74% (cyclonic advection from S+SE) to 77% (cyclone's center or cyclonic trough). Most diverse spatial distribution of cloud cover was observed during the days of central anticyclonic situations and anticyclonic wedge, while least diverse when cyclone's center, cyclonic trough or anticyclonic advection from S+SW occurred. MODIS-derived cloud cover variability can be well explained by circulation influence, e.g. foehn effect associated with anticyclonic E+SE advection, cloud amount increase as a result of S+SW or W+NW cyclonic advection from Norwegian Sea. Although annual course of cloud cover, as determined with satellite information, seems reliable, future studies should emphasise a comparison of MODIS data with surface based observations. Temporal coverage should be also expanded to years 2003-2008 in order to obtain statistically significant results.