Wpływ zmian temperatury wód w Bramie Farero-Szetlandzkiej na temperaturę powietrza w Arktyce (1950-2005)

Praca analizuje związki między wskaźnikiem charakteryzującym zasoby ciepła w wodach atlantyckich wprowadzanych do Prądu Norweskiego, a dalej przez Prąd Zachodniospitsbergeński i Prąd Nordkapski do Arktyki, a roczną temperaturą powietrza w Arktyce. Analizę związków przeprowadzono dla Arktyki jako całości oraz jej sektorów: atlantyckiego, syberyjskiego, pacyficznego kanadyjskiego i sektora Morza Baffina. Wykazano istnienie silnie rozciągniętych w czasie (od 0 do 9 lat opóźnienia) związków z temperaturą powietrza w całej Arktyce, potwierdzających istotny statystycznie wpływ zmian zasobów ciepła w wodach na zmiany temperatury powietrza w Arktyce. Związki regionalne wykazują silne zróżnicowanie - na wzrost zasobów ciepła niemal natychmiastowo reaguje temperatura powietrza w Arktyce Atlantyckiej, z 2-6 letnim opóźnieniem temperatura powietrza w Arktyce Kanadyjskiej. Związki z temperaturą powietrza w sektorach syberyjskim i pacyficznym nie przekraczają progu istotności statystycznej. Zmiany temperatury powietrza w sektorze Morza Baffina wyprzedzają w czasie zmiany zasobów ciepła w wodach atlantyckich wprowadzanych następnie do Arktyki. To ostatnie może stanowić przyczynę okresowości w przebiegu temperatury powietrza w niektórych częściach Arktyki i strefy umiarkowanej.

Styszyńska (2005, 2007) has shown the existence of clear statistical relationships between heat contents in the waters of the Atlantic flowing towards the Arctic via the Norwegian, West Spitsbergen, and North Cape currents and the air temperature in Spitsbergen, Jan Mayen and Hopen between the years 1982 and 2002. These relationships extend in time: following rises in the heat content of the waters of the Norwegian Current, an increase in air temperature follows in the same year and the following year. Heat contents in the Atlantic waters flowing towards the Arctic are assessed according to the average sea surface temperature (SST) in the Faeroe-Shetland Channel (grid 62°N, 004°W) from January to April. These values are used to calculate a determining indicator such as FS1-42L, established as the average of two successive years: data from one year (k) and the year preceding it (k-1). The aim of this work is to investigate whether there are relationships between FS1-42L and the air temperature in both the whole of the Arctic and in individual Arctic sectors and, if so, what the character of these relationships is. The data analysed were a set of yearly air temperatures for the whole of the Arctic and for particular Arctic sectors (fig. 2) according to Przybylak (2007), as well as a set of monthly SST values including values calculated for the FS1-42L indicator (NOAA NCDC ERSST v.1; Smith and Reynolds, 2002). The primary methodology employed was Cross-Correlation Function Analysis. The FS1-42L was established as a first value, with the yearly air temperature used as a lagged value. The analysis was carried out for a 55-year period, from 1951 to 2005. The analysis showed that, taken as a whole, relationships between heat contents leading to the Arctic and air temperature over the whole of the Arctic (calculated from averages of individual sectors) were not particularly significant, though there was marked significance in these relationships from year 0 (fig. 3) to year +9 (fig. 4). The strongest relationships were those from the same year for which the FS1-42L was dated, after which relationships grew gradually weaker, until they finally disappeared in the tenth year. In the Atlantic sector of the Arctic the relationship was strong and almost immediate (fig 5). In the Siberian (fig. 6) and Pacific (fig. 7) sectors there was an absence of statistically significant relationships, and any that did exist were weak, with varying degrees of ?echo? in air temperature reactions. Air temperature in the Canadian sector (fig. 8) reacted to increases in heat contents with a delay of 2 to 6 years, with the strongest relations from FS1-42L being noted with a 5-year delay. The situation in Baffin Bay was entirely different, with air temperature changes preceding changes in the heat contents of the waters of the Faeroe-Shetland Channel by 1 to 6 years. The maximum strengths of these relations were -5 and -4 per year (fig. 9). Analysis of the reasons for these regional variations in the influence of FS1-42L on air temperature allows us to conclude that a major role is played by the bathymetry of the Arctic Ocean. Atlantic waters sinking beneath Arctic Surface Water (ASW) contribute to changes in the temperature of Arctic Intermediate Water (AIW). Independent of the routes taken by the processes, the influence of AIW on the air temperatures in the Siberian and Pacific sectors is limited, with these sectors being isolated by wide shelves from the Arctic Ocean. In the Canadian sector, which is separated by narrow shelves from deep-water parts of the Arctic Ocean and is situated a relatively short distance from the Atlantic sector, the influence of heat contents on the ASW is apparent, with a certain delay. Changes in the air temperature of the Baffin Bay sector are related to the variable activity of the Labrador Current, bringing cold waters to the North from the Gulf Stream delta. The force of strong cooling waters from the Labrador Current, with the appropriate delay, result in a lessening of the heat contents in the Faroe-Shetland Channel. Because of the fact that there is a strong positive correlation between the yearly air temperatures of the Canadian and Baffin Bay sectors, a chain of dependencies emerges: air temperature in the American sectors of the Arctic the flow of Atlantic waters FS1-42L air temperature in the Atlantic Arctic sector Ž air temperature in the Canadian sector should generate quasi-periodic (> 10 years) air temperature courses.