*, %, # - Trends statistically significant at the levels of 0.05; 0.01 and 0.001, respectively; a - Data from 1958; b - data for 1951-1999; c - data from 1957;d - data without 1997-1998; e - data for 1951-1998
Similar to annual totals, positive trends of P dominate in particular seasons of the year (except winter) in the period 1951-2000. In winter, positive and negative trends of P occur in similar areas (Figure 9.10). In comparison to the period 1951-1990 (Figure 6.12), spatial distribution of the areas dominated by negative and positive trends of P in the analysed period (Figure 9.10), changed least in winter and in spring, and most in the two remaining seasons when the areas dominated by positive trends extended to a significant degree.
Similar to the period 1951-1990, negative winter trends dominated mainly in the Russian Arctic and in IARCR. However, they were significantly smaller and rarely exceeded -5 mm/10 years. The other regions with negative trends, primarily the area of CANSRs and the southern part of BAFR, underwent certain territorial modifications; however, they did not influence the area it covered. Statistically significant negative trends occurred at only two stations (Barrow and Iqaluit A) (Table 9.5). Positive trends were greater than negative ones, exceeding 10 mm/10 years, (Table 9.5, Figure 9.10). Apart from the above mentioned areas, statistically significant positive trends of P occurred also in the northern part of the Canadian Arctic. Similar to their annual values, the greatest consistency in the spatial distribution of winter trends of T and P occurs in the Canadian Arctic and in the southern parts of BAFR and PACR.
As has been mentioned above, the spatial distribution of the areas with negative and positive trends in P in spring in the period 1951-2000 (Figure 9.10) is very similar to that in the period 1951-1990 (Figure 6.12), with the exception of PACR and some parts of the Russian Arctic. Both periods were clearly dominated by positive trends in P. The main differences consist in the decrease in negative trends in the Russian Arctic, and the increase in positive trends in the Canadian Arctic. As may be seen from Table 9.5, they were statistically significant in the northern and western parts of the Canadian Arctic. One can also observe a decrease in the values of positive trends in ATLR. In spring, the greatest consistency of the trends of T and P (cf. Figures 9.7 and 9.10) occurred only in the western and northern parts of the Canadian Arctic and in a considerable area of ATLR and PACR.
In comparison to their values in the period 1951-1990 (Figure 6.12), the trends of summer P in the period analysed (Figure 9.10) decreased almost throughout the whole Arctic, especially when their absolute values are taken into account. A statistically significant trend of P was only observed for one station (Bjornoya) (Table 9.5). The area of negative trends, stretching from PACR to ATLR, was reduced significantly. In the period 1951-2000, negative trends, much weaker by that time, occurred only in the area stretching from Alaska through the central part of the Russian Arctic to the Barents Sea. There also appeared an area of negative trends of P encompassing the southern part of BAFR and the southern part of Baffin Island (Figure 9.10) - an area which had not existed in the period 1951-1990. In summer, the relations between the trends of T and P seem to be the clearest and most consistent. Apart from some small areas (mostly in the Central Arctic and the Barents Sea) there is a significant correspondence between the trends of the above meteorological elements. It should be emphasised that this consistency is observed in most of the Russian Arctic only at this time of the year. In the seasons discussed earlier, including autumn, this area was characterised by the greatest discrepancy in the course of the trends of T and P.
In autumn, the direction of changes between the two discussed periods was similar, and this was consistent with the observations for the remaining seasons of the year. The area with positive trends increased considerably. An increase was observed in the size of the area with positive trends during the period 1951-2000, while there was a growth in trends in the areas which had also had positive trends in the period 1951-1990, exceeding, for example, 10 mm/10 years in the southwestern part of the Canadian Arctic (Figure 9.10). Similar to the northern part of the Canadian Arctic, these trends were statistically significant. Negative trends had a limited territory of occurrence and their values decreased. The greatest changes took place in the Central Arctic and in the area of the Russian Arctic, where negative trends occurred only in the eastern part of SIBR. However, no major changes were observed in the other area with negative trends, i.e. the southern parts of Baffin Island, BAFR, and the western part of ATLR.
When compared to the period 1951-2000, the spatial distribution of seasonal and annual totals of P between 1976 and 2000 did not undergo any significant changes, especially when the ratio of regions with positive trends to those with negative ones is taken into consideration. Positive trends also dominated in this period, though they were rarely statistically significant (Table 9.5). The disappearance of negative trends in the Russian Arctic should be counted as one of the greatest changes. There was an increase in P in all the seasons of the period under discussion; however, it was greatest in autumn, when it was statistically significant at two stations. These lasting tendencies, occurring throughout the year, caused the increase in annual totals of P to be statistically significant. In comparison to the period 1951-2000, the region with negative anomalies increased significantly around southern Greenland (the areas from Jan Mayen to Baffin Island). In the latter area (Baffin Island), the decrease in P was statistically significant (Table 9.5), even though the trends of T were positive (except winter) (see Table 9.3).
Was this article helpful?