Relationships between the variability of the North Atlantic oscillation (NAO) indices and the normalized precipitation at 78 stations in Turkey, and the influences of the extreme NAO index (NAOI) episodes and the year-to-year and longer time-scale variations in the NAO on the precipitation conditions were investigated. The results of the study have led to the following main conclusions and evaluations for Turkey:(i) There is a negative relationship between interannual variability of the Turkish precipitation series and the NAO indices. Negative relationships that are particularly strong in winter and partly in autumn are detected to be weaker in spring and almost non-existent in summer. Correlation coefficients are significant at 61 stations in winter, whereas they are significant at 23 and eight stations in autumn and spring respectively. (ii) Composite precipitation means corresponding to the extreme NAOI phases mostly exhibit an apparent opposite anomaly pattern, except in summer, between the negative and positive NAOI phases. (iii) Annual, winter, spring, autumn and partly the summer composite precipitation means are mostly characterized by wetter than the long-term average conditions during the negative NAOI phase, whereas the positive NAOI responses mostly exhibit drier than the long-term average conditions annually and in all seasons except summer. (iv) Spatially coherent and statistically significant changes in the precipitation amounts during the extreme NAOI phases are more apparent in the west and mid Turkey. (v) Low-frequency fluctuations in the circulation over the Atlantic have been closely linked to the coherent large-scale precipitation anomalies that have persisted, particularly in winter, over Turkey since the early 1960s. (vi) There is a great resemblance between the spatial distribution and magnitude of the negative correlation coefficients and the spatial distribution patterns and severity of the wet (dry) conditions with negative (positive) NAOI phase, annually and in the winter and autumn seasons. (vii) The coherent regions characterized with significant correlation coefficients coincide perfectly with the coherent regions characterized by the extreme NAOI signals. These clear associations increase our confidence with the results.

Long-term variability and trends of the annual numbers of summer and tropical days were investigated by using nonlinear (monotonic) and linear trend detection tests for the period 1950-2010 for 97 meteorological stations of Turkey. The results suggest that the numbers of both summer and tropical days indicate a general increasing tendency in Turkey. For the study period, statistically significant increasing trends for summer (tropical) days are detected at 64 (71) stations, of which 51 (58) of these positive trends are significant at the 0.01 significance level. Two periods of changes in summer and tropical days are identified: in the 1950-1975 sub-period with an episode of slight cooling, the annual number of summer and tropical days decreases, whereas in the sub-period of 1976-2010 with an episode of significant increasing trend, the annual number of summer and tropical days increases. The summer of 2010 was exceptional for the number of summer and tropical days at most of the stations in Turkey. Normalized anomalies of summer (tropical) days at 2010 were larger than three standard deviations with respect to 1961-1990 normal at 27 (43) stations. The largest positive anomalies are observed in the northeastern Anatolia sub-region. This increasing trend has significant impacts on agriculture, energy, tourism and natural ecosystems (e.g., forest fires) in Turkey.

No abstract

No abstract

This study examines the climatology of annual frost days, and analyses the size and behaviour of the longterm variability and trends in annual numbers of frost day at the 72 stations over Turkey from 1950 to 2010. The main results are summarized as follows: (1) The annual number of frost days has evidently decreased at most of the stations with some observed regional differences, (2) The decreasing trends are largest over the Eastern Anatolia, the Marmara regions and along the Mediterranean coastline. The meteorological stations located in the continental northeast and the easternmost parts of the Anatolian Peninsula, including Ardahan, Iǧdır and Van, show a negative linear trend with a rate of four days per decade, (3) As in other regions of the Earth, changes in number of frost days are very likely associated with changes in minimum air temperatures and increasing growing season lengths in Turkey, (4) The decreasing trends in number of frost days also indicated considerable decadal-scale variability. This variability is very likely attributable to the large-scale atmospheric circulation and atmospheric oscillations such as the Arctic Oscillation (AO) or North Atlantic Oscillation (NAO) and the North Sea-Caspian Pattern, (5) Consequently, the long-winter (DJFM) composites of number of frost days were examined for extreme phases of the AO index during the period 1950-2010 in order to assess the influence of atmospheric oscillations on year-to-year variability in number of frost days. According to the Cramer's t k test, winter number of frost days tended to increase significantly during the high (positive) index AO phase, while they tended to decrease significantly during the low (negative) index AO phase. These relationships are statistically significant at the 1% level at the majority of stations.