Recent seasonally contrasting and persistent warming trends in Latvia

Abstract

The Baltics is one of the European regions facing the most rapidly increasing air temperatures – a clear signal of climate change. This article comprises an analysis of the spatial distribution and variability of air temperature in Latvia, in the Baltics, comparing the most recent climate normal (1991–2020) with the previous two climate normals (1971–2000; 1981–2010), as well as with a reference period (1961–1990). Compared to the reference period, the annual mean temperature in the last climate normal was 1.2 °C higher across Latvia, corresponding to a warming rate of +0.4 °C decade⁻¹, which is greater than the European average (between 0.17 and 0.22 °C decade⁻¹). Compared to the reference period, a persistent rise of winter (DJF) temperatures by around 2 °C and a rapid increase in the easternmost locations (distant from the sea) were found. Summer (JJA) and spring (MAM) temperatures rose by around 1 °C. The autumn (SON) temperature increase was less pronounced and was only evident during the last two climate normals. Since the reference period, the minimal mean daily air temperatures (5th quantile) in winter, autumn and, especially, spring months have increased at a rate almost double that of the average (50th quantile) temperature (0.72 to 0.39 °C decade⁻¹, respectively). We note that May or June and October experienced little to no average temperature increase, reported elsewhere in the Baltic region. We have confirmed the presence of two seasons in the spatial pattern of air temperature: one from April to July dominated by an N–S gradient and the other from August to March dominated by a W–E gradient. Furthermore, the possible mechanisms and implications of the observed seasonal pattern of the temperature increase are discussed, particularly considering land–atmosphere water and energy flux feedback.

Appendix

Fig. 8

Seasonal mean temperatures (°C) for the six selected meteorological stations and for Latvia (average of 25 meteorological stations) during the four climate normals and the reference period. On the right, differences in the air temperature during a particular climate normal compared to the previous normal (given as absolute values) are shown. *denotes statistically significant (5% level) using the Mann–Kendall test; the background shades of grey represent the same data as numbers with darker tones indicating stronger warming

Fig. 9

The 5th, 50th and 95th quantile regression slope (°C decade−1) for the six stations (Alūksne, Daugavpils, Kolka, Liepāja, Rīga, Saldus) and Latvia (average of 25 stations) estimated using the daily average air temperature over four seasons for the four different climate normals and the full 60-year period (respectively 1961–1990, 1971–2000, 1981–2010, 1991–2020 and 1961–2020); *values are statistically significant using the Mann–Kendall trend test (5% significance level); the background colour shows the same data as numbers – reds are positive, blues – negative and white – no trend