Scalability of projected climate change in Estonia in relation to global warming

Understanding how changes in regional climate scale with global warming is essential for national climate assessments and adaptation planning. Here, we analyse projected changes in Estonia’s climate using a Global Warming Level framework and a CMIP6 multi-model ensemble. The approach expresses regional climate responses as functions of global warming, rather than functions of scenario-specific time horizons.

We analyse annual and seasonal temperature, precipitation, and wind-related indicators, including indicators of extremes, and express their responses as linear scaling relationships per one degree of global warming. Estonia shows clear regional amplification of global warming, with annual mean temperature increasing by about 1.5 °C per 1 °C of global warming. The amplification is strongest in winter (+1.9 °C/°C) and is weaker in summer (+1.3 °C/°C). Cold extremes respond even more strongly: the annual coldest daily minimum temperature rises by about 3.6 °C/°C, indicating a substantial reduction in cold-season severity. Annual precipitation increases by about 31 mm (4.6%) per 1 °C of global warming, and heavy precipitation intensifies across most models. In contrast, changes in summer precipitation show large inter-model spread, while projected wind changes are generally weak and inconsistent.

A central element of the scaling analysis is the assessment of robustness. Trends are calculated separately for each model and then summarised using the ensemble median together with the 10th and 90th empirical percentiles (P10–P90), representing the central 80% model range. A change is classified as robust when the full P10–P90 interval remains on the same side of zero as the median trend, indicating that at least 90% of models agree on the sign of the change. If the P10–P90 interval crosses zero, the signal is considered non-robust because of substantial inter-model disagreement.

The results highlight a clear contrast between changes dominated by thermodynamic controls and those more strongly influenced by changes in atmospheric circulation. Changes in temperature-related indicators and extreme precipitation show high inter-model agreement and robust scaling with global warming, whereas summer precipitation and wind-related changes remain much more uncertain. This indicates that uncertainties are smaller for thermodynamically driven changes than for such changes, where the future behaviour depends more strongly on circulation responses to global warming and internal variability.