As the sub-Arctic and Arctic regions undergoes rapid changes, understanding its hydroclimate history is more critical than ever. A limited availability of moisture-sensitive proxy data severely constrains our insights, underscoring the urgent need for more hydroclimate reconstructions in the region. Here we present a May–June precipitation reconstruction based on ring width from living and dead trees of Scots pine (Pinus sylvestris L.) growing under drought-stressed conditions at 63° N, near Skuleskogen National Park, on the northern part of the east coast of Sweden. The oldest deadwood sample dates back to the mid-11th century, and the Expressed Population Signal (EPS) exceeds 0.85 from 1320 CE until the present (2021 CE), making this the longest tree-ring-based hydroclimate reconstruction from high-latitude Fennoscandia. Unlike typical high-latitude forests in Fennoscandia, the trees at this site thrive under low-pressure conditions and show strong correlations up to approximately |r| = 0.6 over the period 1920–2021 with drought-related variables such as precipitation, cloud cover, solar radiation, potential evapotranspiration and diurnal temperature range. The climate signal is concentrated to a short window between mid-May and early July but reflects climatic conditions over a broad region. Significant spatial correlations extend over most of Sweden as well as large parts of Norway and Finland, indicating sensitivity to large-scale climate systems. The tree-ring chronology also shows clear cyclic patterns, with a particularly strong ∼ 64-year cycle. The modern era stands out for its variability, with 2018 CE emerging as one of the of the driest year in the entire record. There is potential to extend the chronology further back in time, possibly to 1000 CE or earlier, and to extract additional climate information using other tree-ring parameters such as blue intensity.