Potato (Solanum tuberosum L.) ranks as the fourth most important food crop after rice, wheat, and maize. In the literature, models have been developed to predict potato yield evolution due to climate change, projecting a decrease in production across various regions. This study was conducted on field data collected over 31 years in five contrasting sites in the Western Alps, Switzerland. Results show that 24 % of yield variation can be explained by the genotype, and 50 % by the environment. Among the studied meteorological conditions influencing the yield, 1) “total precipitation from tuber initiation to tuber harvest”, 2) “sum of solar irradiation from planting to maturity”, 3) “average temperature from planting to maturity” and, 4) “sum of daily maximal temperature from planting to maturity”, were the most important variables. The third variable exhibits a positive linear relationship with yield up to an average temperature of 16.5 °C during the growth season. Beyond this threshold, the relationship becomes negative and results in yield loss. Using this unprecedented dataset, we estimated potential yield losses in the Western Alps of Switzerland by the end of the century under three different Representative Concentration Pathway (RCP) scenarios (i.e. 2.6, 4.5, 8.5). In the short term, by 2035, yield losses are projected to range from 3.2 % to 15.0 % regardless of the scenarios. By 2060, RCP 4.5 and RCP 8.5 predict the highest losses, reaching 22.7–50.3 % compared to the 1990–2020 average yield. The most significant loss was predicted under the RCP 8.5 long-term scenario, by 2085, with yield losses ranging from 24.2 % to 84.6 %. These losses are attributed to an estimated precipitation decrease of 42 % compared to the average of the past 30 years and a +7.2 °C increase in average temperature during the potato growth season. Except in the case of RCP 2.6, which estimates low yield losses compared to 1990–2020, this study anticipates significant yield losses by the end of the century in Switzerland. To mitigate these losses due to climate change, potential adaptation strategies include the adoption of drought or heat-stress-resistant genotypes, enhancements in irrigation systems, adjustments of planting schedules, and relocating planting sites to higher elevations. In addition, the G x E interaction effect should be considered in breeding strategies, to cope with climate change impacts on potato yield and to grow genotypes better adapted to their environment.
