In Finland early summer droughts are common. They cause yield losses (Y_Loss) of spring cereals, barley, oats and wheat, that cannot be compensated for later in the growing season. To support farmers in deciding whether to switch or not from rainfed to irrigated production, more data and understanding are needed on precipitation, its regional and interannual variation, caused Y_Loss and the cost-effectiveness of irrigation investments. This study aims to assess the spatiotemporal variation in early summer droughts and Y_Loss for spring cereals in the drought-prone Southwest Finland using past weather data (1971–2020). Furthermore, probability of early summer droughts was estimated based on two climate models, MPI-ESM and HadGEM2 and two greenhouse gas concentration scenarios, RCP4.5 and RCP8.5 for 2041–2070. Past data and future estimates of droughts were provided as 10 × 10 km² gridded data. A cost-benefit analysis was used retrospectively to estimate the feasibility of irrigation in 1991–2020. Two irrigation systems were found to be economically feasible for larger farm units and in the case of high farm yield levels. However, projected changes in future precipitation were not substantial for the critical yield determination phase of cereals. Hence, the change in precipitation per se does not necessarily encourage farmers to invest in irrigation in the future but further expanding farm size and higher future cereal yields might act as additional incentives. To conclude, this novel data on precipitation patterns, caused Y_Loss, and economic feasibility may promote irrigation as a key measure to reduce production uncertainties and yield variability in high-latitude conditions, although early summer droughts are not necessarily increasing.

在芬兰，初夏干旱很常见。它们导致春季谷物、大麦、燕麦和小麦的产量损失 ( Y _Loss )，在生长季节后期无法弥补。为了支持农民决定是否从雨育转向灌溉生产，需要更多关于降水、其区域和年际变化、导致Y_损失和灌溉投资成本效益的数据和了解。本研究旨在评估初夏干旱和Y_损失的时空变化使用过去的天气数据（1971-2020 年）在容易干旱的芬兰西南部的春季谷物中进行分析。此外，根据两个气候模型 MPI-ESM 和 HadGEM2 以及 2041-2070 年的两个温室气体浓度情景 RCP4.5 和 RCP8.5，估计了初夏干旱的概率。过去的干旱数据和未来的估计数据以 10 × 10 km ²网格数据形式提供。回顾性地使用成本效益分析来估计 1991-2020 年灌溉的可行性。发现两种灌溉系统对于较大的农场单位和高农场产量水平在经济上是可行的。然而，对于谷物的关键产量确定阶段，未来降水的预测变化并不大。因此，降水本身的变化不一定鼓励农民在未来投资灌溉，但进一步扩大农场规模和更高的未来谷物产量可能会成为额外的激励措施。总而言之，尽管初夏干旱不一定会增加，但这种关于降水模式的新数据、导致Y_损失和经济可行性可能会促进灌溉作为减少高纬度条件下生产不确定性和产量变异的关键措施。