Reindeer herding is a culturally and economically significant livelihood of local communities in the circumpolar North, strongly depending on environmental conditions. Providing climate information for such a target group requires brining together local knowledge and climate model projections.In this study, information collected in interviews with reindeer herders on what makes a year good or bad for them (critical conditions) was used as a basis for defining indices that can be calculated from climate model projections. In this process, we associated the critical condition to meteorological variables, for example “temperatures above 20°C in June and July” were related to an index tasmax20, which counts the number of days in June and July with daily maximum temperatures above 20°C. In this way, we identified three types of critical conditions/indices (1) indices that can be calculated relatively easily, some of them conforming to climate extreme indices common in climate change analysis; (2) indices that need either more specific information from herders (eg to make a condition “July and August should not be cold”, a specific definition of “cold” is needed) or scientific background knowledge (e.g. “abundance of mosquitos” needs information on the necessary meteorological variables and their thresholds) so they can be calculated; and (3) indices that cannot be calculated from the output climate models commonly provide, because they require for example variables like daily soil temperature (rain on frozen ground) or river ice.We have focussed our analysis on Fennoscandia and northwestern Russia, using data from the CMIP6, EURO CORDEX and Polar CORDEX data bases to calculate in a first step indices from category (1) for different RCP futures. For example, exposure multiplication factors reveal that for tasmax20, increases by the end of the century are up to a factor of 10 for RCP4.5 and a factor of 20 for RCP8.5, with obvious latitudinal patterns.