Supplementary Material

Supplementary Figure 1. Hill & Smith analysis of rate of change in catches and environmental variables. The 3 lakes (L. Peipsi, Vänern and Vättern) where trends in catches could be influenced by strong changes in fishing effort were not included in the analysis. A: projection of the seven variables onto the first factorial plane of the Hill & Smith analysis (axis 1 horizontal and axis 2 vertical) with SON-trend, JJA-trend and MAM-trend being for autumn (September-November), summer (July-August) and spring (March-May) air temperature trends respectively. B-C: projection of the lakes onto the first factorial plane of the Hill & Smith analysis and barycenter of the group of lakes showing significant and non-significant increase or decrease (Increase, IncreaseNs, Decrease, DecreaseNs respectively) in their catches. The group of lakes with significant catch declines segregated significantly (Discriminant analysis, p-value<0.05) from the other lakes, indicating that the selected variables were strong predictors of decline in coregonine catches.

Supplementary Figure 2. A. Comparison of rate of changes in catches among four lakes categories (A: nutrient-rich in a fast warming area, B: nutrient-rich in a slow warming area, C: nutrient-poor in a fast warming area, D: nutrient-poor in a slow warming area). B. Comparison of rate of changes between nutrient-rich and nutrient-poor lakes located in fast warming area. The color of the circles indicates the level of significance of the long-term trend in catches (p-value<0.05: dark blue; not significant: light blue).

Supplementary Table 1. Information on the taxonomic composition of coregonines catches (“No”: only one form was recorded; “Yes”: multiple species or forms were recorded; “Not indicated”: no information available), fishing practices and stressors that have been identified as main drivers of coregonine stocks over the last decades.

Supplementary Table 2. Metadata of the datasets used in the analysis and details on data availability.

Supplementary Table 3. A) Mann-Kendal p-values of the long-term trend in catches and Sen’s slopes values for each lake. B) Results of correlation analysis that were performed on the set of lakes showing significant (p-value <0.05) long-term trends in catches.

© EDP Sciences, 2025