Chironomid-climate continentality conundrum

PLOS One | https://doi.org/10.1371/journal.pone.0327780 August 5, 2025 10 / 18 pallens-type, Neozavrelia, Polypedilum sordens-type, and Microchironomus. The continental-transitional morphotype is Chironomus plumosus-type. Oceanic sites are characterised by Paratanytarsus penicillatus-type, Pseudorthocladius, Thienemannimyia. Oceanic-transitional morphotypes are Procladius, Heterotrissocladius marcidus-type, Sergentia coracina-type, Zalutschia, Chironomus anthracinus-type, Heterotanytarsus, and Tanytarsus chinyensis-type. One morpho- type (Cricotopus) is assigned to the continental-oceanic group. Weighted-average regression (Fig 5; S4 File) reveals that the widest tolerance interval has morphotypes with optima in the transitional part of the dataset (Procladius, Limnophyes, Tanytarsus chinyensis-type) and Thienemannimyia from the oceanic part of the dataset. The smallest tolerances are shown by Microchironomus and Glyptotendipes pallens-type, both from the continental part of the dataset. Inference model for continentality reconstructions The KOI-based two-component WA-PLS inference model has an RMSEP of 5.1, RMSE of 4.3, R2 of 0.72, average bias of −0.1, and maximum bias of 14.6. A scatterplot of the cross-validated predicted vs. observed KOI generally follows a 1:1 relationship (Fig 6A). The reconstruction errors (Fig 6B) indicate an increased error in the transitional and oceanic parts of the dataset. Discussion Effect of environmental and climate variables on the chironomid assemblages

July mean air temperature, GDD5, ice-cover are positively correlated with each other (correlation index >0.8; Fig 2B) and are arranged together along the first axis of both the PCA and RDA plots (Fig 2A and 4). Also, their explanatory powers are relatively similar (Table 2). Thus, July mean air temperature, GDD5, ice-cover reflect the longitudinal gradient, and hence, their individual effects on chironomid assemblages are difficult to distinguish reliably. The discussed variables are positively correlated with the first axis. The positive correlation between bedrock and October mean air temperature is most probably an artefact of the dataset design. The climate variables that did not reveal strong (>0.7) correlations are Fig 5. Weighted-average based Kerner Oceanity Index (KOI) optima and tolerances for morphotypes revealed as indicators by INDVAL. The continentality group affiliation identified by INDVAL is marked in brackets: oc – oceanic, tr – transitional, cont – continental; all taxa revealed statistical significance in the corresponding zone based on IndVal. The background is colored according to the KOI: continental (orange) for KOI < 0, transitional (green) for KOI 0–10, oceanic (blue) for KOI > 10. https://doi.org/10.1371/journal.pone.0327780.g005