The response of the single-celled ciliates to increased temperature during global warming is critical for the structure and functioning of freshwater food webs. I conducted a meta-analysis of the literature from field studies and experimental evidence to assess the parameters characterising the thermal response of freshwater ciliates. The shape of the thermal performance curve predicts the ciliates’ survival at supraoptimal temperatures (i.e., the width of the thermal safety margin, TSM). The ciliates’ typical TSM is ~5°C. One-third of the freshwater ciliates dwelling permanently or occasionally in the pelagial cannot survive at temperatures exceeding 30°C. Likewise, cold-stenothermic species, which represent a significant fraction of euplanktonic ciliates, cannot survive by evolutionary adaptation to rapidly warming environments. The statistical analysis revealed that the ciliates’ thermal performance is affected by their planktonic lifestyle (euplanktonic versus tychoplanktonic), ability to form cysts, and nutritional ecology. Bactivorous ciliates have the widest temperature niche, and algivorous ciliates have the narrowest temperature niche. Phenotypic plasticity and genetic variation, favouring the selection of pre-adapted species in a new environment, are widespread among freshwater ciliates. However, the lack of evidence for the temperature optima and imprecisely defined tolerance limits of most species hamper the present analysis. The extent of acclimation and adaptation requires further research with more ciliate species than the few chosen thus far. Recent eco-evolutionary experimental work and modelling approaches demonstrated that the ciliates’ thermal responses follow general trends predicted by the metabolic theory of ecology and mechanistic functions inherent in enzyme kinetics. The present analysis identified current knowledge gaps and avenues for future research that may serve as a model study for other biota. Thermal adaptation may conflict with adaptation to other stressors (predators, food availability, pH), making general predictions on the future role of freshwater ciliates in a warmer environment difficult, if not impossible, at the moment.

# Data from: Thermal response of freshwater ciliates: can they survive at elevated lake temperatures? [https://doi.org/10.5061/dryad.jdfn2z3jr](https://doi.org/10.5061/dryad.jdfn2z3jr) The dataset results from a meta-analysis to assess the parameters characterising the thermal response of freshwater ciliates (i.e., minimum and maximum temperature tolerated, temperature niche breadth). Cyst formation, the nutritional type, and the planktonic lifestyle were considered as factors affecting the ciliates’ thermal performance. ## Description of the data and file structure The main dataset reporting ciliate species and synonyms, taxonomic affiliation, minimum and maximum temperature and the temperature range tolerated, cysts formation, mixotrophic nutrition, food type, and planktonic lifestyle are reported in the 'Dataset_v4.xlsx' file. This is the main document. Taxonomic affiliation (i.e., order) following Adl et al. (2019, reference [65]J, the GBIF Backbone Taxonomy, and Lynn (2008; reference [66]). Details on the references - i.e., authors, publication year, title, journal/book, volume, and page/article numbers used to compile this dataset and some comments can be found in 'References.xlsx'. Empty cells mean that information is unavailable. References A-E are the main sources of the dataset, i.e., comprehensive review articles published by W. Foissner and colleagues in the 1990s. References 1-64 are case studies, published mainly after 1999. References 65 and 66 refer to the taxonomic affiliation of the ciliate species. More details about each column of the main document can be found in the 'Units_table.xlsx' file. ## Sharing/Access information Data was derived from the following sources: * ISI Web of Science (All Data Bases) * Google Scholar ## Code/Software R statistical software (v 4.0.5, R Core Team 2021) with the packages lme4, lmtest, multcomp, AICcmodavg. WebPlotDigitizer (Version 4.6) for data extraction from figures ## Version changes **06-aug-2024**: Taxonomic affiliation (order) corrected according to GBIF. Genus *Tintinnidium* is now in the order Oligotrichida.

I scrutinised the detailed literature compilations by Foissner and colleagues published in the 1990s; these references are listed as primary sources A-E in the Dataset, see References.xlsx and README.txt) to obtain an overview of the thermal performance, resting cyst formation, and nutritional ecology of planktonic freshwater ciliates. I then searched the ISI Web of Science (All Data Bases) for updates and cross-references of Foissner’s works and further temperature records from (mainly) field studies. Search terms (in all fields) for the latter were ciliate* AND temperature NOT marine NOT ocean NOT soil NOT parasit* (1,339 hits). I followed the PRISMA guidelines in combination with EndNote 20 to filter out the records eligible for screening and analysis. Temperature data for assessing the minimum (Tmin) and maximum temperature (Tmax) of occurrence were eventually extracted from 68 publications. However, because Foissner’s works present extensive reviews, the actual number of publications used for the analysis is much higher. The final dataset obtained from field studies comprised 206 ciliate species. Next, I searched the ISI Web of Science for experimental results, using ciliate* AND temperature AND growth rate* NOT marine as search terms (218 records). Removing results from unsuitable research areas (mainly from medical research) reduced the records to 71 publications, which were screened. The combination of ciliate* AND numerical response NOT marine yielded 40 studies, ciliate* AND thermal performance 21 hits. I checked the selected articles for citations and cross-references using Google Scholar to identify any publications that might have slipped my attention. Eventually, I picked experimental results from 18 studies. If the literature data were only shown in figures, I extracted the data from the plots with WebPlotDigitizer (Version 4.6). I analysed the dataset with the R Statistical Software using the packages lme4, lmerTest, stats, multcomp, AICcmodavg and car.