Philippa Kaur

Angilletta, M. J. Thermal Adaptation: A Theoretical and Empirical Synthesis (Oxford Univ. Press, 2009).Cossins, A. R. & Bowler, K. Temperature Biology of Animals (Chapman and Hall, 1987).Sunday, J. M. et al. Thermal-safety margins and the necessity of thermoregulatory behavior across latitude and elevation. Proc. Natl Acad. Sci. USA 111, 5610–5615 (2014).CAS 
PubMed 
PubMed Central 

Google Scholar 
Perry, A. L., Low, P. J., Ellis, J. R. & Reynolds, J. D. Climate change and distribution shifts in marine fishes. Science 308, 1912–1915 (2005).CAS 
PubMed 

Google Scholar 
Kellermann, V. et al. Upper thermal limits of Drosophila are linked to species distributions and strongly constrained phylogenetically. Proc. Natl Acad. Sci. USA 109, 16228–16233 (2012).CAS 
PubMed 
PubMed Central 

Google Scholar 
IPCC. Climate Change 2021: The Physical Science Basis (eds Masson-Delmotte, V. et al.) (Cambridge Univ. Press, 2021).Hofmann, G. E. & Todgham, A. E. Living in the now: physiological mechanisms to tolerate a rapidly changing environment. Annu. Rev. Physiol. 72, 127–145 (2010).CAS 
PubMed 

Google Scholar 
Schulte, P. M. The effects of temperature on aerobic metabolism: towards a mechanistic understanding of the responses of ectotherms to a changing environment. J. Exp. Biol. 218, 1856–1866 (2015).PubMed 

Google Scholar 
Sunday, J. et al. Thermal tolerance patterns across latitude and elevation. Philos. Trans. R. Soc. B 374, 20190036 (2019).
Google Scholar 
Parratt, S. R. et al. Temperatures that sterilize males better match global species distributions than lethal temperatures. Nat. Clim. Change 11, 481–484 (2021).
Google Scholar 
Sunday, J. M., Bates, A. E. & Dulvy, N. K. Thermal tolerance and the global redistribution of animals. Nat. Clim. Change 2, 686–690 (2012).
Google Scholar 
Schmidt-Nielsen, K. Animal physiology: Adaptation and Environment 5th edn (Cambridge Univ. Press, 1997).Dell, A. I., Pawar, S. & Savage, V. M. Systematic variation in the temperature dependence of physiological and ecological traits. Proc. Natl Acad. Sci. USA 108, 10591–10596 (2011).CAS 
PubMed 
PubMed Central 

Google Scholar 
Seebacher, F., White, C. R. & Franklin, C. E. Physiological plasticity increases resilience of ectothermic animals to climate change. Nat. Clim. Change 5, 61–66 (2014).
Google Scholar 
Dillon, M. E., Wang, G. & Huey, R. B. Global metabolic impacts of recent climate warming. Nature 467, 704–706 (2010).CAS 
PubMed 

Google Scholar 
Deutsch, C. A. et al. Increase in crop losses to insect pests in a warming climate. Science 361, 916–919 (2018).CAS 
PubMed 

Google Scholar 
Jørgensen, L. B., Malte, H. & Overgaard, J. How to assess Drosophila heat tolerance: unifying static and dynamic tolerance assays to predict heat distribution limits. Funct. Ecol. 33, 629–642 (2019).
Google Scholar 
Hollingsworth, M. J. Temperature and length of life in Drosophila. Exp. Gerontol. 4, 49–55 (1969).CAS 
PubMed 

Google Scholar 
Fry, F. E. J., Hart, J. S. & Walker, K. F. Lethal Temperature Relations for a Sample of Young Speckled Trout, Salvelinus fontinalis 9–35 (Univ. Toronto, 1946).MacLean, H. J. et al. Evolution and plasticity of thermal performance: an analysis of variation in thermal tolerance and fitness in 22 Drosophila species. Philos. Trans. R. Soc. B 374, 20180548 (2019).
Google Scholar 
Pörtner, H.-O. & Farrell, A. P. Physiology and climate change. Science 322, 690–692 (2008).PubMed 

Google Scholar 
Ørsted, M., Jørgensen, L. B. & Overgaard, J. Finding the right thermal limit: a framework to reconcile ecological, physiological, and methodological aspects of CTmax in ectotherms. J. Exp. Biol. 225, jeb244514 (2022).Brown, J. H., Gillooly, J. F., Alle, A. P., Savage, V. M. & West, G. B. Toward a metabolic theory of ecology. Ecology 85, 1771–1789 (2004).
Google Scholar 
Munch, S. B. & Salinas, S. Latitudinal variation in lifespan within species is explained by the metabolic theory of ecology. Proc. Natl Acad. Sci. USA 106, 13860–13864 (2009).CAS 
PubMed 
PubMed Central 

Google Scholar 
Jørgensen, L. B., Malte, H., Ørsted, M., Klahn, N. A. & Overgaard, J. A unifying model to estimate thermal tolerance limits in ectotherms across static, dynamic and fluctuating exposures to thermal stress. Sci. Rep. 11, 12840 (2021).PubMed 
PubMed Central 

Google Scholar 
Rezende, E. L., Castañeda, L. E. & Santos, M. Tolerance landscapes in thermal ecology. Funct. Ecol. 28, 799–809 (2014).
Google Scholar 
Bowler, K. Heat death in poikilotherms: is there a common cause? J. Therm. Biol. 76, 77–79 (2018).PubMed 

Google Scholar 
Somero, G. N. The physiology of climate change: how potentials for acclimatization and genetic adaptation will determine ‘winners’ and ‘losers’. J. Exp. Biol. 213, 912–920 (2010).CAS 
PubMed 

Google Scholar 
Buckley, L. B., Huey, R. B. & Kingsolver, J. G. Asymmetry of thermal sensitivity and the thermal risk of climate change. Glob. Ecol. Biogeogr. 31, 2231–2244 (2022).Overgaard, J., Kearney, M. R. & Hoffmann, A. A. Sensitivity to thermal extremes in Australian Drosophila implies similar impacts of climate change on the distribution of widespread and tropical species. Glob. Change Biol. 20, 1738–1750 (2014).
Google Scholar 
Pinsky, M. L., Eikeset, A. M., McCauley, D. J., Payne, J. L. & Sunday, J. M. Greater vulnerability to warming of marine versus terrestrial ectotherms. Nature 569, 108–111 (2019).CAS 
PubMed 

Google Scholar 
Huey, R. B. et al. Predicting organismal vulnerability to climate warming: roles of behaviour, physiology and adaptation. Philos. Trans. R. Soc. B 367, 1665–1679 (2012).
Google Scholar 
Kearney, M., Shine, R. & Porter, W. P. The potential for behavioral thermoregulation to buffer ‘cold-blooded’ animals against climate warming. Proc. Natl Acad. Sci. USA 106, 3835–3840 (2009).CAS 
PubMed 
PubMed Central 

Google Scholar 
Woods, H. A., Dillon, M. E. & Pincebourde, S. The roles of microclimatic diversity and of behavior in mediating the responses of ectotherms to climate change. J. Therm. Biol 54, 86–97 (2015).PubMed 

Google Scholar 
Stevenson, R. D. The relative importance of behavioral and physiological adjustments controlling body temperature in terrestrial ectotherms. Am. Nat. 126, 362–386 (1985).
Google Scholar 
Chen, I., Hill, J. K., Ohlemüller, R., Roy, D. B. & Thomas, C. D. Rapid range shifts of species associated with high levels of climate warming. Science 333, 1024–1026 (2011).CAS 
PubMed 

Google Scholar 
Buckley, L. B. & Kingsolver, J. G. Functional and phylogenetic approaches to forecasting species’ responses to climate change. Annu. Rev. Ecol. Evol. Syst. 43, 205–226 (2012).
Google Scholar 
Roeder, K. A., Bujan, J., de Beurs, K. M., Weiser, M. D. & Kaspari, M. Thermal traits predict the winners and losers under climate change: an example from North American ant communities. Ecosphere 12, e03645 (2021).
Google Scholar 
Penick, C. A., Diamond, S. E., Sanders, N. J. & Dunn, R. R. Beyond thermal limits: comprehensive metrics of performance identify key axes of thermal adaptation in ants. Funct. Ecol. 31, 1091–1100 (2017).
Google Scholar 
Deutsch, C. A. et al. Impacts of climate warming on terrestrial ectotherms across latitude. Proc. Natl Acad. Sci. USA 105, 6668–6672 (2008).CAS 
PubMed 
PubMed Central 

Google Scholar 
Huey, R. B. & Stevenson, R. D. Integrating thermal physiology and ecology of ectotherms: a discussion of approaches. Integr. Comp. Biol. 19, 357–366 (1979).
Google Scholar 
Sinclair, B. J. et al. Can we predict ectotherm responses to climate change using thermal performance curves and body temperatures? Ecol. Lett. 19, 1372–1385 (2016).PubMed 

Google Scholar 
Tewksbury, J. J., Huey, R. B. & Deutsch, C. A. Putting the heat on tropical animals the scale of prediction. Science 320, 1296–1297 (2008).CAS 
PubMed 

Google Scholar 
Kingsolver, J. G., Diamond, S. E. & Buckley, L. B. Heat stress and the fitness consequences of climate change for terrestrial ectotherms. Funct. Ecol. 27, 1415–1423 (2013).
Google Scholar 
Kingsolver, J. G. & Woods, H. A. Beyond thermal performance curves: modeling time-dependent effects of thermal stress on ectotherm growth rates. Am. Nat. 187, 283–294 (2016).PubMed 

Google Scholar 
Kingsolver, J. G., Higgins, J. K. & Augustine, K. E. Fluctuating temperatures and ectotherm growth: distinguishing non-linear and time-dependent effects. J. Exp. Biol. 218, 2218–2225 (2015).PubMed 

Google Scholar 
Clusella-Trullas, S., Garcia, R. A., Terblanche, J. S. & Hoffmann, A. A. How useful are thermal vulnerability indices? Trends Ecol. Evol. 36, 1000–1010 (2021).PubMed 

Google Scholar 
Pincebourde, S. & Casas, J. Narrow safety margin in the phyllosphere during thermal extremes. Proc. Natl Acad. Sci. USA 116, 5588–5596 (2019).CAS 
PubMed 
PubMed Central 

Google Scholar 
Fick, S. E. & Hijmans, R. J. WorldClim 2: new 1‐km spatial resolution climate surfaces for global land areas. Int. J. Climatol. 37, 4302–4315 (2017).
Google Scholar 
Moss, R. H. et al. The next generation of scenarios for climate change research and assessment. Nature 463, 747–756 (2010).CAS 
PubMed 

Google Scholar 
Hausfather, Z. & Peters, G. P. Emissions—the ‘business as usual’ story is misleading. Nature 577, 618–620 (2020).CAS 
PubMed 

Google Scholar 
Tollefson, J. How hot will Earth get by 2100? Nature 580, 443–445 (2020).CAS 
PubMed 

Google Scholar 
Assis, J. et al. Bio‐ORACLE v2.0: extending marine data layers for bioclimatic modelling. Glob. Ecol. Biogeogr. 27, 277–284 (2018).
Google Scholar 
Tyberghein, L. et al. Bio-ORACLE: a global environmental dataset for marine species distribution modelling. Glob. Ecol. Biogeogr. 21, 272–281 (2012).
Google Scholar 
Jørgensen, L. B., Ørsted, M., Malte, H., Wang, T. & Overgaard, J. Data from: Extreme escalation of heat failure rates in ectotherms with global warming. Zenodo https://doi.org/10.5281/zenodo.6979789 (2022).Grove, T. J., McFadden, L. A., Chase, P. B. & Moerland, T. S. Effects of temperature, ionic strength and pH on the function of skeletal muscle myosin from a eurythermal fish, Fundulus heteroclitus. J. Muscle Res. Cell Motil. 26, 191–197 (2005).CAS 
PubMed 

Google Scholar 
Doudoroff, P. The resistance and acclimatization of marine fishes to temperature changes. II. Experiments with Fundulus and Atherinops. Biol. Bull. 88, 194–206 (1945).
Google Scholar 
Sirikharin, R., Söderhäll, I. & Söderhäll, K. Characterization of a cold-active transglutaminase from a crayfish, Pacifastacus leniusculus. Fish Shellfish Immunol. 80, 546–549 (2018).CAS 
PubMed 

Google Scholar 
Becker, C. D. & Genoway, R. G. Resistance of crayfish to acute thermal shock: preliminary studies. in Proc. Thermal Ecology NTIS Conf. 730505 (eds Gibbons, J. W. & Sharitz, R. R.) 146–150 (NTIS, 1974).Widdows, J. Effect of temperature and food on the heart beat, ventilation rate and oxygen uptake of Mytilus edulis. Mar. Biol. 20, 269–276 (1973).
Google Scholar 
Wallis, R. L. Thermal tolerance of Mytilus edulis of eastern Australia. Mar. Biol. 30, 183–191 (1975).
Google Scholar 
Gray, J. The mechanism of ciliary movement. III. The effect of temperature. Proc. R. Soc. B 95, 6–15 (1923).CAS 

Google Scholar 
Shertzer, R. H., Hart, R. G. & Pavlick, F. M. Thermal acclimation in selected tissues of the leopard frog Rana pipiens. Comp. Biochem. Physiol. A 51, 327–334 (1975).CAS 
PubMed 

Google Scholar 
Orr, P. R. Heat death. II. Differential response of entire animal (Rana pipiens) and several organ systems. Physiol. Zool. 28, 294–302 (1955).
Google Scholar 
Lighton, J. R. B. & Duncan, F. D. Energy cost of locomotion: validation of laboratory data by in situ respirometry. Ecology 83, 3517–3522 (2002).
Google Scholar 
Heatwole, H. & Harrington, S. Heat tolerances of some ants and beetles from the pre-Saharan steppe of Tunisia. J. Arid Environ. 16, 69–77 (1989).
Google Scholar  More

Carothers, J. H. & Jaksić, F. M. Time as a Niche difference: The role of interference competition. Oikos 42, 403–406 (1984).
Google Scholar 
Kronfeld-Schor, N. & Dayan, T. Partitioning of time as an ecological resource. Annu. Rev. Ecol. Evol. Syst. 34, 153–181 (2003).
Google Scholar 
Lesmeister, D. B., Nielsen, C. K., Schauber, E. M. & Hellgren, E. C. Spatial and temporal structure of a mesocarnivore guild in midwestern North America. Wildl. Monogr. 191, 1–61 (2015).
Google Scholar 
Chmura, H. E. et al. Plasticity and repeatability of activity patterns in free-living Arctic ground squirrels. Anim. Behav. 169, 81–91 (2020).
Google Scholar 
Helm, B. et al. Two sides of a coin: Ecological and chronobiological perspectives of timing in the wild. Philos. Trans. R. Soc. B Biol. Sci. 372, 20160246 (2017).
Google Scholar 
Alós, J., Martorell-Barceló, M. & Campos-Candela, A. Repeatability of circadian behavioural variation revealed in free-ranging marine fish. R. Soc. Open Sci. 4, 160791 (2017).PubMed 
PubMed Central 

Google Scholar 
Schlicht, L. & Kempenaers, B. The effects of season, sex, age and weather on population-level variation in the timing of activity in Eurasian Blue Tits Cyanistes caeruleus. Ibis 162, 1146–1162 (2020).
Google Scholar 
Helm, B. & Visser, M. E. Heritable circadian period length in a wild bird population. Proc. R. Soc. B Biol. Sci. 277, 3335–3342 (2010).
Google Scholar 
Nikhil, K. L., Abhilash, L. & Sharma, V. K. Molecular correlates of circadian clocks in fruit fly drosophila melanogaster populations exhibiting early and late emergence chronotypes. J. Biol. Rhythms 31, 125–141 (2016).CAS 
PubMed 

Google Scholar 
Allebrandt, K. V. et al. CLOCK gene variants associate with sleep duration in two independent populations. Biol. Psychiatry 67, 1040–1047 (2010).CAS 
PubMed 

Google Scholar 
Maukonen, M. et al. Genetic associations of chronotype in the finnish general population. J. Biol. Rhythms 35, 501–511 (2020).CAS 
PubMed 
PubMed Central 

Google Scholar 
Roecklein, K. A. et al. Melanopsin gene variations interact with season to predict sleep onset and chronotype. Chronobiol. Int. 29, 1036–1047 (2012).CAS 
PubMed 
PubMed Central 

Google Scholar 
Steinmeyer, C., Kempenaers, B. & Mueller, J. C. Testing for associations between candidate genes for circadian rhythms and individual variation in sleep behaviour in blue tits. Genetica 140, 219–228 (2012).CAS 
PubMed 

Google Scholar 
Stuber, E. F., Baumgartner, C., Dingemanse, N. J., Kempenaers, B. & Mueller, J. C. Genetic correlates of individual differences in sleep behavior of free-living great tits (Parus major). G3 GenesGenomesGenetics 6, 599–607 (2016).CAS 

Google Scholar 
Cuthill, I. C. & Macdonald, W. A. Experimental manipulation of the dawn and dusk chorus in the blackbird Turdus merula. Behav. Ecol. Sociobiol. 26, 209–216 (1990).
Google Scholar 
Grava, T., Grava, A. & Otter, K. A. Supplemental feeding and dawn singing in black-capped chickadees. Condor 111, 560–564 (2009).
Google Scholar 
Saggese, K., Korner-Nievergelt, F., Slagsvold, T. & Amrhein, V. Wild bird feeding delays start of dawn singing in the great tit. Anim. Behav. 81, 361–365 (2011).
Google Scholar 
Dominoni, D. M. Effects of artificial light at night on daily and seasonal organization of European blackbirds (Turdus merula). https://kops.uni-konstanz.de/handle/123456789/32198 Accessed 23 February 2022 (2013).
Lehmann, M., Spoelstra, K., Visser, M. E. & Helm, B. Effects of temperature on circadian clock and chronotype: An experimental study on a passerine bird. Chronobiol. Int. 29, 1062–1071 (2012).PubMed 

Google Scholar 
Zsebők, S. et al. Short- and long-term repeatability and pseudo-repeatability of bird song: Sensitivity of signals to varying environments. Behav. Ecol. Sociobiol. 71, 154 (2017).
Google Scholar 
Raap, T., Pinxten, R. & Eens, M. Artificial light at night disrupts sleep in female great tits (Parus major) during the nestling period and is followed by a sleep rebound. Environ. Pollut. 215, 125–134 (2016).CAS 
PubMed 

Google Scholar 
Grunst, M. L., Grunst, A. S., Pinxten, R. & Eens, M. Variable and consistent traffic noise negatively affect the sleep behavior of a free-living songbird. Sci. Total Environ. 778, 146338 (2021).CAS 
PubMed 

Google Scholar 
Gaynor, K. M., Hojnowski, C. E., Carter, N. H. & Brashares, J. S. The influence of human disturbance on wildlife nocturnality. Science 360, 1232–1235 (2018).CAS 
PubMed 

Google Scholar 
Stuber, E. F. et al. Perceived predation risk affects sleep behaviour in free-living great tits Parus major. Anim. Behav. 98, 157–165 (2014).
Google Scholar 
Niemelä, P. T. & Dingemanse, N. J. Individual versus pseudo-repeatability in behaviour: Lessons from translocation experiments in a wild insect. J. Anim. Ecol. 86, 1033–1043 (2017).PubMed 

Google Scholar 
Garamszegi, L. Z. & Møller, A. P. Partitioning within-species variance in behaviour to within- and between-population components for understanding evolution. Ecol. Lett. 20, 599–608 (2017).PubMed 

Google Scholar 
Niemelä, P. T. & Dingemanse, N. J. On the usage of single measurements in behavioural ecology research on individual differences. Anim. Behav. 145, 99–105 (2018).
Google Scholar 
Browne, W. J., McCleery, R. H., Sheldon, B. C. & Pettifor, R. A. Using cross-classified multivariate mixed response models with application to life history traits in great tits (Parus major). Stat. Model. 7, 217–238 (2007).MathSciNet 
MATH 

Google Scholar 
Pettifor, R. A., Sheldon, B. C., Browne, W. J., Rasbash, J. & McCleery, R.
H. Partitioning of Phenotypic Variance in Life-history Traits in the Great Tit, Parus major.
https://seis.bristol.ac.uk/~frwjb/materials/phenovar.pdf (2003). Accessed 23 February 2022.Casasole, G. et al. Neither artificial light at night, anthropogenic noise nor distance from roads are associated with oxidative status of nestlings in an urban population of songbirds. Comp. Biochem. Physiol. A 210, 14–21 (2017).CAS 

Google Scholar 
Payevsky, V. A. Mortality rate and population density regulation in the great tit, Parus major L.: A review. Russ. J. Ecol. 37, 180 (2006).
Google Scholar 
Vermeulen, A., Eens, M., Van Dongen, S. & Müller, W. Does baseline innate immunity change with age? A multi-year study in great tits. Exp. Gerontol. 92, 67–73 (2017).CAS 
PubMed 

Google Scholar 
Haftorn, S. Incubation during the egg-laying period in relation to clutch-size and other aspects of reproduction in the great tit Parus major. Ornis Scand. Scand. J. Ornithol. 12, 169–185 (1981).
Google Scholar 
Grunst, M. L., Grunst, A. S., Pinxten, R., Eens, G. & Eens, M. An experimental approach to investigating effects of artificial light at night on free-ranging animals: Implementation, results and directions for future research. J. Vis. Exp. 180, e63381 (2022).

Google Scholar 
Halfwerk, W. et al. Low-frequency songs lose their potency in noisy urban conditions. Proc. Natl. Acad. Sci. 108, 14549–14554 (2011).CAS 
PubMed 
PubMed Central 

Google Scholar 
Specht, R. Avisoft-saslab pro: Sound analysis and synthesis laboratory. Avis. Bioacoustics
http://avisoft.com/SASLab_deutsch.pdf Accessed 23 February 2022 (2002).Iserbyt, A., Griffioen, M., Borremans, B., Eens, M. & Müller, W. How to quantify animal activity from radio-frequency identification (RFID) recordings. Ecol. Evol. 8, 10166–10174 (2018).PubMed 
PubMed Central 

Google Scholar 
Raap, T., Pinxten, R. & Eens, M. Light pollution disrupts sleep in free-living animals. Sci. Rep. 5, 13557 (2015).PubMed 
PubMed Central 

Google Scholar 
Meijdam, M., Müller, W., Thys, B. & Eens, M. No relationship between chronotype and timing of breeding when variation in daily activity patterns across the breeding season is taken into account. Ecol. Evol. 12, e9353 (2022).PubMed 
PubMed Central 

Google Scholar 
R Core Team. R: a language and environment for statistical computing. R Found. Stat. Comput. https://www.R-project.org/ Accessed 23 February 2022 (2013).Rousset, F. & Ferdy, J.-B. Testing environmental and genetic effects in the presence of spatial autocorrelation. Ecography 37, 781–790 (2014).
Google Scholar 
Araya-Ajoy, Y. G., Mathot, K. J. & Dingemanse, N. J. An approach to estimate short-term, long-term and reaction norm repeatability. Methods Ecol. Evol. 6, 1462–1473 (2015).
Google Scholar 
Mitchell, D. J., Dujon, A. M., Beckmann, C. & Biro, P. A. Temporal autocorrelation: A neglected factor in the study of behavioral repeatability and plasticity. Behav. Ecol. 31, 222–231 (2020).
Google Scholar 
Bell, A. M., Hankison, S. J. & Laskowski, K. L. The repeatability of behaviour: A meta-analysis. Anim. Behav. 77, 771–783 (2009).PubMed 
PubMed Central 

Google Scholar 
Graham, J. L., Cook, N. J., Needham, K. B., Hau, M. & Greives, T. J. Early to rise, early to breed: A role for daily rhythms in seasonal reproduction. Behav. Ecol. 28, 1266–1271 (2017).
Google Scholar 
Maury, C., Serota, M. W. & Williams, T. D. Plasticity in diurnal activity and temporal phenotype during parental care in European starlings Sturnus vulgaris. Anim. Behav. 159, 37–45 (2020).
Google Scholar 
Schlicht, L., Valcu, M., Loës, P., Girg, A. & Kempenaers, B. No relationship between female emergence time from the roosting place and extrapair paternity. Behav. Ecol. 25, 650–659 (2014).
Google Scholar 
Steinmeyer, C., Schielzeth, H., Mueller, J. C. & Kempenaers, B. Variation in sleep behaviour in free-living blue tits, Cyanistes caeruleus: Effects of sex, age and environment. Anim. Behav. 80, 853–864 (2010).
Google Scholar 
Stuber, E. F., Dingemanse, N. J., Kempenaers, B. & Mueller, J. C. Sources of intraspecific variation in sleep behaviour of wild great tits. Anim. Behav. 106, 201–221 (2015).
Google Scholar 
Raap, T., Pinxten, R. & Eens, M. Cavities shield birds from effects of artificial light at night on sleep. J. Exp. Zool. Part Ecol. Integr. Physiol. 329, 449–456 (2018).
Google Scholar 
Edelaar, P., Siepielski, A. M. & Clobert, J. Matching habitat choice causes directed gene flow: A neglected dimension in evolution and ecology. Evolution 62, 2462–2472 (2008).PubMed 

Google Scholar 
Gorissen, L. & Eens, M. Interactive communication between male and female great tits (Parus major) during the dawn chorus. Auk 121, 184–191 (2004).
Google Scholar 
Halfwerk, W., Bot, S. & Slabbekoorn, H. Male great tit song perch selection in response to noise-dependent female feedback. Funct. Ecol. 26, 1339–1347 (2012).
Google Scholar 
Steinmeyer, C., Mueller, J. C. & Kempenaers, B. Individual variation in sleep behaviour in blue tits Cyanistes caeruleus: Assortative mating and associations with fitness-related traits. J. Avian Biol. 44, 159–168 (2013).
Google Scholar 
Cain, J. R. & Wilson, W. O. The influence of specific environmental parameters on the circadian rhythms of chickens. Poult. Sci. 53, 1438–1447 (1974).CAS 
PubMed 

Google Scholar 
Zhang, Z. C. et al. Circadian clock genes are rhythmically expressed in specific segments of the hen oviduct. Poult. Sci. 95, 1653–1659 (2016).CAS 
PubMed 

Google Scholar 
Womack, R. J. Clocks in the wild: biological rhythms of great tits and the environment. https://theses.gla.ac.uk/81345/ Accessed 23 February 2022 (2020).Dominoni, D., Smit, J. A. H., Visser, M. E. & Halfwerk, W. Multisensory pollution: Artificial light at night and anthropogenic noise have interactive effects on activity patterns of great tits (Parus major). Environ. Pollut. 256, 113314 (2020).CAS 
PubMed 

Google Scholar 
Matthysen, E., Adriaensen, F. & Dhondt, A. A. Multiple responses to increasing spring temperatures in the breeding cycle of blue and great tits (Cyanistes caeruleus, Parus major). Glob. Change Biol. 17, 1–16 (2011).
Google Scholar  More

Mangrove cover change variables. We used the Global Mangrove Watch (GMW) v2.0 dataset from 1996 to 201656 to calculate four response variables across landscape mangrove geomorphic units24 over two time periods, 1996–2007 and 2007–2016: (1) percent net loss (units that had a net change in mangrove cover of 0), (3) percent gross loss (units that had a decrease in mangrove cover, not accounting for any increase), and (4) percent gross gain (units that had an increase in mangrove cover, not accounting for any decrease). Percent variables were calculated relative to the area at the start of the time period and were log transformed to meet the assumptions of the statistical models. We initially also considered 5 primary response variables (Supplementary Table 3), including net change in mangrove area ranging from negative (loss) to zero (no change) to positive (gain), however, the data did not meet model assumptions of equal variance (Supplementary Table 9). It was therefore necessary to separate areas of net loss and net gain and areas of gross loss and gross gain to remove zeros and log-transform to achieve normal distribution. Area of mangrove change was correlated with size of the mangrove geomorphic unit (higher area of mangrove loss or gain in bigger units), therefore we included geomorphic unit size as an explanatory variable in the models with primary response variables. We selected the transformations of these primary variables – percent net loss, percent net gain, percent gross loss, and percent gross gain to include in the analysis, because the percent changes control for differences in relative sizes of geomorphic units and because net change alone can underestimate the extent of change57.Examining mangrove change across geomorphic settings is likely to be relevant to socioeconomic and environmental conditions. Mangroves occur in the intertidal zone in diverse coastal geomorphic settings (e.g., deltas, estuaries, lagoons) shaped by rivers, tides, and waves58,59. The distribution, structure, and productivity of mangroves varies spatially with regional climate and local geomorphological processes (e.g., river discharge, tidal range, hydroperiod, and wave activity) that control soil biogeochemistry60,61,62,63. These geomorphic settings are defined by natural landscape boundaries (e.g., catchments/bays) which also often delineate boundaries of human settlements. A global mangrove biophysical typology v2.2 dataset64 was used for the delineation of landscape mangrove geomorphic units, which used a composite of the GMW dataset from the 1996, 2007, 2010, and 2016 timesteps to classify the maximal extent of mangrove cover into 4394 units (classified as delta, estuarine, lagoon or open coast). The mangrove geomorphic units do not include non-mangrove patches, unless they have been lost from the unit over time. The mean size of geomorphic units was 33.63 ha. Some splits of geomorphic units were undertaken to reduce size and divide by country boundaries. The four largest deltas (northern Brazil Delta ID 70000, Sundarbans Delta ID 70004, Niger Delta ID 70009, and Papua coast Delta ID 70013) were split into 4, 5, 4, and 2 units, respectively to aid with data processing. Mangrove geomorphic units that overlapped two countries (Peru/Ecuador, Singapore/Malaysia, and Papua New Guinea/Australia) were split by the national boundary.The country governing each geomorphic unit was assigned to match national-level variables to geomorphic units. To capture mangroves that are mapped outside of country coastline boundaries, we did a union of the GADM country shapefile v3.665 and the Exclusive Economic Zones (EEZs) v1166. The following manual country designations were made to resolve overlapping claims in the EEZs: (1) Hong Kong was merged with China as Hong Kong does not have a mapped EEZ; (2) The overlapping claim of Sudan/Egypt was maintained as a joint Sudan/Egypt designation, as this is an area of disputed land called the Halayib Triangle. However, for this study, mangrove units within this area were assigned to Egypt because Egypt currently has military control over the area; (3) Mayotte (claimed by France and Comoros) was assigned to Mayotte as it is a separate overseas territory of France recognised in GADM that has different socioeconomic variables; (4) The protected zone established under the Torres Strait Treaty was assigned to Australia as these islands are Australian territory.Areas of mangrove cover in 1996, 2007, and 2016, and gross losses and gains in each geomorphic units over the two time periods were assessed in ArcMap 10.867. Percent losses and gains were calculated in R 4.0.268. In using the GMW mapping, a minimum mapping unit of 1 ha is recommended for reliable results5, therefore we removed all geomorphic units less than 1 ha from the analysis, which reduced the available sample size from 4394 across 108 countries to 4235 units across 108 countries. In calculating percent net gains, 11 and 12 of the units returned infinity values for 1996–2007 and 2007–2016, respectively, because there was no initial mangrove cover. In these instances, 100% gain was assigned to these units.Socioeconomic variables (Supplementary Table 4)Economic growthPrevious global analyses of mangroves have been limited by data availability on economic activity to national metrics, such as a country’s Gross Domestic Product (GDP)12,18. Night-time lights satellite data provide local measures of economic activity that are comparable through time and available globally9,69. The data improve estimates of GDP in low to middle income countries69 and are strongly correlated with local indicators of human development70 and electricity consumption and GDP at the national-level71. We used the Night-time Lights Time Series v472 stable lights data, where transient lights that are deemed ephemeral, e.g., fires, have been filtered out and non-lit areas set to zero73, choosing the newer satellites where applicable70. As a proxy for local economic growth, we calculated the change in annual average stable lights within a 100 km buffer of the centroid of each geomorphic unit from 1996 to 2007 and 2007 to 2013 (no data available past 2013) using the ‘raster’ package in R74. The 100 km buffer was chosen to account for pressures from human activity within and surrounding the mangrove area, and to avoid bias with larger spatial units70.Market accessibilityTravel time to the nearest major market (national or provincial capital, landmark city, or major population centre) has been shown to be a stronger predictor of fish biomass on coral reefs than population density or linear distance to markets27. We used the global map of travel time to cities for 201575 to estimate the average travel time from each geomorphic unit to the nearest city via surface transport using the ‘raster’ package in R74, as an indicator of access to markets to trade commodities (e.g., rice, shrimp, palm oil).Economic complexityPrevious studies have examined the effect of GDP on mangrove change18, however, this is a blunt measure of country capability. Measuring a country’s economic complexity, that is the diversified capability of a nation’s economy, is preferable. For example, a country with high GDP but low economic complexity can be prone to regulatory capture by high-value natural resource industries and resource corruption26. Therefore, we used the Economic Complexity Index (ECI)76 for countries as an indicator of regulatory independence. The ECI had better coverage of countries in later years (Supplementary Table 4), therefore the ECI for the end of the time periods was used (2007 and 2016), although we recognise this may reduce the detection of trends because of potential time lags in impacts.DemocracyWe used the Varieties of Democracy (VDEM) index v10 which measures a country’s degree of freedom of association, clean elections, freedom of expression, elected executives, and suffrage77, and has been indicated to influence NDC ambition in countries to address climate change78. We adopted the VDEM index for the start of the time periods (1996 and 2007) to account for potential time lags in impacts.Community forestry supportWe determined the extent that community forestry (CF) is implemented across countries through a systematic review of articles returned in the Web of Science database (Core collection; Thomson Reuters, New York, U.S.A.). We used the search terms: TS = (“community forestry” OR “community-based forestry” OR “social forestry”) AND (TI = ”country” OR AB = ”country”) to identify how many CF case studies were reported in each country, and whether any were in mangroves. As scientific literature is biased towards particular regions, we also reviewed relevant FAO global studies79,80,81 and online databases (ICCA registry82 and REDD projects database83) to identify additional case studies (Supplementary Fig. 5). We then generated scores of 0–3 for each country based on summing values assessed using these criteria: +1 (1–50 CF case studies); +2 ( >50 CF case studies); +1 (CF case study in mangroves). There may have been some double counting as we counted the number of case studies in each article, and we will have missed CF projects not published or communicated in English. However, this is likely to have had a limited impact on the scoring method.Indigenous landThe proportion of Indigenous peoples’ land versus other land per country was calculated from national-level data84. Whilst this study involved Indigenous peoples’ land mapping at a global scale, the spatial data was not published, and thus we could only evaluate the influence of Indigenous land at the national level rather than local level.Restoration effortThe number of mangrove restoration sites per country was calculated from combining two datasets collated by C. Lovelock (2020) and Y.M. Gatt and T.A. Worthington (2020) identifying mangrove restoration project locations from web searches in English and for scientific and grey literature using Google Scholar. Duplications were removed and the number of sites was used as an indicator of effort. This will underrepresent effort in countries with few, large sites, and where restoration projects are not published or communicated in English.Climate commitmentsThe Paris Agreement is a global programme for countries to commit to climate action by submitting Nationally Determined Contributions (NDCs) to the United Nations Framework for the Convention of Climate Change (UNFCCC). First, we reviewed NDCs for mangrove-holding nations from the NDC Registry85 submitted as of 07/01/2021 to determine the extent that mangroves or coastal ecosystems were included in national climate policy (scoring method in Supplementary Table 4). We hypothesised that countries with mangrove or coastal ecosystem NDCs may be more likely to promote mangrove conservation or restoration. While the first NDCs were submitted around 2015, at the end of our time series, we suspected higher commitments would point towards a stronger baseline in environmental governance. Most countries submitted updated or second NDCs during 2021 however these were not considered relevant to the time periods assessed. Google Translate was used to interpret NDCs in languages other than English.Ramsar wetlandsThe ecological character of Ramsar wetlands have been found to be significantly better than those of wetlands generally86. The area of Ramsar coastal and marine wetlands from the Ramsar Sites Information Service87 was calculated per country. Thirty-eight mangrove-holding countries are not signatories to the Ramsar Convention, and these countries were assigned a value of 0. The area of Ramsar wetlands per country was scaled by dividing by the country’s area of mangroves in 1996.Environmental governanceWe assessed the Environmental Performance Index (EPI)88 as an indicator of a country’s effectiveness in environmental governance. The biodiversity and habitat (BDH) issue category assesses countries’ actions toward retaining natural ecosystems and protecting the full range of biodiversity within their borders. We took the BDH score for 2020 for the 2007–2016 time period and the BDH score for 2010 for the 1996–2007 time period (calculated by subtracting the ten-year change from BDH 2020). However, due to collinearity with other variables this index was excluded from the analysis (see statistical analysis).Protected area managementWe also assessed Marine Protected Area (MPA) staff capacity as an indicator of the effectiveness of management of protected areas for countries. We used published global marine protected area (MPA) management data14 which is based on the Management Effectiveness Tracking Tool (METT), the World Bank MPA Score Card, and the NOAA Coral Reef Conservation Programme’s MPA Management Assessment Checklist. Adequate staff capacity was the most important factor in explaining fish responses to MPA management globally, followed by budget capacity, but they were significantly correlated14. We, therefore, calculated the mean staff capacity across MPAs per country as our indicator. Mangroves can be included in terrestrial protected areas, which are not represented in this dataset, however, this measure provides an indicator of national governance of protected areas. However, due to collinearity with other variables this indicator was excluded from the analysis (see statistical analysis). The extent of protected areas was not included in the analysis because it has already been found to influence mangrove loss18.Biophysical variables (Supplementary Table 5)Coastal geomorphic typeMangrove extent change likely varies among different coastal geomorphic settings because human activities or environmental changes occur more commonly in some geomorphic settings than others. For example, losses of lagoonal mangroves were nearly twice as large as those in other geomorphic types24. Landscape geomorphic units from the global mangrove typology dataset v2.264 were classified as delta, estuary, lagoon or open coast.Sediment availabilityMangrove expansion and retreat are driven by sediment deposition and erosion, which are influenced by sediment availability from rivers and wave action, and alterations in hydrodynamic regimes47,89. We used the sediment trapping index from the global free-flowing rivers (FFR) dataset90 to indicate sediment availability from rivers within different geomorphic units. A mangrove catchment dataset was created based on the HydroSHEDS database91. River networks that intersected with mangrove geomorphic units were linked to that unit’s ID. Where rivers intersected multiple units, they were manually assigned by visual inspection. River basins that intersected either with the geomorphic units directly or the river networks were also linked to that unit’s ID. The FFR dataset90 was then spatially joined to the mangrove catchment dataset to identify the most downstream (i.e., the coastal outlet) segment of each FFR and its associated sediment trapping index. Not all geomorphic units (n = 3475) were linked to an FFR, however, an individual unit could be linked with several FFRs. Therefore, the unit sediment trapping index was the weighted mean of the river values, with weighting based on each FFR’s average long-term (1971–2000) naturalised discharge (m3s−1), with discharge set to the minimum value for segments with zero flow. Geomorphic units without connecting FFRs were given an index of zero (no sediment trapping). The sediment trapping index represents the percentage of the potential sediment load trapped by anthropogenic barriers along the river section. The focus on river barriers may obscure larger scale oceanic patterns that influence mangrove losses and gains (e.g., movement of mud banks from the Amazon River over 1000’s of kilometres92) or increases in sediment that could be coming from soils with catchment deforestation and erosion.Habitat fragmentationMany countries with high mangrove loss have been associated with elevated fragmentation of mangrove forests, although the relationship is not consistent at the global scale93. We calculated the clumpiness index of mangrove patches within geomorphic units within each time period, as this habitat fragmentation metric is independent of areal extent93. Whilst habitat fragmentation can be human-driven, clumpiness measures the patchy distribution of mangroves, which can also be due to natural factors inducing edge effects. We used a similar approach to Bryan-Brown, et al.86 to quantify the clumpiness index. The ‘landscape’ was defined as the combined extent of the mangrove geomorphic units across four timesteps (1996, 2007, 2010, and 2016) from the GMW dataset56. For the three focal years in this study (1996, 2007, and 2016) each geomorphic unit (n = 4394) was converted into a two-class polygon, where class one represented mangroves present during that time step and class two mangroves present in the other time steps (i.e., areas of mangrove loss). The polygons were transformed to a projected coordinate system (World Cylindrical Equal Area) and converted to rasters with a resolution of 25 m. Each raster was imported into R version 3.6.394, with clumpiness calculated using the package ‘landscapemetrics’ v1.5.095.Clumpiness describes how patches are dispersed across the landscape and ranges between minus one, where patches are maximally disaggregated, to one, where patches are maximally aggregated, a value of zero represents a case whereby patches are randomly distributed across the landscape. The clumpiness index requires that both classes are present in the landscape, therefore a no data value (NA) was returned for units where no loss of mangroves had occurred, or where there was 100% gain of mangroves in a later time period. The number of directions in which patches were connected was set to eight. The following manual fixes were conducted for NA values returned: 1) Where NA was returned for units where no loss of mangroves had occurred in another time period, i.e., class 1 (mangrove present) = 1 and class 2 (mangrove loss) = 0, assume +1 (maximally clumped); and 2) Where NA was returned for units where there was 100% gain of mangroves in a later time period, i.e., class 1 (mangrove present) = 0, class 2 (mangrove present) = 1 (100% gain), assume −1 (maximally disaggregated).Tidal amplitudeIn settings of low tidal range, mangrove vertical accretion is less likely to keep pace with rapid sea level rise3. However, in settings of high tidal range, mangroves may be more extensive and vulnerable to conversion to aquaculture or agriculture because of larger tidal flat extents. The Finite Element Solution global tide model (FES2014)96 is considered one of the most accurate tide models for shallow coastal areas97 and was selected to estimate the mean tidal amplitude within each geomorphic unit using the principal lunar semi-diurnal or M2 tidal amplitude as this is this most dominant tidal constituent98. To account for potential variation in the tidal amplitude across large geomorphic units, the raster pixel value for M2 tidal amplitude96 closest to the centroid of each mangrove patch within each unit was calculated, with the smallest value set at 0.01 m. For each geomorphic unit, the tidal amplitude was calculated as the weighted mean of the patch values, with weighting based on the patch area relative to the total unit area.Antecedent sea-level riseThe distribution of mangroves on shorelines changes over time with sediment accretion, erosion, subsidence, and sea-level rise (SLR)99, and periods of low sea level can cause mangrove dieback100. We used regional mean sea-level trends between January 1993 and December 2015 from the global sea level Essential Climate Variable (ECV) product v.2101,102 to estimate the mean antecedent SLR for each geomorphic unit. Spatial variation in regional sea-level trends generally range between −5 and +5 mm yr−1 (global mean of 3 mm yr−1)13. Extreme values ( >5 mm yr−1) observed in the dataset are subject to high levels of uncertainty (Sea Level CCI team, pers. comm.), and were therefore truncated to 5 mm yr−1. The raster pixel value for SLR102 closest to the centroid of each mangrove patch within each geomorphic unit was calculated. The geomorphic unit antecedent SLR values was calculated as the weighted mean of the patch values within the unit.DroughtWhilst long-term precipitation and temperature influence mangrove distribution globally62, periods of low rainfall have been reported to cause extensive mangrove dieback at regional scales, particularly when combined with high temperatures and low sea levels103. We used the Standardized Precipitation-Evapotranspiration Index (SPEI) from the global SPEI database v.2.6104 as an index of drought severity. SPEI is derived from precipitation and temperature and is considered an improved drought index that allows spatial and temporal comparability105,106. The mean SPEI raster pixel value was calculated for each time period and then averaged across the geomorphic units using the ‘ncdf4’107 and ‘raster’ packages74 in R.Tropical storm frequencyLarge-scale destruction of mangroves across regions have been reported from strong winds, high energy waves, and storm surges associated with tropical storms108. We used the International Best Track Archive for Climate Stewardship (IBTrACS) dataset since 1980 v4109 to calculate the number of tropical cyclone occurrences (points along their paths) within a 200 km buffer of the centroid of geomorphic units within each time period using the sf package110 in R. Maximum wind velocity and surface pressures are likely experienced within 100 km of a cyclone’s eye111, therefore the 200 km buffer zone was selected to cover the average size of geomorphic units (33.63 ha), and all tropical storms potentially influencing mangrove growth. Whilst tropical storms affect only 42% of the world’s mangroves60, they are likely to be important stressors within cyclone-impacted countries.Minimum temperatureExtreme low temperature events were a driver of mangrove loss in subtropical regions, such as Florida and Louisianan of the US, and China28,112. We used the WorldClim bioclimatic variable 6 (minimum temperature of the coldest month averaged for the years 1970–2000)113 to calculate the mean minimum temperature across the geomorphic units using the ‘sf’110 and ‘raster’ packages74 in R. Where NAs were returned due to no overlapping raster layer, the value of the closest raster pixel to the centroid of the geomorphic unit was assigned.Statistical analysisWe used multi-level linear modelling to investigate relationships between mangrove cover change variables and socioeconomic and biophysical variables to consider landscape (level 1) and country (level 2) predictors in a hierarchical approach114. For each response variable, we modelled the response for 1996–2007 and 2007–2016, using explanatory variables specific to the time-period where available. Data inspection revealed that high percent loss or gain was concentrated in small geomorphic units, therefore to avoid bias in our results, we removed geomorphic units less than 100 ha from the analysis, which further reduced the available sample size to 3134 units across 95 countries. Statistical analysis was undertaken in R 4.0.268.The response variables were log-transformed to fit normal distribution. We tested for collinearity between our explanatory variables using Pearson’s correlation coefficient (r  > 0.5) (Supplementary Tables 6 and 7). MPA staff capacity and EPI were excluded from our models because MPA staff capacity was correlated with ECI 2007 and ECI 2016 (both r = 0.54), and EPI 2020 was correlated with VDEM 2016 (r = 0.63). To improve model fit, travel time to the nearest city, mangrove restoration effort and Ramsar wetland area (relative) were log+1-transformed, and tidal amplitude was log-transformed.Two linear multi-level (mixed-effects) models were fitted for each response variable using the lme function in the ‘lme4’ package115 (Supplementary Table 8). First, a random intercept model with intercepts of landscape-level predictors varying by country was fitted. Then a random intercept and slope (coefficients) model with intercepts of landscape-level predictors varying by country, as well as slopes for socioeconomic predictors considered to have between-country variation (travel time to nearest city and night-time lights growth) was fitted, as we expect that mangrove cover change may respond to economic growth and market accessibility depending on national governance. A likelihood ratio test between the null linear model and the null random intercept model for each response variable showed that effects varied across countries and therefore we included country as a random effect (Supplementary Table 9). We also conducted likelihood ratio tests between the random intercept model and the random coefficient model to test whether the effect of travel time and night-time lights on mangrove change varies across countries. If significant, the model including random slopes for travel time and night-time lights was used (Supplementary Table 9). Mixed-effects models were fitted by maximum likelihood and model fit was validated by inspection of residual plots for the four response variables included in the analysis; percent net loss, percent net gain, percent gross loss, and percent gross gain (Supplementary Table 9).To test for spatial autocorrelation we performed spatial autoregressive (SAR) models using the errorsarlm function in the ‘spatialreg’ package116. SAR models were first fitted using a range of neighbourhood distances (50, 500, and 1000 km in 100 km intervals) for the net change variable117. Distance of 500 km showed the smallest AIC and was therefore adopted for all response variables. Neighbourhood lists of the centroid coordinates of the geomorphic units were defined with the row-standardised (‘W’) coding using the ‘spdep’ package118. We then produced Moran’s I correlograms using the correlog function in the ‘ncf’ package119 and the centroid coordinates of the geomorphic units. Correlograms for the multi-level model and SAR model were compared for each response variable (Supplementary Fig. 4). The SAR models did not improve spatial autocorrelation for any of the mangrove cover change variables and therefore the multi-level models were adopted.Hotspot estimatesWe defined hotspots as geomorphic units where raw values of percent net and gross loss and gain between 2007 and 2016 ((gamma)) differed by more than two standard deviations (sd) from the country average ((mu)).$${{{{{{rm{More}}}}}}},{{{{{{rm{loss}}}}}}}/{{{{{{rm{more}}}}}}},{{{{{{rm{gain}}}}}}}=left(gamma -mu right) , > , (2,times {{{{{{rm{sd}}}}}}})$$
(1)
$${{{{{{rm{Less}}}}}}},{{{{{{rm{loss}}}}}}},/,{{{{{{rm{less}}}}}}},{{{{{{rm{gain}}}}}}}=left(gamma -mu right) , < , -(2,times {{{{{{rm{sd}}}}}}})$$ (2) We excluded countries with only one geomorphic unit. Large deviations of the raw value from the country average were found for small units at a threshold below 50 km2, therefore we removed all units smaller than 50 km2 to overcome bias of hotspots towards smaller sites. This likely removed the identification of several hotspots. For example, Myanmar has had some large gains due to river sediments in the Gulf of Martaban (net gain of 100 % in Estuary 5834 and 39 % in Open Coast 62244), however, these areas were small (8 and 2 km2, respectively) and were therefore removed from the hotspot estimates.We analysed the factors contributing to hotspots by spatial investigation of satellite imagery in Google Earth with mangrove specialists from those countries. The hotspots were also assessed against protected area datasets for those countries120,121,122,123.Reporting summaryFurther information on research design is available in the Nature Research Reporting Summary linked to this article. More

Study areaWe conducted the study in the best-preserved stands of the Białowieża Forest, strictly protected within the Białowieża National Park (hereafter BNP; coordinates of Białowieża village: 52°42′N, 23°52′E). The extensive Białowieża Forest (c. 1500 km2) straddles the Polish-Belarusian border, where the climate is subcontinental with annual mean temperatures during May–July of 13–18 °C, and mean annual precipitation of 426–940 mm66,67.The forest provides a unique opportunity to observe animals under conditions that likely prevailed across European lowlands before widespread deforestation and forest exploitation by humans66,68,69. The stands have retained a primeval character distinguished by a multi-layered structure, frequent fallen and standing dead trees, and a high species richness66,70. The stands are composed of about a dozen tree species of various ages, up to several hundred years old. The interspecific interactions and natural processes have been little affected by direct human activity.We conducted observations mostly within the three permanent study plots (MS, N, W), totalling c. 130 ha, and in other nearby fragments of primeval oak-lime-hornbeam Tilio-Carpinetum or mixed deciduous-coniferous Pino-Quercetum stands. However, a small number of observations from adjacent managed deciduous forest stands were also included. For details of the study area see71,72,73.Study speciesOur study system focused on ground-nesting Wood Warblers Phylloscopus sibilatrix, blowflies Protocalliphora azurea, and Myrmica or Lasius ants, which occurred in the birds’ nests.The Wood Warbler is a small (c. 10 g) insectivorous songbird that winters in equatorial Africa and breeds in temperate European forests, typically rearing one or two broods each year74. Wood Warblers build dome-shaped nests for each breeding attempt, composed of woven grass, leaves and moss, and lined with animal hair73. The nests are situated on the ground among moderately sparse vegetation, often under a tussock of vegetation or near a fallen tree-branch or log (see examples in Supplementary Fig. S2)53,75. The breeding season of Wood Warblers begins in late April–early May and ends in July–August, when nestlings from replacement clutches (after initial loss) or second broods leave the nest. The typical clutch size in BNP is 5–7 eggs, and the nestling stage lasts 12–13 days74,76.Wood Warbler nests are inhabited by various arthropods, including Myrmica ruginodis or M. rubra ants, and less often Lasius platythorax, L. niger or L. brunneus. The ants foraged and/or raised their own broods within the Wood Warbler nests52. The Myrmica and Lasius ant species are common in Europe77,78. Their colonies contain from tens to thousands of workers, and can be found on the forest floor, e.g. in soil, within or under fallen dead wood, in patches of moss, or among fallen tree-leaves53,77,78. All of the ant species found in the Wood Warbler nests are predators of other arthropods77,79,80.Blowflies, Protocalliphora spp., are obligatory blood-sucking (hematophagous) ectoparasites that reproduce within bird nests. The occurrence, abundance, and impact of blowflies on Wood Warbler offspring is largely unknown, similar to many other European songbirds that build dome-shaped nests. Adult blowflies emerge in late spring and summer to lay eggs on the birds’ nesting material or directly onto the skin of typically newly hatched nestlings14,26. The blowfly larvae hatch within two–three days, and develop in the structure of warm bird nests for another 6–15 days, during which they emerge intermittently to feed on host blood, before finally pupating within the nests14,25,26,27.Data collectionNest monitoring and measurements of nestlingsWe searched for Wood Warbler nests daily from late April until mid-July in 2018–2020, by following birds mainly during nest-building. Nests were assigned to a deciduous or mixed deciduous-coniferous habitat type, depending on the tree stand where they were found. We inspected nests systematically, according to the protocol described in Wesołowski and Maziarz76. The number of observer visits was kept to a minimum to reduce disruptions for birds or potential risks of nest predation.We aimed to establish the dates of hatching (day 0 ± 1 day), nestlings vacating the nest (fledging; ± 1 day) or nest failure (± 1–2 days). When nestlings hatched asynchronously, the hatching date corresponded to the earliest record of nestling hatching. The dates of fledging or nest failure were the mid-dates between the last visit when the nestlings were present in the nest, and the following visit, when the nest was found empty. Nest failure was primarily due to predation, which is the main cause of the Wood Warbler nest losses in BNP76,81 and elsewhere in Europe82,83.To assess fitness consequences for birds of variable weather conditions, blowfly abundance and/or ant presence, we measured nestling growth and determined brood reduction (i.e. the mortality of chicks in the nest) from hatching until fledging. To define brood reduction, we assessed the number of hatchlings (nestlings up to 4 days old) and the number of fledglings leaving the nests. To ensure accurate counting and avoid premature fledging of nestlings, we established the number of fledglings on the day of measurement, when all nestlings were temporarily extracted from the nest.We measured nestling growth on a single occasion when they were 6–9 days old (median 8 days), almost fully developed but too young to leave the nest. The measurements lasted for less than 10–15 min at each nest to minimise any potential risk of attracting predators. For each nestling we measured (using a ruler) the emerged length of the longest (3rd) primary feather vane (± 0.5 mm) on the left wing84,85, and body mass to the nearest 0.1 g using an electronic balance. The length of the feather vane is closely linked to feather growth86 and is one of the characteristics of nestling growth85,87. We treated the length of the primary feather vane and body mass as indices of nestling growth rate under varying conditions of weather, blood-sucking ectoparasites, or ant presence.Extraction of arthropods from bird nestsTo assess the number of blowflies and to establish the presence of ants, we checked the contents of 129 nests (including 11 nests from the managed forest stands) at which Wood Warbler nestlings had been measured. The sample included 86 successful breeding attempts (where a minimum of one nestling successfully left the nest), 27 failed (predated) nests (remnants of nestlings were found, but the nest structure remained intact), and 16 nests with an unknown fate (nestlings were large, so were capable of leaving the nest, but no family were located or other signs indicating fledging).Due to ethical reasons, we were unable to collect the Wood Warbler nests and extract the ectoparasites and ants from them while they were in use by the birds. Removing the nests and replacing them with dummy nests would cause unacceptable nest desertion by adults. Therefore, we assessed the occurrence and number of blowflies or ant presence after Wood Warbler nestlings fledged or the breeding attempts failed naturally. We retrospectively explored the changes in blowfly infestation14, including the effect of ant presence53 in the same nests.We collected nests from the field as soon as a breeding attempt ended, within approximately five days (median 1 day) following fledging or nest failure (nest structure remained intact). The delay of nest collection would not bias the ectoparasite infestation, as blowfly larvae pupate within bird nests and stay there after the hosts abandon their nests; puparia can be still found in nests collected in autumn or winter14. As the likelihood of finding ant broods (larvae or pupae associated with workers) was rather stable with the delay of nest collection53, the method seemed reliable also for assessing the presence of ant broods (35 of all 71 Wood Warbler nests containing ants). Only the number of nests with lone foraging ant workers could be underestimated, potentially inflating the uncertainty of tested relationships. However, as ants usually re-use rich food resources88, foraging Myrmica or Lasius ant workers might regularly exploit warbler nests, increasing the chances of finding the insects in the collected nests.Wood Warbler nests were collected in one piece, with each placed into a separate sealed and labelled plastic bag. We carefully inspected the leaf litter around the nests, and the soil surface under them, to make sure that all blowfly larvae or pupae were collected. We transported the collected nests to a laboratory, where we stored them in a fridge for up to 5–6 days before the arthropod extraction.To establish the number of blowflies and the presence of ants, in 2018, we carefully pulled apart the nesting material and searched for the arthropods amongst it 52. We gathered all blowfly pupae or larvae and a sample of ant specimens into separate tubes, labelled and filled with 70–80% alcohol, for later species identification. For nests collected in 2019–2020, we extracted the arthropods with a Berlese-Tullgren funnel. During the extraction, which usually lasted for 72 h, each nest was covered with fine metal mesh and placed c. 15 cm under the heat of a 40 W electric lamp. The arthropods were caught in 100 ml plastic bottles containing 30 ml of 70–80% ethanol, installed under each funnel. After the arthropod extraction, we carefully inspected the nesting material in the same way as in 2018, to collect any blowflies that remained within the nests. The quality of information collected on the number of ectoparasites and ant presence should be comparable each year.Weather dataWe obtained the mean daily temperatures and rainfall sums from a meteorological station, operated by the Meteorology and Water Management National Research Institute in the Białowieża village, 1–7 km from the study areas.Data analysesWeather conditions affecting blowfly ectoparasitesTo explore the impact of weather on blowfly ectoparasites, for each Wood Warbler nest we calculated average temperatures from daily means, and total sums of rainfall from daily sums, for the two time-windows in which we assumed the impact of weather would be of greatest importance:

i.

the early nestling stage, when Wood Warbler nestlings were 1–4 days old. During this stage, female blowflies require a minimum temperature of c. 16 °C to become active and oviposit in bird nests27. Thus, cool and wet weather in the early nestling stage should reduce the activity of ovipositing blowflies, leading to less frequent ectoparasite infestation of Wood Warbler nests.

ii.

The late nestling stage, when the warbler nestlings were aged between over four days old and until fledging or nest failure. During this stage, blowfly larvae grow and develop in bird nests after hatching a few days after oviposition14,25,26,27. As the temperature of bird nests strongly depends on ambient temperatures21, mortality of blowfly larvae should increase in cool weather, resulting in fewer ectoparasites in nests collected shortly after the fledging of birds29.

Weather conditions affecting Wood Warbler nestling growthTo explore the impact of weather on nestling growth, for each nest we calculated the average temperatures and total sums of rainfall for the period when nestlings were over four days old and until their measurement, usually on day 8 from hatching (see above). During this stage, nestlings are no longer brooded by a parent74, so must balance their energetic expenditure between growth (feather length and body mass) or thermoregulation89. Thus, we expected that the gain in body mass and the growth of flight feathers would be reduced in nestlings during cool and wet weather, when maintaining a stable body temperature would be costly90.Statistical analysesAll statistical tests were two-tailed and performed in R version 4.1.091.The changes in blowfly infestation of the Wood Warbler nestsTo test the changes in blowfly infestation of warbler nests, we used zero-augmented negative binomial models (package pscl in R;92,93), which deal with the problem of overdispersion and excess of zeros92. In this study, hurdle and zero-inflated models fitted with the same covariates had an almost identical Akaike Information Criterion (AIC). Therefore, we presented only the results of hurdle models, which are easier to interpret than zero-inflated models. Hurdle models consisted of two parts: a left-truncated count with a negative binomial distribution representing the number of blowflies in infested nests, and a zero hurdle binomial estimating the probability of blowfly presence. We used models with a negative binomial distribution, which had a much lower AIC than with a Poisson distribution on a count part.We designed the most complex (global) model that contained a response variable of the number of blowflies in each of the 129 Wood Warbler nests. The covariates were: mean ambient temperature, total sum of rainfall, presence (or absence) of ants in the same nests, habitat type (deciduous vs mixed deciduous-coniferous forest), study year (2018–2020), the number of nestlings hatched (brood size), and nest phenology (the relative hatching date of Wood Warbler nestlings, as days from the median hatching date in a season: 23 May in 2018, 25 May in 2019 and 29 May in 2020). The initial global model also contained the two-way interaction terms that we suspected to be important: between temperature and rainfall, temperature and presence of ants, and rainfall and presence of ants.To explore all potentially meaningful subsets of models, we used the same covariates on both parts (count and binomial) of the global model. We performed automated model selection with the MuMIn package94, starting from the most complex (global) model and using all possible simpler models (i.e. all subsets)95. To attain the minimum sample size of c. 20 data points for each parameter96, we limited the maximum number of parameters to six in each part (count or binomial) of the candidate models.As some of the interaction terms appeared insignificant in the initial model selection, to minimise the risk of over-parametrisation, we included only the significant interaction term on a count part of the final global model. As described above, we performed model selection again. We tested linear relationships, as the quadratic effects of weather variables (presuming temperature or rainfall optima) appeared insignificant.To test whether blowfly infestation changed with weather in the early or late nestling stages, we twice repeated the procedure described above. The first global model included the mean ambient temperature and the total sum of rainfall for the early nestling stage, and the second global model contained weather variables for the late nestling stage. The remaining covariates were the same.A practice of including the same sets of covariates on count and binomial parts has been previously questioned97. However, our approach allowed us to comply with these objections97, as we presented only the most parsimonious models (with ΔAICc  More

Bolliger, J., Kienast, F. & Zimmermann, N. E. Risk of global warming on montane and subalpine forests in Switzerland—A modeling study. Reg. Environ. Change 1, 99–111 (2000).
Google Scholar 
Bugmann, H. & Pfister, Ch. Impacts of interannual climate variability on past and future forest composition. Reg. Environ. Change 1, 112–125 (2000).
Google Scholar 
Becker, A. & Bugmann, H. (eds.) Global change and mountain regions: The Mountain Research Initiative. IHDP Report 13, GTOS Report 28 and IGBP Report 49, Stockholm (2001).Kullman, L. 20th Century climate warming and tree-limit rise in the southern Scandes of Sweden. Ambio 30, 72–80. https://doi.org/10.1579/0044-7447-30.2.72 (2001).CAS 
PubMed 

Google Scholar 
Körner, Ch. & Paulsen, J. A world-wide study of high altitude treeline temperatures. J. Biogeogr. 31, 713–732. https://doi.org/10.1111/j.1365-2699.2003.01043.x (2004).
Google Scholar 
Harsch, M. A. & Bader, M. Y. Treeline form—A potential key to understanding treeline dynamics. Global Ecol. Biogeogr. 20, 582–596. https://doi.org/10.1111/j.1466-8238.2010.00622.x (2011).
Google Scholar 
Tokarczyk, N. Forest encroachment on temperate mountain meadows: scale, drivers, and current research directions. Geogr. Pol. 90, 463–480 (2017).
Google Scholar 
Vitali, A. et al. Pine recolonization dynamics in Mediterranean human-disturbed treeline ecotones. For. Ecol. Manag. 435, 28–37. https://doi.org/10.1016/j.foreco.2018.12.039 (2019).
Google Scholar 
Heikkinen, O., Obrębska-Starkel, B. & Tuhkanen, S. Introduction: the timberline—A changing battlefront. Prace Geograficzne UJ 98, 7–16 (1995).
Google Scholar 
Mattson, J. Human impact on the timberline in the far North of Europe. Zeszyty Naukowe UJ, Prace Geogr. 98, 41–56 (1995).
Google Scholar 
Stanisci, A., Lavieri, D., Acosta, A. & Blasi, C. Structure and diversity trends at Fagus timberline in central Italy. Community Ecol. 1, 133–138 (2000).
Google Scholar 
Gehrig-Fasel, J., Guisan, A. & Zimmermann, N. E. Tree line shifts in the Swiss Alps: Climate change or land abandonment?. J. Veg. Sci. 18, 571–582 (2007).
Google Scholar 
Feurdean, A. et al. Long-term land-cover/use change in a traditional farming landscape in Romania inferred from pollen data, historical maps and satellite images. Reg. Environ. Change 17, 2193–2207. https://doi.org/10.1007/s10113-016-1063-7 (2017).
Google Scholar 
Burga, C. A., Bührer, S. & Klötzli, F. Mountain ash (Sorbus aucuparia) forests of the Central and Southern Alps (Grisons and Ticino, Switzerland-Prov. Verbano-Cusio-Ossola, N-Italy): Plant ecological and phytosociological aspects. Tuexenia 39, 121–138 (2019).
Google Scholar 
Slayter, R. O. & Noble, I. R. Dynamics of Montane Treelines. In Landscape Boundaries, Consequences for Biotic Diversity and Ecological Flows. Ecological Studies Vol. 92 (eds Hansen, A. J. & di Castri, F.) 346–359 (Springer-Verlag, 1992).
Google Scholar 
Bryn, A. Recent forest limit changes in south-east Norway: Effects of climate change or regrowth after abandoned utilisation?. Nor. Geogr. Tidsskr. 62(4), 251–270. https://doi.org/10.1080/00291950802517551 (2008).
Google Scholar 
Lu, X., Liang, E., Wang, Y., Babst, F. & Camarero, J. J. Mountain treelines climb slowly despite rapid climate warming. Glob. Ecol. Biogeogr. 30(1), 305–315. https://doi.org/10.1111/geb.13214 (2021).
Google Scholar 
Armand, A. D. Sharp and Gradual Mountain Timberlines as Result of species Interaction. Landscape Boundaries, Consequences for Biotic Diversity and Ecological Flows. In Ecological Studies Vol. 92 (eds Hansen, A. J. & di Castri, F.) 360–377 (Springer-Verlag, 1992).
Google Scholar 
Kucharzyk, S. Ekologiczne znaczenie drzewostanów w strefie górnej granicy lasu w Karpatach Wschodnich i ich wrażliwość na zmiany antropogeniczne [Ecological importance of stands at the upper forest limit in the Eastern Carpathians and their sensibility to anthropogenic changes]. Roczn. Bieszcz. 14, 15–43 (2006) (in Polish with English summary).
Google Scholar 
Surina, B. & Rakaj, M. Subalpine beech forest with Hairy alpenrose (Polysticho lonchitis-Fagetum Rhododendretosum hirsuti subass. nova) on Mt. Snežnik (Liburnian Karst, Dinaric Mts). Hacquetia 6, 195–208 (2007).
Google Scholar 
Kucharzyk, S. Zmiany przebiegu górnej granicy lasu w pasmie Szerokiego Wierchu w Bieszczadzkim Parku Narodowym [Changes of upper forest limit in the Szeroki Wierch range (Bieszczady National Park)]. Roczn. Bieszcz. 12, 81–102 (2004) (in Polish with English summary).
Google Scholar 
Kucharzyk, S. & Augustyn, M. Dynamika górnej granicy lasu w Bieszczadach Zachodnich – zmiany w ciągu półtora wieku [The upper forest limit dynamics in the Western Bieszczady Mts.—Changes over a century and a half]. Stud. Nat. 54, 133–156 (2008) (in Polish with English summary).
Google Scholar 
Kubijowicz, W. Życie pasterskie w Beskidach Wschodnich [La Vie Pastorale dans les Beskides Orientales]. Prace Instytutu Geograficznego UJ 5, 3–30 (1926) (in Polish).
Google Scholar 
Zarzycki, K. Lasy Bieszczadów Zachodnich [The forests of the Western Bieszczady Mts (Polish Eastern Carpathians)]. Acta Agr. et Silv. Ser. Leśna 3, 1–131 (1963) (in Polish with English summary).
Google Scholar 
Augustyn, M. Połoniny w Bieszczadach Zachodnich [Almen im westlichen Bieszczady-Gebirge]. Materiały Muzeum Budownictwa Ludowego w Sanoku 31, 88–98 (1993) (in Polish with German summary).
Google Scholar 
Winnicki, T. Zbiorowiska roślinne połonin Bieszczadzkiego Parku Narodowego (Bieszczady Zachodnie, Karpaty Wschodnie) [Plant communities of subalpine poloninas in the Bieszczady National Park (Western Bieszczady Mts, Eastern Carpathians)]. Monogr. Bieszczadzkie 4, 1–215 (1999) (in Polish with English summary).
Google Scholar 
Mróz, W. Zróżnicowanie szaty roślinnej przy górnej granicy lasu w Bieszczadach Wschodnich i Zachodnich [The diversity of vegetation near the upper timberline in the Eastern and the Western Bieszczady Mts]. Roczn. Bieszcz. 14, 45–62 (2006) (in Polish with English summary).
Google Scholar 
Augustyn, M. & Kucharzyk, S. Górna granica lasu na terenie wsi Ustrzyki Górne i Wołosate w końcu XVIII wieku [Timberline in the Western Bieszczady Mts.]. Roczn. Bieszcz. 20, 15–27 (2012) (in Polish with English summary).
Google Scholar 
Jeník, J. Succession on the Połonina Balds in the Western Bieszczady, the Eastern Carpathians. Tuexenia 3, 207–216 (1983).
Google Scholar 
Michalik, S. & Szary, A. Zbiorowiska leśne Bieszczadzkiego Parku Narodowego [The forest communities of the Bieszczady National Park]. Monogr. Bieszcz. 1, 1–175 (1997).
Google Scholar 
Zemanek, B. & Winnicki, T. Rośliny naczyniowe Bieszczadzkiego Parku Narodowego [Vascular plants of the Bieszczady National Park]. Monogr. Bieszcz. 3, 1–249 (1999) (in Polish with English summary).
Google Scholar 
Kucharzyk, S. & Augustyn, M. Trwałość polan reglowych w Bieszczadzkim Parku Narodowym [Stability of mountain glades in the Bieszczady National Park]. Roczn. Bieszcz. 18, 45–58 (2010) (in Polish with English summary).
Google Scholar 
Durak, T., Żywiec, M. & Ortyl, B. Rozprzestrzenianie się zarośli drzewiastych w piętrze połonin Bieszczad Zachodnich [Expansion of brushwood in the subalpine zone of the Western Bieszczady Mts]. Sylwan 157, 130–138 (2013) (in Polish with English summary).
Google Scholar 
Durak, T., Żywiec, M., Kapusta, P. & Holeksa, J. Impact of land use and climate changes on expansion of woody species on subalpine meadows in the Eastern Carpathians. For. Ecol. Manag. 339, 127–135. https://doi.org/10.1016/j.foreco.2014.12.014 (2015).
Google Scholar 
Durak, T., Żywiec, M., Kapusta, P. & Holeksa, J. Rapid spread of a fleshy-fruited species in abandoned subalpine meadows—Formation of an unusual forest belt in the eastern Carpathians. iForest – Biogeosci. For. 9, 337–343. https://doi.org/10.3832/ifor1470-008 (2015).
Google Scholar 
Wężyk, P. & Hawryło, P. Analiza struktury 3D drzewostanów Bieszczadzkiego PN na podstawie danych lotniczego skanowania laserowego oraz ortofotomap lotniczych CIR [3D structure analysis of stands of the Bieszczady National Park on the basis of airborne laser scanning data and CIR aerial ortho-photomaps] (ProGea Consulting, 2015) (in Polish).Anselin, L. Local indicators of spatial association—LISA. Geogr. Anal. 27, 93–115. https://doi.org/10.1111/j.1538-4632.1995.tb00338.x (1995).
Google Scholar 
Scott, L. M. & Janikas, M. V. Spatial Statistics in ArcGIS. In Handbook of Applied Spatial Analysis (eds Fischer, M. M. & Getis, A.) 27–41 (Springer, 2010).
Google Scholar 
Cui, H., Wu, L., Hu, S., Lu, R. & Wang, S. Research on the driving forces of urban hot spots based on exploratory analysis and binary logistic regression model. Trans. GIS 25(3), 1522–1541. https://doi.org/10.1111/tgis.12739 (2021).
Google Scholar 
Pierce, K. B., Lookingbill, T. & Urban, D. A simple method for estimating potential relative radiation (PRR) for landscape-scale vegetation analysis. Landsc. Ecol. 20, 137–147 (2005).
Google Scholar 
Riley, S. J., DeGloria, S. D. & Elliot, R. A terrain ruggedness index that quantifies topographic heterogeneity. Int. J. Sc. 5, 23–27 (1999).
Google Scholar 
Böhner, J. & Antonić, O. Land-surface parameters specific to topo-climatology. Geomorphometry – Concepts, Softw. Appl. Dev. Soil Sci. 33, 195–226. https://doi.org/10.1016/S0166-2481(08)00008-1 (2009).
Google Scholar 
R Core Team. R: A Language and Environment for Statistical Computing (R Core Team, 2021).
Google Scholar 
Agresti, A. An Introduction to Categorical Data Analysis 2nd edn. (Wiley & Sons Inc., 2007).MATH 

Google Scholar 
Cottrell, A. Gnu Regression, Econometrics and Time-series Library gretl. http://gretl.sourceforge.net/(2020).Hellevik, O. Linear versus logistic regression when the dependent variable is a dichotomy. Qual. Quant. 43, 59–74 (2009).
Google Scholar 
Azen, R. & Traxel, N. Using dominance analysis to determine predictor importance in logistic regression. J. Educ. Behav. Stat. 34, 319–347. https://doi.org/10.3102/1076998609332754 (2009).
Google Scholar 
Borcard, P., Legendre, P. & Drapeau, P. Partialling out the spatial component of ecological variation. Ecology 73, 1045–1055 (1992).
Google Scholar 
Przybylska, K. & Kucharzyk, S. Skład gatunkowy i struktura lasów Bieszczadzkiego Parku Narodowego [Species composition and structure of forest of the Bieszczady National Park. Monogr. Bieszcz. 6, 1–159 (1999) (in Polish with English summary).
Google Scholar 
Bader, M. Y. et al. A global framework for linking alpine-treeline ecotone patterns to underlying processes. Ecography 44(2), 265–292. https://doi.org/10.1111/ecog.05285 (2021).
Google Scholar 
Nowosad, M. Zarys klimatu Bieszczadzkiego Parku Narodowego i jego otuliny w świetle dotychczasowych badań [Outlines of climate of the Bieszczady National Park and its bufferzone in the light of previous studies]. Roczn. Bieszcz. 4, 163–183 (1995) (in Polish with English summary).
Google Scholar 
Nowosad, M. & Wereski, S. Warunki klimatyczne. Bieszczadzki Park Narodowy–40 lat ochrony [Climatic conditions. Bieszczady National Park–40 years of protection]. In Bieszczadzki Park Narodowy [The Bieszczady National Park] (eds Górecki, A. & Zemanek, B.) 31–38 (Wyd. Bieszczadzki Park Narodowy, 2016) (in Polish with English summary).
Google Scholar 
Kukulak, J. Neotectonics and planation surfaces in the High Bieszczady Mountains (Outer Carpathians, Poland). Ann. Soc. Geol. Pol. 74, 339–350 (2004).
Google Scholar 
Haczewski, G., Kukulak, J. & Bąk, K. Budowa geologiczna i rzeźba Bieszczadzkiego Parku Narodowego [Geology and relief of the Bieszczady National Park]. Prace monograficzne (Akademia Pedagogiczna im. Komisji Edukacji Narodowej w Krakowie) 468, 1–156 (2007) (in Polish with English summary).
Google Scholar 
Skiba, S., Drewnik, M., Kacprzak, A. & Kołodziejczyk, M. Gleby litogeniczne Bieszczadów i Beskidu Niskiego [Lithogenous soils of the Bieszczady and Beskid Niski Mts (Polish Carpathians)]. Roczn. Bieszcz. 7, 387–396 (1998) (in Polish with English summary).
Google Scholar 
Skiba, S. & Winnicki, T. Gleby zbiorowisk roślinnych bieszczadzkich połonin [Soils of the subalpine meadows plant communities in the Bieszczady Mts]. Roczn. Bieszcz. 4, 97–109 (1995) (in Polish with English summary).
Google Scholar 
Musielok, Ł, Drewnik, M., Szymański, W. & Stolarczyk, M. Classification of mountain soils in a subalpine zone—A case study from the Bieszczady Mountains (SE Poland). Soil Sci. Annu. 70, 170–177. https://doi.org/10.2478/ssa-2019-0015 (2019).CAS 

Google Scholar 
Spatz, G. Succession patterns on mountain pastures. Vegetatio 43, 39–41 (1980).
Google Scholar 
Kozak, J. Zmiany powierzchni lasów w Karpatach Polskich na tle innych gór świata [Changes in the Land Cover in the Polish Carpathians at the Turn of the 20th and 21st Century in Relation to Local Development Level]. Wydawnictwo Uniwersytetu Jagiellońskiego, Kraków (2005) (in Polish with English summary).Vitali, A., Urbinati, C., Weisberg, P. J., Urza, A. K. & Garbarino, M. Effects of natural and anthropogenic drivers on land-cover change and treeline dynamics in the Apennines (Italy). J. Veg. Sci. 29(2), 189–199. https://doi.org/10.1111/jvs.12598 (2018).
Google Scholar 
Micu, D. M., Dumitrescu, A., Cheval, S., Nita, I.-A. & Birsan, M.-V. Temperature changes and elevation-warming relationships in the Carpathian Mountains. Int. J. Climatol. 41, 2154–2172. https://doi.org/10.1002/joc.6952 (2020).
Google Scholar 
Rehman, A. Ziemie dawnej Polski. Cz. I. Karpaty [The lands of ancient Poland. Part I. The Carpathians]. (Gubrynowicz i Schmidt, Lwów) (1895) (in Polish).Frey, W. The influence of snow on growth and survival of planted trees. Arct. Alp. Res. 15, 241–251 (1983).
Google Scholar 
Malanson, G. P. et al. Alpine treeline of Western North America: Linking organism-to-landscape dynamics. Phys. Geogr. 28, 378–396. https://doi.org/10.2747/0272-3646.28.5.378 (2007).
Google Scholar 
Holtmeier, F. K. & Broll, G. Wind as an ecological agent at treelines in North America, the Alps, and the European Subarctic. Phys. Geogr. 31, 203–233. https://doi.org/10.2747/0272-3646.31.3.203 (2010).
Google Scholar 
Barclay, A. M. & Crawford, R. M. M. Winter desiccation stress and resting bud viability in relation to high altitude survival in Sorbus aucuparia L. Flora 172, 21–34 (1982).
Google Scholar 
Raspé, O., Findlay, C. & Jacquemart, A. L. Sorbus aucuparia L. J. Ecol. 88, 910–930 (2000).
Google Scholar 
Zerbe, S. On the ecology of Sorbus aucuparia (Rosaceae) with special regard to germination, establishment and growth. Pol. Bot. J. 46, 229–239 (2001).
Google Scholar 
Smith, W. K., Germino, M. J., Hancock, T. E. & Johnson, D. M. Another perspective on altitudinal limits of alpine timberlines. Tree Physiol. 23, 1101–1112 (2003).PubMed 

Google Scholar 
Trant, A., Higgs, E. & Starzomski, B. M. A century of high elevation ecosystem change in the Canadian Rocky Mountains. Sci. Rep. 10, 9698. https://doi.org/10.1038/s41598-020-66277-2 (2020).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Barbeito, I., Dawes, M. A., Rixen, C., Senn, J. & Bebi, P. Factors driving mortality and growth at treeline: A 30-year experiment of 92 000 conifers. Ecology 93(2), 389–401 (2012).PubMed 

Google Scholar 
Kullman, L. A 25-year survey of geoecological change in the scandes mountains of Sweden. Geogr. Ann. Ser. B 79, 139–165 (1997).
Google Scholar 
Pękala, K. Rzeźba Bieszczadzkiego Parku Narodowego [Relief of the Bieszczady National Park]. Roczn. Bieszcz. 6, 19–38 (1997) (in Polish with English summary).
Google Scholar 
Kullman, L. Temporal and spatial aspects of subalpine populations of Sorbus aucuparia in Sweden. Ann. Bot. Fenn. 23, 267–275 (1986).
Google Scholar 
Hoersch, B. Modelling the spatial distribution of montane and subalpine forests in the Central Alps using digital elevation models. Ecol. Model. 168, 267–282 (2003).
Google Scholar 
Resler, L. M., Butler, D. R. & Malanson, G. P. Topographic shelter and conifer establishment and mortality in an alpine environment, Glacier National Park, Montana. Phys. Geogr. 26, 112–125 (2005).
Google Scholar 
Kollmann, J. Regeneration window for fleshy-fruited plants during scrub development on abandoned grassland. Ecoscience 2, 213–222 (1995).
Google Scholar 
Lediuk, K. D., Damascos, M. A., Puntieri, J. G. & de Torres Curth, M. I. Population dynamics of an invasive tree, Sorbus aucuparia, in the understory of a Patagonian forest. Plant Ecol. 217, 899–911 (2016).
Google Scholar 
McCutchan, M. H. & Fox, D. G. Effect of elevation and aspect on wind, temperature and humidity. J. Appl. Meteorol. Climatol. 25(12), 1996–2013 (1986).ADS 

Google Scholar 
Stage, A. R. & Salas, C. Interactions of elevation, aspect, and slope in models of forest species composition and productivity. For. Sci. 53, 486–492 (2007).
Google Scholar 
Pocewicz, A. L., Gessler, P. & Robinson, A. P. The relationship between effective plant area index and Landsat spectral response across elevation, solar insolation, and spatial scales in a northern Idaho forest. Can. J. For. Res. 34, 465–480 (2004).
Google Scholar 
Kucharzyk, S. & Sugiero, D. Zróżnicowanie dynamiki procesów lasotwórczych w buczynach bieszczadzkich w zależności od wystawy i wzniesienia [Variability of the dynamics of forest development processes in the Bieszczady beech forests in relation to exposition and altitude]. Sylwan 7, 29–38 (2007) (in Polish with English summary).
Google Scholar 
Drewnik, M., Musielok, Ł, Stolarczyk, M., Mitka, J. & Gus, M. Effects of exposure and vegetation type on organic matter stock in the soils of subalpine meadows in the Eastern Carpathians. CATENA 147, 167–176. https://doi.org/10.1016/j.catena.2016.07.014 (2016).CAS 

Google Scholar 
Zheng, L. et al. Tree regeneration patterns on contrasting slopes at treeline ecotones in Eastern Tibet. Forests 12, 1605. https://doi.org/10.3390/f12111605 (2021).
Google Scholar  More

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain
the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in
Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles
and JavaScript. More

Roff, D. A. The Evolution of Life Histories: Theory and Analysis (Chapman and Hall, 1992).
Google Scholar 
Sæther, B. E. & Bakke, O. Avian life history variation and contribution of demographic traits to the population growth rate. Ecology 81, 642–653 (2000).
Google Scholar 
Koons, D. N., Pavard, S., Baudisch, A. & Metcalf, J. E. C. Is life-history buffering or lability adaptive in stochastic environments?. Oikos 118, 972–980 (2009).
Google Scholar 
Boyce, M. S., Haridas, C. V. & Lee, C. T. Demography in an increasingly variable world. Trends Ecol. Evol. 21, 141–148 (2006).PubMed 

Google Scholar 
Morris, W. F. & Doak, D. F. Buffering of life histories against environmental stochasticity: Accounting for a spurious correlation between the variabilities of vital rates and their contributions to fitness. Am. Nat. 163, 579–590 (2004).PubMed 

Google Scholar 
Ripple, W. J. et al. Saving the world’s terrestrial megafauna. Bioscience 66(10), 807–812 (2016).PubMed 
PubMed Central 

Google Scholar 
He, F. et al. Disappearing giants: A review of threats to freshwater megafauna. WIREs Water 4, e1208 (2017).
Google Scholar 
Blackburn, T. M., Cassey, P., Duncan, R. P., Evans, K. L. & Gaston, K. J. Avian extinction and mammalian introductions on oceanic islands. Science 305(5692), 1955–1958 (2004).ADS 
CAS 
PubMed 

Google Scholar 
Courchamp, F., Langlais, M. & Sugihara, G. Rabbits killing birds: Modelling the hyperpredation process. J. Anim. Ecol. 69, 154–164 (2000).
Google Scholar 
Roemer, G. W., Coonan, T. J., Garcelon, D. K., Bascompte, J. & Laughrin, L. Feral pigs facilitate hyperpredation by golden eagles and indirectly cause the decline of the island fox. Anim. Conserv. 4, 307–318 (2001).
Google Scholar 
Kristan, W. B. & Boarman, W. I. Spatial patterns of risk of common raven predation on desert tortoises. Ecology 84, 2432–2443 (2003).
Google Scholar 
Whelan, C. J., Brown, J. S. & Maina, G. Search biases, frequency-dependent predation and species co-existence. Evol. Ecol. Res. 5, 329–343 (2003).
Google Scholar 
Moleón, M., Almaraz, P. & Sánchez-Zapata, J. A. An emerging infectious disease triggering large-scale hyperpredation. PLoS ONE 3, e2307 (2008).ADS 
PubMed 
PubMed Central 

Google Scholar 
Moleón, M., Almaraz, P. & Sánchez-Zapata, J. A. Inferring ecological mechanisms from hunting bag data in wildlife management: A reply to blanco-aguiar et al. 2012. Eur. J. Wildl. Res. 59, 599–608 (2013).
Google Scholar 
Bate, A. M. & Hilker, F. M. Rabbits protecting birds: Hypopredation and limitations of hyperpredation. J. Theor. Biol. 297, 103–115 (2012).ADS 
MathSciNet 
PubMed 
MATH 

Google Scholar 
Turner, F. B., Medica, P. A. & Lyons, C. L. Reproduction and survival of the desert tortoise (Scaptochelys agassizii) in Ivanpah Valley California. Copeia 1984(4), 811–820 (1984).
Google Scholar 
Graciá, E. et al. Assessment of the key evolutionary traits that prevent extinctions in human altered habitats using a spatially explicit individual-based model. Ecol. Model. 415, 108823 (2020).
Google Scholar 
Segura, A., Jiménez, J. & Acevedo, P. Predation of young tortoises by rabbits: The effect of habitat structure on tortoise detectability and abundance. Sci. Rep. 10, 1–9 (2020).
Google Scholar 
Watson, J. The golden eagle (Bloomsbury Publishing, 2010).
Google Scholar 
Fischer, W., Zenker, D. & Baumgart, W. Ein beitrag zum bestand und zur ernährung des steinadlers Aquila chrysaetos af der balkanhalbinsel. Beiträge zur Vogelskunde 21, 275–287 (1975).
Google Scholar 
Delibes, M., Calderón, J. & Hiraldo, F. Selección de presa y alimentación en españa del águila real (Aquila chrysaetos). Ardeola 21, 285–303 (1975).
Google Scholar 
Handrinos, G. The Golden Eagle in Greece. Actes 1er Coll. Intern. Aigle Royal en Europe, Arvieux, 1986: 18–22 (1987).Bautista, J., Gil-Sánchez, J. M. & Moleón, M. Dieta del águila real en el sur de españa. Quercus 364, 17–23 (2016).
Google Scholar 
Bautista, J., Castillo, S., Paz, J. L., Llamas, J. & Ellis, D. H. Golden eagles (Aquila chrysaetos) as potential predators of barbary macaques (Macaca sylvanus) in northern Morocco: Evidences of predation. Go-South Bull. 15, 172–179 (2018).
Google Scholar 
Kouzmanov, G., Stoyanov, R. & Todorov, V. Sur la biologie et la Protection de l`Aigle royal Aquila chrysaetos en Bulgarie. In Eagle studies (eds Meyburg, B. & Chancellor, R.) 505–516 (World Working Group on Birds of Prey, 1996).
Google Scholar 
Capper, S. The predation of Testudo spp. By Golden Eagles Aquila chrysaetos in Dadia Forest Reserve, NE Greece. University of Reading (1998).Karyakin, I. V., Kovalenko, A. V., Levin, A. S. & Pazhenkov, A. S. Eagles of the Aral-Caspian region Kazakhstan. Raptors Conserv. 22, 92–152 (2011).
Google Scholar 
Papageorgiou, N., Vlachos, C., Bakaloudis, D. E., Kazaklis, A., Birtsas, P. Study on the biology and management of raptors in Dadia forest–Evros. Thessaloniki, GR (1995).Sidiropoulos, L. et al. Pronounced seasonal diet diversity expansion of golden eagles (Aquila chrysaetos) in Northern Greece during the non-breeding season: The role of tortoises. Diversity 14(2), 135 (2022).
Google Scholar 
IUCN. The IUCN red list of threatened species. Version 2020–3 (2020).Graciá, E. et al. Expansion after expansion: dissecting the phylogeography of the widely distributed spur-thighed tortoise, Testudo graeca (Testudines: Testudinidae). Biol. J. Linn. Soc. 121, 641–654 (2017).
Google Scholar 
Graciá, E. et al. Genetic patterns of a range expansion: The spur-thighed tortoise Testudo graeca graeca in southeastern Spain. Amphib. Reptil. 32, 49–61 (2011).
Google Scholar 
Graciá, E. et al. The uncertainty of late pleistocene range expansions in the western Mediterranean: A case study of the colonization of south-eastern Spain by the spur-thighed tortoise, Testudo graeca.. J. Biogeogr 40, 323–334 (2013).
Google Scholar 
Anadón, J. D., Giménez, A., Perez, I., Martinez, M. & Esteve-Selma, M. A. Habitat selection by the spur-thighed tortoise Testudo graeca in a multisuccessional landscape: implications for habitat management. Biodivers. Conserv. 15, 2287–2299 (2006).
Google Scholar 
Rodríguez-Caro, R. C. et al. Low tortoise abundances in pine forest plantations in forest-shrubland transition areas. PLoS ONE 12, e0173485 (2017).PubMed 
PubMed Central 

Google Scholar 
Rodríguez-Caro, R. C. et al. The limits of demographic buffering in coping with environmental variation. Oikos 130(8), 1346–1358 (2021).
Google Scholar 
Rodríguez-Caro, R. C., Lima, M., Anadón, J. D., Graciá, E. & Giménez, A. Density dependence, climate and fires determine population fluctuations of the spur-thighed tortoise, Testudo graeca. J. Zool. 300, 265–273 (2016).
Google Scholar 
Rodríguez-Caro, R. C. et al. A low cost approach to estimate demographic rates using inverse modeling. Biol. Conserv. 237, 358–365 (2019).
Google Scholar 
Jiménez-Franco, M. V. et al. Sperm storage reduces the strength of the mate-finding allee effect. Ecol. Evol. 10(4), 1938–1948 (2020).PubMed 
PubMed Central 

Google Scholar 
Graciá, E. et al. From troubles to solutions: Conservation of mediterranean tortoises under global change. Basic Appl. Herpetol. 34, 5–16 (2020).
Google Scholar 
Pérez, I. et al. Exurban sprawl increases the extinction probability of a threatened tortoise due to pet collections. Ecol. Model. 245, 19–30 (2012).
Google Scholar 
Del Moral, J. C. El águila real en España. Población reproductora en 2008 y método de censo. SEO/BirdLife. Madrid. pp. 30–50 (2009).Virgós, E., Cabezas-Díaz, S. & Lozano, J. Is the wild rabbit (Oryctolagus cuniculus) a threatened species in Spain? Sociological constraints in the conservation of species. Biodivers. Conserv. 16, 3489–3504 (2007).
Google Scholar 
Fernández, C. Effect of the viral haemorrhagic pneumonia of the wild rabbit on the diet and breeding success of the golden eagle Aquila chrysaetos (L.). Rev. Ecol. Terre et Vie 48, 323–329 (1993).
Google Scholar 
Villafuerte, R., Luco, D. F., Gortázar, C. & Blanco, J. C. Effect on red fox litter size and diet after rabbit haemorrhagic disease in northeastern Spain. J. Zool. 240, 764–767 (1996).
Google Scholar 
Martínez, J. A. & Zuberogoitia, I. The response of eagle owl (Bubo bubo) to an outbreak of the rabbit haemorrhagic disease. J. Ornithol. 142, 204–211 (2001).
Google Scholar 
Moleón, M. et al. Large-scale spatiotemporal shifts in the diet of a predator mediated by an emerging infectious disease of its main prey. J. Biogeogr. 36, 1502–1515 (2009).
Google Scholar 
Adamakopoulos, T., Gatzoyannis, S., Poirazidis, K. Study on the assessment, the enhancement of the legal infrastructure and the management of the protected area in the forest of Dadia. Specific Environmental Study, WWF-Greece, Athens (1995).Delibes, M., Hiraldo, F. The rabbit as prey in the Iberian Mediterranean ecosystem. In Proceedings of the World Lagomorph Conference. Guelph: University of Guelph. 1979: 614–622 (1979).Futuyma, D. J. & Moreno, G. The evolution of ecological specialization. Annu. Rev. Ecol. Syst. 19, 207–233 (1988).
Google Scholar 
Moleón, M. et al. Predator–prey relationships in a mediterranean vertebrate system: Bonelli’s eagles, rabbits and partridges. Oecologia 168, 679–689 (2012).ADS 
PubMed 

Google Scholar 
Fedriani, J. M., Ferreras, P. & Delibes, M. Dietary response of the Eurasian badger, Meles meles, to a decline of its main prey in the Doñana national park. J. Zool. 245, 214–218 (1998).
Google Scholar 
Ferrer, M. & Negro, J. J. The near extinction of two large European predators: Super specialists pay a price. Conserv. Biol. 18, 344–349 (2004).
Google Scholar 
Lozano, J., Moleón, M. & Virgós, E. Biogeographical patterns in the diet of the wildcat, Felis silvestris Schreber, in Eurasia: Factors affecting the trophic diversity. J. Biogeogr. 33, 1076–1085 (2006).
Google Scholar 
Burgos, T. et al. Prey density determines the faecal-marking behaviour of a solitary predator, the Iberian lynx (Lynx pardinus). Ethol. Ecol. Evol. 31, 219–230 (2019).
Google Scholar 
Ontiveros, D. & Pleguezuelos, J. M. Influence of prey densities in the distribution and breeding success of Bonelli’s eagle (Hieraaetus fasciatus): Management implications. Biol. Conserv. 93, 19–25 (2000).
Google Scholar 
Araújo, M. S. & Gonzaga, M. O. Individual specialization in the hunting wasp Trypoxylon (Trypargilum) albonigrum (Hymenoptera, Crabronidae). Behav. Ecol. Sociobiol. 61, 1855–1863 (2007).
Google Scholar 
Stephens, D. W. & Krebs, J. R. Foraging Theory 1st edn. (Monographs in Behavior and Ecology. Princeton University Press, 1986).
Google Scholar 
Heath, J. A. et al. Golden Eagle dietary shifts following wildfire and shrub loss have negative consequences for nestling survivorship. Ornithol. Appl. 123(4), duabo34 (2021).
Google Scholar 
Anadón, J. D., Wiegand, T. & Giménez, A. Individual-based movement models reveal sex-biased effects of landscape fragmentation on animal movement. Ecosphere 3, 1–32 (2012).
Google Scholar 
Sanz-Aguilar, A., Anadón, J. D., Giménez, A., Ballestar, R. & Oro, D. Coexisting with fire: The case of the terrestrial tortoise Testudo graeca in mediterranean shrublands. Biol. Conserv. 144, 1040–1049 (2011).
Google Scholar 
Arroyo, B. Águila real – Aquila chrysaetos. In: Enciclopedia Virtual de los Vertebrados Españoles. Salvador, A., Morales, M. B. (Eds.). Museo Nacional de Ciencias Naturales, Madrid. http://www.vertebradosibericos.org/ (2017).Arroyo, B., Ferreiro, E., Garza, V. El águila real (Aquila chrysaetos) en España. Censo, distribución, reproducción y conservación. Serie Técnica, ICONA. Madrid (1990).Bautista, J., Gil-Sánchez, J. M., González Miras, E., Gómez, G. J. & Sánchez Balsera, J. L. Increase in the population of golden eagle in andalusian baetic system mountain ranges (southern of Spain): evidences of competition with the Bonelli’s eagle. Quercus 332, 16–22 (2013).
Google Scholar 
Rodríguez-Caro, R. C., Graciá, E., Anadón, J. D. & Giménez, A. Maintained effects of fire on individual growth and survival rates in a spur-thighed tortoise population. Eur. J. Wildl. Res. 59, 911–913 (2013).
Google Scholar 
Beissinger, S. R. & McCullough, D. R. Population viability analysis (University of Chicago Press, 2002).
Google Scholar 
Tylianakis, J. M., Didham, R. K., Bascompte, J. & Wardle, D. A. Global change and species interactions in terrestrial ecosystems. Ecol. Lett. 11, 1351–1363 (2008).PubMed 

Google Scholar 
Real, J. Biases in diet study methods in the Bonelli’s eagle. J. Wildl. Manag. 60(3), 632–638 (1996).
Google Scholar 
Moleón, M. et al. Laying the foundations for a human-predator conflict solution: Assessing the impact of Bonelli’s eagle on rabbits and partridges. PLoS ONE 6, e22851 (2011).ADS 
PubMed 
PubMed Central 

Google Scholar 
Esteve-Selma, M. A., et al. Effects of climate change on the potential distribution of Testudo graeca in southeastern Iberian Peninsula. In Graciá E, Rodríguez-Caro RC and Giménez A. Conservation of Mediterranean tortoises under global change. Madrid. Asociación Herpetológica Española. ISBN: 978-84-921999-6-9.Anadón, J. D., Giménez, A., Ballestar, R. & Pérez, I. Evaluation of local ecological knowledge as a method for collecting extensive data on animal abundance. Conserv. Biol. 23, 617–625 (2009).PubMed 

Google Scholar 
Abad, V. Variaciones del Índice corporal en una población de tortuga mora (Testudo graeca) del Sureste Ibérico. MSc thesis, Universidad Miguel Hernández de Elche, Spain (2007).Linden, H., Wikman, M. Goshawk predation on tetraonids: Availability of prey and diet of the predator in the breeding season. J. Anim. Ecol., 953–968 (1983).Fevold, H. R. & Craighead, J. J. Food requirements of the golden eagle. Auk 75, 312–317 (1958).
Google Scholar 
Collopy, M. W. Food consumption and growth energetics of nestling golden eagles. Wilson Bull. 445–458 (1986).Blanco, J. C., Villafuerte, R. Factores ecológicos que influyen sobre las poblaciones de conejos. Efectos de la enfermedad hemorrágico vírica. TRAGSA, Madrid Spain (1993). More

The studied viviparous clausiliids developed four types of morphological adaptations that facilitate the delivery of embryos through the shell aperture: (1) reduction of the clausiliar apparatus, (2) decrease of embryonic shell width, (3) widening of the shell canal, and (4) development of a flexible embryonic shell.Reduction of the clausiliar apparatusMembers of the Reinia genus, arboreal species from Japan (Fig. 1), show the most advanced adaptations to live-bearing compared to hypothetical ancestral Phaedusinae. The shell shape in these species is more conical than fusiform, the number of whorls decreases, and the aperture widens. One of the species, R. variegata, features almost full reduction of the clausiliar apparatus that consists of only vestigial folds (Fig. 1F). This species also lacks the clausilium, so the entrance through the aperture is unprotected.Figure 1Different stages of reduction of apertural barriers in members of genus Reinia: R. ashizuriensis (A–C; upper row) and R. variegata (D–F; lower row). (A,D) Adult shells; (B,C,E,F) adult shells with body whorl cut open dorsally in microCT visualisation. cp clausilium plate, il inferior lamella, pr principal plica, sc subcolumellar lamella, sl superior lamella, sp spiral lamella, upp upper palatal plica.Full size imageDecrease of embryonic shell widthAnother adaptation concerns the shape of the embryonic shell (“protoconch”), which becomes very narrow in some viviparous species. This feature is conspicuous because embryonic whorls remain in the adult shell as apical whorls. For instance in S. addisoni (Fig. 2A–D), the apical part being much narrower than the first whorls of the teleoconch is a clear evidence that the growth trajectory has changed abruptly after birth. Other examples include E. cylindrella and E. steetzneri, in which both the protoconch and the teleoconch are very narrow, yet at the borderline between these parts, the shell axis is slightly bent (Fig. 2E–L). We suppose that this feature develops as a result of obstruction during birth.Figure 2Width difference between protoconch and teleoconch in Stereophaedusa addisoni (A–D, upper row), Euphaedusa cylindrella (E–H, middle row), Euphaedusa steetzneri (I–L, lower row). (A,C,E,G,I,K) Adult shells with very narrow apical whorls; (B,F,J) X-rayed adults; (F,J) with retained embryos inside; (D,H,L) X-rays of apical part of adult shell with schematic drawings of a neonate.Full size imageWidening of the shell canalThe third type of adaptation is the widening of the shell canal in the body whorl, allowing for easier passage of the embryo between the lamellae and plicae of the apertural barriers. In this case, the outline of the shell changes only slightly giving the body whorl a more convex appearance. A substantial difference to egg-laying species concerns the apertural barriers: the clausiliar includes a broad clausilium plate and a spirally ascending inferior lamella (Fig. 3A–D). These modifications result in a spacious shell canal in the body whorl, for example in S. addisoni and E. sheridani, that can accommodate the transfer of a large embryo. Table 1 presents neonatal size in these species (shell width ca. 1.2 mm), which is very similar to their clausilium width (ca. 1.1–1.2 mm).Figure 3Two types of clausiliar apparatus occurring in Phaedusinae in microCT visualisation: with spirally ascending inferior lamella and wide clausilium plate (upper row), and with straight ascending inferior lamella and narrow clausilium plate (lower row). (A) T. sheridani adult shell with the body whorl cut open dorsally; (B) clausilium of T. sheridani; (C) clausilium of S. addisoni; (D) clausilium of R. ashizuriensis; (E) Zaptyx ventriosa adult shell with body whorl cut open dorsally; (F) clausilium of Z. ventriosa; (G,H) clausilia of O. miranda. Note, that all depicted species are viviparous.Full size imageTable 1 Shell size of studied Phaedusinae species.Full size tableMost viviparid clausiliids develop one of these three types of modification; some adaptations co-occur within a single species, for example a wide clausilium accompanies a narrow apex. Interestingly, the Reinia genus includes taxa with a gradual escalation of viviparity-related adaptations: R. ashizurensis, with a stout shell shape and a low number of whorls, has fully developed apertural barriers with a broad clausilium plate (Fig. 1A–C), while its congener, R. variegata, has reduced apertural barriers (Fig. 1D–F).Development of a flexible embryonic shellThe fourth type of adaptation found in Phaedusinae concerns the structure of the embryonic shells. We report this adaptation in O. miranda and Z. ventriosa.Oospira miranda is a dextral, often decollated, ground-dwelling species from Vietnam (Fig. 4A). The species is viviparous: during microCT scanning of museum specimens, we found embryos within a parental shell (Fig. 4B); in laboratory culture, we observed neonates immediately after live birth (Fig. 4C,D). Morphological characters recognized in the adult shell, i.e., a wide apex (= wide embryonic shell), straightly ascending inferior lamella, and a narrow clausilium plate (Fig. 3G,H), seemed to exclude the possibility of live-bearing reproduction, as embryos are too large to pass through the shell canal at the narrowest point. The height and width of the neonatal shell (mean values: 5.19 mm, 3.59 mm) evidently exceeds the width of the clausilium plate in this species (1.97 mm) (Table 1). However, under closer examination, we found the shell to be thin and delicate, which we refer to as a ‘soft shell’. In direct examination, the neonatal shell of O. miranda resembles cellophane, which may keep a given shape for a long time but becomes distorted already under slight pressure.Figure 4Viviparous clausiliids and their ‘soft-shelled’ neonates born in laboratory culture. (A–D) O. miranda: adult shell, X-rayed shell with embryo visible inside, neonates; (E–H) Z. ventriosa: adult shell, X-rayed shell with eggs visible inside, neonates.Full size imageA similar adaptation exists in Z. ventriosa, a Taiwanese species with a very wide apex, never decollated, a straight ascending inferior lamella, and a narrow clausilium plate (Figs. 3E,F, 4E,F). This species produces neonates in laboratory culture (Fig. 4G–H). The dimensions of the neonates (mean values: height 3.37 mm, width 2.51 mm) exceed at last twofold the width of the clausilium plate (1.08 mm). The shells of such freshly delivered juveniles, when gently touched with laboratory tweezers, became dented, but not fractured. More intense and stronger pressing can break this dentation.These initial observations, that we made during the maintenance of the laboratory culture, suggested that the neonatal shells of O. miranda and Z. ventriosa have flexible walls. These ‘soft-shells’ seem to be highly malleable during the entire embryonic development period and delivery through apertural barriers, hardening shortly after birth. We further investigated the physical properties of the embryonic shell by means of microcomputed tomography and scanning electron microscopy.Microcomputed tomographyWe scanned ‘soft-shelled’ neonates of O. miranda and Z. ventriosa, together with ‘hard-shelled’ embryos and neonates of S. addisoni and T. sheridani, in order to compare the density and thickness of the shells (Fig. 5).Figure 5Comparison of embryonic shell thickness in clausiliids: ‘soft-shelled’ neonates of Z. ventriosa (A,B,G,H) and O. miranda (C,D,I,J); “hard-shelled” neonate of S. addisoni (E,K) and embryo of T. sheridani (F,L) scanned inside a parental shell. Upper row—microCT visualisation of shell surface; middle row—microCT sections of those specimens; (M–O) X-ray photographs of S. addisoni (embryo from dissected adult) and Z. ventriosa (neonate) enlarged in (N,O), respectively, showing the difference in shell density and thickness; (P) microCT based volume rendering of O. miranda (left) and S. addisoni (right) neonates, showing difference between relative density of their shells.Full size imagePreliminary observations using the two-dimensional X-ray photographs showed a difference in thickness and density between S. addisoni and Z. ventriosa (Fig. 5M, enlarged in N and O, respectively). The 3D visualization of O. miranda and S. addisoni (the same microCT scanning and reconstruction parameters) confirmed the difference between density and shell thickness of these two species (Fig. 5P).Due to variations in wall thickness within the neonatal shell (e.g., between the first and the second whorls), it is not possible to precisely determine the thickness of the shell wall. The accuracy of the measurement is also limited by the resolution of the microCT scans, especially in the case of the relatively large neonates of O. miranda and Z. ventriosa. When scanning the whole embryonic shell of Z. ventriosa (approximately 3.5 mm in height), the size of the voxel was approximately 1 µm. Thus, we cannot determine the shell thickness down to the nearest micron, but we can estimate it from a few to a dozen microns. A direct comparison between virtual microCT sections of specimens scanned under the same conditions shows a clear difference between the ‘soft-shelled’ and ‘hard-shelled’ taxa (Fig. 5G–L). The ’hard-shelled’ neonates have a shell wall of 30–40 µm thick. We examined the sequence of three ’soft-shelled’ O. miranda specimens that differed in size (the exact time of birth of each of the cultured neonates is unknown, ca. 1–2 days). The larger (older) the neonate was, the thicker the shell. The shell of the largest of the studied O. miranda was up to 20 µm thick. However, the shell wall of this relatively large juvenile (several millimeters in height) still did not reach the thickness of the small ’hard-shelled’ T. sheridani embryo, which was already about 30–40 µm thick, stiff and rigid during the retention in the genital tract. The neonates of O. miranda and Z. ventriosa were much larger than the embryos and neonates of S. addisoni and R. variegata (Table 1), however, the former taxa has much thinner shells.Scanning electron microscopyAfter the non-invasive microCT scan, we scanned embryos and neonates using SEM (Fig. 6). The different properties of the shells of Z. ventriosa and O. miranda vs. S. addisoni and R. variegata were already visible during the preparation of the analysis. Under vacuum conditions, the soft shells of Z. ventriosa and O. miranda shrank and crumpled, creating a cellophane-like surface (Fig. 6A). Embryos and neonates of S. addisoni and R. variegata did not require any special preparation and their shell shape remained unchanged under the vacuum conditions applied during the SEM examination (Fig. 6D,E). To reduce the shell deformations, we freeze-dried the next group of thin-shelled neonates prior to SEM analyses (Fig. 6B,C).Figure 6Neonates of O. miranda (A,B,F,I,L,M,O) and Z. ventriosa (C,G,J,P) in direct comparison with hard-shelled embryos and neonates of R. variegata (D,N,Q) and S. addisoni (E,H,K); SEM microphotographs. The vacuum conditions in SEM led to the shrinkage of the thin O. miranda shell (A); freeze-drying of ‘soft-shelled’ neonates prior to SEM imaging reduced the level of deformity (B,C). Contrastingly, R. variegata and S. addisoni shells do not require special preparation and retain their shape (D,E). (F) The dented surface of O. miranda neonate and SEM-close-up (I) on a cross-section of the shell just a few micrometers thick (arrow in F indicates the region enlarged in I). (G,J) Shell of Z. ventriosa in comparison with similarly ornamented fragment of S. addisoni (H,K); note several times thicker shell in the latter (arrows in G,H indicate the regions enlarged in J,K, respectively). (L,M) Inner surface of intact periostracum which still connects two fragments of broken aragonite shell of O. miranda (the arrow in M indicates the region enlarged in L); note the difference between shell thickness in O. miranda (L,M) and R. variegata (N). All observed specimens have similar crossed-lamellar microstructure (L–Q). However, just as shell thickness, also the number of lamellar layers of alternate orientation within the shell differs (L,M,O,P vs N,Q).Full size imageThe SEM studies allowed for complementary measurements of the shells. In the broken fragments of Z. ventriosa and O. miranda, the thickness of the shell wall ranged from 2–3 µm (Fig. 6F,G,I,J,L,M) to 18 µm in the largest neonate of O. miranda (Fig. 6O). The shells of S. addisoni (Fig. 6H,K) and R. variegata (Fig. 6N) are several times thicker.All analyzed samples have a thin ( More