Ecology

Yeo, D. C. J. et al. Global diversity of crabs (Crustacea: Decapoda: Brachyura). In Freshwater Animal Diversity Assessment (eds Balian, E. V. et al.). Hydrobiologia Vol. 595, 275–286 (2008).Álvarez, F. et al. Revision of the higher taxonomy of Neotropical freshwater crabs of the family Pseudothelphusidae, based on multigene and morphological analyses. Zool. J. Linn. Soc. 193, 973–1001 (2021).Article 

Google Scholar 
Ng, D. J. J. & Yeo, D. C. J. Terrestrial scavenging behaviour of the Singapore freshwater crab, Johora singaporensis (Crustacea: Brachyura: Potamidae). Nat. Singap. 6, 207–210 (2013).
Google Scholar 
Dobson, M. Freshwater crabs in Africa. Freshw. Forum 21, 3–26 (2004).
Google Scholar 
Dobson, M., Magana, A. M., Mathooko, J. M. & Ndegwa, F. K. Distribution and abundance of freshwater crabs (Potamonautes spp.) in rivers draining Mt Kenya, East Africa. Fundam. Appl. Limnol. 168, 271–279 (2007).Article 

Google Scholar 
Cumberlidge, N. et al. Freshwater crabs and the biodiversity crisis: Importance, threats, status, and conservation challenges. Biol. Conserv. 142, 1665–1673 (2009).Article 

Google Scholar 
Jouladeh-Roudbar, A., Ghanavi, H. R. & Doadrio, I. Ichthyofauna from Iranian freshwater: Annotated checklist, diagnosis, taxonomy, distribution and conservation. Assessment 21, 1–303 (2020).
Google Scholar 
Brandis, D., Storch, V. & Türkay, M. Taxonomy and zoogeography of the freshwater crabs of Europe, North Africa, and the Middle East (Crustacea, Decapoda, Potamidae). Senckenberg. Biol. 80, 5–56 (2000).
Google Scholar 
Keikhosravi, A. & Schubart, C. D. Description of a new freshwater crab species of the genus Potamon (Decapoda, Brachyura, Potamidae) from Iran, based on morphological and genetic characters. In Advances in Freshwater Decapod Systematics and Biology 115–133 (BRILL, 2014). https://doi.org/10.1163/9789004207615_008.Keikhosravi, A. & Schubart, C. D. Revalidation and redescription of Potamon elbursi Pretzmann, 1976 (Brachyura, Potamidae) from Iran, based on morphology and genetics. Open Life Sci. 9, 114–123 (2014).Article 

Google Scholar 
Keikhosravi, A., Naderloo, R. & Schubart, C. D. Morphological and molecular diversity in the freshwater crab Potamon ruttneri-P. gedrosianum species complex (Decapoda, Brachyura) indicate the need for taxonomic revision. Crustaceana 89, 129–139 (2016).Article 

Google Scholar 
Parvizi, E., Naderloo, R., Keikhosravi, A., Solhjouy-Fard, S. & Schubart, C. D. Multiple Pleistocene refugia and repeated phylogeographic breaks in the southern Caspian Sea region: Insights from the freshwater crab Potamon ibericum. J. Biogeogr. 45, 1234–1245 (2018).Article 

Google Scholar 
Folmer, O., Black, M., Hoeh, W., Lutz, R. & Vrijenhoek, R. DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Aust. J. Zool. https://doi.org/10.1071/ZO9660275 (1994).Article 

Google Scholar 
Kearse, M. et al. Geneious basic: An integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics 28, 1647–1649 (2012).Article 

Google Scholar 
Katoh, K. MAFFT: A novel method for rapid multiple sequence alignment based on fast Fourier transform. Nucleic Acids Res. 30, 3059–3066 (2002).CAS 
Article 

Google Scholar 
Katoh, K. & Standley, D. M. MAFFT multiple sequence alignment software version 7: Improvements in performance and usability. Mol. Biol. Evol. 30, 772–780 (2013).CAS 
Article 

Google Scholar 
Nguyen, L.-T., Schmidt, H. A., von Haeseler, A. & Minh, B. Q. IQ-TREE: A fast and effective stochastic algorithm for estimating maximum-likelihood phylogenies. Mol. Biol. Evol. 32, 268–274 (2015).CAS 
Article 

Google Scholar 
Kalyaanamoorthy, S., Minh, B. Q., Wong, T. K. F., von Haeseler, A. & Jermiin, L. S. ModelFinder: Fast model selection for accurate phylogenetic estimates. Nat. Methods 14, 587–589 (2017).CAS 
Article 

Google Scholar 
Felsenstein, J. Confidence limits on phylogenies: An approach using the bootstrap. Evolution 39, 783 (1985).Article 

Google Scholar 
Zhang, J., Kapli, P., Pavlidis, P. & Stamatakis, A. A general species delimitation method with applications to phylogenetic placements. Bioinformatics 29, 2869–2876 (2013).CAS 
Article 

Google Scholar 
Tamura, K., Stecher, G. & Kumar, S. MEGA11: Molecular evolutionary genetics analysis version 11. Mol. Biol. Evol. 38, 3022–3027 (2021).CAS 
Article 

Google Scholar 
Masters, B. C., Fan, V. & Ross, H. A. Species delimitation—A geneious plugin for the exploration of species boundaries. Mol. Ecol. Resour. 11, 154–157 (2011).Article 

Google Scholar  More

Our study combining field data and aerial imagery analysis clearly showed that the spotted souslik avoids close coexistence with another burrowing species, i.e. the European mole, in the period of low population abundance. This is the first study on this subject described in the available literature, as attention has been paid mainly to other parameters of the habitat so far14,18,20. The present results can (1) make a new contribution to the knowledge of the ecology of burrowing mammals and their interspecies relationships, (2) contribute to better designs of conservation and assessment of the quality of habitats of endangered burrowing mammals, and (3) indicate new possibilities of using remote sensing and deep learning methods in ecology and conservation. Below we will try to address each of these issues.The interaction between underground animals is not a new idea in ecology (e.g.22); however, this issue has not been analyzed for the mole and the souslik so far. This was probably related to the fact that the potential negative or positive relationships between these species are not intuitively obvious. The spatial distribution of underground tunnels of these animals is completely different: the mole builds an extensive network of horizontal tunnels close to the ground surface, while the souslik usually builds one deep nest burrow with a vertical entrance and possibly a small number of shallow safety burrows near the nest burrow. Moreover, the food preferences of the souslik and the mole differ, i.e. the former is mainly a herbivore, while the latter is an obligatory predator. There are also clear differences in the annual cycle: the mole is active all year round, and the souslik hibernates in an underground nest for about half a year from October to March. Thus, it seems that the emergence of competitive relationships between these two species is unlikely. Our study shows, however, that these species avoid each other in space, which raises the question of the mechanism of this relationship. Based on the knowledge of the biology of both species, some hypothetical mechanisms can be proposed.Although they are colonial animals, sousliks inhabit burrows alone (except for mother and offspring) and they have a strong behavioural trait of a negative reaction to the presence of other animals in their burrows and their close vicinity14,23. The negative reaction to other sousliks is a reflection of the intraspecific competition in the population and the territoriality of individuals. It is regulated by odour signals and the social structure of the population30,31. Koshev32 described aggressive reactions of free-ranging European sousliks to other vertebrate species that appeared near burrows: towards the reptile Lacerta trilineata, the bird Corvus frugilegus, and the mammal Mustela nivalis. Theoretically, the mole can get into the souslik’s burrow unintentionally when digging new tunnels. For souslik, the presence of moles in their nest burrow means a violation of its strictly defended territory and is probably a highly stressful episode. It can therefore be assumed that sousliks should choose places outside areas of frequent occurrence of other burrowing mammals to set up a nest burrow.It remains an open question whether avoidance of areas where the mole is often present may be important for the souslik during winter hibernation. Theoretically, the presence of moles in souslik burrows during hibernation may disturb this process and cause waking up and energy-consuming increases in metabolism, which may reduce winter survival. It is also unknown whether the mole can be a predator for the souslik during winter hibernation. Remains of rodent species were found in the digestive tracts of moles33; therefore, at least theoretically, the mole may use such a food source. On the other hand, remains of vertebrates, including the remains of moles, were sometimes found in the stomachs of sousliks18. The relationship between the souslik and the mole may therefore be more complex and require further research focused on this issue. It is possible that the moles can avoid the souslik colonies as well. This scenario seems also realistic, since the moles home ranges are likely much more dynamic than that of sousliks, that likely benefit from dwelling within an existing colony of the conspecifics.The spotted souslik protection requires the designation of special areas of conservation16. A number of various conservation activities are also routinely undertaken for this species, including regular monitoring of the population size, habitat monitoring, mowing, reduction of predation risk, and application of more invasive methods such as reintroduction. Similar activities are also performed for a closely related species, i.e. the European souslik Spermophilus citellus, in Europe. Importantly, in the current guidelines of souslik conservation, the issue of the competition with other species and its impact on spatial distribution is not considered. In turn, there is evidence in the literature that interspecies interactions may be important for the souslik population21. In periods of low abundance, when the survival of the population is at risk, the sousliks may have different habitat preferences than in periods of the abundant population20. It seems, therefore, that nowadays, when the souslik most often forms small populations, more attention should be paid to a wider range of factors and threats that may determine longer term population trends or the health condition, survival, and abundance of their colonies.Our study indicates that, in the period of low population abundance, the presence of other burrowing species may be an important factor determining the distribution of sousliks. This observation shows that in addition to the assessment of the area and condition of the habitat the presence of other potentially competitive species should also be taken into account in the analysis of population survival. In such a case, the actual area of habitats suitable for sousliks in a given location may turn out to be much lower than assumed. In our study area, the habitat suitable for the souslik was reduced from 105 ha to approx. 65 ha, i.e. by nearly 38%, but it probably is even smaller (compare Fig. 8). This observation has consequences for improvement of the reintroduction methods of sousliks (or other burrowing mammals), which are constantly of scientific interest20,34,35. Our results indicate that the reintroduction of sousliks should be carried out in places where there is the lowest probability of competition for resources including even shelter or space with other burrowing species and where adequate space for the settlement of the population is ensured.So far, investigations of the distribution of small burrowing mammals have been based on laborious field studies involving site inspections by trained observers (e.g.36,37,38). Our results show that, in certain conditions, high-resolution imagery can be successfully used to support studies of the distribution of such animals. As reported by other authors (e.g.7,10,12), however, such animals must produce clear signs of their presence in the environment. Evidence of the presence of the European mole, i.e. mounds of soil, in short vegetation habitats has shown that remote sensing can detect moles and their area of occupancy successfully. The advantage of these markers of the presence of moles is that the mounds are redundant and quite durable and can be visible in the environment for up to several months.By combining field research and remote sensing, it is also possible to study more sophisticated ecological issues, e.g. interspecies interactions. In this work, the remote estimation of the distribution of moles facilitated estimation of the actual habitat available to the souslik and excluded areas with the lowest probability of its occurrence. As a result, the population may be monitored more economically. Since the conservation guidelines recommend monitoring souslik populations by means of laborious inspections of transects, the indication of areas with no burrows may significantly reduce the amount of fieldwork without negative consequences for the accuracy of results. Some areas of the souslik occurrence are large, e.g. Świdnik (105 ha) or Pastwiska nad Huczwą (150 ha), and every 10 ha to be monitored means one day’s work for one observer (according to the calculations presented in the results). Our study showed that when the area of the occurrence of moles is excluded from the monitoring (Fig. 8), the error in estimating the size of the souslik population will be relatively small (0.9–8.7%). At the same time, the time devoted to the research can be limited by 14% or 38%, respectively. This suggests that our method can contribute to improved monitoring and management of these protected species, especially that souslik monitoring requires considerable research effort and has to be carried out twice a year.However, mole mounds may be underestimated by remote sensing, which can be seen in Fig. 7. Small mole mounds that are easily identified during field research may not be noticed by remote sensing. Such underestimation does not constitute a critical threat to the determination of the mole area according to the scheme shown in Fig. 8, since its marks are highly redundant. However, since there is currently little research on this subject, we recommend combining field research and remote sensing in assessments similar to ours. Finally, it is worth noting that, for a better understanding of the issue of the interactions between souslik and other burrowing species, it is advisable to use another remote sensing technique—telemetry. Telemetry studies are successfully conducted in Bulgarian souslik populations34 and their combination with studies of habitat selectivity dependent on other burrowing species may provide new and valuable insight into this issue. More

Brienen, R. J. W. et al. Long-term decline of the Amazon carbon sink. Nature 519, 344–348 (2015).CAS 
PubMed 
Article 

Google Scholar 
Hubau, W. et al. Asynchronous carbon sink saturation in African and Amazonian tropical forests. Nature 579, 80–87 (2020).CAS 
PubMed 
Article 

Google Scholar 
Zuleta, D., Duque, A., Cardenas, D., Muller-Landau, H. C. & Davies, S. J. Drought-induced mortality patterns and rapid biomass recovery in a terra firme forest in the Colombian Amazon. Ecology 98, 2538–2546 (2017).PubMed 
Article 

Google Scholar 
Phillips, O. L. et al. Drought sensitivity of the Amazon rainforest. Science 323, 1344–1347 (2009).CAS 
PubMed 
Article 

Google Scholar 
Powers, J. S. et al. A catastrophic tropical drought kills hydraulically vulnerable tree species. Glob. Chang. Biol. 26, 3122–3133 (2020).PubMed 
Article 

Google Scholar 
Bennett, A. C. et al. Resistance of African tropical forests to an extreme climate anomaly. Proc. Natl Acad. Sci. USA 118, e2003169118 (2021).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Brodribb, T. J., Powers, J., Cochard, H. & Choat, B. Hanging by a thread? Forests and drought. Science 368, 261–266 (2020).CAS 
PubMed 
Article 

Google Scholar 
McDowell, N. G. et al. Pervasive shifts in forest dynamics in a changing world. Science 368, (2020).Pan, Y. et al. A large and persistent carbon sink in the world’s forests. Science 333, 988–993 (2011).CAS 
PubMed 
Article 

Google Scholar 
Matthews, H. D. et al. An integrated approach to quantifying uncertainties in the remaining carbon budget. Commun. Earth Environ. 2, 7 (2021).Article 

Google Scholar 
Girardin, C. A. J. et al. Nature-based solutions can help cool the planet—if we act now. Nature 593, 191–194 (2021).CAS 
PubMed 
Article 

Google Scholar 
Friedlingstein, P. et al. Earth Syst. Sci. Data 14, 1917–2005 (2022)
Google Scholar 
Choat, B. et al. Triggers of tree mortality under drought. Nature 558, 531–539 (2018).CAS 
PubMed 
Article 

Google Scholar 
Rowland, L. et al. Death from drought in tropical forests is triggered by hydraulics not carbon starvation. Nature 528, 119–122 (2015).CAS 
PubMed 
Article 

Google Scholar 
Lloyd, J. & Farquhar, G. D. Effects of rising temperatures and [CO2] on the physiology of tropical forest trees. Phil. Trans. R. Soc. B 363, 1811–1817 (2008).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
O’Sullivan, O. S. et al. Thermal limits of leaf metabolism across biomes. Glob. Chang. Biol. 23, 209–223 (2017).PubMed 
Article 

Google Scholar 
Grossiord, C. et al. Plant responses to rising vapor pressure deficit. New Phytol. 226, 1550–1566 (2020).PubMed 
Article 

Google Scholar 
Rifai, S. W., Li, S. & Malhi, Y. Coupling of El Niño events and long-term warming leads to pervasive climate extremes in the terrestrial tropics. Environ. Res. Lett. 14, 105002 (2019).CAS 
Article 

Google Scholar 
Rifai, S. W. et al. ENSO drives interannual variation of forest woody growth across the tropics. Phil. Trans. R. Soc. B 373, 20170410 (2018).PubMed 
PubMed Central 
Article 

Google Scholar 
Smith, M. N. et al. Empirical evidence for resilience of tropical forest photosynthesis in a warmer world. Nat. Plants 6, 1225–1230 (2020).CAS 
PubMed 
Article 

Google Scholar 
Malhi, Y. et al. Exploring the likelihood and mechanism of a climate-change-induced dieback of the Amazon rainforest. Proc. Natl Acad. Sci. USA 106, 20610–20615 (2009).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
McDowell, N., Allen, C. D. & Anderson‐Teixeira, K. Drivers and mechanisms of tree mortality in moist tropical forests. New Phytol. 219, 851–869 (2018).PubMed 
Article 

Google Scholar 
McDowell, N. et al. Mechanisms of plant survival and mortality during drought: why do some plants survive while others succumb to drought? New Phytol. 178, 719–739 (2008).PubMed 
Article 

Google Scholar 
Bauman, D. et al. Tropical tree growth sensitivity to climate is driven by species intrinsic growth rate and leaf traits. Glob. Chang. Biol. 28, 1414–1432 (2022).PubMed 
Article 

Google Scholar 
Esquivel-Muelbert, A. et al. Tree mode of death and mortality risk factors across Amazon forests. Nat. Commun. 11, 5515 (2020).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Anderegg, W. R. L., Anderegg, L. D. L., Kerr, K. L. & Trugman, A. T. Widespread drought-induced tree mortality at dry range edges indicates that climate stress exceeds species’ compensating mechanisms. Glob. Chang. Biol. 25, 3793–3802 (2019).PubMed 
Article 

Google Scholar 
Aguirre-Gutiérrez, J. et al. Drier tropical forests are susceptible to functional changes in response to a long-term drought. Ecol. Lett. 22, 855–865 (2019).PubMed 
Article 

Google Scholar 
Aguirre-Gutiérrez, J. et al. Long-term droughts may drive drier tropical forests towards increased functional, taxonomic and phylogenetic homogeneity. Nat. Comm. 11, 3346 (2020).Article 

Google Scholar 
Meir, P., Mencuccini, M. & Dewar, R. C. Drought-related tree mortality: addressing the gaps in understanding and prediction. New Phytol. 207, 28–33 (2015).PubMed 
Article 

Google Scholar 
Sullivan, M. J. P. et al. Long-term thermal sensitivity of Earth’s tropical forests. Science 368, 869–874 (2020).CAS 
PubMed 
Article 

Google Scholar 
Yuan, W. et al. Increased atmospheric vapor pressure deficit reduces global vegetation growth. Sci. Adv. 5, eaax1396 (2019).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
McMahon, S. M., Arellano, G. & Davies, S. J. The importance and challenges of detecting changes in forest mortality rates. Ecosphere 10, e02615 (2019).Article 

Google Scholar 
Trugman, A. T. et al. Tree carbon allocation explains forest drought-kill and recovery patterns. Ecol. Lett. 21, 1552–1560 (2018).CAS 
PubMed 
Article 

Google Scholar 
Trugman, A. T., Anderegg, L. D. L., Anderegg, W. R. L., Das, A. J. & Stephenson, N. L. Why is tree drought mortality so hard to predict? Trends Ecol. Evol. 36, 520–532 (2021).PubMed 
Article 

Google Scholar 
Phillips, O. L. et al. Drought–mortality relationships for tropical forests. New Phytol. 187, 631–646 (2010).PubMed 
Article 

Google Scholar 
Aleixo, I. et al. Amazonian rainforest tree mortality driven by climate and functional traits. Nat. Clim. Change 9, 384–388 (2019).Article 

Google Scholar 
Lingenfelder, M. & Newbery, D. M. On the detection of dynamic responses in a drought-perturbed tropical rainforest in Borneo. Plant Ecol. 201, 267–290 (2009).Article 

Google Scholar 
McDowell, N. G. et al. The interdependence of mechanisms underlying climate-driven vegetation mortality. Trends Ecol. Evol. 26, 523–532 (2011).PubMed 
Article 

Google Scholar 
Zuleta, D. et al. Individual tree damage dominates mortality risk factors across six tropical forests. New Phytol. 233, 705–721 (2022).PubMed 
Article 

Google Scholar 
Fontes, C. G. et al. Dry and hot: the hydraulic consequences of a climate change-type drought for Amazonian trees. Phil. Trans. R. Soc. B 373, 20180209 (2018).Chave, J. et al. Towards a worldwide wood economics spectrum. Ecol. Lett. 12, 351–366 (2009).PubMed 
Article 

Google Scholar 
Peters, J. M. R. et al. Living on the edge: a continental-scale assessment of forest vulnerability to drought. Glob. Chang. Biol. 27, 3620–3641 (2021).PubMed 
Article 

Google Scholar 
Yang, J., Cao, M. & Swenson, N. G. Why functional traits do not predict tree demographic rates. Trends Ecol. Evol. 33, 326–336 (2018).PubMed 
Article 

Google Scholar 
Espírito-Santo, F. D. B. et al. Size and frequency of natural forest disturbances and the Amazon forest carbon balance. Nat. Commun. 5, 3434 (2014).PubMed 
Article 

Google Scholar 
Chambers, J. Q. et al. The steady-state mosaic of disturbance and succession across an old-growth Central Amazon forest landscape. Proc. Natl Acad. Sci. USA 110, 3949–3954 (2013).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Rifai, S. W. et al. Landscape-scale consequences of differential tree mortality from catastrophic wind disturbance in the Amazon. Ecol. Appl. 26, 2225–2237 (2016).PubMed 
Article 

Google Scholar 
López, J., Way, D. A. & Sadok, W. Systemic effects of rising atmospheric vapor pressure deficit on plant physiology and productivity. Glob. Chang. Biol. 27, 1704–1720 (2021).PubMed 
PubMed Central 
Article 

Google Scholar 
Brando, P. M. et al. Abrupt increases in Amazonian tree mortality due to droughttextendashfire interactions. Proc. Natl Acad. Sci. USA 111, 6347–6352 (2014).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Phillips, O. L. et al. Pattern and process in Amazon tree turnover, 1976–2001. Phil. Trans. R. Soc. Lond. B 359, 381–407 (2004).CAS 
Article 

Google Scholar 
Harris, R. M. B. et al. Biological responses to the press and pulse of climate trends and extreme events. Nat. Clim. Change 8, 579–587 (2018).Article 

Google Scholar 
Andrus, R. A., Chai, R. K., Harvey, B. J., Rodman, K. C. & Veblen, T. T. Increasing rates of subalpine tree mortality linked to warmer and drier summers. J. Ecol. 109, 2203–2218 (2021).Article 

Google Scholar 
Murphy, H. T., Bradford, M. G., Dalongeville, A., Ford, A. J. & Metcalfe, D. J. No evidence for long-term increases in biomass and stem density in the tropical rain forests of Australia. J. Ecol. 101, 1589–1597 (2013).Article 

Google Scholar 
Bennett, A. C., McDowell, N. G., Allen, C. D. & Anderson-Teixeira, K. J. Larger trees suffer most during drought in forests worldwide. Nat. Plants 1, 15139 (2015).PubMed 
Article 

Google Scholar 
Chitra-Tarak, R. et al. Hydraulically-vulnerable trees survive on deep-water access during droughts in a tropical forest. New Phytol. 231, 1798–1813 (2021).PubMed 
PubMed Central 
Article 

Google Scholar 
Anderegg, W. R. L. et al. Meta-analysis reveals that hydraulic traits explain cross-species patterns of drought-induced tree mortality across the globe. Proc. Natl Acad. Sci. USA 113, 5024–5029 (2016).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Taylor, T. C., Smith, M. N., Slot, M. & Feeley, K. J. The capacity to emit isoprene differentiates the photosynthetic temperature responses of tropical plant species. Plant Cell Environ. 42, 2448–2457 (2019).CAS 
PubMed 
Article 

Google Scholar 
Arellano, G., Zuleta, D. & Davies, S. J. Tree death and damage: a standardized protocol for frequent surveys in tropical forests. J. Veg. Sci. 32, e12981 (2021).Article 

Google Scholar 
Bradford, M. G., Murphy, H. T., Ford, A. J., Hogan, D. L. & Metcalfe, D. J. Long-term stem inventory data from tropical rain forest plots in Australia. Ecology 95, 2362 (2014).Article 

Google Scholar 
Johnson, D. J. et al. Climate sensitive size-dependent survival in tropical trees. Nat. Ecol. Evol. 2, 1436–1442 (2018).PubMed 
Article 

Google Scholar 
Needham, J., Merow, C., Chang-Yang, C.-H., Caswell, H. & McMahon, S. M. Inferring forest fate from demographic data: from vital rates to population dynamic models. Proc. Biol. Sci. 285, 20172050 (2018).PubMed 
PubMed Central 

Google Scholar 
Lewis, S. L. et al. Tropical forest tree mortality, recruitment and turnover rates: calculation, interpretation and comparison when census intervals vary. J. Ecol. 92, 929–944 (2004).Article 

Google Scholar 
Reeves, J., Chen, J., Wang, X. L., Lund, R. & Lu, Q. Q. A review and comparison of changepoint detection techniques for climate data. J. Appl. Meteorol. Climatol. 46, 900–915 (2007).Article 

Google Scholar 
Clark, J. S., Bell, D. M., Kwit, M. C. & Zhu, K. Competition-interaction landscapes for the joint response of forests to climate change. Glob. Chang. Biol. 20, 1979–1991 (2014).PubMed 
Article 

Google Scholar 
Oliva, J., Stenlid, J. & Martínez-Vilalta, J. The effect of fungal pathogens on the water and carbon economy of trees: implications for drought-induced mortality. New Phytol. 203, 1028–1035 (2014).CAS 
PubMed 
Article 

Google Scholar 
Franklin, J. F., Shugart, H. H. & Harmon, M. E. Tree death as an ecological process. Bioscience 37, 550–556 (1987).Article 

Google Scholar 
Yanoviak, S. P. et al. Lightning is a major cause of large tree mortality in a lowland neotropical forest. New Phytol. 225, 1936–1944 (2020).PubMed 
Article 

Google Scholar 
Preisler, Y., Tatarinov, F., Grünzweig, J. M. & Yakir, D. Seeking the ‘point of no return’ in the sequence of events leading to mortality of mature trees. Plant Cell Environ. 44, 1315–1328 (2020).PubMed 
Article 

Google Scholar 
Aragão, L. E. O. C. et al. Spatial patterns and fire response of recent Amazonian droughts. Geophys. Res. Lett. 34, L07701 (2007).Article 

Google Scholar 
Malhi, Y. et al. The linkages between photosynthesis, productivity, growth and biomass in lowland Amazonian forests. Glob. Chang. Biol. 21, 2283–2295 (2015).PubMed 
Article 

Google Scholar 
Hutchinson, M. F., Xu, T., Kesteven, J. L., Marang, I. J. & Evans, B. J.ANUClimate v2.0, NCI Australia. https://doi.org/10.25914/60a10aa56dd1b (2021).Abatzoglou, J. T., Dobrowski, S. Z., Parks, S. A. & Hegewisch, K. C. TerraClimate, a high-resolution global dataset of monthly climate and climatic water balance from 1958–2015. Sci. Data 5, 170191 (2018).PubMed 
PubMed Central 
Article 

Google Scholar 
Carscadden, K. A. et al. Niche breadth: causes and consequences for ecology, evolution, and conservation. Q. Rev. Biol. 95, 179–214 (2020).Article 

Google Scholar 
Swenson, N. G. et al. A reframing of trait–demographic rate analyses for ecology and evolutionary biology. Int. J. Plant Sci. 181, 33–43 (2020).Article 

Google Scholar 
Morueta-Holme, N. et al. Habitat area and climate stability determine geographical variation in plant species range sizes. Ecol. Lett. 16, 1446–1454 (2013).PubMed 
PubMed Central 
Article 

Google Scholar 
Brum, M. et al. Hydrological niche segregation defines forest structure and drought tolerance strategies in a seasonal Amazon forest. J. Ecol. 107, 318–333 (2019).Article 

Google Scholar 
Chitra-Tarak, R. et al. The roots of the drought: hydrology and water uptake strategies mediate forest-wide demographic response to precipitation. J. Ecol. 106, 1495–1507 (2018).Article 

Google Scholar 
Boria, R. A., Olson, L. E., Goodman, S. M. & Anderson, R. P. Spatial filtering to reduce sampling bias can improve the performance of ecological niche models. Ecol. Modell. 275, 73–77 (2014).Article 

Google Scholar 
Farquhar, G. D., von Caemmerer, S. & Berry, J. A. A biochemical model of photosynthetic CO2 assimilation in leaves of C3 species. Planta 149, 78–90 (1980).CAS 
PubMed 
Article 

Google Scholar 
Duursma, R. A. Plantecophys—an R package for analysing and modelling leaf gas exchange data. PLoS ONE 10, e0143346 (2015).PubMed 
PubMed Central 
Article 

Google Scholar 
De Kauwe, M. G. et al. A test of the ‘one-point method’ for estimating maximum carboxylation capacity from field-measured, light-saturated photosynthesis. New Phytol. 210, 1130–1144 (2016).PubMed 
Article 

Google Scholar 
Bloomfield, K. J. et al. The validity of optimal leaf traits modelled on environmental conditions. New Phytol. 221, 1409–1423 (2019).CAS 
PubMed 
Article 

Google Scholar 
McElreath, R. Statistical Rethinking: A Bayesian Course with Examples in R and STAN (CRC Press, 2020).“RStan: the R interface to Stan.” R package version 2.21.2. http://mc-stan.org/ (Stan Development Team, 2020).Bürkner, P.-C. brms: An R package for Bayesian multilevel models using Stan. J. Stat. Softw. 80, 1–28 (2017).Article 

Google Scholar 
R Core Team. R: a language and environment for statistical computing. https://www.R-project.org/ (R Foundation for Statistical Computing, 2021).Dinerstein, E. et al. An ecoregion-based approach to protecting half the terrestrial realm. Bioscience 67, 534–545 (2017).PubMed 
PubMed Central 
Article 

Google Scholar  More

Hughes, C. E. & Atchinson, G. W. The ubiquity of alpine plant radiations: from the Andes to the Hengduan Mountains. N. Phytol. 207, 275–282 (2015).Article 

Google Scholar 
Rahbek, C. et al. Humboldt’s enigma: what causes global patterns of mountain biodiversity? Science 365, 1108–1113 (2019).ADS 
CAS 
PubMed 
Article 

Google Scholar 
Antonelli, A. et al. Geological and climatic influences on mountain biodiversity. Nat. Geosci. 11, 718–725 (2018).ADS 
CAS 
Article 

Google Scholar 
Quintero, I. & Jetz, W. Global elevational diversity and diversification of birds. Nature 555, 246–250 (2018).ADS 
CAS 
PubMed 
Article 

Google Scholar 
Merckx, V. S. F. T. et al. Evolution of endemism on a young tropical mountain. Nature 524, 347–350 (2015).ADS 
CAS 
PubMed 
Article 

Google Scholar 
Körner, C. Alpine Plant Life (Springer, 1999).Smyčka, J. et al. Reprint of: Disentangling drivers of plant endemism and diversification in the European Alps – a phylogenetic and spatially explicit approach. Perspect. Plant Ecol. Evol. Syst. 30, 31–40 (2018).Article 

Google Scholar 
Schönswetter, P., Stehlik, I., Holderegger, R. & Tribsch, A. Molecular evidence for glacial refugia of mountain plants in the European Alps. Mol. Ecol. 14, 3547–3555 (2005).PubMed 
Article 
CAS 

Google Scholar 
Haller, A. von. Enumeratio Methodica Stirpium Helvetiae indigenarum. (Officina Academica Abrami Vandenhoek, 1742).de Candolle, A. Sur les causes de l’inégale distribution des plantes rares dans la chaîne des Alpes. Atti del Congr. Internazionale Bot. Tenuto Firenze. 92–104 (1875).Boucher, F. C., Zimmermann, N. E. & Conti, E. Allopatric speciation with little niche divergence is common among alpine Primulaceae. J. Biogeogr. 43, 591–602 (2016).Article 

Google Scholar 
Schneeweiss, G. M. et al. Molecular phylogenetic analyses identify Alpine differentiation and dysploid chromosome number changes as major forces for the evolution of the European endemic Phyteuma (Campanulaceae). Mol. Phylogenet. Evol. 69, 634–652 (2013).CAS 
PubMed 
Article 

Google Scholar 
Tkach, N. et al. Molecular phylogenetics, morphology and a revised classification of the complex genus Saxifraga (Saxifragaceae). Taxon 64, 1159–1187 (2015).Article 

Google Scholar 
Favre, A. et al. Out-of-Tibet: the spatio-temporal evolution of Gentiana (Gentianaceae). J. Biogeogr. 43, 1967–1978 (2016).Article 

Google Scholar 
Kadereit, J. W., Griebeler, E. M. & Comes, H. Quaternary diversification in European alpine plants: pattern and process. Philos. Trans. R. Soc. Lond. B. Biol. Sci. 359, 265–274 (2004).PubMed 
PubMed Central 
Article 

Google Scholar 
Xing, Y. & Ree, R. H. Uplift-driven diversification in the Hengduan Mountains, a temperate biodiversity hotspot. Proc. Natl Acad. Sci. 114, E3444–E3451 (2017).CAS 
PubMed 
PubMed Central 

Google Scholar 
Lagomarsino, L. P., Condamine, F. L., Antonelli, A., Mulch, A. & Davis, C. C. The abiotic and biotic drivers of rapid diversification in Andean bellflowers (Campanulaceae). N. Phytol. 210, 1430–1442 (2016).Article 

Google Scholar 
Ding, W. N., Ree, R. H., Spicer, R. A. & Xing, Y. W. Ancient orogenic and monsoon-driven assembly of the world’s richest temperate alpine flora. Science 369, 578–581 (2020).ADS 
CAS 
PubMed 
Article 

Google Scholar 
Roquet, C., Boucher, F. C., Thuiller, W. & Lavergne, S. Replicated radiations of the alpine genus Androsace (Primulaceae) driven by range expansion and convergent key innovations. J. Biogeogr. 40, 1874–1886 (2013).PubMed 
PubMed Central 
Article 

Google Scholar 
Luebert, F. & Muller, L. A. H. Biodiversity from mountain building. Front. Genet. 6, (2015).Zachos, J. C., Dickens, G. R. & Zeebe, R. E. An early Cenozoic perspective on greenhouse warming and carbon-cycle dynamics. Nature 451, 279–283 (2008).ADS 
CAS 
PubMed 
Article 

Google Scholar 
Haffer, J. Speciation in Colombian forest birds west of the Andes. Am. Museum Novit. 2294, 1–58 (1967).Aguilée, R., Claessen, D. & Lambert, A. Adaptive radiation driven by the interplay of eco-evolutionary and landscape dynamics. Evolution 67, 1291–1306 (2013).PubMed 
Article 

Google Scholar 
Feng, G., Mao, L., Sandel, B., Swenson, N. G. & Svenning, J. C. High plant endemism in China is partially linked to reduced glacial-interglacial climate change. J. Biogeogr. 43, 145–154 (2016).Article 

Google Scholar 
Molina-Venegas, R., Aparicio, A., Lavergne, S. & Arroyo, J. Climatic and topographical correlates of plant palaeo- and neoendemism in a Mediterranean biodiversity hotspot. Ann. Bot. 119, 229–238 (2017).PubMed 
Article 

Google Scholar 
Saladin, B. et al. Rapid climate change results in long-lasting spatial homogenization of phylogenetic diversity. Nat. Commun. 11, 1–8 (2020).Article 
CAS 

Google Scholar 
Hughes, C. & Eastwood, R. Island radiation on a continental scale: exceptional rates of plant diversification after uplift of the Andes. Proc. Natl Acad. Sci. 103, 10334–10339 (2006).ADS 
CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Pouchon, C. et al. Phylogenomic analysis of the explosive adaptive radiation of the Espeletia complex (Asteraceae) in the tropical Andes. Syst. Biol. 67, 1041–1060 (2018).CAS 
PubMed 
Article 

Google Scholar 
Kadereit, J. W. The role of in situ species diversification for the evolution of high vascular plant species diversity in the European Alps—a review and interpretation of phylogenetic studies of the endemic flora of the Alps. Perspect. Plant Ecol. Evol. Syst. 26, 28–38 (2017).Article 

Google Scholar 
Escobar García, P. et al. Extensive range persistence in peripheral and interior refugia characterizes Pleistocene range dynamics in a widespread Alpine plant species (Senecio carniolicus, Asteraceae). Mol. Ecol. 21, 1255–1270 (2012).PubMed 
PubMed Central 
Article 

Google Scholar 
Lohse, K., Nicholls, J. A. & Stone, G. N. Inferring the colonization of a mountain range-refugia vs. nunatak survival in high alpine ground beetles. Mol. Ecol. 20, 394–408 (2011).CAS 
PubMed 
Article 

Google Scholar 
Stehlik, I. Resistance or emigration? Response of alpine plants to the ice ages. Taxon 52, 499–510 (2003).Article 

Google Scholar 
Schneeweiss, G. M. & Schönswetter, P. A re-appraisal of nunatak survival in arctic-alpine phylogeography. Mol. Ecol. 20, 190–192 (2011).PubMed 
Article 

Google Scholar 
Westergaard, K. B. et al. Glacial survival may matter after all: Nunatak signatures in the rare European populations of two west-arctic species. Mol. Ecol. 20, 376–393 (2011).PubMed 
Article 

Google Scholar 
Bettin, O., Cornejo, C., Edwards, P. J. & Holderegger, R. Phylogeography of the high alpine plant Senecio halleri (Asteraceae) in the European Alps: In situ glacial survival with postglacial stepwise dispersal into peripheral areas. Mol. Ecol. 16, 2517–2524 (2007).CAS 
PubMed 
Article 

Google Scholar 
Tomasello, S., Karbstein, K., Hodač, L., Paetzold, C. & Hörandl, E. Phylogenomics unravels Quaternary vicariance and allopatric speciation patterns in temperate-montane plant species: a case study on the Ranunculus auricomus species complex. Mol. Ecol. 29, 2031–2049 (2020).PubMed 
Article 

Google Scholar 
Ozenda, P. L’endémisme au niveau de l’ensemble du Système alpin. Acta Bot. Gall. 142, 753–762 (1995).Article 

Google Scholar 
Rolland, J., Lavergne, S. & Manel, S. Combining niche modelling and landscape genetics to study local adaptation: A novel approach illustrated using alpine plants. Perspect. Plant Ecol. Evol. Syst. 17, 491–499 (2015).Article 

Google Scholar 
Alvarez, N. et al. History or ecology? Substrate type as a major driver of spatial genetic structure in Alpine plants. Ecol. Lett. 12, 632–640 (2009).PubMed 
Article 

Google Scholar 
Gao, Y.-D., Gao, X.-F. & Harris, A. Species boundaries and parapatric speciation in the complex of alpine shrubs, Rosa sericea (Rosaceae), based on population genetics and ecological tolerances. Front. Plant Sci. 10, 1–16 (2019).Article 

Google Scholar 
Knox, E. B. Adaptive radiation of African montane plants. In Adaptive Speciation (eds. Dieckmann, U., Doebeli, M., Metz, J. A. J. & Tautz, D.) 345–361 (Cambridge University Press, 2004).Segar, S. T. et al. Speciation in a keystone plant genus is driven by elevation: a case study in New Guinean Ficus. J. Evol. Biol. 30, 512–523 (2017).CAS 
PubMed 
Article 

Google Scholar 
Pouchon, C. et al. Phylogenetic signatures of ecological divergence and leapfrog adaptive radiation in Espeletia. Am. J. Bot. 108, 113–128 (2021).CAS 
PubMed 
Article 

Google Scholar 
Luebert, F. & Weigend, M. Phylogenetic insights into Andean plant diversification. Front. Ecol. Evol. 2, 1–17 (2014).Article 

Google Scholar 
Nagy, L. & Grabherr, G. The Biology of Alpine Habitats (Oxford University Press, 2009).Louca, S. & Pennell, M. W. Extant timetrees are consistent with a myriad of diversification histories. Nature 580, 502–505 (2020).ADS 
CAS 
PubMed 
Article 

Google Scholar 
Goldberg, E. E., Lancaster, L. T. & Ree, R. H. Phylogenetic inference of reciprocal effects between geographic range evolution and diversification. Syst. Biol. 60, 451–465 (2011).PubMed 
Article 

Google Scholar 
Goldberg, E. E. & Igić, B. Tempo and mode in plant breeding system evolution. Evolution 66, 3701–3709 (2012).PubMed 
Article 

Google Scholar 
Gitzendanner, M., Soltis, P., Yi, T.-S., Li, D.-Z. & Soltis, D. Plastome Phylogenetics: 30 years of inferences into plant evolution. In Advances in Botanical Research 293–313 (Elsevier, 2018).Birks, H. H. The late-quaternary history of arctic and alpine plants. Plant Ecol. Divers. 1, 135–146 (2008).Article 

Google Scholar 
Mai, D. Tertiäre Vegetationsgeschichte Europas—Metoden und Ergebnisse. (Gustav Fischer Verlag, 1995).Svenning, J. C. Deterministic Plio-Pleistocene extinctions in the European cool-temperate tree flora. Ecol. Lett. 6, 646–653 (2003).Article 

Google Scholar 
Fauquette, S. et al. The Alps: a geological, climatic and human perspective on vegetation history and modern plant diversity. In Mountains, Climate and Biodiversity (eds. Hoorn, C., Perrigo, A. & Antonelli, A.) 413 (Wiley-Blackwell, 2018).Mráz, P. et al. Vascular plant endemism in the Western Carpathians: spatial patterns, environmental correlates and taxon traits. Biol. J. Linn. Soc. 119, 630–648 (2016).Article 

Google Scholar 
Puşcaş, M. et al. Post-glacial history of the dominant alpine sedge Carex curvula in the European Alpine System inferred from nuclear and chloroplast markers. Mol. Ecol. 17, 2417–2429 (2008).PubMed 
Article 
CAS 

Google Scholar 
Puşcaş, M., Taberlet, P. & Choler, P. No positive correlation between species and genetic diversity in European alpine grasslands dominated by Carex curvula. Divers. Distrib. 14, 852–861 (2008).Article 

Google Scholar 
Magyari, E. K. et al. Late Pleniglacial vegetation in eastern-central Europe: are there modern analogues in Siberia? Quat. Sci. Rev. 95, 60–79 (2014).ADS 
Article 

Google Scholar 
Prodon, R., Thibault, J. C. & Dejaifve, P. A. Expansion vs. compression of bird altitudinal ranges on a Mediterranean island. Ecology 83, 1294–1306 (2002).Article 

Google Scholar 
Moen, D. & Morlon, H. Why does diversification slow down? Trends Ecol. Evol. 29, 190–197 (2014).PubMed 
Article 

Google Scholar 
Aguilée, R., Gascuel, F., Lambert, A. & Ferriere, R. Clade diversification dynamics and the biotic and abiotic controls of speciation and extinction rates. Nat. Commun. 9, 1–13 (2018).Article 
CAS 

Google Scholar 
Vargas, P. Molecular evidence for multiple diversification patterns of alpine plants in Mediterranean Europe. Taxon 52, 463–476 (2003).Article 

Google Scholar 
Kruckeberg, A. R. An essay: the stimulus of unusual geologies for plant speciation. Syst. Bot. 11, 455–463 (1986).Article 

Google Scholar 
Cowling, R. M. & Holmes, P. M. Endemism and speciation in a lowland flora from the Cape Floristic Region. Biol. J. Linn. Soc. 47, 367–383 (1992).Article 

Google Scholar 
Lexer, C. et al. Genomics of the divergence continuum in an African plant biodiversity hotspot, I: drivers of population divergence in Restio capensis (Restionaceae). Mol. Ecol. 23, 4373–4386 (2014).CAS 
PubMed 
Article 

Google Scholar 
Anacker, B. L. & Strauss, S. Y. The geography and ecology of plant speciation: range overlap and niche divergence in sister species. Proc. R. Soc. B Biol. Sci. 281, 20132980 (2014).Article 

Google Scholar 
Moore, A. J. & Kadereit, J. W. The evolution of substrate differentiation in Minuartia series Laricifoliae (Caryophyllaceae) in the European Alps: in situ origin or repeated colonization? Am. J. Bot. 100, 2412–2425 (2013).PubMed 
Article 

Google Scholar 
Guggisberg, A. et al. The genomic basis of adaptation to calcareous and siliceous soils in Arabidopsis lyrata. Mol. Ecol. 27, 5088–5103 (2018).PubMed 
Article 

Google Scholar 
Gigon, A. Vergleich alpiner Rasen auf Silikat- und auf Karbonatboden—Konkurrenz—und Stickstofformenversuche sowie standortskundliche Untersuchungen im Nardetum und im Seslerietum bei Davos. (ETH Zuerich, 1971).Davies, M. S. & Snaydon, R. W. Physiological differences among populations of Anthoxanthum odoratum L. collected from the park grass experiment, Rothamsted. I. Response to calcium. J. Appl. Ecol. 10, 33–45 (1973).Article 

Google Scholar 
Snaydon, R. W. Rapid population differentiation in mosaic environment. I. The response of Anthoxantum odoratum populations to soils. Evolution 24, 257–269 (1970).CAS 
PubMed 
Article 

Google Scholar 
Zohlen, A. & Tyler, G. Soluble inorganic tissue phosphorus and calcicole-calcifuge behaviour of plants. Ann. Bot. 94, 427–432 (2004).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Kassen, R., Llewellyn, M. & Rainey, P. B. Ecological contraints on diversification in a model adaptive radiation. Nature 431, 984–988 (2004).ADS 
CAS 
PubMed 
Article 

Google Scholar 
MacLean, R. C., Bell, G. & Rainey, P. B. The evolution of a pleiotropic fitness tradeoff in Pseudomonas fluorescens. Proc. Natl Acad. Sci. USA 101, 8072–8077 (2004).ADS 
CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Rabosky, D. L. & Goldberg, E. E. Model inadequacy and mistaken inferences of trait-dependent speciation. Syst. Biol. 64, 340–355 (2015).CAS 
PubMed 
Article 

Google Scholar 
Kolář, F. et al. Northern glacial refugia and altitudinal niche divergence shape genome-wide differentiation in the emerging plant model Arabidopsis arenosa. Mol. Ecol. 25, 3929–3949 (2016).PubMed 
Article 

Google Scholar 
Dentant, C. & Lavergne, S. Plantes de haute montagne: état des lieux, évolution et analyse diachronique dans le massif des Écrins (France). Bull. Soc. linn. Provence 64, 83–98 (2013).
Google Scholar 
Dentant, C. The highest vascular plants on Earth. Alp. Bot. 128, 97–106 (2018).Article 

Google Scholar 
Boucher, F. C. et al. Reconstructing the origins of high‐alpine niches and cushion life form in the genus Androsace sl (Primulaceae). Evolution 66, 1255–1268 (2012).PubMed 
Article 

Google Scholar 
Boucher, F. C., Lavergne, S., Basile, M., Choler, P. & Aubert, S. Evolution and biogeography of the cushion life form in angiosperms. Perspect. Plant Ecol. Evol. Syst. 20, 22–31 (2016).Article 

Google Scholar 
Schönswetter, P. & Schneeweiss, G. M. Is the incidence of survival in interior Pleistocene refugia (nunataks) underestimated? Phylogeography of the high mountain plant Androsace alpina (Primulaceae) in the European Alps revisited. Ecol. Evol. 9, 4078–4086 (2019).PubMed 
PubMed Central 
Article 

Google Scholar 
Aeschimann, D., Rasolofo, N. & Theurillat, J. P. Analyse de la flore des Alpes. 2: Diversité et chorologie. Candollea 66, 225–253 (2011).Article 

Google Scholar 
Ebersbach, J. et al. In and out of the Qinghai-Tibet Plateau: divergence time estimation and historical biogeography of the large arctic-alpine genus Saxifraga L. J. Biogeogr. 44, 900–910 (2017).Article 

Google Scholar 
Hannon, G. FASTX. http://hannonlab.cshl.edu/fastx_toolkit/ (2014).Coissac, E. The ORGanelle ASseMbler 1.0.3. https://git.metabarcoding.org/org-asm/org-asm/wikis/home (2016).Shaw, J. et al. Chloroplast DNA sequence utility for the lowest phylogenetic and phylogeographic inferences in angiosperms: the tortoise and the hare IV. Am. J. Bot. 101, 1987–2004 (2014).PubMed 
Article 

Google Scholar 
Mansion, G. et al. How to handle speciose clades? Mass taxon-sampling as a strategy towards illuminating the natural history of Campanula (Campanuloideae). PLoS ONE 7, e50076 (2012).Rossi, M. Taxonomy, phylogeny and reproductive ecology of Gentiana lutea L (University in Bologna, 2011).Hämmerli, M. Molecular Aspects in Systematics of Gentiana Sect. Calathianae Froel (Université de Neuchâtel, 2007).Hungerer, K. B. & Kadereit, J. W. The phylogeny and biogeography of Gentiana L. sect. Ciminalis (Adans.) Dumort.: A historical interpretation of distribution ranges in the European high mountains. Perspect. Plant Ecol. Evol. Syst. 1, 121–135 (1998).Article 

Google Scholar 
Ranwez, V., Harispe, S., Delsuc, F. & Douzery, E. J. P. MACSE: Multiple alignment of coding SEquences accounting for frameshifts and stop codons. PLoS One 6, e22594 (2011).Castresana, J. Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis. Mol. Biol. Evol. 17, 540–552 (2000).CAS 
PubMed 
Article 

Google Scholar 
Kück, P. & Meusemann, K. FASconCAT: convenient handling of data matrices. Mol. Phylogenet. Evol. 56, 1115–1118 (2010).PubMed 
Article 
CAS 

Google Scholar 
Katoh, K., Kuma, K. I., Toh, H. & Miyata, T. MAFFT version 5: Improvement in accuracy of multiple sequence alignment. Nucleic Acids Res. 33, 511–518 (2005).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Bouckaert, R. et al. BEAST 2: a software platform for Bayesian evolutionary analysis. PLoS Comput. Biol. 10, 1–6 (2014).Article 
CAS 

Google Scholar 
Bouckaert, R. R. & Drummond, A. J. bModelTest: Bayesian phylogenetic site model averaging and model comparison. BMC Evol. Biol. 17, 1–11 (2017).Article 

Google Scholar 
Morlon, H. Phylogenetic approaches for studying diversification. Ecol. Lett. 17, 508–525 (2014).PubMed 
Article 

Google Scholar 
Aeschimann, D., Lauber, K., Moser, D. M. & Theurillat, J. P. Flora Alpina (Editions Belin, 2004).Castroviejo, S. Flora Iberica (Real Jardin Botanico CSIC, 2012).Goliášová, K. & Michalková, E. Flóra Slovenska (Vydavateľstvo Slovenskej akadémie vied, 2012).Speta, E. & Rákosy, L. Wildpflanzen Siebenbürgen (Naturhistorisches Museum Wien, 2010).Sarić, M. Flora Srbije (Srpska akademija nauka i umetnosti, 1992).Schönswetter, P. & Schneeweiss, G. M. Androsace komovensis sp. nov., a long mistaken local endemic from the southern Balkan Peninsula with biogeographic links to the Eastern Alps. Taxon 58, 544–549 (2009).Article 

Google Scholar 
Schönswetter, P., Magauer, M. & Schneeweiss, G. M. Androsace halleri subsp. nuria Schönsw. & Schneew. (Primulaceae), a new taxon from the eastern Pyrenees (Spain, France). Phytotaxa 201, 227–232 (2015).Article 

Google Scholar 
Schneeweiss, G. M. & Schonswetter, P. The wide but disjunct range of the European mountain plant Androsace lactea L. (Primulaceae) reflects Late Pleistocene range fragmentation and post-glacial distributional stasis. J. Biogeogr. 37, 2016–2025 (2010).
Google Scholar 
Webb, D. A. & Gornall, R. J. Saxifrages of Europe (Timber Press, 1989).GBIF. https://www.gbif.org/ (2018).Körner, C. et al. A global inventory of mountains for bio-geographical applications. Alp. Bot. 127, 1–15 (2017).Article 

Google Scholar 
Anacker, B. L., Whittall, J. B., Goldberg, E. E. & Harrison, S. P. Origins and consequences of serpentine endemism in the California flora. Evolution 65, 365–376 (2011).PubMed 
Article 

Google Scholar 
Morlon, H. et al. RPANDA: An R package for macroevolutionary analyses on phylogenetic trees. Methods Ecol. Evol. 7, 589–597 (2016).Article 

Google Scholar 
Burnham, K. & Anderson, D. Model Selection and Multimodel Inference (Springer, 2002).Fitzjohn, R. G., Maddison, W. P. & Otto, S. P. Estimating trait-dependent speciation and extinction rates from incompletely resolved phylogenies. Syst. Biol. 58, 595–611 (2009).PubMed 
Article 

Google Scholar 
O’Meara, B. C. & Beaulieu, J. M. Past, future, and present of state-dependent models of diversification. Am. J. Bot. 103, 792–795 (2016).PubMed 
Article 

Google Scholar 
Beaulieu, J. M. & O’Meara, B. C. Detecting hidden diversification shifts in models of trait-dependent speciation and extinction. Syst. Biol. 65, 583–601 (2016).PubMed 
Article 

Google Scholar 
Herrera-Alsina, L., Van Els, P. & Etienne, R. S. Detecting the dependence of diversification on multiple traits from phylogenetic trees and trait data. Syst. Biol. 68, 317–328 (2019).PubMed 
Article 

Google Scholar 
Onstein, R. E. et al. To adapt or go extinct? The fate of megafaunal palm fruits under past global change. Proc. R. Soc. B Biol. Sci. 285, (2018).Rabosky, D. L. & Goldberg, E. E. FiSSE: a simple nonparametric test for the effects of a binary character on lineage diversification rates. Evolution 71, 1432–1442 (2017).PubMed 
Article 

Google Scholar 
Holland, B. R., Ketelaar-Jones, S., O’Mara, A. R., Woodhams, M. D. & Jordan, G. J. Accuracy of ancestral state reconstruction for non-neutral traits. Sci. Rep. 10, 1–10 (2020).Article 
CAS 

Google Scholar 
Ree, R. H. & Sanmartín, I. Conceptual and statistical problems with the DEC+J model of founder-event speciation and its comparison with DEC via model selection. J. Biogeogr. 45, 741–749 (2018).Article 

Google Scholar 
Schoener, T. W. Nonsynchronous spatial overlap of lizards in patchy habitats. Ecology 51, 408–418 (1970).Article 

Google Scholar 
Zhang, J. spaa: SPecies Association Analysis 0.2.2. https://cran.r-project.org/package=spaa (2016).Smyčka, J. Tempo and drivers of plant diversification in the European mountain system. multidiv, https://doi.org/10.5281/zenodo.6341727 (2022). More

Spatial standardisation workflowTo produce spatially-standardised fossil occurrence datasets which remain geographically consistent through time, we designed a subsampling algorithm which enforces consistent spatial distribution of occurrence data between time bins, while maximising data retention and permitting highly flexible regionalisation (Fig. 8). Our method was developed in light of, and takes some inspiration from, the spatial standardisation procedure of Close et al.2. This method provides, within a given time bin, subsamples of occurrence data with threshold MST lengths. An average diversity estimate can be taken from this ‘forest’ of MSTs, selecting only those of a target tree length to ensure spatially-standardised measurements. It does not produce a single dataset across time bins, however; rather a series of discontinuous, bin-specific datasets which cannot then simply be concatenated as the spatial extents of each bin-specific forest are not standardised (despite each individual MST being so), even when MSTs are assigned to a specific geographic region, e.g. a continent or to a particular latitudinal band. This prevents estimation of rates, because such analyses require datasets that span multiple time bins and remain geographically consistent and spatially standardised through the time span of interest. This is the shortcoming that our method overcomes. The workflow consists of three main steps.Fig. 8: Component steps of our spatial standardisation workflow.A A spatial window (dotted lines) is used to demarcate the spatial region of interest, which may shift in a regular fashion through time to track that region. Data captured in each window is clipped to a target longitude–latitude range (orange lines). B The data forming the longitude–latitude extent is marked, then masked from further subsampling. C Data are binned using a hexagonal grid, the tally of occurrences in each grid cell taken, and a minimum spanning tree constructed from the grid cell centres. D The cells with the smallest amount of data are iteratively removed from the minimum spanning tree until a target tree length is reached.Full size image1. First, the user demarcates a spatially discrete geographic area (herein the spatial window) and a series of time bins into which fossil occurrence data is subdivided. Occurrence data falling outside the window in each time bin are dropped from the dataset, leaving a spatially restricted subsample (Fig. 8A). Spatial polygon demarcation is a compromise between the spatial availability of data to subsample and the region of interest to the user but allows creation of a dataset where regional nuances of biodiversity may be targeted. Careful choice of window extent can even aid subsequent steps by targeting regions that have a consistently sampled fossil record through time, even if the extent of that record fluctuates. To account for spatially non-random changes in the spatial distribution of occurrence data arising from the interlinked effects of continental drift, preservation potential and habitat distribution17, the spatial polygon may slide to track the location of the available sampling data through time. This drift is performed with two conditions. First, the drift is unidirectional so that the sampling of data remains consistent relative to global geography, rather than allowing the window to hop across the globe solely according to data availability and without biogeographic context. Second, spatial window translation is performed in projected coordinates so that its sampling area remains near constant between time bins, avoiding changes in spatial window area that could induce sampling bias from the species-area effect.2. Next, subsampling routines are applied to the data to standardise its spatial extent to a common threshold across all time bins using two metrics: the length of the MST required to connect the locations of the occurrences; and the longitude–latitude extent of the occurrences. MST length has been shown to measure spatial sampling robustly as it captures not just the absolute extent of the data but also the intervening density of points, and so is highly correlated with multiple other geographic metrics16. MSTs with different aspect ratios may show similar total lengths but could sample over very different spatial extents, inducing a bias by uneven sampling across spatially organised diversity gradients16; standardising longitude–latitude extent accounts for this possibility. The standardisation methods can be applied individually or serially if both MST length and longitude–latitude range show substantial fluctuations through time. Data loss is inevitable during subsampling and may risk degrading the signals of origination, extinction and preservation. To address this issue, subsampling is performed to retain the greatest amount of data possible. During longitude–latitude standardisation, the range containing the greatest amount of data is preserved. During MST standardisation, occurrences are spatially binned using a hexagonal grid to reduce computational burden and to permit assessment of spatial density (Fig. 8B). The grid cells containing the occurrences that define the longitude–latitude extent of the data are first masked from the subsampling procedure so that this property of the dataset is unaffected, and then the occurrences within the grid cells at the tips of the MST are tabulated. Tip cells with the least data are iteratively removed (removal of non-tip cells may have little to no effect on the tree topology) until the target MST length is achieved (Fig. 8D), with tree length iteratively re-calculated to include the branch lengths added by the masked grid cells.For both methods, the solution with the smallest difference to the target is selected and so both metrics may fluctuate around this target from bin to bin, with the degree of fluctuation depending upon the availability of data to exclude—larger regions that capture more data are more amenable to the procedure than smaller regions. Similarly, the serial application of both metrics reduces the pool of data available to the second method, although longitude–latitude standardisation is always applied first in the serial case so that the resultant extent will be retained during MST standardisation. Consequently, the choice of standardisation procedure and thresholds must be tailored to the availability and extent of data within the sampling region through time, along with the resulting degree of data loss. This places further emphasis on the careful construction of the spatial window in the first step. Threshold choice is also a compromise between data loss and consistency of standardisation across the dataset and so it may be necessary to choose targets that standardise spatial extent well for the majority of the temporal range of a dataset, rather than imposing a threshold that spans the entire data range but causes unacceptable data loss in some bins.3. Once the time-binned, geographically restricted data have been spatially standardised, the relationship between diversity and spatial extent is scrutinised. After standardisation, it is expected that residual fluctuations in spatial extent should induce little or no change in apparent diversity. Bias arising from temporal variation in sampling intensity may still be present, so diversity is calculated using coverage-based rarefaction (also referred to as shareholder quorum subsampling13,62,63), with a consistent coverage quorum from bin to bin. While coverage-based rarefaction has known biases, it remains the most accurate non-probabilistic means of estimating fossil diversity14. As such, we consider it the most appropriate method to assess the diversity of a region-level fossil dataset. The residual fluctuations in spatial extent may then be tested for correlation with spatially standardised, temporally corrected diversity. If a significant relationship is found, then the user must go back and alter the standardisation parameters, including the spatial window geometry and drift, the longitude–latitude threshold, and the MST threshold. Otherwise, the dataset is considered suitable for further analysis.We implement our subsample standardisation workflow in R with a custom algorithm, spacetimestand, along with a helper function spacetimewind to aid the initial construction of spatial window. spacetimestand can then accept any fossil occurrence data with temporal constraints in millions of years before present and longitude–latitude coordinates in decimal degrees. Spatial polygon construction and binning is handled using the sp library64, MST manipulation using the igraph and ape libraries65,66, spatial metric calculation using the sp, geosphere and GeoRange libraries67,68, hexagonal gridding using the icosa library69, and diversity calculation by coverage-based rarefaction using the estimateD function from the iNEXT library70. Next, we apply our algorithm to marine fossil occurrence data from the Late Permian to Early Triassic.Data acquisition and cleaningFossil occurrence data for the Late Permian (260 Ma) to Early Jurassic (190 Ma) were downloaded from the PBDB on 28/04/21 with the default major overlap setting applied (an occurrence is treated as within the requested time span if 50% or more of its stratigraphic duration intersects with that time span), in order to minimise edge effects resulting from incomplete sampling of taxon ranges within our study interval of interest (the Permo-Triassic to Triassic-Jurassic boundaries). Other filters in the PBDB API were not applied during data download to minimise the risk of data exclusion. Occurrences from terrestrial facies were excluded, along with plant, terrestrial-freshwater invertebrate and terrestrial tetrapod occurrences (as these may still occur in marine deposits from transport) and occurrences from several minor and poorly represented phyla. Finally, non-genus level occurrences were removed, leaving 104,741 occurrences out of the original 168,124. Based on previous findings2, siliceous occurrences were not removed from the dataset, despite their variable preservation potential compared to calcareous fossils. To increase the temporal precision of the dataset, occurrences with stratigraphic information present were revised to substage- or stage-level precision using a stratigraphic database compiled from the primary literature. To increase the spatial and taxonomic coverage of the dataset, the PBDB data were supplemented by an independently compiled genus-level database of Late Permian to Late Triassic marine fossil occurrences36. Prior to merging, occurrences from the same minor phyla were excluded, along with a small number lacking modern coordinate data, leaving 47,661 occurrences out of an original 51,054. Absolute numerical first appearance and last appearance data (FADs and LADs) were then assigned to the occurrences from their first and last stratigraphic intervals, based on the ages given in A Geologic Timescale 202071. Palaeocoordinates were calculated from the occurrence modern-day coordinates and midpoint ages using the Getech plate rotation model. Finally, occurrences with a temporal uncertainty greater than 10 million years and occurrences for which palaeocoordinate reconstruction was not possible were removed from the composite dataset, leaving 145,701 occurrences out of the original 152,402.In the total dataset, we note that the age uncertainty for occurrences is typically well below their parent stage duration, aside for the Wuchiapingian and Rhaetian where the mean and quartile ages are effectively the same as the stage length (Fig. S44). This highlights the chronostratigraphic quality of our composite dataset, particularly for the Norian stage (~18-million-year duration) which has traditionally been an extremely coarse and poorly resolved interval in Triassic-aged macroevolutionary analyses. Taxonomically, most occurrences are molluscs (Fig. 8), which is unsurprising given the abundance of ammonites, gastropods and bivalves in the PBDB, but introduces the caveat that downstream results will be driven primarily by these clades. Foraminiferal and radiolarian occurrences together comprise the next most abundant element of the composite dataset, demonstrating that we nonetheless achieve good coverage of both the macrofossil and microfossil records, along with broad taxonomic coverage in the former despite the preponderance of molluscs.Spatiotemporal standardisationWe chose a largely stage-level binning scheme when applying our spatial standardisation procedure for several reasons. First, the volume of data in each bin is greater than in a substage bin, providing a more stable view of occurrence distributions through time and increasing the availability of data for subsampling. Spatial variation at substage level might still affect the sampling of diversity, but the main goal of this study is to analyse origination and extinction rates where taxonomic ranges are key rather than pointwise taxonomic observations. Consequently, substage level variation in taxon presences likely amounts to noise when examining taxonomic ranges, making stage-level bins preferable in order maximise signal.During exploratory standardisation trials, we found a large crash in diversity and spatial sampling extent during the Hettangian (201.3–199.3 Mya). No significant relationships with spatially-standardised diversity were found when the Hettangian bin was excluded from correlation tests, indicating its disproportionate effect in otherwise well-standardised time series. Standardising the data to the level present in the Hettangian would have resulted in unacceptable data loss so we instead accounted for this issue by merging the Hettangian bin with the succeeding Sinemurian bin, where sampling returns to spatial extents consistent with older intervals. While this highlights a limitation of our method, as the Hettangian is  2: positive support, log BF  > 6: strong support)85 using the -plotRJ function of PyRate.Probabilistic diversity estimationTraditional methods of estimating diversity do not directly address uneven sampling arising from variation in preservation, collection and description rates, and their effectiveness is highly dependent on the structure of the dataset. We present an alternative method to infer corrected diversity trajectories based on the sampled occurrences and on the preservation rates through time and across lineages as inferred by PyRate, which we term mcmcDivE. The method implements a hierarchical Bayesian model to estimate corrected diversity across arbitrarily defined time bins. The method estimates two classes of parameters: the number of unobserved species for each time bin and a parameter quantifying the volatility of the diversity trajectory.We assume the sampled number of taxa (i.e. the number of fossil taxa, here indicated with xt) in a time bin to be a random subset of an unknown total taxon pool, which we indicate with Dt. The goal of mcmcDivE is to estimate the true diversity trajectory ({{{{{bf{D}}}}}},=,left{{D}_{1},{D}_{2},ldots ,{D}_{t}right}), of which the vector of sampled diversity ({{{{{bf{x}}}}}},=,{{x}_{1},{x}_{2},ldots ,{x}_{t}}) is a subset. The sampled diversity is modelled as a random sample from a binomial distribution86 with sampling probability pt:$${x}_{t}, ,sim, {{{{{rm{Bin}}}}}}({D}_{t},{p}_{t})$$
(1)
We obtain the sampling probability from the preservation rate (qt) estimated in the initial PyRate analysis. If the PyRate model assumes no variation across lineages the sampling probability based on a Poisson process is ({p}_{t},=,1,-,{{{{{rm{exp }}}}}}({-q}_{t},times, {delta }_{t})), where δt is the duration of the time bin. When using a Gamma model in PyRate, however, the qt parameter represents the mean rate across lineages at time t and the rate is heterogeneous across lineages based on a gamma distribution with shape and rate parameters equal to an estimated value α.To account for rate heterogeneity across lineages in mcmcDivE, we draw an arbitrarily large vector of gamma-distributed rate multipliers g1, …, gR ~ Γ(α,α) and compute the mean probability of sampling in a time bin as:$${p}_{t},=,frac{1}{R}mathop{sum }limits_{i,=,R}^{R}1,-,{{{{{rm{exp }}}}}}(-{q}_{t},,times, {g}_{i},times, {delta }_{t})$$
(2)
We note that while qt quantifies the mean preservation rate in PyRate (i.e. averaged among taxa in a time bin t), the mean sampling probability pt will be lower than (1,-,{{{{{rm{exp }}}}}}({-q}_{t},times, {delta }_{t})) (i.e. the probability expected under a constant preservation rate equal to qt) especially for high levels of rate heterogeneity, due to the asymmetry of the gamma distribution and the non-linear relationship between rates and probabilities. We sample the corrected diversity from its posterior through MCMC. The likelihood of the sampled number of taxa is computed as the probability mass function of a binomial distribution with Di as the ‘number of trials’ and pi as the ‘success probability’. To account for the expected temporal autocorrelation of a diversity trajectory87, we use a Brownian process as a prior on the log-transformed diversity trajectory through time. Under this model, the prior probability of Dt is:$$Pleft({{{{{rm{log }}}}}}left({D}_{t}right)right),{{{{{mathscr{ sim }}}}}},{{{{{mathscr{N}}}}}}({{{{{rm{log }}}}}}left({D}_{t,-,1}right),,sqrt{{sigma }^{2},,times, ,{delta }_{t}})$$
(3)
where σ2 is the variance of the Brownian process. For the first time bin in the series, Dt = 0, we use a vague prior ({{{{{mathscr{U}}}}}}(0,infty )). Because the variance of the process is itself unknown and may vary among clades as a function of their diversification history, we assign it an exponential hyperprior Exp(1) and estimate it using MCMC. Thus, the full posterior of the mcmcDivE model is:$$underbrace{P(D,{sigma }^{2}|x,q,alpha )}_{{{{{rm{posterior}}}}}}propto underbrace{P(x|D,q,alpha )}_{{{{{rm{likelihood}}}}}}times underbrace{P(D|{sigma }^{2})}_{{{{{rm{prior}}}}}}times underbrace{P({sigma }^{2})}_{{{{{rm{hyperprior}}}}}}$$
(4)
where ({{{{{bf{D}}}}}},=,{{D}_{0},{D}_{1},ldots ,{D}_{t}}) and ({{{{{bf{q}}}}}},=,{{q}_{0},{q}_{1},ldots ,{q}_{t}}) are vectors of estimated diversity, sampled diversity, and preservation rates for each of T time bins. We estimate the parameters D and σ2 using MCMC to obtain samples from their joint posterior distribution. To incorporate uncertainties in q and α we randomly resample them during the MCMC from their posterior distributions obtained from PyRate analyses of the fossil occurrence data. While in mcmcDivE we use a posterior sample of qt and α precomputed in PyRate for computational tractability of the problem, a joint estimation of all PyRate and mcmcDivE parameters is in principle possible, particularly for smaller datasets. mcmcDivE is implemented in Python v.3 and is available as part of the PyRate software package.Simulated and empirical diversity analysesWe assessed the performance of the mcmcDivE method using 600 simulated datasets obtained under different birth-death processes and preservation scenarios. The settings of the six simulations (A–F) are summarised in Table S65 and we simulated 100 datasets from each setting. Since the birth-death process is stochastic and can generate a wide range of outcomes, we only accepted simulations with 100 to 500 species, although the resulting number of sampled species decreased after simulating the preservation process. From each birth-death simulation we sampled fossil occurrences based on a heterogeneous preservation process. Each simulation included six different preservation rates which were drawn randomly within the boundaries 0.25 and 2.5, with rate shifts set to 23, 15, 8, 5.3 and 2.6 Ma. To ensure that most rates were small (i.e. reflecting poor sampling), we randomly sampled preservation rates as:$$q, sim ,exp left({{{{{mathscr{U}}}}}}left(log left(0.25right),,log left(2.5right)right)right)$$
(5)
In two of the five scenarios (D, F), we included strong rate heterogeneity across lineages (additionally to the rate variation through time), by assuming that preservation rates followed a gamma distribution with shape and rate parameters set to 0.5. This indicates that if the mean preservation rate in a time bin was 1, the preservation rate varied across lineages between More

Study areaWith an area of about 2.33 × 105 km2, the Western Sichuan Plateau (27.11°–34.31°N and 97.36°–104.62°E) is located in the transition zone between the Qinghai-Tibet Plateau and the Sichuan Basin, including all of Garze Prefecture and Aba Prefecture, and parts of Liangshan Yi Autonomous Prefecture28 (Fig. 1). With an altitude of 780–7556 m, this area is dominated by mountain and ravine areas and high mountain and plateau areas, and the terrain is high in the west and low in the east. The climate belongs to the subtropical plateau monsoon climate, with large temperature difference between day and night and abundant sunshine. The annual average temperature is about 9.01–10.5 °C, and the precipitation is about 556.8–730 mm28. The study area is rich in water resources, including the Yalong River, Minjiang River and other important river systems in the upper reaches of the Yangtze River, and the Baihe River, Heihe river and other river systems of the Yellow River. The main types of soil are plateau meadow soil dark brown soil, brown soil, cold frozen soil and cinnamon soil, and the main vegetation types are alpine meadow and scrub. With rich and diverse soil vegetation types and distinctive vertical zonal distribution characteristics, it is one of the global biodiversity conservation hotspots29.Figure 1Location of the study area. The map is created in the support of ArcGIS 10.2 (ESRI). The China map and Western Sichuan Plateau boundary data were collected from Resources and Environmental Science and Data Center (http://www.resdc.cn/). The Qinghai-Tibetan Plateau boundary data were collected from the Global Change Research Data Publishing & Repository (http://www.geodoi.ac.cn/WebCn/Default.aspx).Full size imageData source and processingMultisource archival data were used in this study (Table 1). The land use remote sensing monitoring data, administrative boundary data and geological disaster vector data were obtained from Resources and Environmental Science and Data Center. The spatial resolution of land use remote sensing monitoring data is 30 × 30 m, including 6 first-level classification and 26s-level classification. The first-level classification includes cropland, woodland, grassland, water body, built-up land, and unused land. The accuracy of remote sensing classification is not less than 95% for cropland and built-up land, not less than 90% for grassland, woodland, and water body, and not less than 85% for unused land, which meets the need of the research. Landsat remote sensing monitoring data is used as the main information resources, among which Landsat-TM/ETM remote sensing monitoring data is used in 2000, 2005, 2010 and Landsat 8 remote sensing monitoring data is used in 2015 and 2020. In light of actual conditions and the implementation of policies and philosophies including the natural forest protection project, return of farmland to forest, land remediation, ecological civilization, the period from 2000 to 2020 is selected as the study period, and the land use data of each period is cropped using ArcGIS 10.2 to reclassify the 26 secondary classifications into cropland, woodland, grassland, water body, built-up land and unused land.Table 1 Characteristics of data used for the study.Full size tableThe DEM data were obtained from SRTM (Shuttle Radar Topography Mission) of Resources and Environmental Science and Data Center, the spatial resolution of 30 × 30 m, absolute horizontal accuracy ± 20 m, absolute elevation accuracy ± 16 m, elevation and slope are extracted from the downloaded DEM. The Qinghai-Tibetan Plateau boundary data were collected from the Global Change Research Data Publishing & Repository. Data of carbon density of different land types were obtained from Chinese Ecosystem Research Network Data Center (http://www.nesdc.org.cn/).A total of 29,284 evaluation units were collected for spatial grid processing of the Western Sichuan Plateau according to 3 km × 3 km by ArcGIS 10.2. The impact factors obtained in this study include grid data per kilometer of GDP spatial distribution, grid data per kilometer of population spatial distribution, annual mean temperature spatial interpolation data, annual mean rainfall spatial interpolation data, long-term normalized difference vegetation index (NDVI) comes from Resources and Environmental Science and Data Center with a resolution of 1 km × 1 km. The Human Active Index (HAI), with a resolution of 30 m × 30 m, can be calculated by formula30,31, and the factors are discretized into the data type required for the geodetector by the natural breakpoint method.MethodsThe InVEST modelThe InVEST model was developed by Stanford University, the University of Minnesota, the Nature Conservancy and the World Wide Fund for Nature (WWF). The model’s terrestrial ecosystem services assessment includes four modules: soil conservation, water retention, carbon storage and biodiversity assessment, and provides an overall measurement of regional ecosystem services32. The carbon storage model of the InVEST model divides the carbon storage of the ecosystem into 4 basic carbon pools, namely above-ground carbon, underground carbon, soil carbon, dead organic matter carbon7.The calculation formula of total carbon storage in the Western Sichuan Plateau is as follows7:$$C_{total} = C_{above} + C_{below} + C_{soil} + C_{dead}$$
(1)
In formula (1), Ctotal is the total carbon storage; Cabove is the above-ground carbon storage; Cbelow is the underground carbon storage; Csoil is the soil carbon storage, and Cdead is the dead organic matter carbon storage.Based on the carbon density and land use data of different land use type, the carbon storage of each land use type in the Western Sichuan Plateau is calculated by the formula7:$$C_{{text{total}}i} = (C_{{text{above}}i} + C_{{text{below}}i} + C_{{text{soil}}i} + C_{{text{dead}}i}) times A_{i}$$
(2)
In formula (2), i is the average carbon density of each land use, and Ai is the area of this land used.The carbon density data of different land use types in this study were obtained from the shared date of the National Ecological Science Data Center and some documents33,34,35,36,37. Since the carbon density data were collected from the results of studies in different parts of China, the selected documents should be close to or similar to the study area as far as possible to avoid excessive data gap. At the same time, the carbon density varies with climate, soil properties and land use38, so the carbon density should be modified according to the climate characteristics and land use types of the Western Sichuan Plateau. Existing research results show that the carbon density is positively correlated with annual precipitation and weakly correlated with annual average temperature. The quantitative expression of the relationship between carbon density and temperature and precipitation is as follows39,40,41,42:$$C_{SP} = 3.3968 times P + 3996.1;;left( {{text{R}}^{{2}} = 0.{11}} right)$$
(3)
$$C_{BP} = 6.7981e^{0.00541p};;;left( {{text{R}}^{{2}} = 0.{7}0} right)$$
(4)
$$C_{BT} = 28 times {text{T}} + 398;;left( {{text{R}}^{{2}} = 0.{47,};{text{P}} < 0.0{1}} right)$$ (5) In these formula, CSP is the soil carbon density (kg m−2) based on the annual precipitation; CBP is the biomass carbon density (kg m−2) based on the annual precipitation; CBT is the biomass carbon density (kg m−2) based on annual average temperature; P is the average annual precipitation (mm), and T is the annual average temperature (°C). According to the data of China Meteorological Data Service Centre (http://data.cma.cn/), in the past 20 years, the average annual temperature of China and the Western Sichuan Plateau was 9.0 °C and 6.3 °C, and the average annual precipitation was 643.50 mm and 812.65 mm respectively.The modified formula of carbon density in the Western Sichuan Plateau is as follows7:$$K_{BP} = frac{C^{prime}{_{BP}}}{{C^{primeprime}{_{BP}}}}$$ (6) $$K_{BT} = frac{C^{prime}{_{BT}}}{{C^{primeprime}{_{BT}}}}$$ (7) $$C_{BT} = 28 times T + 398;;left( {{text{R}}^{{2}} = 0.{47,};{text{P}} < 0.0{1}} right)$$ (8) $$K_{S} = frac{C^{prime}{_{SP}}}{{C^{primeprime}{_{SP}}}}$$ (9) In these formula, KBP is the modified indices of precipitation factor in biomass carbon density; KBT is the modified indices of temperature factor; C'BP and C''BP are the biomass carbon density obtained from annual precipitation in the Western Sichuan Plateau and the whole country respectively. C'BT and C''BT are the biomass carbon density obtained from annual average temperature; C'SP and C''SP are the soil carbon density data obtained from annual average temperature; KB and KS are the biomass carbon density modified indices and soil carbon density modified indices respectively. The carbon density values of each land use type after modified in the Western Sichuan Plateau are shown in Table 2.Table 2 Carbon density values of different land use types in the Western Sichuan plateau (t hm−2).Full size tableExploratory spatial analysis methodGlobal spatial autocorrelationGlobal Moran’s I was used to describe the spatial differentiation characteristics of carbon storage in the study area, and the expression formula is as follows43:$$I = frac{{nsumnolimits_{i = 1}^{n} {sumnolimits_{j = 1}^{n} {w_{i,j} left( {x_{i} - overline{x} } right)left( {x_{j} - overline{x} } right)} } }}{{sumnolimits_{i = 1}^{n} {sumnolimits_{j = 1}^{n} {omega_{ij} } } sumnolimits_{i = 1}^{n} {left( {x_{i} - overline{x} } right)^{2} } }}$$ (10) wij is the spatial weight; x is the attribute mean; xi and xj are the attribute values of elements i, j, respectively; n is the number of cells, and the correlation is considered significant when |Z|  > 1.96.Local indications of spatial association (LISA)LISA reveals the local cluster characteristics of spatial unit attributes by analyzing the difference and significance between spatial units and surrounding units, and the expression formula is as follows42:$$I_{i} (d) = frac{{n(x_{i} – overline{x} )sumnolimits_{j = 1}^{n} {w_{ij} (x_{j} – overline{x} )} }}{{sumnolimits_{i = 1}^{n} {(x_{j} – overline{x} )^{2} } }}$$
(11)
Correlation analysisIn order to evaluate the influence of natural factors and socioeconomic factors on carbon storage in the study area, the correlation coefficients of temperature, rainfall, NDVI, GDP, population density (PD), HAI and carbon storage were calculated according to the Pearson correlation coefficient method. The calculation formula is as follows44:$$r_{xy} = frac{{sumnolimits_{i = 1}^{n} {(M_{i} – overline{x} )(y_{i} – overline{y} )} }}{{sqrt {sumnolimits_{i = 1}^{n} {(M_{i} – overline{x} )^{2} sumnolimits_{i = 1}^{n} {(y_{i} – overline{y} )} } } }}$$
(12)
rxy represents the correlation coefficient between x and y; Mi represents the carbon storage in the ith year; yi represents the value of the impact factor Y in the ith year, and ({overline{text{x}}}) and ({overline{text{y}}}) respectively represents the average value of carbon storage and impact factor in the research period over several years.Human influence index analysis methodLand use is significantly spatially clustered in the study area31, and LUCC changes will have a certain impact on the structure and process of the ecosystem. HAI has the characteristics of spatial variability, which can reflect the impact of human activities on land use and landscape composition changes. In this study, Human Influence Index Analysis Method (HAI) index was used to analyze the correlation between carbon storage and human interference intensity in the Western Sichuan Plateau. The calculation formula is as follows30,$$HAI = sumlimits_{i = 1}^{n} {left( {A_{i} P_{i} /TA} right)}$$
(13)
HAI is Human Active Index; Ai is the total area of the ith land use type; Pi The intensity parameter of human impact reflected by type i land use type; TA is the total final surface area of land use type in evaluation unit; n is the number of land use types. Combined with the land use type of this study, Pi is assigned by Delphi method, in which cropland is 0.67, woodland is 0.13, grassland is 0.12, water body is 0.10, built-up land is 0.96, and unused land is 0.0530,45.GeodetectorGeodetector is an algorithm that uses spatial heterogeneity principle to detect driving factors of carbon storage, which can quantitatively detect the influence of impact factors on carbon storage and explore the interaction between driving factors. Geodetector includes factor detection, risk detection, interaction detection and ecological detection46.Differentiation and factor detection: the influence factors were discretized, and then the significance test of the difference in the mean values of the impact factors was conducted to detect the relative importance among the factors. The statistical quantity q is used to measure the explanatory power of impact factors on the carbon storage spatial differentiation and the value range of q is between 0 and 1. The larger the value, the stronger the explanatory power of the factor47.$$q = 1{ – }frac{{sumnolimits_{h = 1}^{L} {N_{h} sigma_{h}^{2} } }}{{Nsigma^{2} }}$$
(14)
In this formula, h = 1, 2…, L is the classification or partition of variable (Y) or factor (X); Nh and N are layer h and regional number units respectively; and (sigma_{h}^{2}) and (sigma_{{}}^{2}) are the variance of the layer h and regional value Y respectively.The variance of the regional value Y is calculated as follows,$$sigma^{2} = frac{{sumnolimits_{i = 1}^{n} {(Y_{i} – overline{Y} )^{2} } }}{N – 1}$$
(15)
where, Yi and (overline{Y}) are the mean value of sample j and the region Y, respectively.$$sigma^{2} = frac{{sumnolimits_{i = 1}^{{n_{h} }} {(Y_{h,i} – overline{{Y_{h} }} )^{2} } }}{{N_{h} – 1}}$$
(16)
where, Y and (overline{Y}) are the value and mean value of sample i in layer h, respectively.Interaction detection: it is used to identify the interaction between different impact factors Xs, that is, to evaluate whether the combined action of X1 and X2 will increase or weaken the explanatory power of vegetation coverage Y, or the influence of these factors on Y is independent of each other. The evaluation method is to first calculate the value q of the two factors X1 and X2 for Y respectively: q(X1) and q(X2), and calculate the value q of their interaction (the new polygon distribution formed by the tangent of the two layers of the superimposed variables X1 and X2) : q(X1 ∩ X2) and compare q(X1) and q(X2) with q(X1 ∩ X2)46. More

Barbier, E. B. et al. The value of estuarine and coastal ecosystem services. Ecol. Monogr. 81, 169–193 (2011).Article 

Google Scholar 
Lotze, H. K. Depletion, degradation, and recovery potential of estuaries and coastal seas. Science 312, 1806–1809 (2006).ADS 
CAS 
PubMed 
Article 

Google Scholar 
Davis, J. & Kidd, I. M. Identifying major stressors: The essential precursor to restoring cultural ecosystem services in a degraded estuary. Estuar. Coast. 35, 1007–1017 (2012).Article 

Google Scholar 
Copeland, B. Effects of decreased river flow on estuarine ecology. J. Water Pollut. Control Fed. 66, 1831–1839 (1966).
Google Scholar 
Mcowen, C. J. et al. A global map of saltmarshes. Biodivers. Data J. https://doi.org/10.3897/BDJ.5.e11764 (2017).Article 
PubMed 
PubMed Central 

Google Scholar 
Jackson, J. B. C. Historical overfishing and the recent collapse of coastal ecosystems. Science 293, 629–637 (2001).CAS 
PubMed 
Article 

Google Scholar 
McAfee, D. & Connell, S. D. The global fall and rise of oyster reefs. Front. Ecol. Environ. 19, 118–125 (2021).Article 

Google Scholar 
Powers, R. P. & Jetz, W. Global habitat loss and extinction risk of terrestrial vertebrates under future land-use-change scenarios. Nat. Clim. Chang. 9, 323–329 (2019).ADS 
Article 

Google Scholar 
Cardinale, B. J. et al. Biodiversity loss and its impact on humanity. Nature 486, 59–67 (2012).ADS 
CAS 
PubMed 
Article 

Google Scholar 
Hoekstra, J. M., Boucher, T. M., Ricketts, T. H. & Roberts, C. Confronting a biome crisis: Global disparities of habitat loss and protection. Ecol. Lett. 8, 23–29 (2005).Article 

Google Scholar 
Goldewijk, K. K. Estimating global land use change over the past 300 years: The HYDE database. Glob. Biogeochem. 15, 417–433 (2001).ADS 
Article 

Google Scholar 
Munday, P. L. Habitat loss, resource specialization, and extinction on coral reefs. Glob. Chang. Biol. 10, 1642–1647 (2004).ADS 
Article 

Google Scholar 
Tilman, D., May, R. M., Lehman, C. L. & Nowak, M. A. Habitat destruction and the extinction debt. Nature 371, 65–66 (1994).ADS 
Article 

Google Scholar 
Elliott, M., Burdon, D., Hemingway, K. L. & Apitz, S. E. Estuarine, coastal and marine ecosystem restoration: Confusing management and science—A revision of concepts. Estuar. Coast. Shelf Sci. 74, 349–366 (2007).ADS 
Article 

Google Scholar 
Benayas, J. M. R., Newton, A. C., Diaz, A. & Bullock, J. M. Enhancement of biodiversity and ecosystem services by ecological restoration: A meta-analysis. Science 325, 1121–1124 (2009).ADS 
CAS 
Article 

Google Scholar 
Hobbs, R. J. & Norton, D. A. Towards a conceptual framework for restoration ecology. Restor. Ecol. 4, 93–110 (1996).Article 

Google Scholar 
Jones, C. G., Lawton, J. H. & Shachak, M. Organisms as Ecosystem Engineers in Ecosystem Management 130–147 (Springer, 1994).Tolley, S. G. & Volety, A. K. The role of oysters in habitat use of oyster reefs by resident fishes and decapod crustaceans. J. Shellf. Res. 24, 1007–1012 (2005).Article 

Google Scholar 
Carroll, J. M., Keller, D. A., Furman, B. T. & Stubler, A. D. Rough around the edges: Lessons learned and future directions in marine edge effects studies. Curr. Landsc. Ecol. Rep. 4, 91–102 (2019).Article 

Google Scholar 
Harwell, H. D., Posey, M. H. & Alphin, T. D. Landscape aspects of oyster reefs: Effects of fragmentation on habitat utilization. J. Exp. Mar. Biol. Ecol. 409, 30–41 (2011).Article 

Google Scholar 
Shervette, V. R. & Gelwick, F. Seasonal and spatial variations in fish and macroinvertebrate communities of oyster and adjacent habitats in a Mississippi estuary. Estuar. Coast. 31, 584–596 (2008).Article 

Google Scholar 
Gain, I. E. et al. Macrofauna using intertidal oyster reef varies in relation to position within the estuarine habitat mosaic. Mar. Biol. 164, 1–16 (2017).MathSciNet 
Article 

Google Scholar 
Wong, M., Peterson, C. & Piehler, M. Evaluating estuarine habitats using secondary production as a proxy for food web support. Mar. Ecol. Prog. Ser. 440, 11–25 (2011).ADS 
Article 

Google Scholar 
Meyer, D. L. Habitat partitioning between the xanthid crabs Panopeus herbstii and Eurypanopeus depressus on intertidal oyster reefs (Crassostrea virginica) in southeastern North Carolina. Estuaries 17, 674–679 (1994).Article 

Google Scholar 
McDonald, J. Divergent life history patterns in the co-occurring intertidal crabs Panopeus herbstii and Eurypanopeus depressus (Crustacea: Brachyura: Xanthidae). Mar. Ecol. Prog. Ser. 8, 173–180 (1982).ADS 
Article 

Google Scholar 
Grabowski, J. H. & Peterson, C. H. Restoring oyster reefs to recover ecosystem services in Theoretical Ecology Series vol. 4 281–298 (Elsevier, 2007).Beck, M. W. et al. Oyster reefs at risk and recommendations for conservation, restoration, and management. Bioscience 61, 107–116 (2011).Article 

Google Scholar 
Reeves, S. E. et al. Facilitating better outcomes: how positive species interactions can improve oyster reef restoration. Front. Mar. Sci. 7, 656 (2020).Article 

Google Scholar 
Jud, Z. R. & Layman, C. A. Changes in motile benthic faunal community structure following large-scale oyster reef restoration in a subtropical estuary. Food Webs 25, e00177 (2020).Article 

Google Scholar 
Pinnell, C. M., Ayala, G. S., Patten, M. V. & Boyer, K. E. Seagrass and oyster reef restoration in living shorelines: effects of habitat configuration on invertebrate community assembly. Diversity 13, 246 (2021).Article 

Google Scholar 
La Peyre, M. K., Humphries, A. T., Casas, S. M. & La Peyre, J. F. Temporal variation in development of ecosystem services from oyster reef restoration. Ecol. Eng. 63, 34–44 (2014).Article 

Google Scholar 
Geraldi, N. R., Powers, S. P., Heck, K. L. & Cebrian, J. Can habitat restoration be redundant? Response of mobile fishes and crustaceans to oyster reef restoration in marsh tidal creeks. Mar. Ecol. Prog. Ser. 389, 171–180 (2009).ADS 
Article 

Google Scholar 
Humphries, A. T. & La Peyre, M. K. Oyster reef restoration supports increased nekton biomass and potential commercial fishery value. PeerJ 3, e1111 (2015).PubMed 
PubMed Central 
Article 

Google Scholar 
Ziegler, S. L., Grabowski, J. H., Baillie, C. J. & Fodrie, F. Effects of landscape setting on oyster reef structure and function largely persist more than a decade post-restoration. Restor. Ecol. 26, 933–942 (2018).Article 

Google Scholar 
Rodriguez, A. B. et al. Oyster reefs can outpace sea-level rise. Nat. Clim. Change 4, 493–497 (2014).ADS 
Article 

Google Scholar 
Hanke, M. H., Posey, M. H. & Alphin, T. D. The influence of habitat characteristics on intertidal oyster Crassostrea virginica populations. Mar. Ecol. Prog. Ser. 571, 121–138 (2017).ADS 
Article 

Google Scholar 
Barber, A., Walters, L. & Birch, A. Potential for restoring biodiversity of macroflora and macrofauna on oyster reefs in Mosquito Lagoon, Florida. Fla. Sci. 66, 47–62 (2010).
Google Scholar 
Boudreaux, M. L., Stiner, J. L. & Walters, L. J. Biodiversity of sessile and motile macrofauna on intertidal oyster reefs in Mosquito Lagoon, Florida. J. Shellf. Res. 25, 1079–1089 (2006).Article 

Google Scholar 
Desmond, J. S., Deutschman, D. H. & Zedler, J. B. Spatial and temporal variation in estuarine fish and invertebrate assemblages: Analysis of an 11-year data set. Estuaries 25, 552–569 (2002).Article 

Google Scholar 
Xu, Y., Xian, W. & Li, W. Spatial and temporal variations of invertebrate community in the Yangtze River Estuary and its adjacent waters. Biodivers. Sci. 22, 311 (2014).Article 

Google Scholar 
Nichols, F. H. Abundance fluctuations among benthic invertebrates in two pacific estuaries. Estuaries 8, 136 (1985).Article 

Google Scholar 
Van Horn, J. & Tolley, S. G. Patterns of distribution along a salinity gradient in the flatback mud crab Eurypanopeus depressus. Gulf Mex. Sci. 26, 66 (2008).
Google Scholar 
Costlow, J. D., Bookhout, C. G. & Monroe, R. Salinity-temperature effects on the larval development of the crab, Panopeus herbstii Milne-Edwards, reared in the laboratory. Physiol. Zool. 35, 79–93 (1962).Article 

Google Scholar 
Sulkin, S., Heukelem, W. & Kelly, P. Behavioral basis for depth regulation in the hatching and post larval stages of the mud crab Eurypanopeus depresus. Mar. Ecol. Prog. Ser. 11, 157–164 (1983).ADS 
Article 

Google Scholar 
Phlips, E. J., Badylak, S. & Grosskopf, T. Factors affecting the abundance of phytoplankton in a restricted subtropical lagoon, the Indian River Lagoon, Florida, USA. Estuar. Coast. Shelf. Sci. 55, 385–402 (2002).ADS 
CAS 
Article 

Google Scholar 
Menendez, R. J. Vertical zonation of the xanthid mud crabs Panopeus obesus and Panopeus simpsoni on oyster reefs. Bull. Mar. Sci. 40, 73–77 (1987).
Google Scholar 
Barshaw, D. E. & Lavalli, K. L. Predation upon postlarval lobsters Homarus americanus by cunners Tautogolabrus adspersus and mud crabs Neopanope sayi on three different substrates: Eelgrass, mud and rocks. Mar. Ecol. Prog. Ser. 48, 119–123 (1988).ADS 
Article 

Google Scholar 
Key, P. B., Wirth, E. F. & Fulton, M. H. A review of grass shrimp, Palaemonetes spp., as a bioindicator of anthropogenic impacts. Environ. Bioindic. 1, 115–128 (2006).CAS 
Article 

Google Scholar 
Kneib, R. T. & Weeks, C. A. Intertidal distribution and feeding habits of the mud crab, Eurytium limosum. Estuaries 13, 462 (1990).Article 

Google Scholar 
Hsueh, P.-W., McClintock, J. B. & Hopkins, T. S. Comparative study of the diets of the blue crabs Callinectes similis and C. sapidus from a mud-bottom habitat in Mobile Bay, Alabama. J. Crust. Biol. 12, 615–619 (1992).Article 

Google Scholar 
King, S. P. & Sheridan, P. Nekton of new seagrass habitats colonizing a subsided salt marsh in Galveston Bay, Texas. Estuar. Coast. 29, 286–296 (2006).Article 

Google Scholar 
Zupo, V. & Nelson, W. Factors influencing the association patterns of Hippolyte zostericola and Palaemonetes intermedius (Decapoda: Natantia) with seagrasses of the Indian River Lagoon, Florida. Mar. Biol. 134, 181–190 (1999).Article 

Google Scholar 
Weber, J. C. & Epifanio, C. E. Response of mud crab (Panopeus herbstii) megalopae to cues from adult habitat. Mar. Biol. 126, 655–661 (1996).Article 

Google Scholar 
Harris, K. P. Oyster Reef Restoration: Impacts on Infaunal Communities in a Shallow Water Estuary. Honors Thesis. University of Central Florida (2018).Shaffer, M., Donnelly, M. & Walters, L. Does intertidal oyster reef restoration affect avian community structure and behavior in a shallow estuarine system? A post-restoration analysis. Fla. Field Nat. 47, 37–59 (2019).
Google Scholar 
Puckett, B. J. et al. Integrating larval dispersal, permitting, and logistical factors within a validated habitat suitability index for oyster restoration. Front. Mar. Sci. 5, 76 (2018).Article 

Google Scholar 
Kim, C., Park, K. & Powers, S. P. Establishing restoration strategy of eastern oyster via a coupled biophysical transport model. Restor. Ecol. 21, 353–362 (2013).Article 

Google Scholar 
Rodriguez-Perez, A., James, M. A. & Sanderson, W. G. A small step or a giant leap: Accounting for settlement delay and dispersal in restoration planning. PLoS ONE 16, e0256369 (2021).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Yeager, L. A. & Layman, C. A. Energy flow to two abundant consumers in a subtropical oyster reef food web. Aquat. Ecol. 45, 267–277 (2011).Article 

Google Scholar 
Rodney, W. S. & Paynter, K. T. Comparisons of macrofaunal assemblages on restored and non-restored oyster reefs in mesohaline regions of Chesapeake Bay in Maryland. J. Exp. Mar. Biol. Ecol. 335, 39–51 (2006).Article 

Google Scholar 
Macreadie, P. I., Geraldi, N. R. & Peterson, C. H. Preference for feeding at habitat edges declines among juvenile blue crabs as oyster reef patchiness increases and predation risk grows. Mar. Ecol. Prog. Ser. 466, 145–153 (2012).ADS 
Article 

Google Scholar 
Fodrie, F. J. et al. Measuring individuality in habitat use across complex landscapes: Approaches, constraints, and implications for assessing resource specialization. Oecologia 178, 75–87 (2015).ADS 
PubMed 
Article 

Google Scholar 
Grabowski, J. H. et al. Regional environmental variation and local species interactions influence biogeographic structure on oyster reefs. Ecology 101, e02921 (2020).PubMed 
Article 

Google Scholar 
Peterson, C., Grabowski, J. & Powers, S. Estimated enhancement of fish production resulting from restoring oyster reef habitat: Quantitative valuation. Mar. Ecol. Prog. Ser. 264, 249–264 (2003).ADS 
Article 

Google Scholar 
Garvis, S. K., Sacks, P. E. & Walters, L. J. Formation, movement, and restoration of dead intertidal oyster reefs in Canaveral National Seashore and Mosquito Lagoon, Florida. J. Shellf. Res. 34, 251–258 (2015).Article 

Google Scholar 
Gilmore, G. R. Environmental and biogeographic factors influencing ichthyofaunal diversity: Indian River Lagoon. Bull. Mar. Sci. 57, 153–170 (1995).
Google Scholar 
Swain, H. M. Reconciling rarity and representation: A review of listed species in the Indian River Lagoon. Bull. Mar. Sci. 57, 252–266 (1995).ADS 

Google Scholar 
Tremain, D. M. & Adams, D. H. Seasonal variations in species diversity, abundance, and composition of fish communities in the northern Indian River Lagoon, Florida. Bull. Mar. Sci. 57, 171–192 (1995).
Google Scholar 
Paperno, R., Mille, K. & Kadison, E. Patterns in species composition of fish and selected invertebrate assemblages in estuarine subregions near Ponce de Leon Inlet, Florida. Estuar. Coast. Shelf. Sci. 52, 117–130 (2001).ADS 
Article 

Google Scholar 
Smithsonian Marine Station (SMS) at Fort Pierce. Indian River Lagoon Species Inventory https://naturalhistory2.si.edu/smsfp/irlspec/Walters, L. J. et al. A negative association between recruitment of the eastern oyster Crassostrea virginica and the brown tide Aureoumbra lagunensis in Mosquito Lagoon, Florida. Fla. Sci. 84, 81–91 (2021).
Google Scholar 
Walters, L. J., Sacks, P. E. & Campbell, D. E. Boating impacts and boat-wake resilient restoration of the eastern oyster Crassostrea virginica in Mosquito Lagoon, Florida, USA. Fla. Sci. 84, 173–199 (2021).
Google Scholar 
Hanke, M. H., Posey, M. H. & Alphin, T. D. The effects of intertidal oyster reef habitat characteristics on faunal utilization. Mar. Ecol. Prog. Ser. 581, 57–70 (2017).ADS 
Article 

Google Scholar 
Crabtree, R. E. & Dean, J. M. The structure of two South Carolina estuarine tide pool fish assemblages. Estuaries 5, 2–9 (1982).Article 

Google Scholar 
Baggett, L. P. et al. Guidelines for evaluating performance of oyster habitat restoration: Evaluating performance of oyster restoration. Restor. Ecol. 23, 737–745 (2015).Article 

Google Scholar 
Chambers, L. G. et al. How well do restored intertidal oyster reefs support key biogeochemical properties in a coastal lagoon?. Estuar. Coast. 41, 784–799 (2018).Article 

Google Scholar 
Shannon, C. & Wiener, W. The Mathematical Theory of Communication (Illinois Press, 1963).
Google Scholar 
Nagendra, H. Opposite trends in response for the Shannon and Simpson indices of landscape diversity. Appl. Geogr. 22, 175–186 (2002).Article 

Google Scholar 
Pinheiro, J., Bates, D., DebRoy, S., Sarkar, D., & R Core Team. nlme: Linear and Nonlinear Mixed Effects Models (2021).Hulbert, J. Pseudoreplication and the design of field experiments in ecology. Ecol. Monogr. 54, 187–211 (1984).Article 

Google Scholar 
Wang, Z. & Goonewardene, L. A. The use of MIXED models in the analysis of animal experiments with repeated measures data. Can. J. Anim. Sci. 84, 1–11 (2004).Article 

Google Scholar 
Bates, D., Mächler, M., Bolker, B. & Walker, S. Fitting Mixed-Effects Models Using the lme4 Package in R (2008).Lenth, R. V. emmeans: Estimated Marginal Means, aka Least-Squares Means (2021).Clarke, K. R., Somerfield, P. J. & Chapman, M. G. On resemblance measures for ecological studies, including taxonomic dissimilarities and a zero-adjusted Bray–Curtis coefficient for denuded assemblages. J. Exp. Mar. Biol. Ecol. 330, 55–80 (2006).Article 

Google Scholar 
Bray, J. R. & Curtis, J. T. An ordination of the upland forest communities of southern Wisconsin. Ecol. Monogr. 27, 325–349 (1957).Article 

Google Scholar 
Anderson, M. J. & Willis, T. J. Canonical analysis of principal coordinates: A useful method of constrained ordination for ecology. Ecology 84, 511–525 (2003).Article 

Google Scholar 
Clarke, K. R., Gorley, R., Somerfield, P. J. & Warwick, R. Change in Marine Communities: An Approach to Statistical Analysis and Interpretation (Primer-E Ltd, 2014).Akaike, H. A new look at the statistical model identification. IEEE Trans. Autom. Control 19, 716–723 (1974).ADS 
MathSciNet 
MATH 
Article 

Google Scholar 
Venables, W. N. & Ripley, B. D. Modern Applied Statistics with S. (Springer, 2002). 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