Ecology

Transforming our World: The 2030 Agenda for Sustainable Development (UN, 2015).May, R. M. Will a large complex system be stable? Nature 238, 413–414 (1972).CAS 
PubMed 
Article 

Google Scholar 
Pimm, S. L. The complexity and stability of ecosystems. Nature 307, 321–326 (1984).Article 

Google Scholar 
McCann, K. S. The diversity–stability debate. Nature 405, 228–233 (2000).CAS 
PubMed 
Article 

Google Scholar 
Ives, A. R. & Carpenter, S. R. Stability and diversity of ecosystems. Science 317, 58–62 (2007).CAS 
PubMed 
Article 

Google Scholar 
Donohue, I. et al. Navigating the complexity of ecological stability. Ecol. Lett. 19, 1172–1185 (2016).PubMed 
Article 

Google Scholar 
Yachi, S. & Loreau, M. Biodiversity and ecosystem productivity in a fluctuating environment: the insurance hypothesis. Proc. Natl Acad. Sci. USA 96, 1463–1468 (1999).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Isbell, F. et al. Biodiversity increases the resistance of ecosystem productivity to climate extremes. Nature 526, 574–577 (2015).CAS 
PubMed 
Article 

Google Scholar 
Tilman, D., Reich, P. B. & Knops, J. M. Biodiversity and ecosystem stability in a decade-long grassland experiment. Nature 441, 629–632 (2006).CAS 
PubMed 
Article 

Google Scholar 
Loreau, M. et al. Biodiversity as insurance: from concept to measurement and application. Biol. Rev. 96, 2333–2354 (2021).PubMed 
Article 

Google Scholar 
Thibaut, L. M. & Connolly, S. R. Understanding diversity–stability relationships: towards a unified model of portfolio effects. Ecol. Lett. 16, 140–150 (2013).PubMed 
Article 

Google Scholar 
Xu, Q. et al. Consistently positive effect of species diversity on ecosystem, but not population, temporal stability. Ecol. Lett. 24, 2256–2266 (2021).PubMed 
Article 

Google Scholar 
Hector, A. et al. General stabilizing effects of plant diversity on grassland productivity through population asynchrony and overyielding. Ecology 91, 2213–2220 (2010).CAS 
PubMed 
Article 

Google Scholar 
Hautier, Y. et al. Anthropogenic environmental changes affect ecosystem stability via biodiversity. Science 348, 336–340 (2015).CAS 
PubMed 
Article 

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

Google Scholar 
Tilman, D., Isbell, F. & Cowles, J. M. Biodiversity and ecosystem functioning. Annu. Rev. Ecol. Evol. Syst. 45, 471–493 (2014).Article 

Google Scholar 
Isbell, F. et al. Linking the influence and dependence of people on biodiversity across scales. Nature 546, 65–72 (2017).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Gonzalez, A. et al. Scaling-up biodiversity–ecosystem functioning research. Ecol. Lett. 23, 757–776 (2020).PubMed 
PubMed Central 
Article 

Google Scholar 
Hooper, D. U. et al. Effects of biodiversity on ecosystem functioning: a consensus of current knowledge. Ecol. Monogr. 75, 3–35 (2005).Article 

Google Scholar 
Wang, S. & Loreau, M. Ecosystem stability in space: alpha, beta and gamma variability. Ecol. Lett. 17, 891–901 (2014).PubMed 
Article 

Google Scholar 
Wang, S. & Loreau, M. Biodiversity and ecosystem stability across scales in metacommunities. Ecol. Lett. 19, 510–518 (2016).PubMed 
PubMed Central 
Article 

Google Scholar 
Wang, S. et al. Biotic homogenization destabilizes ecosystem functioning by decreasing spatial asynchrony. Ecology 102, e03332 (2021).Zhang, Y., He, N., Loreau, M., Pan, Q. & Han, X. Scale dependence of the diversity–stability relationship in a temperate grassland. J. Ecol. 106, 1227–1285 (2018).PubMed 
Article 

Google Scholar 
Wang, S., Lamy, T., Hallett, L. M. & Loreau, M. Stability and synchrony across ecological hierarchies in heterogeneous metacommunities: linking theory to data. Ecography 42, 1200–1211 (2019).Article 

Google Scholar 
Hautier, Y. et al. General destabilizing effects of eutrophication on grassland productivity at multiple spatial scales. Nat. Commun. 11, 5375 (2020).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Liang, M., Liang, C., Hautier, Y., Wilcox, K. R. & Wang, S. Grazing-induced biodiversity loss impairs grassland ecosystem stability at multiple scales. Ecol. Lett. 24, 2054–2064 (2021).PubMed 
Article 

Google Scholar 
Qiao, X. et al. Spatial asynchrony matters more than alpha stability in stabilizing ecosystem productivity in a large temperate forest region. Glob. Ecol. Biogeogr. 31, 1133–1146 (2022).Article 

Google Scholar 
Catano, C. P., Fristoe, T. S., LaManna, J. A. & Myers, J. A. Local species diversity, beta-diversity and climate influence the regional stability of bird biomass across North America. Proc. R. Soc. B 287, 20192520 (2020).PubMed 
PubMed Central 
Article 

Google Scholar 
Patrick, C. J. et al. Multi‐scale biodiversity drives temporal variability in macrosystems. Front. Ecol. Environ. 19, 47–56 (2021).Article 

Google Scholar 
Wilcox, K. R. et al. Asynchrony among local communities stabilises ecosystem function of metacommunities. Ecol. Lett. 20, 1534–1545 (2017).PubMed 
PubMed Central 
Article 

Google Scholar 
Zhang, Y. et al. Nitrogen addition does not reduce the role of spatial asynchrony in stabilising grassland communities. Ecol. Lett. 22, 563–571 (2019).PubMed 
PubMed Central 
Article 

Google Scholar 
Garcia, R. A., Cabeza, M., Rahbek, C. & Araujo, M. B. Multiple dimensions of climate change and their implications for biodiversity. Science 344, 1247579 (2014).PubMed 
Article 
CAS 

Google Scholar 
García-Palacios, P., Gross, N., Gaitán, J. & Maestre, F. T. Climate mediates the biodiversity–ecosystem stability relationship globally. Proc. Natl Acad. Sci. USA 115, 8400–8405 (2018).PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Hillebrand, H. On the generality of the latitudinal diversity gradient. Am. Nat. 163, 192–211 (2004).PubMed 
Article 

Google Scholar 
Qian, H. & Ricklefs, R. E. A latitudinal gradient in large-scale beta diversity for vascular plants in North America. Ecol. Lett. 10, 737–744 (2007).PubMed 
Article 

Google Scholar 
Kraft, N. J. B. et al. Disentangling the drivers of β diversity along latitudinal and elevational gradients. Science 333, 1755–1758 (2011).CAS 
PubMed 
Article 

Google Scholar 
Ma, Z. et al. Climate warming reduces the temporal stability of plant community biomass production. Nat. Commun. 8, 15378 (2017).PubMed 
PubMed Central 
Article 

Google Scholar 
Song, J. et al. A meta-analysis of 1,119 manipulative experiments on terrestrial carbon-cycling responses to global change. Nat. Ecol. Evol. 3, 1309–1320 (2019).PubMed 
Article 

Google Scholar 
Valencia, E. et al. Synchrony matters more than species richness in plant community stability at a global scale. Proc. Natl Acad. Sci. USA 117, 24345–24351 (2020).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Gilbert, B. et al. Climate and local environment structure asynchrony and the stability of primary production in grasslands. Glob. Ecol. Biogeogr. 29, 1177–1188 (2020).Article 

Google Scholar 
Hallett, L. M. et al. Biotic mechanisms of community stability shift along a precipitation gradient. Ecology 95, 1693–1700 (2014).PubMed 
Article 

Google Scholar 
Hong, P. et al. Biodiversity promotes ecosystem functioning despite environmental change. Ecol. Lett. 25, 555–569 (2022).PubMed 
Article 

Google Scholar 
. Plant presence and percent cover, RELEASE-2021. NEON (National Ecological Observatory Network) https://doi.org/10.48443/abge-r811 (2021).Barnett, D. T. et al. The plant diversity sampling design for The National Ecological Observatory. Netw. Ecosphere 10, e02603 (2019).
Google Scholar 
Lasky, J. R., Uriarte, M. & Muscarella, R. Synchrony, compensatory dynamics, and the functional trait basis of phenological diversity in a tropical dry forest tree community: effects of rainfall seasonality. Environ. Res. Lett. 11, 115003 (2016).Article 

Google Scholar 
Inchausti, P. & Halley, J. Investigating long-term ecological variability using the global population dynamics database. Science 293, 655–657 (2001).CAS 
PubMed 
Article 

Google Scholar 
Luo, M. et al. The effects of dispersal on spatial synchrony in metapopulations differ by timescale. Oikos 130, 1762–1772 (2021).Article 

Google Scholar 
Pimm, S. L. & Redfearn, A. The variability of population densities. Nature 334, 613–614 (1988).Article 

Google Scholar 
Craven, D. et al. Multiple facets of biodiversity drive the diversity–stability relationship. Nat. Ecol. Evol. 2, 1579–1587 (2018).PubMed 
Article 

Google Scholar 
Peet, R. K., Wentworth, T. R. & White, P. S. A flexible, multipurpose method for recording vegetation composition and structure. Castanea 63, 262–274 (1998).
Google Scholar 
Hijmans, R. J., Cameron, S. E., Parra, J. L., Jones, P. G. & Jarvis, A. Very high resolution interpolated climate surfaces for global land areas. Int. J. Climatol. 25, 1965–1978 (2005).Article 

Google Scholar 
Loreau, M. & de Mazancourt, C. Species synchrony and its drivers: neutral and nonneutral community dynamics in fluctuating environments. Am. Nat. 172, E48–E66 (2008).PubMed 
Article 

Google Scholar 
Lefcheck, J. S. piecewiseSEM: piecewise structural equation modelling inr for ecology, evolution, and systematics. Methods Ecol. Evol. 7, 573–579 (2016).Article 

Google Scholar 
Pinheiro, J., Bates, D., DebRoy, S., Sarkar, D. & R Core Team. nlme: Linear and nonlinear mixed effects models. R package v.3.1–152 https://CRAN.R-project.org/package=nlme (2021).R Core Team. R: A Language and Environment for Statistical Computing (R Foundation for Statistical Computing, 2019). More

Beer, C. et al. Terrestrial gross carbon dioxide uptake: global distribution and covariation with climate. Science 329, 834 (2010).Article 

Google Scholar 
Anderegg, W. R. L. et al. Pervasive drought legacies in forest ecosystems and their implications for carbon cycle models. Science 349, 528–532 (2015).Article 

Google Scholar 
Ciais, P. et al. Europe-wide reduction in primary productivity caused by the heat and drought in 2003. Nature 437, 529–533 (2005).Article 

Google Scholar 
Peñuelas, J. et al. Shifting from a fertilization-dominated to a warming-dominated period. Nat. Ecol. Evol. 1, 1438–1445 (2017).Article 

Google Scholar 
Morin, X. et al. Temporal stability in forest productivity increases with tree diversity due to asynchrony in species dynamics. Ecol. Lett. 17, 1526–1535 (2014).Article 

Google Scholar 
Isbell, F. et al. Biodiversity increases the resistance of ecosystem productivity to climate extremes. Nature 526, 574 (2015).Article 

Google Scholar 
De Boeck, H. J. et al. Patterns and drivers of biodiversity–stability relationships under climate extremes. J. Ecol. 106, 890–902 (2018).Article 

Google Scholar 
Grossiord, C. Having the right neighbors: how tree species diversity modulates drought impacts on forests. N. Phytol. 228, 42–49 (2020).Article 

Google Scholar 
O’Brien, M. J. et al. Resistance of tropical seedlings to drought is mediated by neighbourhood diversity. Nat. Ecol. Evol. 1, 1643–1648 (2017).Article 

Google Scholar 
Gazol, A. & Camarero, J. J. Functional diversity enhances silver fir growth resilience to an extreme drought. J. Ecol. 104, 1063–1075 (2016).Article 

Google Scholar 
Pretzsch, H., Schütze, G. & Uhl, E. Resistance of European tree species to drought stress in mixed versus pure forests: evidence of stress release by inter-specific facilitation. Plant Biol. 15, 483–495 (2013).Article 

Google Scholar 
Grossiord, C. et al. Tree diversity does not always improve resistance of forest ecosystems to drought. P. Natl Acad. Sci. USA 111, 14812–14815 (2014).Article 

Google Scholar 
Grossiord, C. et al. Does drought influence the relationship between biodiversity and ecosystem functioning in boreal forests. Ecosystems 17, 394–404 (2014).Article 

Google Scholar 
Loreau, M., Mouquet, N. & Gonzalez, A. Biodiversity as spatial insurance in heterogeneous landscapes. P. Natl Acad. Sci. USA 100, 12765 (2003).Article 

Google Scholar 
Lloret, F. et al. Woody plant richness and NDVI response to drought events in Catalonian (northeastern Spain) forests. Ecology 88, 2270–2279 (2007).Article 

Google Scholar 
He, Q. & Bertness, M. D. Extreme stresses, niches, and positive species interactions along stress gradients. Ecology 95, 1437–1443 (2014).Article 

Google Scholar 
Hafner, B. D. et al. Hydraulic redistribution under moderate drought among English oak, European beech and Norway spruce determined by deuterium isotope labeling in a split-root experiment. Tree Physiol. 37, 950–960 (2017).Article 

Google Scholar 
Forrester, D. I. & Bauhus, J. A review of processes behind diversity–productivity relationships in forests. Curr. For. Rep. 2, 45–61 (2016).Article 

Google Scholar 
Vitali, V., Forrester, D. I. & Bauhus, J. Know your neighbours: drought response of Norway spruce, silver fir and Douglas fir in mixed forests depends on species identity and diversity of tree neighbourhoods. Ecosystems 21, 1215–1229 (2018).Article 

Google Scholar 
The State of the World’s Forests 2020: Forests, Biodiversity and People (FAO and UNEP, 2020).Zhang, J., Fu, B., Stafford-Smith, M., Wang, S. & Zhao, W. Improve forest restoration initiatives to meet Sustainable Development Goal 15. Nat. Ecol. Evol. https://doi.org/10.1038/s41559-020-01332-9 (2020).Schulze, K., Malek, Ž. & Verburg, P. H. Towards better mapping of forest management patterns: a global allocation approach. For. Ecol. Manage. 432, 776–785 (2019).Article 

Google Scholar 
Harris, N. L. et al. Global maps of twenty-first century forest carbon fluxes. Nat. Clim. Change https://doi.org/10.1038/s41558-020-00976-6 (2021).Maxwell, S. L. et al. Area-based conservation in the twenty-first century. Nature 586, 217–227 (2020).Article 

Google Scholar 
Williams, A. P. et al. Large contribution from anthropogenic warming to an emerging North American megadrought. Science 368, 314–318 (2020).Article 

Google Scholar 
Blackman, C. et al. Leaf hydraulic vulnerability is related to conduit dimensions and drought resistance across a diverse range of woody angiosperms. N. Phytol. 188, 1113–1123 (2010).Article 

Google Scholar 
Liang, J. et al. Positive biodiversity–productivity relationship predominant in global forests. Science 354, aaf8957 (2016).Article 

Google Scholar 
Wieczynski, D. J. et al. Climate shapes and shifts functional biodiversity in forests worldwide. P. Natl Acad. Sci. USA 116, 587–592 (2019).Article 

Google Scholar 
Tomppo, E. et al. National Forest Inventories: Pathways for Common Reporting (Springer, 2010).Chirici, G. et al. National Forest Inventories: Contributions to Forest Biodiversity Assessments (Springer, 2011).Magnussen, S., Smith, B. & Uribe, S. National Forest inventories in North America for monitoring forest tree species diversity. Plant Biosyst. 141, 113–122 (2007).Article 

Google Scholar 
Lesiv, M. et al. Global forest management data for 2015 at a 100 m resolution. Sci. Data 9, 199 (2022).Article 

Google Scholar 
Vicente-Serrano, S. M., Beguería, S. & López-Moreno, J. I. A. Multiscalar drought index sensitive to global warming: the Standardized Precipitation Evapotranspiration Index. J. Clim. 23, 1696–1718 (2010).Article 

Google Scholar 
Hansen, M. C. et al. High-resolution global maps of 21st-century forest cover change. Science 342, 850 (2013).Article 

Google Scholar 
Forest Resources Assessment 2015 (FAO, 2015).Lyapustin, A. et al. Scientific impact of MODIS C5 calibration degradation and C6+ improvements. Atmos. Meas. Tech. 7, 4353–4365 (2014).Article 

Google Scholar 
Didan, K. & Brreto, A. NASA MEaSUREs Vegetation Index and Phenology (VIP) Phenology EVI2 Yearly Global 0.05Deg CMG, NASA EOSDIS Land Processes DAAC, https://doi.org/10.5067/MEaSUREs/VIP/VIPPHEN_EVI2.004 (2016).Olson, D. M. et al. Terrestrial Ecoregions of the World: A New Map of Life on Earth: a new global map of terrestrial ecoregions provides an innovative tool for conserving biodiversity. Bioscience 51, 933–938 (2001).Article 

Google Scholar 
Kline, T. J. B. Sample issues, methodological implications, and best practices. Can. J. Behav. Sci. 49, 71–77 (2017).Article 

Google Scholar 
Gourlet-Fleury, S. et al. Tropical forest recovery from logging: a 24 year silvicultural experiment from Central Africa. Phil. Trans. R. Soc. B 368, 20120302 (2013).Article 

Google Scholar 
Obiang, N. L. E. et al. Spatial pattern of central African rainforests can be predicted from average tree size. Oikos 119, 1643–1653 (2010).Article 

Google Scholar 
Plotkin, J. B. et al. Predicting species diversity in tropical forests. P. Natl Acad. Sci. USA 97, 10850–10854 (2000).Article 

Google Scholar 
Graham, M. H. Confronting multicollinearity in ecological multiple regression. Ecology 84, 2809–2815 (2003).Article 

Google Scholar 
Tukey, J. W. Exploratory Data Analysis (Addison-Wesley,1977). More

Species occurrence recordsWe compiled data on the distribution of 21,252 endemic species of any of the twelve megadiverse countries from four tetrapod (5,757) and four vascular plant groups (15,389) (amphibians, reptiles, birds, mammals, lycophytes, ferns, gymnosperms, and flowering plants). Species occurrence records were obtained from the Global Biodiversity Information Facility (GBIF)27, the International Union of Conservation of Nature (IUCN)28, and BirdLife60,61. We only modeled species with at least 25 unique records at a 5 arc-minute resolution (~10 km at the equator). In many cases, the processing of the IUCN polygons resulted in species with thousands of occurrence records. In these cases, we randomly chose a maximum of 500 records per species. The greater the number of observed records, more problems can be associated with spatial bias in the modeling62. In the case of records coming from IUCN polygons, more records require more computing time and these do not necessarily provide more information into the modeling given that their distribution is quite homogeneous.For tetrapods, we first explored the possibility of using occurrence records from GBIF, but data for megadiverse countries were scarce. Consequently, we decided to use the distribution polygons provided by the IUCN for amphibians, reptiles, and mammals (terrestrial and freshwater species)28, and the distribution polygons provided by BirdLife60. We based this decision on the fact that ecological niche modeling using IUCN polygons has been proven to give robust results20. For the IUCN polygons, we retained species that have been categorized as “extant”, “possibly extinct”, “probably extant”, “possibly extant”, and “presence uncertain”, discarding species considered to be “extinct”. In addition, we did not model species reported by the IUCN as “introduced”, “vagrant”, or those in the “assisted colonization” category; for mammals and birds, we only considered the distribution of “resident” species. Depending on the taxonomic group, and given the information available, we used different approaches to identify species endemic to any of twelve megadiverse countries: Australia, Brazil, China, Colombia, Ecuador, India, Indonesia, Madagascar, Mexico, Peru, Philippines, and Venezuela. For birds, we used BirdLife to identify species listed as “breeding endemic” and then choose the corresponding IUCN polygons. To identify the rest of endemic species in the other groups, we used a 0.08333° buffer around each country to select the IUCN polygons that fall completely within the country limits. We converted all selected species polygons into unique records at a 5 min resolution (~10 km at the equator).For vascular plants, we used geographic occurrence data obtained from the Global Biodiversity Information Facility by querying all records under “Tracheophyta” (we only considered “Preserved Specimens” in our search). Plants records were taxonomically homogenized and cleaned following the procedures described in ref. 63 using Kew’s Plants of the World database64 as the source of taxonomic information. Mostly, we identified endemic species as those with all occurrence records restricted to any given megadiverse country. For countries in which data for vascular plants were scarce or absent (e.g., India), we complemented occurrence information with polygons from the IUCN (although IUCN data for plants remains limited) following the procedure described for tetrapods.Climatic dataWe used the 19 bioclimatic variables available at WorldClim v.2 (Fick 2017) as the baseline (present-day) climatic conditions (1970–2000) (annual mean temperature, mean diurnal range, isothermality, temperature seasonality, the maximum temperature of the warmest month, minimum temperature of the coldest month, temperature annual range, mean annual range, mean temperature of wettest quarter, mean temperature of driest quarter, mean temperature of warmest quarter, mean temperature of coldest quarter, annual precipitation, precipitation of wettest month, precipitation of driest month, precipitation seasonality, precipitation of wettest quarter, precipitation of driest quarter, precipitation of warmest quarter and precipitation of coldest quarter). From this baseline scenario, bioclimatic variables start to vary because of climate change. We used bioclimatic variables derived from the IPSL-CM5-LR ocean-atmospheric model under five scenarios: (i) the high-emissions RCP 8.5 W/m2; and (ii) melting scenarios consisting of four different experiments of freshwater discharge into the North Atlantic from Greenland’s meltwater (see DeFrance16 for details). We acknowledge that using a single GCM does not allow us to estimate inter-GCM variability in the resulting distribution models; however, the melting scenarios do only exist for IPSL-CM5-LR GCM. We applied as control scenario RCP 8.5 because melting scenarios would have been more complicated to support with lower emission scenarios. In addition, we are using well-designed opportunity experiments from ref. 11 and wanted to be consistent with their choice of RCP 8.5. Also, these experiments are based on CMIP5, which shows similar climate impact fingerprints than CMIP665. This might be explained by the fact that CMIP5 and CMIP6 are still relatively close, and that the main climatic effects of the AMOC are already well-represented by the climate dynamics in CMIP5.The four melting scenarios are equivalent to a sea-level rise of 0.5, 1.0, 1.5, and 3.0 meters above the current sea level, and these are named accordingly: Melting 0.5, Melting 1.0, Melting 1.5., and Melting 3.0. These AMOC scenarios are experiments that were superimposed to the RCP 8.5 scenario adding 0.11, 0.22, 0.34, and 0.68 Sv (1 Sv = 106 m3/s) coming from a freshwater release that starts in 2020 and finishes in 2070 (Anthoff et al.14). We obtained debiased bioclimatic variables11 under the five future scenarios for three consecutive time horizons: T1: 2030 (2030–2060); T2: 2050 (2050–2080); and T3: 2070 (2070–2100). The time horizons evaluated represent short, medium, and long terms in order to help decision-makers order conservation priorities.Ecological niche modelingAt their most basic, the algorithms used to construct species distribution models relate species occurrence records with climatic variables to create a climatic profile that can be projected onto other time periods and geographic regions66. The resulting models have proven useful in evaluating the impacts of climate change on biodiversity and to identify varying levels of vulnerability among species32,67,68. Here, we employed a multi-algorithm (ensemble) approach to construct species distribution models as implemented in the “biomod2” package67 in R69 (Supplementary Fig. 33). The underlying philosophy of ensemble modeling is that each model carries a true “signal” about the climate-occurrence relationships we aim to capture, but it also carries “noise” created by biases and uncertainties in the data and model structure32,67. By combining models created with different algorithms, ensemble models aim at capturing the true “signal” while controlling for algorithm-derived model differences; therefore, model uncertainty is accounted for during model construction (see Supplementary Material for further detail).Prior to modeling, we reduced the number of bioclimatic variables per species by estimating collinearity among present-day bioclimatic variables. We employed the “corrSelect” function of the package fuzzySim70 in R69, using a Pearson correlation threshold of 0.8 and variance inflation factors as criteria to select variables. Given the number of species evaluated and the ecological information scarcity, we did not select a set of variables based on ecological knowledge by each of the species modeled. Instead, for the variables pre-selection, we used the statistical approach described above that has been proven to give models with good performance71,72. We used seven algorithms with a good predictive performance (evaluated with the TSS and ROC statistics; Supplementary Fig. 1): Maxent (MAXENT.Phillips), Generalized Additive Models (GAM), Classification Trees Analysis (CTA), Artificial Neural Networks (ANN), Surface Range Envelope (SRE), Flexible Discriminant Analysis (FDA), and Random Forest (RF). Because occurrence datasets consisted of presence-only data, for each model, we randomly generated 10,000 pseudo-absences within the model calibration area; we gave presences and absences the same importance during the calibration process (BIOMOD’s prevalence = 0.5). For each species, we selected a calibration area (i.e., the accessible area or M)73 using a spatial intersection between a 4° buffer around species occurrences and the terrestrial ecoregions occupied by the species73 (Supplementary Fig. 33). The projected M (i.e., the area accessible for species in future scenarios) was defined using a 2° buffer around the present-day calibration area (M). By limiting the M, we incorporated information about dispersal and ecological limitations of each species into the modeling66. We did this to take into account a more realistic dispersal scenario given the velocity with which climatic changes are happening and because there are geographic and ecological barriers, which is the reason why we used ecoregions to limit our M. We assumed climatic niche conservatism across time; and inside the projected M we also assumed full dispersal. Consequently, inside the projected M, the evaluated species can win or lose suitable climatic conditions.We calibrated each algorithm using a random sample of 70% of occurrence records and evaluated the resulting models using the remaining 30% of records. To validate the predictive power of the ecological niche models, we used the True Skill Statistics (TSS) and the Receiver Operating Characteristics (ROC) and performed 10 replicates for every model, providing a tenfold internal cross-validation. To account for uncertainty, we constructed the ensemble models (seven algorithms × ten replicates) using a total consensus rule, where models from different algorithms were assembled using a weighted mean of replicates with an evaluation threshold of AUC  > 0.7 (Supplementary Fig. 1). However, as shown by the distribution of validation statistic in Supplementary Fig. 1, most ensemble models presented a very good predictive power (AUC  > 0.8). In some cases, modeling issues in some insular species required that we change the calibration area (M) to the entire country.We used the resulting ensemble models to project the potential distribution of each species under both current and future climatic conditions (Supplementary Fig. 34). We then examined the frequency in which different bioclimatic variables appeared to have the highest contribution during model construction for each species. The algorithms used (Maxent, GAM, CTA, ANN, SRE, FDA, and RF) identify these variables by iteratively testing combinations of all the available variables (i.e., those selected based on low correlation values) until reaching a set of variables that was most informative on the distribution of species; this set of variables had the highest predictive power of species occurrence. For every species, we retrieved the two variables with the largest model contribution (Supplementary Figs. 34 and 35).Species geographic rangeWe converted ensemble probability maps into binary maps of presence/absence using the TSS threshold; these binary maps reflect the distribution of climatic suitability of species, where values of 0 and 1 represent grid cells with non-suitable and suitable climates, respectively. In order to approximate the vulnerability of individual species to climate change, we estimated the temporal changes in the extent of the area of climatic suitability (geographic range) for every species relative to the present-day distribution. We estimated species’ geographic ranges by identifying and counting those grid cells with suitable climatic conditions (values of 1) in the present-day and under future scenarios. We then estimated the proportion of range changes through time, quantifying the proportion of grid cells either lost or gained for each species. This allowed us to estimate the proportion of species (by country and group) projected to have a complete loss of geographic ranges in the future.Species richness, differences in species richness, potential species hotspots (PSH), and temporal dissimilarityWe used binary maps to construct presence-absence matrices (PAM), which contain information on the presence (values of 1) or absence (values of 0) of species across grid cells. Using these PAMs, we estimated species richness (SR) as the sum of species present in each grid cell; to visualize SR across space, we generated 16 species richness maps corresponding to the present-day and the four future scenarios at each of the three temporal horizons. We used these maps to estimate and visualize temporal differences in species richness (ΔSR) over time by subtracting the estimated SR in the future from the current SR, for every grid cell; for visualization, we standardized SR per country to the range 0–1. We assumed full dispersal ability of species in all analyses, meaning that all suitable areas in the future had the same probability of being occupied, irrespective of the distance to the present-day distribution.By calculating species richness (SR) across grid cells, we defined Potential Species Hotspots (PSH) within each country as those grid cells with the highest levels of SR. For this, we defined the PSH by calculating the maximum present-day species richness (maxSR) observed in each country and then identified grid cells with richness values above a threshold of maxSR*0.6. Considering only those grid cells with a SR above this threshold, we estimated the geographic extent of PSH across time periods and scenarios and estimated changes to the extent of PSH relative to present-day conditions. Given that we use the threshold to define PSHs, we tested two additional thresholds (20 and 90%) to define and quantify the extent of PSHs. However, these additional results agree with the general trend. We chose not to base our threshold on the distribution of SR values (i.e., quantiles, median) due to the high proportion of grid cells with SR  More

Nelson G. From Candolle to croizat: comments on the history of biogeography. J Hist Biol. 1978;11:269–305.PubMed 
Article 
CAS 

Google Scholar 
Lomolino MV, Riddle BR, Whittaker RJ, Brown JH. Biogeography. Sunderland, MA: Sinauer Associates; 2005. p. 752Wang J, Soininen J, Zhang Y, Wang B, Yang X, Shen J. Contrasting patterns in elevational diversity between microorganisms and macroorganisms. J Biogeogr. 2011;38:595–603.Article 

Google Scholar 
Treseder KK, Maltz MR, Hawkins BA, Fierer N, Stajich JE, Mcguire KL. Evolutionary histories of soil fungi are reflected in their large-scale biogeography. Ecol Lett. 2014;17:1086–93.PubMed 
Article 

Google Scholar 
Meyer KM, Memiaghe H, Korte L, Kenfack D, Alonso A, Bohannan BJM. Why do microbes exhibit weak biogeographic patterns? ISME J. 2018;12:1404–13.PubMed 
PubMed Central 
Article 

Google Scholar 
Lindström ES, Langenheder S. Local and regional factors influencing bacterial community assembly. Environ Microbiol Rep. 2012;4:1–9.PubMed 
Article 

Google Scholar 
Ghiglione JF, Galand PE, Pommier T, Pedrós-Alió C, Maas EW, Bakker K, et al. Pole-to-pole biogeography of surface and deep marine bacterial communities. Proc Natl Acad Sci USA 2012;109:17633–8.PubMed 
PubMed Central 
Article 

Google Scholar 
Sul WJ, Oliver TA, Ducklow HW, Amaral-Zettlera LA, Sogin ML. Marine bacteria exhibit a bipolar distribution. Proc Natl Acad Sci USA 2013;110:2342–7.PubMed 
PubMed Central 
Article 

Google Scholar 
Sunagawa S, Coelho LP, Chaffron S, Kultima JR, Labadie K, Salazar G, et al. Structure and function of the global ocean microbiome. Science. 2015;348:1261359.PubMed 
Article 
CAS 

Google Scholar 
de Vargas C, Audic S, Henry N, Decelle J, Mahé F, Logares R, et al. Eukaryotic plankton diversity in the sunlit ocean. Science. 2015;348:1261605.PubMed 
Article 
CAS 

Google Scholar 
Milici M, Tomasch J, Wos-Oxley ML, Decelle J, Jáuregui R, Wang H. et al. Bacterioplankton biogeography of the Atlantic ocean: a case study of the distance-decay relationship. Front Microbiol. 2016;7:Article 590.PubMed 

Google Scholar 
Raes EJ, Bodrossy L, Van De Kamp J, Bissett A, Ostrowski M, Brown MV, et al. Oceanographic boundaries constrain microbial diversity gradients in the south pacific ocean. Proc Natl Acad Sci USA 2018;115:8266–75.Article 
CAS 

Google Scholar 
Wu W, Lu HP, Sastri A, Yeh YC, Gong GC, Chou WC, et al. Contrasting the relative importance of species sorting and dispersal limitation in shaping marine bacterial versus protist communities. ISME J. 2018;12:485–94.PubMed 
Article 

Google Scholar 
Vellend M. Conceptual synthesis in community ecology. Q Rev Biol. 2010;85:183–206.PubMed 
Article 

Google Scholar 
Hanson CA, Fuhrman JA, Horner-Devine MC, Martiny JBH. Beyond biogeographic patterns: Processes shaping the microbial landscape. Nat Rev Microbiol. 2012;10:497–506.PubMed 
Article 
CAS 

Google Scholar 
Nemergut DR, Schmidt SK, Fukami T, O’Neill SP, Bilinski TM, Stanish LF, et al. Patterns and processes of microbial community assembly. Microbiol Mol Biol Rev. 2013;77:342–56.PubMed 
PubMed Central 
Article 

Google Scholar 
Stegen JC, Lin X, Fredrickson JK, Chen X, Kennedy DW, Murray CJ, et al. Quantifying community assembly processes and identifying features that impose them. ISME J. 2013;7:2069–79.PubMed 
PubMed Central 
Article 

Google Scholar 
Schmidt TSB, Matias Rodrigues JF, Von Mering C. A family of interaction-adjusted indices of community similarity. ISME J. 2017;11:791–807.PubMed 
Article 

Google Scholar 
Zhou J, Ning D. Stochastic community assembly: does it matter in microbial ecology? Microbiol Mol Biol Rev. 2017;81:1–32.Article 

Google Scholar 
Djurhuus A, Port J, Closek CJ, Yamahara KM, Romero-maraccini O, Walz KR. et al. Evaluation of filtration and DNA extraction methods for environmental DNA biodiversity assessments across multiple trophic levels. Front Mar Sci. 2017;4:Article 314.Article 

Google Scholar 
Wang ZB, Sun YY, Li Y, Chen XL, Wang P, Ding HT, et al. Significant bacterial distance-decay relationship in continuous, well-connected southern ocean surface water. Micro Ecol. 2020;80:73–80.Article 
CAS 

Google Scholar 
Dlugosch L, Pohlein A, Wemheuer B, Pfeiffer B, Badewien T, Daniel R, et al. Significance of gene variants for the functional biogeography of the near-surface Atlantic Ocean microbiome. Nat Commun. 2022;13:456.PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Lozupone C, Knight R. UniFrac: A new phylogenetic method for comparing microbial communities. Appl Environ Microbiol. 2005;71:8228–35.PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Logares R, Deutschmann IM, Junger PC, Giner CR, Krabberød AK, Schmidt TSB, et al. Disentangling the mechanisms shaping the surface ocean microbiota. Microbiome. 2020;8:55.PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Doblin MA, Petrou K, Sinutok S, Seymour JR, Messer LF, Brown MV, et al. Nutrient uplift in a cyclonic eddy increases diversity, primary productivity and iron demand of microbial communities relative to a western boundary current. PeerJ. 2016;4:e1973.PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Polovina JJ, Howell E, Kobayashi DR, Seki MP. The transition zone chlorophyll front, a dynamic global feature defining migration and forage habitat for marine resources. Prog Oceanogr. 2001;49:469–83.Article 

Google Scholar 
Karl DM, Church MJ. Ecosystem structure and dynamics in the north pacific subtropical gyre: new views of an old ocean. Ecosystems. 2017;20:433–57.Article 

Google Scholar 
Mestre M, Ruiz-González C, Logares R, Duarte CM, Gasol JM, Sala MM. Sinking particles promote vertical connectivity in the ocean microbiome. Proc Natl Acad Sci USA 2018;115:6799–807.Article 
CAS 

Google Scholar 
Balmonte JP, Simon M, Giebel HA, Arnosti C. A sea change in microbial enzymes: Heterogeneous latitudinal and depth-related gradients in bulk water and particle-associated enzymatic activities from 30°S to 59°N in the Pacific Ocean. Limnol Oceanogr. 2021;66:3489–507.Article 
CAS 

Google Scholar 
Giebel H-A, Arnosti C, Badewien TH, Bakenhus I, Balmonte JP, Billerbeck S. et al. Microbial growth and organic matter cycling in the Pacific Ocean along a latitudinal transect between subarctic and subantarctic waters. Front Mar Sci. 2021;8:Article 764383.Article 

Google Scholar 
Milici M, Tomasch J, Wos-Oxley ML, Wang H, Jáuregui R, Camarinha-Silva A, et al. Low diversity of planktonic bacteria in the tropical ocean. Sci Rep. 2016;6:19054.PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Longhurst AR. Ecological geography of the sea. San Diego, USA: Academic Press; 2007.Parada AE, Needham DM, Fuhrman JA. Every base matters: Assessing small subunit rRNA primers for marine microbiomes with mock communities, time series and global field samples. Environ Microbiol. 2016;18:1403–14.PubMed 
Article 
CAS 

Google Scholar 
Milke F, Sanchez-Garcia S, Dlugosch L, McNichol J, Fuhrman J, Simon M. et al. Composition and biogeography of pro- and eukaryotic communities in the Atlantic Ocean: primer choice matters. Front Microbiol. 2022;13:Article 895875.PubMed 
Article 

Google Scholar 
Vaulot D, Geisen S, Mahé F, Bass D. pr2-primers: An 18S rRNA primer database for protists. Mol Ecol Resour. 2022;22:168–79.PubMed 
Article 
CAS 

Google Scholar 
Yeh YC, McNichol J, Needham DM, Fichot EB, Berdjeb L, Fuhrman JA. Comprehensive single-PCR 16S and 18S rRNA community analysis validated with mock communities, and estimation of sequencing bias against 18S. Environ Microbiol. 2021;23:3240–50.PubMed 
Article 
CAS 

Google Scholar 
Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, et al. The SILVA ribosomal RNA gene database project: Improved data processing and web-based tools. Nucleic Acids Res. 2013;41:590–6.Article 
CAS 

Google Scholar 
Guillou L, Bachar D, Audic S, Bass D, Berney C, Bittner L, et al. The Protist Ribosomal Reference database (PR2): a catalog of unicellular eukaryote Small Sub-Unit rRNA sequences with curated taxonomy. Nucleic Acids Res. 2013;41:597–604.Article 
CAS 

Google Scholar 
Bolyen E, Rideout JR, Dillon MR, Bokulich NA, Abnet CC, Al-Ghalith GA, et al. Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2 (Nature Biotechnology, (2019), 37, 8, (852-857), 10.1038/s41587-019-0209-9). Nat Biotechnol. 2019;37:1091.PubMed 
Article 
CAS 

Google Scholar 
Callahan BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AJA, Holmes SP. DADA2: High-resolution sample inference from Illumina amplicon data. Nat Methods. 2016;13:581–3.PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Friedman J, Alm EJ. Inferring correlation networks from genomic survey data. PLoS Comput Biol. 2012;8:e1002687.PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Bodenhofer U, Bonatesta E, Horejš-Kainrath C, Hochreiter S. Msa: an R package for multiple sequence alignment. Bioinformatics. 2015;31:3997–9.PubMed 
CAS 

Google Scholar 
Kaufman L, Rousseeuw PJ. Finding groups in data: an introduction to cluster analysis. Hoboken NJ, USA: John Wiley & Sons; 2009.Pruesse E, Peplies J, Glöckner FO. SINA: Accurate high-throughput multiple sequence alignment of ribosomal RNA genes. Bioinformatics. 2012;28:1823–9.PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Price MN, Dehal PS, Arkin AP. FastTree 2 – approximately maximum-likelihood trees for large alignments. PLoS ONE. 2010;5:e9490.PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Losos JB. Phylogenetic niche conservatism, phylogenetic signal and the relationship between phylogenetic relatedness and ecological similarity among species. Ecol Lett. 2008;11:995–1003.PubMed 
Article 

Google Scholar 
Stegen JC, Lin X, Konopka AE, Fredrickson JK. Stochastic and deterministic assembly processes in subsurface microbial communities. ISME J. 2012;6:1653–64.PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Fine PVA, Kembel SW. Phylogenetic community structure and phylogenetic turnover across space and edaphic gradients in western Amazonian tree communities. Ecography. 2011;34:552–65.Article 

Google Scholar 
Chase JM, Kraft NJB, Smith KG, Vellend M, Inouye BD. Using null models to disentangle variation in community dissimilarity from variation in α-diversity. Ecosphere. 2011;2:1–11.Article 

Google Scholar 
NASA Goddard Space Flight Center, Ocean Ecology Laboratory OBPG. Moderate-resolution Imaging Spectroradiometer (MODIS) aqua chlorophyll data. https://oceancolor.gsfc.nasa.gov/data/10.5067/AQUA/MODIS/L3B/CHL/2018/. Accessed 13 Nov 2020.Pommier T, Douzery EJP, Mouillot D. Environment drives high phylogenetic turnover among oceanic bacterial communities. Biol Lett. 2012;8:562–6.PubMed 
PubMed Central 
Article 

Google Scholar 
Giovannoni SJ, Cameron Thrash J, Temperton B. Implications of streamlining theory for microbial ecology. ISME J. 2014;8:1553–65.PubMed 
PubMed Central 
Article 

Google Scholar 
Sañudo-Wilhelmy SA, Gómez-Consarnau L, Suffridge C, Webb EA. The role of B vitamins in marine biogeochemistry. Ann Rev Mar Sci. 2014;6:339–67.PubMed 
Article 

Google Scholar 
Morris JJ, Lenski RE, Zinser ER. The black queen hypothesis: evolution of dependencies through adaptive gene loss. MBio. 2012;3:e00036–12.PubMed 
PubMed Central 
Article 

Google Scholar 
Carini P, Campbell EO, Morré J, Sañudo-Wilhelmy SA, Cameron Thrash J, Bennett SE, et al. Discovery of a SAR11 growth requirement for thiamin’s pyrimidine precursor and its distribution in the Sargasso Sea. ISME J. 2014;8:1727–38.PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Wienhausen G, Bruns S, Sultana S, Dlugosch L, Groon L, Wilkes H, et al. The overlooked role of a biotin precursor for marine bacteria – desthiobiotin as an escape route for biotin auxotrophy. ISME J. 2022. https://doi.org/10.1038/s41396-022-01304-w.Biller SJ, Coe A, Chisholm SW. Torn apart and reunited: Impact of a heterotroph on the transcriptome of Prochlorococcus. ISME J. 2016;10:2831–43.PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Sokolovskaya OM, Shelton AN, Taga ME. Sharing vitamins: cobamides unveil microbial interactions. Science. 2020;369:eaba0165.PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Wienhausen G, Dlugosch L, Jarling R, Wilkes H, Giebel H-A, Simon M. Availability of vitamin B12 and its lower ligand intermediate a-ribazole impact prokaryotic and protist communities in oceanic systems. ISME J. 2022;16:2002–14.PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Reintjes G, Arnosti C, Fuchs B, Amann R. Selfish, sharing and scavenging bacteria in the Atlantic Ocean: a biogeographical study of bacterial substrate utilisation. ISME J. 2019;13:1119–32.PubMed 
Article 
CAS 

Google Scholar 
Bertrand EM, McCrow JP, Moustafa A, Zheng H, McQuaid JB, Delmont TO, et al. Phytoplankton-bacterial interactions mediate micronutrient colimitation at the coastal Antarctic sea ice edge. Proc Natl Acad Sci USA 2015;112:9938–43.PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Amin SA, Hmelo LR, Van Tol HM, Durham BP, Carlson LT, Heal KR, et al. Interaction and signalling between a cosmopolitan phytoplankton and associated bacteria. Nature. 2015;522:98–101.PubMed 
Article 
CAS 

Google Scholar 
Shibl AA, Isaac A, Ochsenkühn MA, Cárdenas A, Fei C, Behringer G, et al. Diatom modulation of select bacteria through use of two unique secondary metabolites. Proc Natl Acad Sci USA 2020;117:27445–55.PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Villarino E, Watson JR, Chust G, Woodill AJ, Klempay B, Jonsson B, et al. Global beta diversity patterns of microbial communities in the surface and deep ocean. Glob Ecol Biogeogr. 2022;00:1–14.
Google Scholar 
Cravatte S, Kestenare E, Marin F, Dutrieux P, Firing E. Subthermocline and intermediate zonal currents in the tropical Pacific Ocean: Paths and vertical structure. J Phys Oceanogr. 2017;47:2305–24.Article 

Google Scholar 
Cho BC, Azam F. Major role of bacteria in biogeochemical fluxes in the ocean’s interior. Nature. 1988;332:441–3.Article 
CAS 

Google Scholar 
Salazar G, Cornejo-Castillo FM, Benítez-Barrios V, Fraile-Nuez E, Álvarez-Salgado XA, Duarte CM, et al. Global diversity and biogeography of deep-sea pelagic prokaryotes. ISME J. 2016;10:596–608.PubMed 
Article 

Google Scholar 
Delmont TO, Kiefl E, Kilinc O, Esen OC, Uysal I, Rappé MS, et al. Single-amino acid variants reveal evolutionary processes that shape the biogeography of a global SAR11 subclade. Elife. 2019;8:e46497.PubMed 
PubMed Central 
Article 

Google Scholar 
Hillebrand H. On the generallity of the latutinal diversity gradient. Am Nat. 2004;163:192–211.PubMed 
Article 

Google Scholar  More

Logan, C. A., Dunne, J. P., Ryan, J. S., Baskett, M. L. & Donner, S. D. Nat. Clim. Chang. 11, 537–542 (2021).Article 

Google Scholar 
Cook, C. N. & Sgrò, C. M. Conserv. Biol. 31, 501–512 (2017).Article 

Google Scholar 
Gonzalez, A., Ronce, O., Ferriere, R. & Hochberg, M. E. Phil. Trans. R. Soc. Lond. B 368, 1–8 (2013).
Google Scholar 
Kovach, R. P., Gharrett, A. J. & Tallmon, D. A. Proc. R. Soc. Lond. B 279, 3870–3878 (2012).
Google Scholar 
Bonnet, T. et al. Science 376, 1012–1016 (2022).CAS 
Article 

Google Scholar 
Norberg, J. et al. Nat. Clim. Chang. 2, 747–751 (2012).Article 

Google Scholar 
Torda, G. et al. Nat. Clim. Chang. 7, 627–636 (2017).Article 

Google Scholar 
Catullo, R. A., Llewelyn, J., Phillips, B. L. & Moritz, C. C. Curr. Biol. 29, R996–R1007 (2019).CAS 
Article 

Google Scholar 
Keppel, G. et al. Glob. Ecol. Biogeogr. 21, 393–404 (2012).Article 

Google Scholar 
Vos, C. C. et al. J. Appl. Ecol. 45, 1722–1731 (2008).Article 

Google Scholar 
Isaak, D. J. et al. Glob. Change Biol. 21, 2540–2553 (2015).Article 

Google Scholar 
Beyer, H. L. et al. Conserv. Lett. 11, e12587 (2018).Article 

Google Scholar 
Tingley, M. W., Estes, L. D. & Wilcove, D. S. Nature 500, 271–272 (2013).CAS 
Article 

Google Scholar 
Schindler, D. E., Armstrong, J. B. & Reed, T. E. Front. Ecol. Environ. 13, 257–263 (2015).Article 

Google Scholar 
Cornwell, B. et al. eLife 10, e64790 (2021).CAS 
Article 

Google Scholar 
National Academies. of Sciences Engineering & Medicine. A Decision Framework for Interventions to Increase the Persistence and Resilience of Coral Reefs. (The National Academies Press, 2019).Palumbi, S. R., Barshis, D. J., Traylor-Knowles, N. & Bay, R. A. Science 344, 895–898 (2014).CAS 
Article 

Google Scholar 
Matz, M. V., Treml, E. A. & Haller, B. C. Glob. Change Biol. 26, 3473–3481 (2020).Article 

Google Scholar 
Bay, R. A. & Palumbi, S. R. Curr. Biol. 24, 2952–2956 (2014).CAS 
Article 

Google Scholar 
Donovan, M. K. et al. Science 372, 977–980 (2021).CAS 
Article 

Google Scholar 
Anthony, K. et al. Nat. Ecol. Evol. 1, 1420–1422 (2017).Article 

Google Scholar 
Morrison, T. H. et al. Nature 573, 333–336 (2019).CAS 
Article 

Google Scholar 
van Oppen, M. J. H., Oliver, J. K., Putnam, H. M. & Gates, R. D. Proc. Natl Acad. Sci. USA 112, 2307–2313 (2015).Article 

Google Scholar 
DeFilippo, L. B. et al. Ecol. Appl. https://doi.org/10.1002/eap.2650 (2022).Steneck, R. S. et al. Front. Mar. Sci. 6, 265 (2019).Article 

Google Scholar 
Dixon, G. B. et al. Science 348, 1460–1462 (2015).CAS 
Article 

Google Scholar 
McManus, L. C. et al. Glob. Change Biol. 27, 4307–4321 (2021).CAS 
Article 

Google Scholar 
Kleypas, J. A. et al. Glob. Change Biol. 22, 3539–3549 (2016).Article 

Google Scholar 
McManus, L. C. et al. Ecology 102, e03381 (2021).Article 

Google Scholar 
Walsworth, T. E. et al. Nat. Clim. Chang. 9, 632–636 (2019).Article 

Google Scholar  More

Urine spraying and scraping as potential scent-markingThe urine spraying and the scraping were reported in other felids6,20,21. In this study, only half of the other excretion instances were accompanied by sniffing, whereas almost all urine spraying and scraping events were accompanied by sniffing, indicating that these are scent-markings. The sniffing was also often observed immediately before urine spraying and scraping. Given the significant association of sniffing before excretion, especially with regard to the scraping, the presence or absence of a scent on the object was thought to be a trigger.Furthermore, during the scraping, liquid secretions thought to originate from the anal glands, were released. Domestic cats have scent glands in the anal sac22. The presence of secretions from the anal sac has also been confirmed in not only tigers, lions (Panthera leo), and bobcats (Lynx rufus), but also in cheetahs1,6,23; however, this study was the first to investigate their role in excretion. Generally, secretions are considered to be caused by health problems or estrus, but in this study, none of the individuals had health problems, and all secretions were observed only in males. Therefore, it was thought that the secretion was produced by the scent glands and contributed to a stronger smell than only urine and feces.Variations based on sexUrine spraying was observed only in adult males and females, and was more frequent in males, as reported in other felids4,5,6,9,24. In wild cheetahs, although urine spraying and scraping have been observed as scent-making, the frequency of scent-marking is known to be substantially higher in territorial than in non-territorial males and in females15,16,25, and the marking locations are concentrated in the core area of the male territories16. The territories of a single male cheetah or a male group are relatively small and exclusive, whereas the relatively large home ranges of non-territorial males (also known as “floaters”) overlap with each other and with those of females15,16. A male’s home range is also larger than that of a female15,16,26,27. Male cheetahs rarely encounter other males because they communicate via marking posts28. Given these reports, the frequent urine spraying by males may help prevent encounters between males. In addition, observations of captive cheetahs have shown a significantly positive correlation between urinary spraying frequency and fecal estradiol content in female cheetahs19. Therefore, as Cornhill and Kerley24 mentioned, female urine spraying is caused by estrus, and male urine spraying is intended as a home range marker for other males or as a sign for females.The action of scraping using the hind paws has been reported to occur in both males and females in servals, lions, tigers, black-footed cats, etc.2,5,6,7,29; however, this behavior was only observed in adult males in this study. Sunquist and Sunquist3 reported that female cheetahs also perform the scraping. In this study, we only recorded observations when the cheetahs were released in the outdoor enclosures, and not when they were in the indoor facilities. In 43.6% of the scraping events, the males excreted feces. During the observation period, the females defecated in the indoor facilities, and no defecation was observed in the outdoor enclosures. It is possible that no scraping action was observed among the females because defecation was not observed in the outdoor enclosure. In indoor facilities, the cheetahs were in a completely monopolized enclosure; hence, the females defecated in their own spaces. There was a difference in the defecation sites and frequency of scraping between the males and females; this was attributed to the sex difference in scent-marking.Differences in target height for each behaviorUrine spraying was frequently done on objects approximately 170 cm or higher, such as walls or fences, standing trees, and stumps, whereas scraping was observed on low-lying objects on the ground, such as a straw pile approximately 3 cm high and a fallen tree that was 10–50 cm high. In other words, the cheetah engaged in urine spraying and scraping depending on the object nearby. This might indicate the functional role of these behaviors. This is consistent with previous findings of urine spraying by tigers being more frequent in wooded forests than in grasslands, with few prominent objects, and scraping being more common in the latter6. In addition, in a study that investigated the place where the smell of the urine of domestic cats is likely to remain, the smell persisted for a long time on rough surfaces, areas covered with moss, and overhanging slopes30. Even for cheetahs living in the savanna woodlands, where there are comparatively fewer upright objects than in the habitat of felids living in the forest, increasing the chances of transmitting information via not only urine spraying but also by the scraping might be more important. On the other hand, in their natural habitat, there are some large carnivores like lions and leopards (Panthera pardus). Wild cheetahs tend not to visit the sites where such carnivores’ scent-mark is present31, suggesting that they might confine their marking to specific sites devoid of other carnivores’ scent. Further research is needed to determine how wild cheetahs use urine spraying and scraping. In this study, scraping was frequently observed even on tall stumps and rocks if they were within the cheetahs’ reach. Scraping by wild cheetahs has been also observed on trees32. Zoos other than Zoo C had few prominent horizontal objects. Therefore, the presence of straw piles, fallen trees, stumps, and rocks may have elicited the scraping.Differences in housing conditionsIn zoos C and D, where animals shared enclosures, the frequency of both urine spraying and scraping by males was higher than in the males in the monopolized enclosures. They possibly showed a more frequent scent-marking to strengthen their home range claims when sharing the exhibition space15. Regarding the scraping, Zoo C had at least four low and horizontal objects (straw piles, fallen tree, stones, and rocks), and scraping was frequently observed. As mentioned above, the placement of objects might have elicited the scraping.In this study, the frequency of urine spraying decreased when the submissive individual (Male 17) was released in the enclosure where the dominant individual (Male 13) was previously released. Among wild cheetahs, territorial males have been reported to mark their territories more often than non-territorial males17,25. Therefore, the difference in the number of markings is considered to be related to whether or not the target individual is within the territory, and it is highly possible that the dominant/submissive relationship between males at that location has an effect on marking.Function of scraping using hind pawsOther felid studies have reported scraping in tigers, pumas, jaguars, clouded leopards, and small felids6,10,20,21,32,33; however, there are fewer studies on different types of scraping. In certain species, such as jaguars and pumas, scraping using hind paws is more frequent than urine spraying33. From this study, the use of secretions was confirmed in the scraping, and it was considered to be a significant marking of the cheetah.The possible functions of scraping include: (1) dispersing the smell of excrement, (2) placing the smell of excrement on the hindlegs, (3) smearing the objects with excrement, and (4) adding the scent of the hind paws. Domestic cats are known to cover their feces with soil34; however, in this study, the cheetahs did not cover the feces with soil and were not observed to scrap only after excretion. Therefore, scraping using hind paws was not meant for concealing urine and/or feces. The results of this study suggest that the scrapings were mostly performed during and after excretion for any of the aforementioned functions. However, 43.2% of the observed scraping events were performed before excretion, and in these cases, the functions 1–3 did not apply, since we did not observe the feces being crushed by scraping the hind paws. As for function 4, domestic cats have sweat glands on the soles of their feet that are thought to retain their smell35. Therefore, the sweat glands on the soles of the feet of the cheetahs possible retain the smell of the hind paws as well. Schaller36 reported that among tigers, scraping on the grassland was exhibited by scratches in the grass and exposure of the ground, creating a visual effect. In the case of cheetahs, scraping may have the function of creating grooves and ridges on the ground to enhance the visual effect; however, the formation of grooves and ridges were not observed in this study. In certain cases, they scraped against trees and stones. Because trees and stones are not easily deformed, it is hard to say whether the visual effect was enhanced by scraping with their hind paws.Scraping has been reported in other felids; however, the movement of the hindlimbs is not uniform. For example, in the case of bobcats, behaviors such as kicking back on the ground with no surrounding objects and scattering of soil have been observed during scrapings20. The snow leopard slowly moves its hindlimbs on the ground near the rocks, exposing the ground; in fact, Schaller29 observed a tiger scraping its hind paws to create a pile of soil [37; Kinoshita, personal communication: Online Resource 3; Scraping of snow leopard]. The movement of urine spraying also varies among species. For example, bobcats sometimes squat and urinate on the ground20, and snow leopards rub their cheeks against the target object and then spray urine9, but cheetahs do not rub their face before urine spraying. Hence, even in the same behavior of “spraying/scraping,” the actions differ. Because felids are widely distributed in various environments, such differences in movements are possibly related to differences in habitat and behavioral functions.In conclusion, urine spraying and scraping using hind paws were considered scent-markings because they were more strongly associated with sniffing than other excretion. Both behaviors were also observed only in adults; however, urine spraying was confirmed in both sexes and was more frequent in males than in females, whereas scraping was observed only in males. Also, the frequencies of both behaviors were significantly higher in males kept in shared enclosures containing other individuals than in males kept in monopolized enclosures, while there was no difference in the frequencies among females. Hence, there were sex differences in these scent-markings possibly because of the difference in the sociality between the sexes even in captivity; the frequency of scent-markings was affected by the living environment including the number of target objects; urine spraying was frequently done on tall objects such as walls or fences, whereas scraping was more commonly done on low-lying objects near the ground, such as straw piles. To our knowledge, this study is the first to confirm that during the scraping a liquid other than feces and urine was secreted, presumably from the anal glands. Taken together, the results can serve to enhance our knowledge regarding the behavior of cheetahs, help improve management of these animals in captivity as well as breeding and animal welfare ex situ conservation, and help elucidate the kind of habitat that should be preserved for the in situ conservation of cheetahs. More

Plants increase their freezing resistance upon exposure to low temperatureThe freezing resistance (LT50 values) was found to vary ranging from − 6.9 °C (14-August-2017) to − 31.7 °C (04-November-2018) over the course of study period. The freezing resistance of leaves recorded during the 12 sampling time-points has been provided in Table 1 (also see39). The overlap of confidence intervals around the mean was examined for comparison of LT50 values for the different sampling time-points. Significant differences in freezing resistance were observed across the sampling time-points (Table 1). Leaves of R. anthopogon collected during summer [July and August (Air temperature and photoperiod was about 9.6 °C and 13 h day−1 respectively)] showed marginal resistance to freezing (LT50: − 7 °C) and thus, are more susceptible to freezing damage. Further, as the ambient air temperature and photoperiod decreased towards the end of growing season (i.e., October and November 2017 with air temperature and photoperiod of about − 1.1 °C and 10.5 h day−1 respectively), the plants acquired the highest freezing resistance (LT50: − 30 °C). Interestingly, a sharp increase in freezing resistance (− 29.4 °C) was observed in September 2018, when the daily mean air temperature decreased below 0 °C due to sudden snowfall (Supplementary Fig. S2). Comparison of LT50 values of all the leaf samples of R. anthopogon showed that cold de-acclimation occurred after the snowmelt during early spring in June (LT50: − 13.4 °C) with an increase in air temperature and photoperiod. These results demonstrated that R. anthopogon plants exhibit lowered freezing resistance during the warmer months [hence, these time-periods were referred as non-acclimation (NA)], progressively develop greater freezing resistance during the onset of winter season (hence, referred as cold acclimation) followed by an intermediate level of freezing resistance during the spring [hence, these time-periods were referred as de-acclimation (DA)].Table 1 The estimates of LT50, calculated by fitting sigmoidal curve to electrolyte leakage values of temperature treatments, recorded for leaves collected during the different sampling time-points (from August 22, 2017 to September 18, 2018).Full size tableDuring the acclimation period (i.e., late in the growing season), plants acquired the highest resistance to freezing (Fig. 1). The low electrolyte leakage (= high freezing resistance) observed during this period might be due to changes in cell wall properties (such as increase in lignification and suberization of cell walls), which provide resistance to diffusion of electrolytes from cells of the leaves to the extracellular water47. Moreover, high freezing resistance may also be attributed to high leaf toughness and sclerophyllous habit of this evergreen species48. Further, it was found that freezing resistance was the lowest during mid-summer period. This pattern could be explained by a trade-of between plant growth rates and freezing resistance, where warmer temperatures favour plant allocation to growth49. These observations corroborated well with earlier reports that showed a rapid increase in ‘freezing resistance’ during the transition from summer to early winter and vice versa50.Figure 1LT50 [black point (with solid fill) on the curve] calculated by fitting sigmoidal curve to relative electrolyte leakage (REL %) values recorded during the three different acclimation phases. GOF indicates ‘goodness of fit’ test values for the fitted sigmoidal curves.Full size imagePhotosynthetic rates are higher during non-acclimation and de-acclimation periodIt was found that PN of R. anthopogon varied in the range from 8.336 to 17.64 μmol(CO2)m−2 s−1 and E from 2.281 to 4.912 mol(H2O)m−2 s−1, throughout its growing season. The Gs of leaves was estimated to be in the range from 0.110 to 0.265 mol (H2O) m−2 s−1. WUE, a ratio of PN and E, varied between 52.21 and 87.68 (Table 2). The gas exchange parameters of R. anthopogon varied significantly among the sampling time-points [referred to here as different acclimation phases of the growing period of evergreen shrub (Fig. 2, Table 3)]. In particular, PN was significantly lower on 18-September-2018 (referred as cold acclimation phase), whereas it was higher on 31-August-2018 and 15-June-2018 (referred as NA and DA phases, respectively). Similarly, Gs of leaves was significantly lower during cold acclimation in comparison to the rest of the acclimation phases (i.e., NA and DA). Further, WUE was significantly higher during cold acclimation, while it was lower during both NA and DA (p ≤ 0.05) (Fig. 2).Table 2 Variability in leaf gas exchange parameters of R. anthopogon during the different acclimation phases (NA = Non-acclimation, LA = Late cold acclimation and DA = De-acclimation).Full size tableFigure 2Variability in leaf gas exchange parameters of R. anthopogon during the three acclimation phases [i.e., Non-acclimation (31 August, 2018), Cold acclimation (18 September, 2018) and De-acclimation (15 June, 2018)]. Different alphabets (a, b, c) represent statistically significant values (p  More

Costanza, R. et al. Changes in the global value of ecosystem services. Glob. Environ. Change 26, 152–158 (2014).Article 

Google Scholar 
Mittermeier, R. A. et al. Wilderness and biodiversity conservation. Proc. Natl. Acad. Sci. USA 100, 10309–10313 (2003).CAS 
Article 

Google Scholar 
Dirzo, R. & Raven, P. H. Global state of biodiversity and loss. Annu. Rev. Env. Resour. 28, 137–167 (2003).Article 

Google Scholar 
Marengo, J. A. et al. Changes in climate and land use over the amazon region: current and future variability and trends. Front. Earth Sci. 6, 1–21 (2018).Article 

Google Scholar 
Anderson-Teixeira, K. J. et al. Climate-regulation services of natural and agricultural ecoregions of the Americas. Nat. Clim. Change 2, 177–181 (2012).Article 

Google Scholar 
Marengo, J. A. et al. The drought of Amazonia in 2005. J. Clim. 21, 495–516 (2008).Article 

Google Scholar 
Lewis, S. L., Brando, P. M., Phillips, O. L., Van Der Heijden, G. M. F. & Nepstad, D. The 2010 Amazon drought. Science 331, 554 (2011).CAS 
Article 

Google Scholar 
Jiménez-Muñoz, J. C. et al. Record-breaking warming and extreme drought in the Amazon rainforest during the course of El Niño 2015–2016. Sci. Rep. 6, 33130 (2016).Article 
CAS 

Google Scholar 
Phillips, O. L. et al. Drought sensitivity of the amazon rainforest. Science 323, 1344–1347 (2009).Koren, G. et al. Widespread reduction in sun-induced fluorescence from the Amazon during the 2015/2016 El Niño. Philos. Trans. R. Soc. Lond. B Biol. Sci 373, 20170408 (2018).Article 
CAS 

Google Scholar 
Feldpausch, T. R. et al. Amazon forest response to repeated droughts. Glob. Biogeochem. Cycles 30, 964–982 (2016).CAS 
Article 

Google Scholar 
Sousa, T. R. et al. Palms and trees resist extreme drought in Amazon forests with shallow water tables. J. Ecol. 108, 2070–2082 (2020).CAS 
Article 

Google Scholar 
Barros, F. et al. Hydraulic traits explain differential responses of Amazonian forests to the 2015 El Niño-induced drought. New Phytol. 223, 1253–1266 (2019).CAS 
Article 

Google Scholar 
Magney, T. S. et al. Mechanistic evidence for tracking the seasonality of photosynthesis with solar-induced fluorescence. Proc. Natl Acad. Sci. USA https://doi.org/10.1073/pnas.1900278116 (2019).Ciemer, C. et al. Higher resilience to climatic disturbances in tropical vegetation exposed to more variable rainfall. Nat. Geosci. 12, 174–179 (2019).CAS 
Article 

Google Scholar 
Gloor, E. et al. Tropical land carbon cycle responses to 2015/16 El Niño as recorded by atmospheric greenhouse gas and remote sensing data. Philos. Trans. R. Soc. B 373, 20170302 (2018).Article 
CAS 

Google Scholar 
Jiménez-Muñoz, J. C., Sobrino, J. A., Mattar, C. & Malhi, Y. Spatial and temporal patterns of the recent warming of the Amazon forest. J. Geophys. Res. Atmos. 118, 5204–5215 (2013).Article 

Google Scholar 
Choat, B. et al. Global convergence in the vulnerability of forests to drought. Nature 491, 752–755 (2012).CAS 
Article 

Google Scholar 
Esquivel-Muelbert, A. et al. Seasonal drought limits tree species across the Neotropics. Ecography 60, 12 (2016).
Google Scholar 
Fisher, R. A., Williams, M., de Lourdes Ruivo, M., de Costa, A. L. & Meir, P. Evaluating climatic and soil water controls on evapotranspiration at two Amazonian rainforest sites. Agric. For. Meteorol. 148, 850–861 (2008).Article 

Google Scholar 
Marthews, T. R. et al. High-resolution hydraulic parameter maps for surface soils in tropical South America. Geosci. Model Dev. 7, 711–723 (2014).Article 

Google Scholar 
Esteban, E. J. L., Castilho, C. V., Melgaço, K. L. & Costa, F. R. C. The other side of droughts: wet extremes and topography as buffers of negative drought effects in an Amazonian forest. New. Phytol. 229, 1995–2006 (2021).CAS 
Article 

Google Scholar 
Castro, A. O. et al. OCO-2 solar-induced chlorophyll fluorescence variability across ecoregions of the amazon basin and the extreme drought effects of El Niño (2015–2016). Remote Sens. 12, 1202 (2020).Article 

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

Google Scholar 
Sombroek, W. Spatial and temporal patterns of amazon rainfall. Ambio 30, 388–396 (2001).CAS 
Article 

Google Scholar 
Quesada, C. A. et al. Basin-wide variations in Amazon forest structure and function are mediated by both soils and climate. Biogeosciences 9, 2203–2246 (2012).Fan, Y., Li, H. & Miguez-Macho, G. Global patterns of groundwater table depth. Science 339, 940–943 (2013).CAS 
Article 

Google Scholar 
Joetzjer, E., Douville, H., Delire, C. & Ciais, P. Present-day and future Amazonian precipitation in global climate models: CMIP5 versus CMIP3. Clim. Dyn. 41, 2921–2936 (2013).Article 

Google Scholar 
Schietti, J. et al. Vertical distance from drainage drives floristic composition changes in an Amazonian rainforest. Plant. Ecol. Divers. 7, 241–253 (2014).Oliveira, R. S. et al. Embolism resistance drives the distribution of Amazonian rainforest tree species along hydro‐topographic gradients. New Phytol. 221, 1457–1465 (2018).Fyllas, N. M. et al. Basin-wide variations in foliar properties of Amazonian forest: phylogeny, soils and climate. Biogeosciences 6, 2677–2708 (2009).Sterck, F., Markesteijn, L., Schieving, F. & Poorter, L. Functional traits determine trade-offs and niches in a tropical forest community. PNAS 108, 20627–20632 (2011).CAS 
Article 

Google Scholar 
Oliveira, R. S. et al. Linking plant hydraulics and the fast–slow continuum to understand resilience to drought in tropical ecosystems. New Phytol. 230, 904–923 (2021).Article 

Google Scholar 
Guillemot, J. et al. Small and slow is safe: On the drought tolerance of tropical tree species. Glob. Chang. Biol. 28, 2622–2638 (2022).CAS 
Article 

Google Scholar 
DeSoto, L. et al. Low growth resilience to drought is related to future mortality risk in trees. Nat. Commun. 11, 545 (2020).CAS 
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 
Article 

Google Scholar 
de Almeida Castanho, A. D. et al. Changing Amazon biomass and the role of atmospheric CO2 concentration, climate, and land use. Glob. Biogeochem. Cycles 30, 18–39 (2016).Article 
CAS 

Google Scholar 
Friedlingstein, P. et al. Global carbon budget 2020. Earth Syst. Sci. Data 12, 3269–3340 (2020).Article 

Google Scholar 
Sitch, S. et al. Evaluation of ecosystem dynamics, plant geography and terrestrial carbon cycling in the LPJ dynamic global vegetation model. Glob. Chang. Biol. 9, 161–185 (2003).Article 

Google Scholar 
Lathière, J. et al. Impact of climate variability and land use changes on global biogenic volatile organic compound emissions. Atmos. Chem. Phys. 6, 2129–2146 (2006).Article 

Google Scholar 
Galbraith, D. et al. Multiple mechanisms of Amazonian forest biomass losses in three dynamic global vegetation models under climate change. New Phytol. 187, 647–65 (2010).Article 

Google Scholar 
Johnson, M. O. et al. Variation in stem mortality rates determines patterns of above-ground biomass in Amazonian forests: implications for dynamic global vegetation models. Glob. Chang. Biol. 22, 3996–4013 (2016).Article 

Google Scholar 
Thonicke, K. et al. Simulating functional diversity of European natural forests along climatic gradients. J. Biogeogr. 47, 1069–1085 (2020).Article 

Google Scholar 
Feldpausch, T. R. et al. Height-diameter allometry of tropical forest trees. Biogeosciences 8, 1081–1106 (2011).Article 

Google Scholar 
Feldpausch, T. R. et al. Tree height integrated into pantropical forest biomass estimates. Biogeosciences 9, 3381–3403 (2012).Article 

Google Scholar 
Potapov, P. et al. The last frontiers of wilderness: tracking loss of intact forest landscapes from 2000 to 2013. Sci. Adv. 3, 1–14 (2017).Article 

Google Scholar 
Olson, D. M. et al. Terrestrial ecoregions of the world: a new map of life on Earth. BioScience 51, 933–938 (2001).Article 

Google Scholar 
Running, Steve, Mu, Qiaozhen & Zhao, Maosheng. MOD16A2 MODIS/Terra net evapotranspiration 8-day L4 global 500m. https://doi.org/10.5067/MODIS/MOD16A2.006 (2017).van Schaik, E. et al. Improved SIFTER v2 algorithm for long-term GOME-2A satellite retrievals of fluorescence with a correction for instrument degradation. https://doi.org/10.5194/amt-2019-384 (2020).Kooreman, M. L. et al. GOME-2A SIFTER v2 (2007-2018) [Data set]. SIFTER sun-induced vegetation fluorescence data from GOME-2A (Version 2.0) [Data set]. Royal Netherlands Meteorological Institute (KNMI). https://doi.org/10.21944/gome2a-sifter-v2-sun-induced-fluorescence.Hoese, D. et al. pytroll/pyresample: Version 1.23.0. Zenodo, https://doi.org/10.5281/zenodo.6375741 (2022).Kooreman, M., Tuinder, O., Boersma, K. F. & van Schaik, E. Algorithm Theoretical Basis Document for the GOME-2 NRT, Offline and Data Record Sun-Induced Fluorescence Products. (2019).Wigneron, J.-P. et al. Tropical forests did not recover from the strong 2015–2016 El Niño event. Sci. Adv. 6, eaay4603 (2020).CAS 
Article 

Google Scholar 
Gatti, L. V. et al. Drought sensitivity of Amazonian carbon balance revealed by atmospheric measurements. Nature 506, 76–80 (2014).CAS 
Article 

Google Scholar 
Doughty, R. et al. TROPOMI reveals dry-season increase of solar-induced chlorophyll fluorescence in the Amazon forest. Proc. Natl. Acad. Sci. USA 116, 22393–22398 (2019).CAS 
Article 

Google Scholar 
Porcar-Castell, A. et al. Chlorophyll a fluorescence illuminates a path connecting plant molecular biology to Earth-system science. Nat. Plants 7, 998–1009 (2021).CAS 
Article 

Google Scholar 
Sun, Y. et al. OCO-2 advances photosynthesis observation from space via solar-induced chlorophyll fluorescence. Science 358, eaam5747 (2017).Article 
CAS 

Google Scholar 
Wood, J. D. et al. Multiscale analyses of solar-induced florescence and gross primary production. Geophys. Res. Lett. 44, 533–541 (2017).Article 

Google Scholar 
Verma, M. et al. Effect of environmental conditions on the relationship between solar-induced fluorescence and gross primary productivity at an OzFlux grassland site. J. Geophys. Res. Biogeosci. 122, 716–733 (2017).Article 

Google Scholar 
Parazoo, N. C. et al. Terrestrial gross primary production inferred from satellite fluorescence and vegetation models. Glob. Chang Biol. 20, 3103–3121 (2014).Article 

Google Scholar 
Copernicus Climate Change Service (C3S). ERA5: Fifth generation of ECMWF atmospheric reanalyses of the global climate. https://cds.climate.copernicus.eu/cdsapp#!/home (2017).Goddard Earth Sciences Data and Information Services Center (GES DISC). Tropical Rainfall Measuring Mission (TRMM) – TRMM (TMPA/3B43) Rainfall Estimate L3 1 month 0.25 degree x 0.25 degree V7. https://doi.org/10.5067/TRMM/TMPA/MONTH/7 (2011).Aragão, L. E. O. C. et al. Spatial patterns and fire response of recent Amazonian droughts. Geophys. Res. Lett. 34 (2007).Paca, V. H. et al. The spatial variability of actual evapotranspiration across the Amazon River Basin based on remote sensing products validated with flux towers. Ecol. Process. 8, 6 (2019).Phillips, O. L. et al. Drought–mortality relationships for tropical forests. New Phytol. 187, 631–646 (2010).Article 

Google Scholar 
Maeda, E. E. et al. Evapotranspiration seasonality across the Amazon Basin. Earth Syst. Dyn. 8, 439–454 (2017).Article 

Google Scholar 
Hengl, T. et al. SoilGrids250m: Global gridded soil information based on machine learning. PLoS One 12, e0169748 (2017).Article 
CAS 

Google Scholar 
Costa, F. R. C., Schietti, J., Stark, S. C. & Smith, M. N. The other side of tropical forest drought: do shallow water table regions of Amazonia act as large-scale hydrological refugia from drought? New Phytol. https://nph.onlinelibrary.wiley.com/doi/10.1111/nph.17914 .Walsh, R. P. D. & Lawler, D. M. Rainfall seasonality: description, spatial patterns and change through time. Weather 36, 201–208 (1981).Article 

Google Scholar 
Gorelick, N. et al. Google Earth Engine: Planetary-scale geospatial analysis for everyone. Remote Sens. Environ. https://doi.org/10.1016/j.rse.2017.06.031 (2017).Heinze, G., Wallisch, C. & Dunkler, D. Variable selection – a review and recommendations for the practicing statistician. Biom. J. 60, 431–449 (2018).Article 

Google Scholar 
Wickham, H. ggplot2: Elegant Graphics for Data Analysis (Springer-Verlag New York, 2016).QGIS.org. QGIS Geographic Information System (QGIS Association, 2022).Fancourt, M. Repository for Code, Data and Figures. https://zenodo.org/badge/latestdoi/514231211 (2022). More