Ecology

Soil physicochemical propertiesTable 1 shows the basic chemical characteristics of the 81 soil samples. The pH values of most of the soil samples from the rape fields and the paddy fields ranged between 6.5 and 7.5, while the pH values of most of the soil samples from the wheat fields were less than 6.0. The relatively low pH values for soils from the wheat fields could be due to traditional farming practices adopted in the farms, including continuous cropping. In addition, acidic soil is conducive for wheat growth. Organic matter contents were generally low in all the soils, and soils from the rape fields had the lowest organic matter contents. Generally, the mean soil total N contents in the fields were in the order of rape fields (756 mg/kg)  Zn  > Cd, and the concentrations of metals in most soil samples collected from the wheat and paddy fields were in the order of Fe  > Mn  > Pb  > Cu  > Zn  > Cd. However, independent of the farmland use patterns, the concentrations of all the metals extracted using NH4OAC and NH4NO3 were much lower than those extracted with the other three types of extractants.Table 3 Extractable metal concentrations in the soils (mg/kg).Full size tableThe concentrations of 0.1 mol/L HCl-extracted heavy metals in soils from the rape fields at the R21, R23, and R32 sites, in the wheat fields at the W21 site, and in the paddy fields at the P1 and P2 sites were extremely low, especially in the cases of Fe and Pb. For example, 0.1 mol/L HCl-extracted Fe concentrations at the R21, R23, and R32 sites were 0.2 mg/kg, 0.6 mg/kg, and below the detection limit, respectively, whereas the range of 0.1 mol/L HCl-extracted Fe concentration in the other sites in the rape fields was 22.4–533 mg/kg. Therefore, 0.1 mol/L HCl-extracted heavy metal concentrations at the R21, R23, R32, W21, P1, and P2 sites were omitted from subsequent analyses to ensure homogeneity of variance. The basic pH values at the sites (pH values at R21, R23, R32, W21, P1, and P2 were 8.0, 8.0, 7.7, 8.0, 8.0, and 8.0, respectively) could explain this low HCl-extractability and reduced mobility35,36. Although HCl is considered a universal extractant, it may not be suitable under alkaline soil conditions.Heavy metal concentrations in rape, wheat and rice grainsFigure 4 illustrates the concentrations of six heavy metals in rape, wheat, and rice grains. Similar to the order of the soil metal concentrations, the Fe, Mn, Zn, and Cu concentrations in the grains were much higher than the Pb and Cd concentrations.Figure 4Concentrations of heavy metals ((a) Cu, (b) Zn, (c)Pb, (d) Cd, (e) Fe, and (f) Mn) in the grain of three different crops. (Rape grains (n = 36); Wheat grains (n = 25); Rice grains (n = 20), on dry weight basis).Full size imageThe results are similar to those reported in previous studies with regard to heavy metal concentration trends in rice grains16,37. The reason could be Fe, Mn, Zn, and Cu are all essential for crop growth as micronutrients, leading to the higher levels in soils. Among the studied crops, rice grains accumulated relatively lower amounts of Fe, Mn, and Zn than rape and wheat grains, where Pb and Cd concentrations exhibited opposite trends. The trends are consistent with the findings of Liu et al.38 and Du et al.1, and indicate that rice has a stronger Cd uptake capacity from soil. Williams et al.39 also reported higher rice Cd concentrations than wheat, barley, and maize Cd concentrations. The results could also be attributed to water management and soil oxidation–reduction status in soil-rice systems. Paddy fields are irrigated considerably more than wheat and rape fields. In addition, the water in the study area was contaminated by Cd, Cu, As, and Zn27. Previous studies have also reported that water management practices influence Cd uptake by rice and its bioavailability in soils19,20,40. In the cases of Zn and Pb, according to Feng et al.23, rice grains accumulated lower amounts of Zn than wheat grains, whereas rice grains accumulated higher Pb amounts than wheat grains. Although soil Cd concentrations were generally high in the rape fields, the concentrations of Cd in rape grains were lower than those in the wheat and rice grains.On the other hand, Cu concentrations in grains in all the three crops were below the maximum allowable Cu levels in food (10 mg/kg (GB 15199-94)). Similarly, Zn concentrations in rice grains were below the maximum allowable Zn levels in food (50 mg/kg (GB 13106-91)), while 25% of the rape grain samples and 20% of the wheat grain samples exceeded the threshold value. Although total Pb concentrations in soils were generally low in the three farmland use patterns, Pb concentrations in 70% of the rice grain samples exceeded the maximum allowable Pb levels in food (0.2 mg/kg (GB 2762-2005)). Only four rape and two wheat grain samples exceeded the Pb threshold values, respectively. The varying Pb trends are potentially linked to physical contamination from direct atmospheric deposition8, differences in physiological activities among the crops, and the fruit structures of the studied crops23. Similar to the soil Cd contamination, 96% and 60% of wheat and rice grain samples, respectively exceeded the maximum allowable Cd levels in food (0.1 mg/kg and 0.2 mg/kg for wheat and rice, respectively (GB 2762-2005)). Furthermore, Cd concentrations in approximately 10% of the rice grain samples exceeded 1.0 mg/kg. Conversely, only 14% of rape grain samples exceeded the maximum allowable Cd levels in food (Cd: 0.1 mg/kg for rape (GB 2762-2005)), and the maximum Cd concentration in rape grains was 0.18 mg/kg. According to the results, rape grains were generally safe for consumption whereas wheat and rice grains posed health threats in the study area.Soil to grain bioaccumulation factorsBAF values have been used widely to evaluate the capacity of crop grains to accumulate metals from soil30,41. Similar to a previous study on food crops37, Fe and Pb had the lowest BAF values (Fig. 5). Generally, metal accumulation in crop grains did not increase considerably with an increase in total concentrations of metals in soil. Heavy metal accumulation could have been regulated by crops, so that only low amounts were accumulated into grains. In addition, there were significant differences in some BAF values of the same metal across different crop species (Fig. 5). Different crop species have different accumulation capacities for the same metal42. Overall, the average BAF values of Cu (0.22), Zn (0.37), and Mn (0.14) in wheat grains were significantly higher than those in rape and rice grains. The average BAF value of Pb (0.005) in rice grains was significantly higher than those in rape and wheat grains, and the average BAF values of Cu (0.07) and Cd (0.06) in rape grains were significantly lower than those in wheat and rice grains. The results indicated that rape grains have lower heavy metal accumulation capacity than wheat and rice grains, except in the cases of Zn and Fe. However, the finding is not consistent with the results of a previous study43, which report that grasses have lower accumulation capacity than dicotyledonous plants. Nevertheless, as mentioned above, numerous factors could influence the accumulation capacity of metals in crop grains.Figure 5Bioaccumulation factors (BAF) of heavy metals ((a) Cu, (b) Zn, (c) Pb, (d) Cd, (e) Fe, and (f) Mn) from soil to the grains of three crop species. The error bars indicate the standard deviation. Different letters on bars indicate significant difference (p  More

Field, C. B., Behrenfeld, M. J., Randerson, J. T. & Falkowski, P. Primary production of the biosphere: integrating terrestrial and oceanic components. Science 281, 237 (1998).CAS 
PubMed 

Google Scholar 
Anbar, A. D. & Knoll, A. H. Proterozoic ocean chemistry and evolution: a bioinorganic bridge? Science 297, 1137–1142 (2002).CAS 
PubMed 

Google Scholar 
Saito, M. A., Sigman, D. M. & Morel, F. M. M. The bioinorganic chemistry of the ancient ocean: the co-evolution of cyanobacterial metal requirements and biogeochemical cycles at the Archean–Proterozoic boundary? Inorg. Chim. Acta 356, 308–318 (2003).CAS 

Google Scholar 
Morel, F. M. M., Lam, P. J. & Saito, M. A. Trace metal substitution in marine phytoplankton. Annu. Rev. Earth Planet Sci. 48, 491–517 (2020).CAS 

Google Scholar 
Morel, F. M. & Price, N. M. The biogeochemical cycles of trace metals in the oceans. Science 300, 944–947 (2003).CAS 
PubMed 

Google Scholar 
Twining, B. S. & Baines, S. B. The trace metal composition of marine phytoplankton. Annu. Rev. Mar. Sci. 5, 191–215 (2013).
Google Scholar 
Ho, T.-Y. et al. The elemental composition of some marine phytoplankton. J. Phycol. 39, 1145–1159 (2003).CAS 

Google Scholar 
Ellwood, M. J. Wintertime trace metal (Zn, Cu, Ni, Cd, Pb and Co) and nutrient distributions in the subantarctic zone between 40–52°S; 155–160°E. Mar. Chem. 112, 107–117 (2008).CAS 

Google Scholar 
Zhao, Y., Vance, D., Abouchami, W. & de Baar, H. J. W. Biogeochemical cycling of zinc and its isotopes in the Southern Ocean. Geochim. Cosmochim. Acta 125, 653–667 (2014).CAS 

Google Scholar 
John, S. G., Helgoe, J. & Townsend, E. Biogeochemical cycling of Zn and Cd and their stable isotopes in the Eastern Tropical South Pacific. Mar. Chem. 201, 256–262 (2018).CAS 

Google Scholar 
Middag, R., de Baar, H. J. W. & Bruland, K. W. The relationships between dissolved zinc and major nutrients phosphate and silicate along the GEOTRACES GA02 transect in the West Atlantic Ocean. Glob. Biogeochem. Cy. 33, 63–84 (2019).CAS 

Google Scholar 
Sunda, W. G. & Huntsman, S. A. Feedback interactions between zinc and phytoplankton in seawater. Limnol. Oceanogr. 37, 25–40 (1992).CAS 

Google Scholar 
Sunda, W. G. & Huntsman, S. A. Cobalt and zinc interreplacement in marine phytoplankton: biological and geochemical implications. Limnol. Oceanogr. 40, 1404–1417 (1995).CAS 

Google Scholar 
Vance, D. et al. Silicon and zinc biogeochemical cycles coupled through the Southern Ocean. Nat. Geosci. 10, 202 (2017).CAS 

Google Scholar 
Weber, T., John, S., Tagliabue, A. & DeVries, T. Biological uptake and reversible scavenging of zinc in the global ocean. Science 361, 72 (2018).CAS 
PubMed 

Google Scholar 
Roshan, S., DeVries, T., Wu, J. & Chen, G. The internal cycling of zinc in the ocean. Glob. Biogeochem. Cy. 32, 1833–1849 (2018).CAS 

Google Scholar 
Scott, C. et al. Bioavailability of zinc in marine systems through time. Nat. Geosci. 6, 125–128 (2012).
Google Scholar 
Mock, T. et al. Evolutionary genomics of the cold-adapted diatom Fragilariopsis cylindrus. Nature 541, 536–540 (2017).CAS 
PubMed 

Google Scholar 
Blaby-Haas, C. E. & Merchant, S. S. Comparative and functional algal genomics. Annu. Rev. Plant Biol. 70, 605–638 (2019).CAS 
PubMed 

Google Scholar 
Zhang, Z. H. et al. Adaptation to extreme Antarctic environments revealed by the genome of a sea ice green alga. Curr. Biol. 30, 3330–3341 (2020).CAS 
PubMed 

Google Scholar 
Clarke, A. et al. The Southern Ocean benthic fauna and climate change: a historical perspective. Philos. Trans. R. Soc. Lond. B 338, 299–309 (1992).
Google Scholar 
Klug, A. The discovery of zinc fingers and their applications in gene regulation and genome manipulation. Annu. Rev. Biochem. 79, 213–231 (2010).CAS 
PubMed 

Google Scholar 
Krishna, S. S., Majumdar, I. & Grishin, N. V. Structural classification of zinc fingers: survey and summary. Nucleic Acids Res. 31, 532–550 (2003).CAS 
PubMed 
PubMed Central 

Google Scholar 
Barlow, P. N. et al. Structure of the C3HC4 domain by 1H-nuclear magnetic resonance spectroscopy: a new structural class of zinc-finger. J. Mol. Biol. 237, 201–211 (1994).CAS 
PubMed 

Google Scholar 
Stephens, T. G. et al. Genomes of the dinoflagellate Polarella glacialis encode tandemly repeated single-exon genes with adaptive functions. BMC Biol. 18, 56 (2020).Aranda, M. et al. Genomes of coral dinoflagellate symbionts highlight evolutionary adaptations conducive to a symbiotic lifestyle. Sci. Rep. 6, 39734 (2016).CAS 
PubMed 
PubMed Central 

Google Scholar 
Liu, H. et al. Symbiodinium genomes reveal adaptive evolution of functions related to coral–dinoflagellate symbiosis. Commun. Biol. 1, 95 (2018).PubMed 
PubMed Central 

Google Scholar 
Shoguchi, E. et al. Draft assembly of the Symbiodinium minutum nuclear genome reveals dinoflagellate gene structure. Curr. Biol. 23, 1399–1408 (2013).CAS 
PubMed 

Google Scholar 
Shoguchi, E. et al. Two divergent Symbiodinium genomes reveal conservation of a gene cluster for sunscreen biosynthesis and recently lost genes. BMC Genomics 19, 458 (2018).PubMed 
PubMed Central 

Google Scholar 
Hoppe, C. J. M., Flintrop, C. M. & Rost, B. The Arctic picoeukaryote Micromonas pusilla benefits synergistically from warming and ocean acidification. Biogeosciences 15, 4353–4365 (2018).CAS 

Google Scholar 
Ferguson, R. E. et al. Housekeeping proteins: a preliminary study illustrating some limitations as useful references in protein expression studies. Proteomics 5, 566–571 (2005).CAS 
PubMed 

Google Scholar 
Aslam, S. N. et al. Identifying metabolic pathways for production of extracellular polymeric substances by the diatom Fragilariopsis cylindrus inhabiting sea ice. ISME J. 12, 1237–1251 (2018).CAS 
PubMed 
PubMed Central 

Google Scholar 
Valenzuela, J. J. et al. Ocean acidification conditions increase resilience of marine diatoms. Nat. Commun. 9, 2328 (2018).PubMed 
PubMed Central 

Google Scholar 
Martin, K. et al. The biogeographic differentiation of algal microbiomes in the upper ocean from pole to pole. Nat. Commun. 12, 5483 (2021).CAS 
PubMed 
PubMed Central 

Google Scholar 
Mock, Thomas. Sea of Change: Eukaryotic Phytoplankton Communities in the Arctic Ocean. United States. https://doi.org/10.25585/1488054Duncan, A. et al. Metagenome-assembled genomes of phytoplankton communities across the Arctic Circle and Atlantic Oceans. Microbiome 10 https://doi.org/10.1186/s40168-022-01254-7 (2022).Persi, E., Wolf, Y. I. & Koonin, E. V. Positive and strongly relaxed purifying selection drive the evolution of repeats in proteins. Nat. Commun. 7, 13570 (2016).CAS 
PubMed 
PubMed Central 

Google Scholar 
Mock, T. & Gradinger, R. Determination of Arctic ice algal production with a new in situ incubation technique. Mar. Ecol. Prog. Ser. 177, 15–26 (1999).CAS 

Google Scholar 
Rühle, T., Hemschemeier, A., Melis, A. & Happe, T. A novel screening protocol for the isolation of hydrogen producing Chlamydomonas reinhardtii strains. BMC Plant Biol. 8, 107 (2008).PubMed 
PubMed Central 

Google Scholar 
Crawford, D. W. et al. Influence of zinc and iron enrichments on phytoplankton growth in the northeastern subarctic Pacific. Limnol. Oceanogr. 48, 1583–1600 (2003).CAS 

Google Scholar 
Provasoli, L. Media and prospects for the cultivation of marine algae. In Cultures and Collections of Algae. Proc. US-Japan Conference, Hakone, 12-15 September 1966 (eds Watanabe, A & Hattori, A.) 63–75 (Japanese Society of Plant Physiology, 1968).Ranallo-Benavidez, T. R., Jaron, K. S. & Schatz, M. C. GenomeScope 2.0 and Smudgeplot for reference-free profiling of polyploid genomes. Nat. Commun. 11, 1432 (2020).Bolger, A. M., Lohse, M. & Usadel, B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30, 2114–2120 (2014).CAS 
PubMed 
PubMed Central 

Google Scholar 
Ye, C. X. et al. DBG2OLC: efficient assembly of large genomes using long erroneous reads of the third generation sequencing technologies. Sci. Rep. 6, 31900 (2016).CAS 
PubMed 
PubMed Central 

Google Scholar 
Chin, C. S. et al. Phased diploid genome assembly with single-molecule real-time sequencing. Nat. Methods 13, 1050–1054 (2016).CAS 
PubMed 
PubMed Central 

Google Scholar 
Qin, M. et al. LRScaf: improving draft genomes using long noisy reads. BMC Genomics 20, 955 (2019).CAS 
PubMed 
PubMed Central 

Google Scholar 
Ruan, J. & Li, H. Fast and accurate long-read assembly with wtdbg2. Nat. Methods 17, 155–158 (2020).CAS 
PubMed 

Google Scholar 
Servant, N. et al. HiC-Pro: an optimized and flexible pipeline for Hi-C data processing. Genome Biol. 16, 259 (2015).PubMed 
PubMed Central 

Google Scholar 
Ellinghaus, D., Kurtz, S. & Willhoeft, U. LTRharvest, an efficient and flexible software for de novo detection of LTR retrotransposons. BMC Bioinf. 9, 18 (2008).
Google Scholar 
Xu, Z. & Wang, H. LTR_FINDER: an efficient tool for the prediction of full-length LTR retrotransposons. Nucleic Acids Res. 35, W265–W268 (2007).PubMed 
PubMed Central 

Google Scholar 
Ou, S. J. & Jiang, N. LTR_retriever: a highly accurate and sensitive program for identification of long terminal repeat retrotransposons. Plant Physiol. 176, 1410–1422 (2018).CAS 
PubMed 

Google Scholar 
Hyatt, D. et al. Prodigal: prokaryotic gene recognition and translation initiation site identification. BMC Bioinf. 11, 119 (2010).
Google Scholar 
Stanke, M., Schöffmann, O., Morgenstern, B. & Waack, S. Gene prediction in eukaryotes with a generalized hidden Markov model that uses hints from external sources. BMC Bioinf. 7, 62 (2006).
Google Scholar 
Kim, D. et al. TopHat2: accurate alignment of transcriptomes in the presence of insertions, deletions and gene fusions. Genome Biol. 14, R36 (2013).PubMed 
PubMed Central 

Google Scholar 
Trapnell, C. et al. Differential gene and transcript expression analysis of RNA-seq experiments with TopHat and Cufflinks. Nat. Protoc. 7, 562–578 (2012).CAS 
PubMed 
PubMed Central 

Google Scholar 
Haas, B. J. et al. Automated eukaryotic gene structure annotation using EVidenceModeler and the program to assemble spliced alignments. Genome Biol. 9, R7 (2008).PubMed 
PubMed Central 

Google Scholar 
Mitchell, A. L. et al. InterPro in 2019: improving coverage, classification and access to protein sequence annotations. Nucleic Acids Res. 47, D351–D360 (2018).PubMed Central 

Google Scholar 
Rice, P., Longden, I. & Bleasby, A. EMBOSS: The European Molecular Biology Open Software Suite. Trends Genet. 16, 276–277 (2000).CAS 
PubMed 

Google Scholar 
Kimura, M. A simple method for estimating evolutionary rates of base substitutions through comparative studies of nucleotide sequences. J. Mol. Evol. 16, 111–120 (1980).CAS 
PubMed 

Google Scholar 
Emms, D. M. & Kelly, S. OrthoFinder: phylogenetic orthology inference for comparative genomics. Genome Biol. 20, 238 (2019).PubMed 
PubMed Central 

Google Scholar 
Zhang, Z. et al. KaKs_Calculator: calculating Ka and Ks through model selection and model averaging. Genom. Proteom. Bioinf. 4, 259–263 (2006).CAS 

Google Scholar 
Kumar, S., Stecher, G., Suleski, M. & Hedges, S. B. TimeTree: a resource for timelines, timetrees, and divergence times. Mol. Biol. Evol. 34, 1812–1819 (2017).CAS 
PubMed 

Google Scholar 
Yang, Z. H. PAML 4: phylogenetic analysis by maximum likelihood. Mol. Biol. Evol. 24, 1586–1591 (2007).CAS 
PubMed 

Google Scholar 
Wiśniewski, J. R., Hein, M. Y., Cox, J. & Mann, M. A ‘proteomic ruler’ for protein copy number and concentration estimation without spike-in standards. Mol. Cell Proteom. 13, 3497–3506 (2014).
Google Scholar 
Huntemann, M. et al. The standard operating procedure of the DOE-JGI Metagenome Annotation Pipeline (MGAP v.4). Stand. Genomic Sci. 10, 86 (2016).
Google Scholar 
Fu, L. et al. CD-HIT: accelerated for clustering the next-generation sequencing data. Bioinformatics 28, 3150–3152 (2012).CAS 
PubMed 
PubMed Central 

Google Scholar 
Bushnell, B. BBMap: A Fast, Accurate, Splice-aware Aligner (Lawrence Berkeley National Laboratory, 2014).Löytynoja, A. Phylogeny-aware Alignment with PRANK: Multiple Sequence Alignment Methods (Humana Press, 2014).Castresana, J. Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis. Mol. Biol. Evol. 17, 540–552 (2000).CAS 
PubMed 

Google Scholar  More

Willcox, G., Fornite, S. & Herveux, L. Early Holocene cultivation before domestication in northern Syria. Veg. Hist. Archaeobot. 17, 313–325 (2008).Article 

Google Scholar 
Fuller, D. Q., Willcox, G. & Allaby, R. G. Cultivation and domestication had multiple origins: arguments against the core area hypothesis for the origins of agriculture in the Near East. World Archaeol. 43, 628–652 (2011).Article 

Google Scholar 
Ibáñez, J. J., Anderson, P. C., González-Urquijo, J. & Gibaja, J. Cereal cultivation and domestication as shown by microtexture analysis of sickle gloss through confocal microscopy. J. Archaeol. Sci. 73, 62–81 (2016).Article 

Google Scholar 
Weiss, E., Kislev, M. E. & Hartmann, A. Autonomous cultivation before domestication. Science 312, 1608–1610 (2006).CAS 
PubMed 
Article 

Google Scholar 
Willcox, G. Measuring grain size and identifying Near Eastern cereal domestication: evidence from the Euphrates Valley. J. Archaeol. Sci. 31, 145–150 (2004).Article 

Google Scholar 
White, C. E. & Makarewicz, C. A. Harvesting practices and early Neolithic barley cultivation at el-Hemmeh, Jordan. Veg. Hist. Archaeobot. 21, 85–94 (2012).Article 

Google Scholar 
Colledge, S., Conolly, J., Finlayson, B. & Kuijt, I. New insights on plant domestication, production intensification, and food storage: the archaeobotanical evidence from PPNA Dhra‘. Levant 50, 14–31 (2018).Article 

Google Scholar 
Kuijt, I. & Finlayson, B. Evidence for food storage and predomestication granaries 11,000 years ago in the Jordan Valley. Proc. Natl Acad. Sci. USA 106, 10966–10970 (2009).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Willcox, G. & Stordeur, D. Large-scale cereal processing before domestication during the tenth millennium cal bc in northern Syria. Antiquity 86, 99–114 (2012).Article 

Google Scholar 
Colledge, S. in The Origins of Agriculture and Crop Domestication (eds Damania, A. B. et al.) 121–131 (ICARDA, 1998).Hillman, G. C., Hedges, R., Moore, A. M. T., Colledge, S. & Pettitt, P. New evidence of Lateglacial cereal cultivation at Abu Hureyra on the Euphrates. Holocene 11, 383–393 (2001).Article 

Google Scholar 
Willcox, G. Searching for the origins of arable weeds in the Near East. Veg. Hist. Archaeobot. 21, 163–167 (2012).Article 

Google Scholar 
Snir, A. et al. The origin of cultivation and proto-weeds, long before neolithic farming. PLoS ONE 10, e0131422 (2015).PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Harris, D. R. & Fuller, D. Q. in Encyclopedia of Global Archaeology (ed. Smith, C.) 104–113 (Springer, 2014).Grime, J. P., Hodgson, J. G. & Hunt, R. Comparative Plant Ecology: A Functional Approach to Common British Species (Springer, 2014).Harlan, J. R., de Wet, J. M. J. & Price, E. G. Comparative evolution of cereals. Evolution 27, 311–325 (1973).PubMed 
Article 

Google Scholar 
Fuller, D. Q. Contrasting patterns in crop domestication and domestication rates: recent archaeobotanical insights from the Old World. Ann. Bot. 100, 903–924 (2007).PubMed 
PubMed Central 
Article 

Google Scholar 
Asouti, E. in Neolithic Corporate Identities. Studies in Early Near Eastern Production, Subsistence and Environment 20 (eds Benz, M. et al.) 21–53 (Ex oriente, 2017).Harris, D. R. in Foraging and Farming: the Evolution of Plant Exploitation (eds Harris, D. R. & Hillman, G.) 11–26 (Unwin Hyman, 1989).Smith, B. D. Low-level food production. J. Archaeol. Res. 9, 1–43 (2001).Article 

Google Scholar 
Rindos, D. The Origins of Agriculture: an Evolutionary Perspective (Academic, 1984).Weide, A. Towards a socio-economic model for southwest Asian cereal domestication. Agronomy 11, 2432 (2021).Article 

Google Scholar 
Hillman, G. C. & Davies, M. S. Measured domestication rates in wild wheats and barley under primitive cultivation, and their archaeological implications. J. World Prehist. 4, 157–222 (1990).Article 

Google Scholar 
Kislev, M. E., Hartmann, A. & Weiss, E. Impetus for sowing and the beginning of agriculture: ground collecting of wild cereals. Proc. Natl Acad. Sci. USA 101, 2692–2695 (2004).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Weide, A. et al. The association of arable weeds with modern wild cereal habitats: implications for reconstructing the origins of plant cultivation in the Levant. Environ. Archaeol. https://doi.org/10.1080/14614103.2021.1882715 (2021).Zohary, M. The segetal plant communities of Palestine. Vegetatio 2, 387–411 (1950).Article 

Google Scholar 
Abbo, S., Lev-Yadun, S. & Gopher, A. Plant domestication and crop evolution in the Near East: on events and processes. Crit. Rev. Plant Sci. 31, 241–257 (2012).Article 

Google Scholar 
Wood, D. & Lenné, J. M. A natural adaptive syndrome as a model for the origins of cereal agriculture. Proc. R. Soc. Lond. B 285, 20180277 (2018).
Google Scholar 
Bogaard, A., Palmer, C., Jones, G., Charles, M. & Hodgson, J. G. A FIBS approach to the use of weed ecology for the archaeobotanical recognition of crop rotation regimes. J. Archaeol. Sci. 26, 1211–1224 (1999).Article 

Google Scholar 
Jones, G., Bogaard, A., Charles, M. & Hodgson, J. G. Distinguishing the effects of agricultural practices relating to fertility and disturbance: a functional ecological approach in archaeobotany. J. Archaeol. Sci. 27, 1073–1084 (2000).Article 

Google Scholar 
Díaz, S. et al. The global spectrum of plant form and function. Nature 529, 167–171 (2016).PubMed 
Article 
CAS 

Google Scholar 
Garnier, E., Navas, M.-L. & Grigulis, K. Plant Functional Diversity: Organism Traits, Community Structure, and Ecosystem Properties (Oxford Univ. Press, 2016).Bogaard, A. Neolithic Farming in Central Europe (Routledge, 2004).Bogaard, A. et al. From traditional farming in Morocco to early urban agroecology in northern Mesopotamia: combining present-day arable weed surveys and crop isotope analysis to reconstruct past agrosystems in (semi-)arid regions. Environ. Archaeol. 23, 303–322 (2018).Article 

Google Scholar 
Hamerow, H. et al. An integrated bioarchaeological approach to the medieval ‘agricultural revolution’: a case study from Stafford, England, c. ad 800–1200. Eur. J. Archaeol. 23, 585–609 (2020).Article 

Google Scholar 
Green, L., Charles, M. & Bogaard, A. Exploring the agroecology of Neolithic Çatalhöyük, Central Anatolia: an archaeobotanical approach to agricultural intensity based on functional ecological analysis of arable weed flora. Paléorient 44, 29–44 (2018).
Google Scholar 
Green, L. Assessing the Nature of Early Farming in Neolithic Western Asia: A Functional Ecological Approach to Emerging Arable Weeds. Univ. of Oxford (2017).Atran, S. Hamula organisation and masha’a tenure in Palestine. Man 21, 271–295 (1986).Article 

Google Scholar 
Palmer, C. ‘Following the plough’: the agricultural environment of northern Jordan. Levant 30, 129–165 (1998).Article 

Google Scholar 
Håkansson, S. in Biology and Ecology of Weeds (eds Holzner, W. & Numata, M.) 123–135 (Springer Netherlands, 1982).Charles, M., Bogaard, A., Jones, G., Hodgson, J. & Halstead, P. Towards the archaeobotanical identification of intensive cereal cultivation: present-day ecological investigation in the mountains of Asturias, northwest Spain. Veg. Hist. Archaeobot. 11, 133–142 (2002).Article 

Google Scholar 
Hartmann-Shenkman, A., Kislev, M. E., Galili, E., Melamed, Y. & Weiss, E. Invading a new niche: obligatory weeds at Neolithic Atlit-Yam, Israel. Veg. Hist. Archaeobot. 24, 9–18 (2015).Article 

Google Scholar 
Kuijt, I. in The Neolithic Demographic Transition and its Consequences (eds Bocquet-Appel, J.-P. & Bar-Yosef, O.) 287–313 (Springer Netherlands, 2008).Bogaard, A. et al. Private pantries and celebrated surplus: storing and sharing food at Neolithic Çatalhöyük, Central Anatolia. Antiquity 83, 649–668 (2009).Article 

Google Scholar 
Jones, G. et al. The origins of agriculture: intentions and consequences. J. Archaeol. Sci. 125, 105290 (2021).Article 

Google Scholar 
Weiss, E., Kislev, M. E., Simchoni, O., Nadel, D. & Tschauner, H. Plant-food preparation area on an Upper Paleolithic brush hut floor at Ohalo II, Israel. J. Archaeol. Sci. 35, 2400–2414 (2008).Article 

Google Scholar 
Kluyver, T. A., Charles, M., Jones, G., Rees, M. & Osborne, C. P. Did greater burial depth increase the seed size of domesticated legumes? J. Exp. Bot. 64, 4101–4108 (2013).CAS 
PubMed 
Article 

Google Scholar 
Preece, C., Jones, G., Rees, M. & Osborne, C. P. Fertile Crescent crop progenitors gained a competitive advantage from large seedlings. Ecol. Evol. 11, 3300–3312 (2021).PubMed 
PubMed Central 
Article 

Google Scholar 
Halstead, P. Two Oxen Ahead: Pre-mechanized Farming in the Mediterranean (Wiley, 2014).Anderson, P. C. in The Origins of Agriculture and Crop Domestication (eds Damania, A. B. et al.) 145–159 (ICARDA, 1998).Mercuri, A. M., Fornaciari, R., Gallinaro, M., Vanin, S. & di Lernia, S. Plant behaviour from human imprints and the cultivation of wild cereals in Holocene Sahara. Nat. Plants 4, 71–81 (2018).PubMed 
Article 

Google Scholar 
Spengler, R. N. & Mueller, N. G. Grazing animals drove domestication of grain crops. Nat. Plants 5, 656–662 (2019).PubMed 
Article 

Google Scholar 
Smith, B. D. General patterns of niche construction and the management of ‘wild’ plant and animal resources by small-scale pre-industrial societies. Phil. Trans. R. Soc. Lond. B 366, 836–848 (2011).Article 

Google Scholar 
Bogaard, A. et al. Reconsidering domestication from a process archaeology perspective. World Archaeol. https://doi.org/10.1080/00438243.2021.1954990 (2021).Coqueugniot, E. in Espace Naturel, Espace Habité En Syrie Du Nord (10e–2e millénaires av. J.-C.) (eds M. Fortin & O. Aurenche) 109–114 (Maison de l’Orient et de la Méditerranée, 1998).Douché, C. Émergence et développement des sociétés agricoles au Néolithique acéramique (Xe-VIIIe millénaires av. n. ère) étude archéobotanique de Dja’de El-Mughara et Tell Aswad, Syrie. PhD thesis (Archaeological Mission of Dja’de el Mughara, 2018).Noy, T. Gilgal I: a pre-pottery Neolithic site, Israel. The 1985–1987 seasons. Paléorient 15, 11–18 (1989).Article 

Google Scholar 
Bar-Yosef, O. & Gopher, A. in An Early Neolithic Village in the Jordan Valley (eds Bar-Yosef, O. & Gopher, A.) 41–69 (Harvard Univ., 1997).Wright, K. I. The social origins of cooking and dining in early villages of western Asia. Proc. Prehist. Soc. 66, 89–121 (2000).Article 

Google Scholar 
Finlayson, B. Egalitarian societies and the earliest Neolithic of southwest Asia. Prehist. Archaeol. J. Interdiscip. Stud. 3, 27–43 (2020).Article 

Google Scholar 
Bowles, S. & Choi, J.-K. The Neolithic agricultural revolution and the origins of private property. J. Polit. Econ. 127, 2186–2228 (2019).Article 

Google Scholar 
Kuijt, I. The Neolithic refrigerator on a Friday night: how many people are coming to dinner and just what should I do with the slimy veggies in the back of the fridge? Environ. Archaeol. 20, 321–336 (2015).Article 

Google Scholar 
Danin, A. Flora and vegetation of Israel and adjacent areas. Zoogeogr. Isr. 30, 251–276 (1988).
Google Scholar 
Noy-Meir, I., Gutman, M. & Kaplan, Y. Responses of Mediterranean grassland plants to grazing and protection. J. Ecol. 77, 290–310 (1989).Article 

Google Scholar 
Noy-Meir, I. The effect of grazing on the abundance of wild wheat, barley and oat in Israel. Biol. Conserv. 51, 299–310 (1990).Article 

Google Scholar 
Jones, G., Bogaard, A., Halstead, P., Charles, M. & Smith, H. Identifying the intensity of crop husbandry practices on the basis of weed floras. Annu. Br. Sch. Athens 94, 167–189 (1999).Article 

Google Scholar 
Sternberg, M., Gutman, M., Perevolotsky, A., Ungar, E. D. & Kigel, J. Vegetation response to grazing management in a Mediterranean herbaceous community: a functional group approach. J. Appl. Ecol. 37, 224–237 (2000).Article 

Google Scholar 
Sternberg, M. et al. Testing the limits of resistance: a 19-year study of Mediterranean grassland response to grazing regimes. Glob. Change Biol. 21, 1939–1950 (2015).Article 

Google Scholar 
Calev, A. et al. High-intensity thinning treatments in mature Pinus halepensis plantations experiencing prolonged drought. Eur. J. For. Res. 135, 551–563 (2016).Article 

Google Scholar 
Osem, Y., Perevolotsky, A. & Kigel, J. Grazing effect on diversity of annual plant communities in a semi-arid rangeland: interactions with small-scale spatial and temporal variation in primary productivity. J. Ecol. 90, 936–946 (2002).Article 

Google Scholar 
Temper, L. Creating facts on the ground: agriculture in Israel and Palestine (1882–2000). Hist. Agrar. 48, 75–110 (2009).
Google Scholar 
Dan, J., Yaalon, D., Koyumdjisky, H. & Raz, Z. The soil association map of Israel (1:1,000,000). Isr. J. Earth Sci. 21, 29–49 (1970).
Google Scholar 
Sans, F. X. & Masalles, R. M. Phenological patterns in an arable land weed community related to disturbance. Weed Res. 35, 321–332 (1995).Article 

Google Scholar 
Zohary, M. & Feinbrun-Dothan, N. Flora Palaestina Vol. 1–4 (Israel Academy of Sciences and Humanities, 1966).Davis, P. Flora of Turkey and the East Aegean Islands Vol. 1–10 (Edinburgh Univ. Press, 1965).Mortimer, A. M. in Weed Control Handbook: Principles (eds Hance, R. J. & Holly, K.) 1–42 (Blackwell, 1990).Douché, C. & Willcox, G. New archaeobotanical data from the Early Neolithic sites of Dja’de el-Mughara and Tell Aswad (Syria): a comparison between the northern and the southern Levant. Paléorient 44, 45–58 (2018).
Google Scholar 
Jones, G. The application of present-day cereal processing studies to charred archaeobotanical remains. Circaea 6, 91–96 (1990).
Google Scholar 
Bogaard, A., Jones, G. & Charles, M. The impact of crop processing on the reconstruction of crop sowing time and cultivation intensity from archaeobotanical weed evidence. Veg. Hist. Archaeobot. 14, 505–509 (2005).Article 

Google Scholar 
Bogaard, A. et al. in Humans and Landscapes of Çatalhöyük: Reports from the 2000–2008 Seasons (ed. Hodder, I.) 93–128 (Cotsen Institute of Archaeology/British Institute at Ankara, 2013).Filipović, D. Early Farming in Central Anatolia: an Archaeobotanical Study of Crop Husbandry, Animal Diet and Land Use at Neolithic Çatalhöyük (British Archaeological Reports, 2014).Helmer, D. et al. in New Methods and the First Steps of Mammal Domestication (eds Vigne, J.-D. et al.) 86–95 (Oxbow Books, 2005).Charles, M. Fodder from dung: the recognition and interpretation of dung-derived plant material from archaeological sites. Environ. Archaeol. 1, 111–122 (1998).Article 

Google Scholar 
Kislev, M. E. in An Early Neolithic Village in the Jordan Valley (eds Ofer Bar-Yosef & Avi Gopher) 209–236 (Harvard Univ., 1997).Kislev, M. E. et al. in Gilgal: Early Neolithic Occupations in the Lower Jordan Valley. The Excavations of Tamar Noy (eds Bar-Yosef, O. et al.) 251–257 (Oxbow Books, 2010).Snir, A., Nadel, D. & Weiss, E. Plant-food preparation on two consecutive floors at Upper Paleolithic Ohalo II, Israel. J. Archaeol. Sci. 53, 61–71 (2015).Article 

Google Scholar 
Jones, G., Charles, M., Bogaard, A. & Hodgson, J. Crops and weeds: the role of weed functional ecology in the identification of crop husbandry methods. J. Archaeol. Sci. 37, 70–77 (2010).Article 

Google Scholar 
Šmilauer, P. & Lepš, J. Multivariate Analysis of Ecological Data Using CANOCO 5 (Cambridge Univ. Press, 2014).Galili, E. et al. Atlit-Yam: a Prehistoric site on the sea floor off the Israeli coast. J. Field Archaeol. 20, 133–157 (1993).
Google Scholar 
Brenet, M., Sanchez-Priego, J. & Ibáñez-Estévez, J. J. in Préhistoire et Approche Expérimentale (eds Bourguignon, L. et al.) 121–164 (Monique Mergoil, 2001).Bar-Yosef, O., Gopher, A., Goring-Morris, A. N. & Kozlowski, S. K. in Gilgal: Early Neolithic Occupations in the Lower Jordan Valley. The Excavations of Tamar Noy (eds Bar-Yosef, O. et al.) 11–26 (Oxbow Books, 2010). More

Lattice data geoprocessing and temporal extentWe latticed the data49 using a worldwide grid composed of 18,874 hexagonal 7774 km2 units, built using Discrete Global for R (https://github.com/r-barnes/dggridR)50. All the information we processed on yellow fever cases, on urban and sylvatic vectors presences, and on zoogeographic, spatial and environmental variables (see details on this information below) was aggregated at this spatial resolution. We used zonal statistics to calculate average variable values using ArcMAP 10.7.The temporal extent for our analysis was divided into three periods: the late 20th century (1970–2000), the early 21st century (2001–2017), and the period 2018–2020. Predictions estimated by the late 20th century models were validated using cases reported in the early 21st century, and predictions from the early 21st century models were validated with records from 2018‒2020. Although the limit between periods at the turn of the century is arbitrary, it reflects: 1) Distributional changes in the ranges of the Ae. aegypti and Ae. albopictus vectors51; 2) after 1999, the yellow fever genotype I has spread outside the endemic regions, and the genotype I modern-lineage has caused all major yellow fever outbreaks detected in non-endemic regions of South America since 200013; 3) the maximum potential of globalization was realised at the beginning of the 21st century with the opening of international borders, the widespread access to the Internet and to cell phones, and the generalization of online travel booking and of low-cost flights34. The end of the second period, 2017, was chosen in order to include three years with occurrence of yellow fever cases in south-western Brazil (and two since its occurrence in Angola and the DRC), while retaining three later years for predictive testing purposes (details on this testing are given below).Yellow fever datasetsWe used georeferenced cases of yellow fever in humans for a period of 51 years (from 1970 to 2020). This study period starts immediately after the suspension of the use of DDT due to to the appearance of resistance of Ae. aegypti in the late 1960s in several countries, after 50 years of eradication efforts10. We took from Shearer et al.6 the distribution of yellow fever cases for the period 1970–2016. We extracted additional cases for the period 1970–2020 from various sources (Supplementary data 1), including ProMED-mail: Program of International society for infectious diseases; World Health Organization (WHO): Yellow fever outbreak weekly situation reports, Rapport de situation fievre jaune en RD Congo and Weekly epidemiological record; Health Ministry of different countries: Epidemiological Bulletins of yellow fever in Brazil, Peru, Colombia, and Paraguay; Pan American Health Organization (PAHO): Epidemiological Update Yellow Fever; European Centre for Disease Prevention and Control (ECDC): Communicable disease threats report and Rapid risk assessment report; Nigeria Centre for Disease Control (NCDC): Situation report, yellow fever outbreak in Nigeria and Global Infectious Disease and Epidemiology Online Network (GIDEON). The reported cases were complemented with publications available since 2016 with geo-referenced information on case location (Supplementary data 1). In addition, information was also sought on cases reported in French and Portuguese from local news reports in Africa.We only represented in the hexagonal lattice the reported cases of yellow fever that had a precise location or that were referred to administrative unit was smaller than or of similar size to the hexagons. This dataset was transformed into a binary variable per study period representing the presence (n = 218 hexagons in the late 20th century; 493 hexagons in the early 21st century, see Supplementary data 2) or absence (n = 18,656 hexagons in the late 20th century; 18,381 hexagons in the early 21st century), hereafter the distribution of reported cases of yellow fever.Vector datasetThe georeferenced presences of vectors involved in the urban cycle of yellow fever (i.e., the mosquitoes Ae. aegypti and Ae. albopictus) were taken from “The global compendium of the Ae. aegypti and Ae. Albopictus occurrence”26 for the period 1970–2014. We complemented these records with georeferenced data scientifically validated for the period 2014–2017, taken from VectorBase (https://www.vectorbase.org/) and Mosquito Alert (http://www.mosquitoalert.com/). We included both species because, although Ae. Aegypti is the main vector of yellow fever, Ae. albopictus can also transmit the yellow fever virus to humans4,52.In addition, we included georeferenced occurrence data of sylvatic vectors (Haemagogus janthinomys, H. leucocelaenus and Sabethes chloropterus in South America; Ae. africanus and Ae. vittatus in Africa), which were obtained from Vectormap (vectormap.si.edu) and Gbif (https://gbif.org).We represented in the hexagonal lattice the reported occurrence of mosquitoes that had a precise location or were located in administrative smaller than or of similar size to the hexagons. With this information, we built binary variables representing the presence or absence of each mosquito species in each hexagon. For species involved in the urban cycle, we built two binary variables per species: one for the late 20th century, and another for the early 21st century. For species involved in the sylvatic cycle, we merged the data of late 20th century and early 21st century in order to build a binary variable per species, due the scarcity of data and under the assumption that their distributions have been stable during the four last decades53,54,55.Zoogeographic, spatial and environmental variablesWe built zoogeographic variables based on chorotypes, or types of distribution ranges, of all non-human primate species, as all are potentially vulnerable to yellow fever56. A chorotype is a distribution pattern shared by a group of species57. For obtaining these zoogeographic variables, we proceeded in 4 steps: (1) Distribution maps of non-human primates were obtained from the IUCN for South-America and Africa; (2) the species ranges were classified hierarchically using the classification algorithm UPGMA according to the Baroni-Urbani & Buser´s similarity index58; (3) we evaluated the statistical significance of all clusters obtained as a result of the classification using RMacoqui 1.0 software (http://rmacoqui.r-forge.r-project.org/)59; (4) in each hexagon, the number of species belonging to each chorotype was quantified. We generated a zoogeographic model based on the non-human primates chorotypes by running a forward-backward stepwise logistic regression using presence/absence of yellow fever cases and the number of species of each chorotype as dependent and predictor variables, respectively. This procedure was made for two periods: late 20th century and early 21st century. Henceforth, only the selected chorotype variables were considered in the baseline disease favourability models explained below.We built a yellow fever spatial variable for each continent (South-America and Africa), which were calculated through the trend surface approach, by performing a backward-stepwise logistic regression of the distribution of yellow fever cases on a ensemble of variables defined for polynomial combinations of longitude (X) and latitude (Y) up to the third degree: X, Y, XY, X2, Y2, X2Y, XY2, X3, and Y3. We transformed probability values derived from logistic regression into spatial favourability values by applying the Favourability Function60,61, using the following equation:$$F=frac{P}{1-P}Big/left(frac{{n}_{1}}{{n}_{0}}+frac{P}{1-P}right)$$
(1)
where P is the spatial probability of occurrence of at least a case of yellow fever at each hexagon, and n1 and n0 are the numbers of hexagons with presence and absence of yellow fever cases, respectively. We built a different spatial variable for each continent and time period.We used environmental variables related to the following factors: climate, human activity, topography, hydrography, biome, ecosystem type, and forest loss. For details about the source and description of the environmental variables selected, see Supplementary Table 3.Pathogeographical approach to transmission risk modellingOur objectives were to construct a global yellow fever transmission risk map, and to identify areas where primates contribute to increased risk, using the methodology previously used to analyse the worldwide dynamic biogeography of zoonotic and anthroponotic dengue34 (see flowchart in Fig. 1 and Supplementary Methods). We produced a transmission model focused on the late 20th century and another for the early 21st century.The risk of transmission was assessed by combining a first model describing areas favourable to the presence of yellow fever, i.e., the “baseline disease model”; and another model describing areas favourable to the presence of mosquitoes known to act as vectors, i.e., the “vector model”. For this combination, we used the fuzzy intersection62, i.e., the risk of transmission at each hexagon was valued at the minimum between favourability in the baseline disease model and favourability in the vector model.In this way, we considered that the vectors are a limiting factor, and that the risk of transmission derives from the degree to which the environmental conditions are simultaneously favourable for the presence of vectors and for disease cases to occur63. In order to analyze the spatio-temporal dynamic of yellow fever, we made comparable models for the late 20th century and the early 21st century, using predictor variables that are available for both periods. Later, we made a 21st-century enhanced model that optimized the predictive capacity of availabe information in the search for current risk areas. For this purpose, we included, in the variable set, predictors that are only accessible for the 21st century (e.g., high-resolution population density, livestock, irrigation, infrastructures, intact forest, and GlobCover land cover classes; see Supplementary Table 3).Baseline disease modelsThe baseline disease model in the late 20th century was expressed in terms of favourability values, using the Eq. (1) (see above). This time, P was calculated through a multivariable forward-backward stepwise logistic regression of the 20th-century yellow fever presences/absences on a set of zoogeographic, environmental and spatial variables. This was made in two blocks: 1) a stepwise selection of environmental and spatial variables; 2) a later stepwise addition of chorotypes whose presence contribute to improve significantly the likelihood of the model based only on the first block. Variables for each block were preselected using RAO´s score tests (which estimated the significance of its association to the distribution of yellow fever cases), and Benjamini and Hochberg´s (1995) false discovery rate (FDR) to control for Type I errors, which could pass due to the number of variables analysed. We also avoided excesive multicollinearity by preventing that variables with Spearman correlation values >0.8 were included in the same model. In case this happened, only the variable with the most significant RAO´s score-test value was retained, and the multivariable model was re-run. The parameters in the models were estimated using a gradient ascent machine learning algorithm, and the significance of these paremeters was assessed using the test of Wald. The goodness of fit of the models was established using the test of Hosmer and Lemeshow, which checks the significance of the difference between expected and observed values, so that non significant differences mean that the fit is good. We used IBM-SPSS Statistics 24 software package to perform the models and all the associated tests.We subsequently updated the baseline disease model to explain the distribution of yellow fever cases in the early 21st century. Compared to the procedure described for the 20th-century model, we included a new block before the two ones mentioned above. Thus, the methodological sequence was as follows: (1) forcing the input, as a predictor variable, of the logit of the late 20th century baseline disease model (the logit being the linear combination of variables in the 20th-century model); (2) making a later stepwise selection of spatial and environmental variables; and (3) a stepwise addition of chorotypes that contribute to improving the model’s likelihood. In this way, we took into account that the current spread of yellow fever is influenced by the inertia of previous situations. This is equivalent to assuming that there is temporal autocorrelation (i.e., disease cases in the early 21st century are more probable to occur in areas where they already occurred in the late 20th century). In the 21st-century model, the variables entering in blocks (2) and (3) represent the drivers potentially favouring the spread34. The preselection of variables for blocks (2) and (3) and the control for excessive multicollinearity between environmental variables were made as explained for the late 20th-century model.Vector modelsWe produced a favourabuility model for each vector species for the late 20th century and for the early 21st century separately. We built multivariable favourability models for urban vectors using the distribution of each urban mosquito species in the late 20th century and the spatial and environmental variables for the late 20th century, following the same procedure used for block (1) in the 20th-century baseline disease model. We also updated each urban vector model for the early 21st century as in the baseline disease model, using the procedure described for blocks (1) and (2).A single model, referred to both the late 20th and the early 21st centuries, was made for sylvatic vectors, for the reasons explained above. Finally, we built up the vector models for the late 20th century and for the early 21st century by joining all individual vector models of each period using the fuzzy union64 (i.e., considering for each hexagon the maximum value shown by any of the species models). This criterion was taken into account because, if the pathogen were present, the occurrence of a single vector species would involve potential for yellow fever transmission.Model fit assessment and validation of its predictive capacityFavourability models were assessed according to their classification and discrimination capacities respect to the training data set (i.e., to the observations used for model training). The classification capacity was based on two classification thresholds: F = 0.5, which represents the neutral favourability, and F = 0.2, below which the risk of transmission was considered to be low61. Six classification assessment indices were used65: (1) sensitivity (i.e., proportion of presences correctly classified in favourable hexagons), (2) specificity (i.e., proportion of absences correctly classified in unfavourable hexagons), (3) CCR (i.e., proportion of presences and absences correctly classified in favourable hexagons respectively), (4) TSS (that is sensitivity + specifity – 1), (5) underprediction rate (i.e., proportion of favourable areas that are recorded to have presences), and (6) overprediction rate (i.e., proportion of favourable areas that are not recorded to have presences). The discrimination capacity was assessed using the area under the receiver operating characteristic (ROC) curve (AUC)66.The validation of the predictive capacity of the late 20th century disease and transmission-risk models was done by evaluating, with the same indices used above, classification and discrimination capacities with respect to the cases of the period 2001‒2020. The predictive capacity of the models for the early 21st century was validated with respect to the yellow fever cases reported in the period 2018‒2020.Relative importance of the zoogeographical factorWe estimated the pure contribution of non-human primates to the baseline disease model, i.e., how much of the variation in favourability for yellow fever cases was explained exclusively by the zoogeographical factor, by performing a variation partitioning analysis67. This implied the use of the zoogeographic model and a spatio-environmental model constructed with the environmental and spatial variables that entered the baseline disease model. This approach also allowed us to calculate how much is the variation of the baseline disease model attributable simultaneously to the zoogeographical and other factors. We built maps identifying the zones where the non-human primates could increase yellow fever cases in humans, that is, where the presence of primates could favour the occurrence of yellow fever regardless of correlations with other factors. To map these areas we identified the hexagons that fulfilled these conditions: 1) favourability values for the baseline disease model were ≥ 0.2; and 2) the difference between the favourability values provided by the baseline disease model and the spatio-environmental model was positive and ≥ 0.01.Reporting summaryFurther information on research design is available in the Nature Research Reporting Summary linked to this article. More

Alieva, N. O. et al. Diversity and evolution of coral fluorescent proteins. PLoS ONE 3, e2680 (2008).PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Dove, S., Hoegh-Guldberg, O. & Ranganathan, S. Major colour patterns of reef-building corals are due to a family of GFP-like proteins. Coral Reefs 19, 197–204 (2001).Article 

Google Scholar 
Shimomura, O., Johnson, F. H. & Saiga, Y. Extraction, purification, and properties of aequorin, a bioluminescent protein from the luminous hydromedusan, Aequorea. J. Cell. Physiol. 59, 223–239 (1962).CAS 
Article 

Google Scholar 
Salih, A., Larkum, A., Cox, G., Kühl, M. & Hoegh-Guldberg, O. Fluorescent pigments in corals are photoprotective. Nature 408, 850–853 (2000).CAS 
PubMed 
Article 

Google Scholar 
Kawaguti, S. Effect of the green fluorescent pigment on the productivity of the reef corals. Micronesica 5, 121 (1969).
Google Scholar 
Gittins, J. R., D’Angelo, C., Oswald, F., Edwards, R. J. & Wiedenmann, J. Fluorescent protein-mediated colour polymorphism in reef corals: multicopy genes extend the adaptation/acclimatization potential to variable light environments. Mol. Ecol. 24, 453–465 (2015).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Roth, M. S., Latz, M. I., Goericke, R. & Deheyn, D. D. Green fluorescent protein regulation in the coral Acropora yongei during photoacclimation. J. Exp. Biol. 213, 3644–3655 (2010).CAS 
PubMed 
Article 

Google Scholar 
Quick, C., D’Angelo, C. & Wiedenmann, J. Trade-offs associated with photoprotective green fluorescent protein expression as potential drivers of balancing selection for color polymorphism in reef corals. Front. Mar. Sci. 5, 11 (2018).Article 

Google Scholar 
Schlichter, D., Fricke, H. W. & Weber, W. Light harvesting by wavelength transformation in a symbiotic coral of the Red Sea twilight zone. Mar. Biol. 91, 403–407 (1986).Article 

Google Scholar 
Bollati, E., Plimmer, D., D’Angelo, C. & Wiedenmann, J. FRET-mediated long-range wavelength transformation by photoconvertible fluorescent proteins as an efficient mechanism to generate orange-red light in symbiotic deep water corals. Int. J. Mol. Sci. 18, 1174 (2017).PubMed Central 
Article 
CAS 

Google Scholar 
Palmer, C. V., Modi, C. K. & Mydlarz, L. D. Coral fluorescent proteins as antioxidants. PLoS ONE 4, e7298 (2009).PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Bou-Abdallah, F., Chasteen, N. D. & Lesser, M. P. Quenching of superoxide radicals by green fluorescent protein. Biochim. Biophys. Biochim. Biophys. Acta Gen. Subj. 1760, 1690–1695 (2006).CAS 
Article 

Google Scholar 
Matz, M. V., Marshall, N. J. & Vorobyev, M. Are corals colorful? Photochem. Photobiol. 82, 345–350 (2006).CAS 
PubMed 
Article 

Google Scholar 
Aihara, Y. et al. Green fluorescence from cnidarian hosts attracts symbiotic algae. Proc. Natl Acad. Sci. USA 116, 2118–2123 (2019).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Yamashita, H., Koike, K., Shinzato, C., Jimbo, M. & Suzuki, G. Can Acropora tenuis larvae attract native Symbiodiniaceae cells by green fluorescence at the initial establishment of symbiosis? PLoS ONE 16, e0252514 (2021).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
D’Angelo, C. et al. Blue light regulation of host pigment in reef-building corals. Mar. Ecol. Prog. Ser. 364, 97–106 (2008).Article 
CAS 

Google Scholar 
Ben-Zvi, O., Eyal, G. & Loya, Y. Light-dependent fluorescence in the coral Galaxea fascicularis. Hydrobiologia 759, 15–26 (2014).Article 

Google Scholar 
Muscatine, L., Porter, J. & Kaplan, I. Resource partitioning by reef corals as determined from stable isotope composition. Mar. Biol. 100, 185–193 (1989).Article 

Google Scholar 
Smith, E. G., D’Angelo, C., Sharon, Y., Tchernov, D. & Wiedenmann, J. Acclimatization of symbiotic corals to mesophotic light environments through wavelength transformation by fluorescent protein pigments. Proc. R. Soc. Lond. Ser. B: Biol. Sci. 284, 20170320 (2017).Schlichter, D., Meier, U. & Fricke, H. Improvement of photosynthesis in zooxanthellate corals by autofluorescent chromatophores. Oecologia 99, 124–131 (1994).CAS 
PubMed 
Article 

Google Scholar 
Gilmore, A. M. et al. Simultaneous time resolution of the emission spectra of fluorescent proteins and zooxanthellar chlorophyll in reef-building corals. Photochem. Photobiol. 77, 515–523 (2003).CAS 
PubMed 
Article 

Google Scholar 
Mazel, C. H. et al. Green-fluorescent proteins in Caribbean corals. Limnol. Oceanogr. 48, 402–411 (2003).CAS 
Article 

Google Scholar 
Dubinsky, Z. & Falkowski, P. Light as a Source of Information and Energy in Zooxanthellate Corals. In Coral Reefs: An Ecosystem in Transition (eds Dubinsky, Z. & Stambler, N.) 107–118 (Springer Science & Business Media, 2011).Kahng, S. E. et al. Light, Temperature, Photosynthesis, Heterotrophy, and the Lower Depth Limits of Mesophotic Coral Ecosystemsin. In Mesophotic Coral Ecosystems. Ch. 42 (eds Loya, Y., Puglise, K. A. & Bridge, T. C. L.) 801–828 (Springer International publishing, 2019).Loya, Y., Poglise, K. & Bridge, T. C. L. Mesophotic Coral Ecosystems (Springer International Publishing, 2019).Eyal, G. et al. Spectral diversity and regulation of coral fluorescence in a mesophotic reef habitat in the Red Sea. PLoS ONE 10, e0128697 (2015).PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Roth, M. S. et al. Fluorescent proteins in dominant mesophotic reef-building corals. Mar. Ecol. Prog. Ser. 521, 63–79 (2015).CAS 
Article 

Google Scholar 
Ben-Zvi, O., Wangpraseurt, D., Bronstein, O., Eyal, G. & Loya, Y. Photosynthesis and bio-optical properties of fluorescent mesophotic corals. Front. Mar. Sci. 8, 651601 (2021).Article 

Google Scholar 
Ben-Zvi, O., Eyal, G. & Loya, Y. Response of fluorescence morphs of the mesophotic coral Euphyllia paradivisa to ultra-violet radiation. Sci. Rep. 9, 5245 (2019).PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Houlbrèque, F. & Ferrier-Pagès, C. Heterotrophy in tropical scleractinian corals. Biol. Rev. Camb. Philos. Soc. 84, 1–17 (2009).PubMed 
Article 

Google Scholar 
Goreau, T. F., Goreau, N. I. & Yonge, C. M. Reef corals: autotrophs or heterotrophs? Biol. Bull. 141, 247–260 (1971).Article 

Google Scholar 
Price, J. T., McLachlan, R. H., Jury, C. P., Toonen, R. J. & Grottoli, A. G. Isotopic approaches to estimating the contribution of heterotrophic sources to Hawaiian corals. Limnol. Oceanogr. 66, 2393–2407 (2021).CAS 
Article 

Google Scholar 
Anthony, K. R. N. Coral suspension feeding on fine particulate matter. J. Exp. Mar. Biol. Ecol. 232, 85–106 (1999).Article 

Google Scholar 
Ferrier-Pagès, C., Rottier, C., Beraud, E. & Levy, O. Experimental assessment of the feeding effort of three scleractinian coral species during a thermal stress: effect on the rates of photosynthesis. J. Exp. Mar. Biol. Ecol. 390, 118–124 (2010).Article 

Google Scholar 
Palardy, E. J., Grottoli, G. A. & Matthews, A. K. Effects of upwelling, depth, morphology and polyp size on feeding in three species of Panamanian corals. Mar. Ecol. Prog. Ser. 300, 79–89 (2005).Article 

Google Scholar 
Mies, M. et al. In situ shifts of predominance between autotrophic and heterotrophic feeding in the reef-building coral Mussismilia hispida: an approach using fatty acid trophic markers. Coral Reefs 37, 677–689 (2018).Article 

Google Scholar 
Jerlov, N. G. Optical Oceanography Vol. 5 (Elsevier, 1968).Crandall, J. B., Teece, M. A., Estes, B. A., Manfrino, C. & Ciesla, J. H. Nutrient acquisition strategies in mesophotic hard corals using compound specific stable isotope analysis of sterols. J. Exp. Mar. Biol. Ecol. 474, 133–141 (2016).CAS 
Article 

Google Scholar 
Martinez, S. et al. Energy sources of the depth-generalist mixotrophic coral Stylophora pistillata. Front. Mar. Sci. 7, 566663 (2020).Article 

Google Scholar 
Williams, G. J. et al. Biophysical drivers of coral trophic depth zonation. Mar. Biol. 165, 60 (2018).Article 

Google Scholar 
Lesser, M. P. et al. Photoacclimatization by the coral Montastraea cavernosa in the mesophotic zone: light, food, and genetics. Ecology 91, 990–1003 (2010).PubMed 
Article 

Google Scholar 
Mass, T. et al. Photoacclimation of Stylophora pistillata to light extremes: metabolism and calcification. Mar. Ecol. Prog. Ser. 334, 93–102 (2007).CAS 
Article 

Google Scholar 
Sturaro, N., Hsieh, Y. E., Chen, Q., Wang, P. L. & Denis, V. Trophic plasticity of mixotrophic corals under contrasting environments. Funct. Ecol. 35, 2841–2855 (2021).Article 

Google Scholar 
Lewis, J. B. & Price, W. S. Feeding mechanisms and feeding strategies of Atlantic reef corals. J. Zool. 176, 527–544 (1975).Article 

Google Scholar 
Levy, O., Mizrahi, L., Chadwick-Furman, N. E. & Achituv, Y. Factors controlling the expansion behavior of Favia favus (Cnidaria: Scleractinia): Effects of light, flow, and planktonic prey. Biol. Bull. 200, 118–126 (2001).CAS 
PubMed 
Article 

Google Scholar 
Levy, O., Dubinsky, Z. & Achituv, Y. Photobehavior of stony corals: responses to light spectra and intensity. J. Exp. Biol. 206, 4041–4049 (2003).CAS 
PubMed 
Article 

Google Scholar 
Turak, E. & DeVantier, L. Reef-Building Corals of the Upper Mesophotic Zone of the Central Indo-West Pacificin. In Mesophotic Coral Ecosystems. Ch. 34 (eds Loya, Y., Puglise, K. A. & Bridge, T. C. L.) 621–651 (Springer International Publishing, 2019).Haddock, S. H. D. & Dunn, C. W. Fluorescent proteins function as a prey attractant: experimental evidence from the hydromedusa Olindias formosus and other marine organisms. Biol. Open 4, 1094–1104 (2015).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Eyal, G. et al. Euphyllia paradivisa, a successful mesophotic coral in the northern Gulf of Eilat/Aqaba, Red Sea. Coral Reefs 35, 91–102 (2016).Article 

Google Scholar 
Cronin, T. W. Invertebrate Vision in the Water. In Invertebrate Vision (eds Warran, E. & Nilsson, D.-E.) 6, 211–249 (Cambridge University Press, 2006).Bradley, D. J. & Forward, R. B. Jr. Phototaxis of adult brine shrimp Artemia salina. Can. J. Zool. 62, 2357–2359 (1984).Article 

Google Scholar 
Audzijonytė, A., Pahlberg, J., Väinölä, R. & Lindström, M. Spectral sensitivity differences in two Mysis sibling species (Crustacea, Mysida): adaptation or phylogenetic constraints? J. Exp. Mar. Biol. Ecol. 325, 228–239 (2005).Article 

Google Scholar 
Beeton, A. M. Photoreception in the opossum shrimp, Mysis relicta Loven. Biol. Bull. 116, 204–216 (1959).Article 

Google Scholar 
Lindström, M. Eye function of Mysidacea (Crustacea) in the northern Baltic Sea. J. Exp. Mar. Biol. Ecol. 246, 85–101 (2000).PubMed 
Article 

Google Scholar 
Marshall, N. J. & Vorobyev, M. The Design of Color Signals and Color Vision in Fishes. In Sensory Processing in Aquatic Environments. Ch. 10 (eds Collin, S. P. & Marshall, N. J.) 10, 194–222 (Springer, 2003).Denton, E. J. & Warren, F. J. The photosensitive pigments in the retinae of deep-sea fish. J. Mar. Biol. Assoc. UK 36, 651–662 (1957).CAS 
Article 

Google Scholar 
Kelber, A. Invertebrate Colour Vision. In Invertebrate Vision (eds Warran, E. & Nilsson, D.-E.) 250–290 (Cambridge University Press, 2006).Kim, H. J., Araki, T., Suematsu, Y. & Satuito, C. G. Ontogenic phototactic behaviors of larval stages in intertidal barnacles. Hydrobiologia 849, 747–761 (2021).Cohen, J. H. & Forward, R. B. Jr. Spectral sensitivity of vertically migrating marine copepods. Biol. Bull. 203, 307–314 (2002).PubMed 
Article 

Google Scholar 
Su, Z., Huang, L., Yan, Y. & Li, H. The effect of different substrates on pearl oyster Pinctada martensii (Dunker) larvae settlement. Aquaculture 271, 377–383 (2007).Article 

Google Scholar 
Marangoni, R., Puntoni, S., Favati, L. & Colombetti, G. Phototaxis in Fabrea salina I. Action spectrum determination. J. Photochem. Photobiol. B: Biol. 23, 149–154 (1994).CAS 
Article 

Google Scholar 
Hollingsworth, L. L., Kinzie, R. A., Lewis, T. D., Krupp, D. A. & Leong, J. A. C. Phototaxis of motile zooxanthellae to green light may facilitate symbiont capture by coral larvae. Coral Reefs 24, 523–523 (2005).Article 

Google Scholar 
Smith, F. E. & Taylor, E. R. B. Color responses in the Cladocera and their ecological significance. Am. Nat. 87, 49–55 (1953).Article 

Google Scholar 
Feller, K. D. & Cronin, T. W. Spectral absorption of visual pigments in stomatopod larval photoreceptors. J. Comp. Physiol. A 202, 215–223 (2016).CAS 
Article 

Google Scholar 
Pietsch, T. W. Bioluminescence and Luring. In Oceanic Anglerfishes: Extraordinary Diversity in the Deep Sea (ed. Pietsch, T. W.) 6, 229–252 (Berkeley: University of California Press, 2009).Johnsen, S., Balser, E. J., Fisher, E. C. & Widder, E. A. Bioluminescence in the deep-sea cirrate octopod Stauroteuthis syrtensis Verrill (Mollusca: Cephalopoda). Biol. Bull. 197, 26–39 (1999).CAS 
PubMed 
Article 

Google Scholar 
Robison, B. H., Reisenbichler, K. R., Hunt, J. C. & Haddock, S. H. Light production by the arm tips of the deep-sea cephalopod Vampyroteuthis infernalis. Biol. Bull. 205, 102–109 (2003).PubMed 
Article 

Google Scholar 
Haddock, S. H. D., Dunn, C. W., Pugh, P. R. & Schnitzler, C. E. Bioluminescent and red-fluorescent lures in a deep-sea siphonophore. Science 309, 263 (2005).CAS 
PubMed 
Article 

Google Scholar 
Hastings, J. & Nealson, K. H. Bacterial bioluminescence. Annu. Rev. Microbiol. 31, 549–595 (1977).CAS 
PubMed 
Article 

Google Scholar 
Zarubin, M., Belkin, S., Ionescu, M. & Genin, A. Bacterial bioluminescence as a lure for marine zooplankton and fish. Proc. Natl Acad. Sci. USA 109, 853–857 (2012).CAS 
PubMed 
Article 

Google Scholar 
Nakaema, S. & Hidaka, M. Fluorescent protein content and stress tolerance of two color morphs of the coral Galaxea fascicularis. Galaxea 17, 1–11 (2015).Article 

Google Scholar 
Vermeij, M. J. A., Delvoye, L., Nieuwland, G. & Bak, R. P. M. Patterns in fluorescence over a Caribbean reef slope: the coral genus. Madracis. Photosynthetica 40, 423–429 (2002).CAS 
Article 

Google Scholar 
Kahng, S. & Salih, A. Localization of fluorescent pigments in a nonbioluminescent, azooxanthellate octocoral suggests a photoprotective function. Coral Reefs 24, 435–435 (2005).Article 

Google Scholar 
Glynn, P. W. Ecology of a Caribbean coral reef. The Porites reef-flat biotope: Part II. Plankton community with evidence for depletion. Mar. Biol. 22, 1–21 (1973).Article 

Google Scholar 
Holzman, R., Reidenbach, M. A., Monismith, S. G., Koseff, J. R. & Genin, A. Near-bottom depletion of zooplankton over a coral reef II: relationships with zooplankton swimming ability. Coral Reefs 24, 87–94 (2005).Article 

Google Scholar 
Mazel, C. H. Spectral measurements of fluorescence emission in Caribbean cnidarians. Mar. Ecol. Prog. Ser. 120, 185–191 (1995).Article 

Google Scholar 
R Core Team. R: a language and environment for statistical computing (R Foundation for Statistical Computing, 2013).Bates, D., Mächler, M., Bolker, B. & Walker, S. Fitting linear mixed-effects models using lme4. J. Stat. Softw. 1 1–48 (2015).Kleiman, E. EMAtools: data management tools for real-time monitoring/ecological momentary assessment data. R package version 0.1.4 (2021). More

Von Humboldt, A. Cosmos: A Sketch of a Physical Description of the Universe Vol. 5 (H.G. Bohn Press, 1895).Körner, C. Alpine Treelines: Functional Ecology of the Global High Elevation Tree Limits (Springer, 2012).Peñuelas, J., Ogaya, R., Boada, M. & Jump, A. S. Migration, invasion and decline: changes in recruitment and forest structure in a warming‐linked shift of European beech forest in Catalonia (NE Spain). Ecography 30, 829–837 (2007).Article 

Google Scholar 
Körner, C. Alpine Plant Life: Functional Plant Ecology of High Mountain Ecosystems (Springer, 2021).Körner, C. A re-assessment of high elevation treeline positions and their explanation. Oecologia 115, 445–459 (1998).PubMed 
Article 

Google Scholar 
Körner, C. The cold range limit of trees. Trends Ecol. Evol. 36, 979–989 (2021).PubMed 
Article 

Google Scholar 
Körner, C. & Paulsen, J. A world-wide study of high altitude treeline temperatures. J. Biogeogr. 31, 713–732 (2004).Article 

Google Scholar 
Paulsen, J. & Körner, C. A climate-based model to predict potential treeline position around the globe. Alp. Bot. 124, 1–12 (2014).Article 

Google Scholar 
Feeley, K. J. & Rehm, E. M. Downward shift of montane grasslands exemplifies the dual threat of human disturbances to cloud forest biodiversity. Proc. Natl Acad. Sci. USA 112, E6084–E6084 (2015).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Lenoir, J. et al. A significant upward shift in plant species optimum elevation during the 20th century. Science 320, 1768–1771 (2008).CAS 
PubMed 
Article 

Google Scholar 
Macias Fauria, M. & Johnson, E. A. Warming-induced upslope advance of subalpine forest is severely limited by geomorphic processes. Proc. Natl Acad. Sci. USA 110, 8117–8122 (2013).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Morueta Holme, N. et al. Strong upslope shifts in Chimborazo’s vegetation over two centuries since Humboldt. Proc. Natl Acad. Sci. USA 112, 12741–12745 (2015).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Greenwood, S. & Jump, A. S. Consequences of treeline shifts for the diversity and function of high altitude ecosystems. Arct. Antarct. Alp. Res. 46, 829–840 (2014).Article 

Google Scholar 
Körner, C. & Hiltbrunner, E. Why is the alpine flora comparatively robust against climatic warming? Diversity 13, 383 (2021).Article 

Google Scholar 
Miehe, G. et al. Highest treeline in the northern hemisphere found in southern Tibet. Mt. Res. Dev. 27, 169–173 (2007).Article 

Google Scholar 
Myers, N. et al. Biodiversity hotspots for conservation priorities. Nature 403, 853–858 (2000).CAS 
PubMed 
Article 

Google Scholar 
Wang, F. et al. Add Himalayas’ Grand Canyon to China’s first national parks. Nature 592, 353–353 (2021).CAS 
PubMed 
Article 

Google Scholar 
Zhu, L. et al. Regional scalable priorities for national biodiversity and carbon conservation planning in Asia. Sci. Adv. 7, eabe4261 (2021).PubMed 
PubMed Central 
Article 

Google Scholar 
Yao, T. et al. Different glacier status with atmospheric circulations in Tibetan Plateau and surroundings. Nat. Clim. Change 2, 663–667 (2012).Article 

Google Scholar 
Dirnböeck, T., Essl, F. & Rabitsch, W. Disproportional risk for habitat loss of high-altitude endemic species under climate change. Glob. Change Biol. 17, 990–996 (2011).Article 

Google Scholar 
Schickhoff, U. et al. Do Himalayan treelines respond to recent climate change? An evaluation of sensitivity indicators. Earth Syst. Dynam. 6, 245–265 (2015).Article 

Google Scholar 
Singh, S., Sharma, S. & Dhyani, P. Himalayan arc and treeline: distribution, climate change responses and ecosystem properties. Biodivers. Conserv. 28, 1997–2016 (2019).Article 

Google Scholar 
Schickhoff, U. The Upper Timberline in the Himalayas, Hindu Kush and Karakorum: A Review of Geographical and Ecological Aspects (Springer, 2005).Liang, E. et al. Species interactions slow warming-induced upward shifts of treelines on the Tibetan Plateau. Proc. Natl Acad. Sci. USA 113, 4380–4385 (2016).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Lu, X. et al. Mountain treelines climb slowly despite rapid climate warming. Glob. Ecol. Biogeogr. 30, 305–315 (2021).Article 

Google Scholar 
Harsch, M. A., Hulme, P. E., McGlone, M. S. & Duncan, R. P. Are treelines advancing? A global meta-analysis of treeline response to climate warming. Ecol. Lett. 12, 1040–1049 (2009).PubMed 
Article 

Google Scholar 
Hansen, M. C. et al. High-resolution global maps of 21st-century forest cover change. Science 342, 850–853 (2013).CAS 
PubMed 
Article 

Google Scholar 
Wan, Z. & Li, Z. A physics-based algorithm for retrieving land-surface emissivity and temperature from EOS/MODIS data. IEEE Trans. Geosci. Remote Sens. 35, 980–996 (1997).Article 

Google Scholar 
Fick, S. E. & Hijmans, R. J. WorldClim 2: new 1-km spatial resolution climate surfaces for global land areas. Int. J. Climatol. 37, 4302–4315 (2017).Article 

Google Scholar 
Breiman, L. Random forests. Mach. Learn. 45, 5–32 (2001).Article 

Google Scholar 
Venter, O. et al. Global terrestrial Human Footprint maps for 1993 and 2009. Sci. Data 3, 160067 (2016).PubMed 
PubMed Central 
Article 

Google Scholar 
Sigdel, S. R. et al. Moisture-mediated responsiveness of treeline shifts to global warming in the Himalayas. Glob. Change Biol. 24, 5549–5559 (2018).Article 

Google Scholar 
Dolezal, J. et al. Sink limitation of plant growth determines tree line in the arid Himalayas. Funct. Ecol. 33, 553–565 (2019).Article 

Google Scholar 
Dolezal, J. et al. Annual and intra-annual growth dynamics of Myricaria elegans shrubs in arid Himalaya. Trees 30, 761–773 (2016).Article 

Google Scholar 
Malcolm, J. R. et al. Global warming and extinctions of endemic species from biodiversity hotspots. Conserv. Biol. 20, 538–548 (2006).PubMed 
Article 

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

Google Scholar 
Pirnat, J. Conservation and management of forest patches and corridors in suburban landscapes. Landsc. Urban Plan. 52, 135–143 (2000).Article 

Google Scholar 
Potapov, P. V. et al. Quantifying forest cover loss in Democratic Republic of the Congo, 2000–2010, with Landsat ETM plus data. Remote Sens. Environ. 122, 106–116 (2012).Article 

Google Scholar 
Paulsen, J. & Körner, C. GIS-analysis of tree-line elevation in the Swiss Alps suggests no exposure effect. J. Veg. Sci. 12, 817–824 (2001).Article 

Google Scholar 
FAO. FRA 2000: On Definitions of Forest and Forest Change. Forest Resource Assessment Programme Working Paper, Rome (Food and Agriculture Organization, 2000).Luedeling, E., Siebert, S. & Buerkert, A. Filling the voids in the SRTM elevation model—a TIN-based delta surface approach. ISPRS-J. Photogramm. Remote Sens. 62, 283–294 (2007).Article 

Google Scholar 
Canny, J. Collision detection for moving polyhedra. IEEE Trans. Pattern Anal. Mach. Intell. 8, 200–209 (1986).CAS 
PubMed 
Article 

Google Scholar 
More, J. J. & Sorensen, D. C. Computing a trust region step. SIAM J. Sci. Comput. 4, 553–572 (1983).Article 

Google Scholar 
Theobald, D. M., Harrison-Atlas, D., Monahan, W. B. & Albano, C. M. Ecologically-relevant maps of landforms and physiographic diversity for climate adaptation planning. PLoS ONE 10, e0143619 (2015).PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Hersbach, H. et al. The ERA5 global reanalysis. Q. J. R. Meteorol. Soc. 146, 1999–2049 (2020).Article 

Google Scholar 
Muñoz-Sabater, J. et al. ERA5-Land: a state-of-the-art global reanalysis dataset for land applications. Earth Syst. Sci. Data 13, 4349–4383 (2021).Article 

Google Scholar 
Liang, E., Wang, Y., Eckstein, D. & Luo, T. Little change in the fir tree-line position on the southeastern Tibetan Plateau after 200 years of warming. New Phytol. 190, 760–769 (2011).PubMed 
Article 

Google Scholar 
Anderegg, W. R. L. et al. Tree mortality predicted from drought-induced vascular damage. Nat. Geosci. 8, 367–371 (2015).CAS 
Article 

Google Scholar 
Abatzoglou, J. T. et al. 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 
Case, B. S. & Buckley, H. L. Local-scale topoclimate effects on treeline elevations: a country-wide investigation of New Zealand’s southern beech treelines. PeerJ 3, e1334 (2015).PubMed 
PubMed Central 
Article 

Google Scholar 
Bush, M. B. et al. Fire and climate: contrasting pressures on tropical Andean timberline species. J. Biogeogr. 42, 938–950 (2015).Article 

Google Scholar 
Herrero, A., Zamora, R., Castro, J. & Hodar, J. A. Limits of pine forest distribution at the treeline: herbivory matters. Plant Ecol. 213, 459–469 (2012).Article 

Google Scholar 
Wang, Y. et al. The stability of spruce treelines on the eastern Tibetan Plateau over the last century is explained by pastoral disturbance. For. Ecol. Manag. 442, 34–45 (2019).Article 

Google Scholar 
Wei, Y. et al. Chinese caterpillar fungus (Ophiocordyceps sinensis) in China: current distribution, trading, and futures under climate change and overexploitation. Sci. Total Environ. 755, 142548 (2021).CAS 
PubMed 
Article 

Google Scholar 
Miehe, G. et al. How old is the human footprint in the world’s largest alpine ecosystem? A review of multiproxy records from the Tibetan Plateau from the ecologists’ viewpoint. Quat. Sci. Rev. 86, 190–209 (2014).Article 

Google Scholar 
Willemann, R. J. & Storchak, D. A. Data collection at the international seismological centre. Seismol. Res. Lett. 72, 440–453 (2001).Article 

Google Scholar 
Chen, A., Huang, L., Liu, Q. & Piao, S. Optimal temperature of vegetation productivity and its linkage with climate and elevation on the Tibetan Plateau. Glob. Change Biol. 27, 1942–1951 (2021).Article 

Google Scholar 
Lehmkuhl, F. & Owen, L. A. Late Quaternary glaciation of Tibet and the bordering mountains: a review. Boreas 34, 87–100 (2005).Article 

Google Scholar 
Owen, L. A. & Dortch, J. M. Nature and timing of Quaternary glaciation in the Himalayan–Tibetan orogen. Quat. Sci. Rev. 88, 14–54 (2014).Article 

Google Scholar 
Strobl, C. et al. Conditional variable importance for random forests. BMC Bioinform. 9, 307 (2008).Article 
CAS 

Google Scholar 
Zuur, A. F., Ieno, E. N. & Elphick, C. S. A protocol for data exploration to avoid common statistical problems. Methods Ecol. Evol. 1, 3–14 (2010).Article 

Google Scholar 
Vassallo, D., Krishnamurthy, R. & Fernando, H. J. S. Decreasing wind speed extrapolation error via domain-specific feature extraction and selection. Wind Energy Sci. 5, 959–975 (2020).Article 

Google Scholar 
Ramirez-Villegas, J. & Jarvis, A. Downscaling Global Circulation Model Outputs: The Delta Method Decision and Policy Analysis Working Paper No. 1 (CIAT, 2010).Wu, Z. & Raven, P. Flora of China (Science Press and Missouri Botanical Garden Press, 1994–2006).Wu, Z. Flora of Tibet (Science Press, 1987).Maclean, I. M. D. et al. Microclimates buffer the responses of plant communities to climate change. Glob. Ecol. Biogeogr. 24, 1340–1350 (2015).Article 

Google Scholar 
Randin, C. F. et al. Climate change and plant distribution: local models predict high-elevation persistence. Glob. Change Biol. 15, 1557–1569 (2009).Article 

Google Scholar 
Scherrer, D. & Körner, C. Topographically controlled thermal-habitat differentiation buffers alpine plant diversity against climate warming. J. Biogeogr. 38, 406–416 (2011).Article 

Google Scholar  More

Munday, L. P., White, W. J. & Warner, R. R. A social basis for the development of primary males in a sex-changing fish. Proc. R. Soc. B Biol. Sci. 273(1603), 2845–2851. https://doi.org/10.1098/rspb.2006.3666 (2006).Article 

Google Scholar 
Kuwamura, T., Sunobe, T., Sakai, Y., Kadota, T. & Sawada, S. Hermaphroditism in fishes: An annotated list of species, phylogeny, and mating system. Ichthyol. Res. 67, 341–360. https://doi.org/10.1007/s10228-020-00754-6 (2020).Article 

Google Scholar 
Sadovy, Y. & Shapiro, D. Y. Criteria for the diagnosis of hermaphroditism in fishes. Copeia 1987, 136–156 (1987).Article 

Google Scholar 
Andersson, M. Sexual Selection (Princeton University Press, 1994).Book 

Google Scholar 
Wootton, R. J. & Smith, C. Reproductive Biology of Teleost Fishes (Wiley, 2014).Book 

Google Scholar 
Pla, S., Benvenuto, C., Capellini, I. & Piferrer, F. A phylogenetic comparative analysis on the evolution of sequential hermaphroditism in seabreams (Teleostei: Sparidae). Sci. Rep. UK 10, 3606 (2020).ADS 
CAS 
Article 

Google Scholar 
Nikolsky, G. V. The Ecology of Fishes (Academic Press Inc., 1963).
Google Scholar 
Moe, M. A. Biology of the Red Grouper Epinephelus morio (Valenciennes) from the Eastern Gulf of Mexico. (Florida Det. Natur. Resources Lab. Professional Papers no. 10, 1969).Avise, C. J. & Mank, E. J. Evolutionary perspectives on hermaphroditism in fishes. Sex. Dev. 3, 152–163 (2008).Article 

Google Scholar 
Smith, C. L. Contribution to a theory of hermaphroditism. J. Theor. Biol. 17(1), 76–90. https://doi.org/10.1016/0022-5193(67)90021-5 (1967).ADS 
CAS 
Article 
PubMed 

Google Scholar 
Leonard, L. J. Sexual selection: Lessons from hermaphrodite mating systems. Integr. Comp. Biol. 46(4), 349–367. https://doi.org/10.1093/icb/icj041 (2006).Article 
PubMed 

Google Scholar 
Goikoetxea, A., Todd, V. E. & Gemmell, J. N. Stress and sex: Does cortisol mediate sex change in fish?. Reproduction 154(6), REP-17-0408 (2017).Article 

Google Scholar 
Goikoetxea, A. et al. A new experimental model for the investigation of sequential hermaphroditism. Sci. Rep. UK 11(1), 22881 (2021).ADS 
CAS 
Article 

Google Scholar 
Lodi, E. Sex inversion in domesticated strains of the swordtail, Xiphophorus helleri Heckel (Pisces, Osteichthyes). B. Zool. 47, 1–8 (1980).Article 

Google Scholar 
Huber, J. H. Procatopus websteri: A new species of Lampeye Killifish from Akaka camp, western Gabon (Teleostei: Poeciliidae: Aplocheilichthyinae), exhibiting Similarities of Pattern and Morphology with another sympatric Lampeye species, Aplocheilichthys spilauchen. Trop. Fish Hobbyist 55(1), 110–114 (2007).
Google Scholar 
Cauvet, C. Pseudepiplatys annulatus. Killi Revue 45(2), 2–20 (2019).
Google Scholar 
Domínguez-Castanedo, O., Mosqueda-Cabrera, M. A. & Valdesalici, S. First observations of annualism in Millerichthys robustus (Cyprinodontiformes: Rivulidae). Ichthyol. Explor. Freshw. 24, 15–20 (2013).
Google Scholar 
Furness, A. I., Lee, K. & Reznick, D. N. Adaptation in a variable environment: Phenotypic plasticity and bet-hedging during egg diapause and hatching in an annual killifish. Evolution 69(6), 1461–1475 (2015).Article 

Google Scholar 
Furness, A. I. The evolution of an annual life cycle in killifish: Adaptation to ephemeral aquatic environments through embryonic diapause. Biol. Rev. 91(3), 796–812. https://doi.org/10.1111/brv.12194 (2015).Article 
PubMed 

Google Scholar 
Wourms, J. P. Developmental biology of annual fishes. I. Stages in the normal development of Austrofundulus myersi. Dahl. J. Exp. Zool. 182, 143–168 (1972).CAS 
Article 

Google Scholar 
Murphy, W. J. & Collier, E. G. A molecular phylogeny for aplocheiloid fishes (Atherinomorpha: Cyprinodontiformes): The role of vicariance and the origins of annualism. Mol. Biol. Evol. 14(8), 790–799 (1994).Article 

Google Scholar 
Berois, N., Arezo, J. M., Papa, G. N. & Clivo, A. G. Annual fish: Developmental adaptations for an extreme environment. Wires. Dev. Biol. 1(4), 595–602. https://doi.org/10.1002/wdev.39 (2012).Article 

Google Scholar 
Podrabsky, E. J. & Wilson, E. N. Hypoxia and anoxia tolerance in the annual killifish Austrofundulus limnaeus. Integr. Comp. Biol. 56(4), 500–509. https://doi.org/10.1093/icb/icw092 (2016).CAS 
Article 
PubMed 

Google Scholar 
Reichard, M., Polačik, M. & Sedláček, O. Distribution, colour polymorphism and habitat use of the African killifish, Nothobranchius furzeri, the vertebrate with the shortest lifespan. J. Fish. Biol. 74, 198–212 (2009).CAS 
Article 

Google Scholar 
Lanés, K. E. J., Wolfgang, K. F. & Maltchik, L. Abundance variations and life history traits of two sympatric species of Neotropical annual fish (Cyprinodontiformes: Rivulidae) in temporary ponds of southern Brazil. J. Nat. Hist. 48, 31–32. https://doi.org/10.1080/00222933.2013.862577 (2014).Article 

Google Scholar 
Hass, R. Sexual selection in Nothobranchius guentheri (Pisces: Cyprinodontidae). Evolution 30, 614–622 (1976).Article 

Google Scholar 
Reichard, M. Male-male strategies. In Encyclopedia of Evolutionary Psychological Science (eds. Shackelford, T. K. & Weekes-Shackelford, V. A.) (Springer, 2016).Reichard, M. et al. Lifespan and telomere length variation across populations of wild-derived African killifish. Mol. Ecol. https://doi.org/10.1111/MEC.16287 (2022).Article 

Google Scholar 
Vrtílek, M., Žák, J., Polačik, M., Blažek, R. & Reichard, M. Longitudinal demographic study of wild populations of African annual killifish. Sci. Rep. UK 8, 4774 (2018).ADS 
Article 

Google Scholar 
Passos, C., Tassino, B., Loureiro, M. & Rosenthal, G. G. Intra-and intersexual selection on male body size in the annual killifish Austrolebias charrua. Behav. Process. 96, 20–26 (2013).Article 

Google Scholar 
Parenti, L. R. The phylogeny of atherinomorphs: Evolution of a novel fish reproductive system. In 1. Viviparous Fishes: International Symposia on Livebearing Fishes (eds. Uribe, M. C. & Grier, J. H.) (New Life Publications, 2005).Loureiro, M. et al. Review of the family Rivulidae (Cyprinodontiformes, Aplocheiloidei) and a molecular and morphological phylogeny of the annual fish genus Austrolebias Costa 1998. Neotrop. Ichthyol. 16(3), e180007 (2018).MathSciNet 
Article 

Google Scholar 
Miller, R. R. & Hubbs, L. C. Rivulus robustus, a new cyprinodontid fish from southeastern México. Copeia 1974(4), 865–868. https://doi.org/10.2307/1442584 (1974).Article 

Google Scholar 
Miller, R. R. Peces dulceacuícolas de México. (CONABIO, Sociedad Ictiológica Mexicana, El Colegio de la Frontera Sur, Consejo de Peces del Desierto, 2009).Domínguez-Castanedo, O., Uribe, M. C. & Rosales-Torres, A. M. Life history strategies of annual killifish Millerichthys robustus (Cyprinodontiformes: Cynolebiidae) in a seasonal ephemeral water body in Veracruz, México. Environ. Biol. Fishes. 100(8), 995–1006. https://doi.org/10.1007/s10641-017-0617-y (2017).Article 

Google Scholar 
Domínguez-Castanedo, O., Muñoz-Campos, T. M., Valdesalici, S., Valdez-Carbajal, S. & Passos, C. First description of color variations in the annual killifish Millerichthys robustus, and preliminary observations about its geographical distribution. Environ. Biol. Fishes. 104(3), 293–307. https://doi.org/10.1007/s10641-021-01076-w (2021).Article 

Google Scholar 
Muñoz-Campos, M. T., Valdez-Carbajal, S. & Domínguez-Castanedo, O. Feeding ecology and coexistence dynamics in a community of fishes in a temperate temporary water body. Ecol. Freshw. Fish. 31(1), 1–16 (2021).
Google Scholar 
Jobling, S., Nolan, M., Tyler, C. R., Brighty, G. & Sumpter, J. P. Widespread sexual disruption in wild fish. Environ. Sci. Technol. 32(17), 2498–2506 (1998).ADS 
CAS 
Article 

Google Scholar 
Wong, B. B. M. & Candolin, U. How is female mate choice affected by male competition?. Biol. Rev. Cam. Philos. 80, 559–571. https://doi.org/10.1017/S1464793105006809 (2005).Article 

Google Scholar 
Domínguez-Castanedo, O. Agonistic interactions with asymmetric body size in two adult-age groups of the annual killifish Millerichthys robustus (Miller & Hubbs, 1974). J. Fish. Biol. 99, 1631. https://doi.org/10.1111/jfb.14757 (2021).Article 

Google Scholar 
Polačik, M. & Reichard, M. Asymmetric reproductive isolation between two sympatric annual killifish with extremely short lifespans. PLoS One 6, e22684 (2011).ADS 
Article 

Google Scholar 
Passos, C., Tassino, B., Reyes, F. & Rosenthal, G. G. Seasonal variation in female mate choice and operational sex ratio in wild populations of an annual fish, Austrolebias reicherti. PLoS ONE 9(7), e101649 (2014).ADS 
Article 

Google Scholar 
Warner, R. R. The adaptive significance of sequential hermaphroditism in animals. Am. Nat. 109, 61–82 (1975).Article 

Google Scholar 
Ghiselin, M. T. The evolution of hermaphroditism among animals. Quat. Rev. Biol. 44, 189–208 (1969).CAS 
Article 

Google Scholar 
Forsgren, E., Amundsen, T., Borg, Å. A. & Bjelvenmark, J. Unusually dynamic sex roles in a fish. Nature 429(6991), 551–554 (2004).ADS 
CAS 
Article 

Google Scholar 
Reichard, M., Polačik, M., Blažek, R. & Vrtílek, M. Female bias in the adult sex ratio of African annual fishes: Interspecific differences, seasonal trends and environmental predictors. Evol. Ecol. 28, 1105–1120 (2014).Article 

Google Scholar 
Lindström, K. Effects of resource distribution on sexual selection and the cost of reproduction in sand gobies. Am. Nat. 158(1), 64–74 (2001).Article 

Google Scholar 
Bartáková, V. et al. Strong population genetic structuring in an annual fish, Nothobranchius furzeri, suggests multiple savannah refugia in southern Mozambique. BMC Evol. Biol. 13, 196 (2013).Article 

Google Scholar 
Domínguez-Castanedo, O. Perceived mate competition risk influences the female mate choice and increases the reproductive effort in the annual killifish Millerichthys robustus. Ethol. Ecol. Evol. https://doi.org/10.1080/03949370.2021.1893827 (2021).Article 

Google Scholar 
Harrington, R. W. Oviparous hermaphroditic fish with internal self-fertilization. Science 134(3492), 1749–1750 (1961).ADS 
Article 

Google Scholar 
Tatarenkov, A. et al. Genetic subdivision and variation in selfing rates among Central American populations of the mangrove Rivulus, Kryptolebias marmoratus. J. Hered. 106(3), 276–284 (2015).CAS 
Article 

Google Scholar 
Black, P. M., Balthazart, J., Baillen, M. & Grober, S. M. Socially induced and rapid increases in aggression are inversely related to brain aromatase activity in a sex-changing fish, Lythrypnus dalli. Proc. R. Soc. B Biol. Sci. 275(1579), 2435–2440. https://doi.org/10.1098/rspb.2005.3210 (2005).Article 

Google Scholar 
Escobar, M. L. et al. Involvement of pro-apoptotic and pro-autophagic proteins in granulosa cell death. J. Cell. Biol. 1, 9–17 (2013).Article 

Google Scholar 
Matsuda-Minehata, F., Inoue, N., Goto, Y. & Manabe, N. The regulation of ovarian granulosa cell death by pro and anti-apoptotic molecules. J. Reprod. Dev. 52(6), 695–705 (2006).CAS 
Article 

Google Scholar 
Evans, A. G. et al. Identification of genes involved in apoptosis and dominant follicle development during follicular waves in cattle. Biol. Reprod. 70, 1475–1484 (2004).CAS 
Article 

Google Scholar 
Cardwell, R. J. & Liley, R. N. Hormonal control of sex and color change in the stoplight parrotfish, Sparisoma viride. Gen. Comp. Endocr. 81, 7–20 (1991).CAS 
Article 

Google Scholar 
Domínguez-Castanedo, O., Uribe, M. C. & Muñóz-Campos, T. M. Morphological patterns of cell death in ovarian follicles of primary and secondary growth and post-ovulatory follicle complex of the annual killifish Millerichthys robustus (Cyprinodontiformes: Cynolebiidae). J. Morp. 280(11), 1668–1681. https://doi.org/10.1002/jmor.21056 (2019).Article 

Google Scholar 
De Mitcheson, S. Y. & Liu, M. Functional hermaphroditism in teleosts. Fish Fish. 9, 1–43 (2008).Article 

Google Scholar 
Valdesalici, S., Domínguez-Castanedo, O. & Mosqueda-Cabrera, M. A. Patterns of reproductive behaviour in Millerichthys robustus (Cyprinodontiformes: Cynolebiidae). Int. J. Ichthyol. 22(4), 177–180 (2016).
Google Scholar 
Aguilar-Morales, M., Coutiño, B. B. & Rosales, S. P. Manual general de técnicas histológicas y citoquímicas. México, D. F. (Facultad de Ciencias, UNAM, 1996).Grier, H. J., Uribe, MC. & Patiño, R. The ovary, folliculogenesis, and oogenesis in teleosts. In 2. Reproductive Biology and Physiology of Fishes (Agnathans and Bony Fishes) (ed. Jamieson, B. G. M.). (Science Publishers, 2009).Domínguez-Castanedo, O. & Uribe, M. C. Ovarian structure, folliculogenesis and oogenesis of the annual killifish Millerichthys robustus (Cyprinodontiformes: Cynolebiidae). J. Morp. 280(3), 316–328. https://doi.org/10.1002/jmor.20945 (2019).Article 

Google Scholar 
Domínguez-Castanedo, O. & Uribe, M. C. Reproductive biology in males of the annual killifish Millerichthys robustus (Cyprinodontiformes: Cynolebiidae). Environ. Biol. Fish. 102(11), 1365–1375 (2019).Article 

Google Scholar 
R Core Team. R: A Language and Environment for Statistical Computing (R Foundation for Statistical Computing, 2019). http://www.R-project.org.Brooks, E. M. et al. glmmTMB balances speed and flexibility among packages for zero-inflated generalized linear mixed modeling. R J. 9(2), 378–400 (2017).Article 

Google Scholar 
Hartig, F. DHARMa: Residual diagnostics for hierarchical (multi-level/mixed) regression models. In Theoretical Ecology, Germany. (University of Regensburg, 2021). More

Species DataWe analyse long-term high-quality monitoring data on 1,478 species of birds, mammals, small rodents, butterflies, moths, forest understory vascular plants and freshwater phytoplankton sampled across 6,504 individual sites along an ~1,200 km latitudinal gradient in Finland. Because of differences in sampling methods and in spatial and temporal coverage, each dataset was analysed separately. We note that most species have a distributional range larger than Finland and that for the current purposes, niche space was estimated based on empirical data compiled within Finland alone.BirdsBird data have been collected using line transect censuses in Finland since the 1970s21. The data are collected yearly based on a one‐visit census in which birds are counted along transects with lengths typically 3–6 km. Transects are previously established (that is, with known locations) and not all transects are sampled every year. The census period is June, with observations typically carried between 3:00 and 9:00 am, when the singing activity of birds is highest in dry weather conditions. The observer walks alone at a speed of approximately 1 km h−1 depending on the density of birds along the transect using a map, compass or global positioning system. The census is carried out earlier in southern Finland (June 1–20) compared with northern Finland (June 10–30) due to later breeding phenology in northern latitudes. The line transect is divided into a main belt and a supplementary belt. The main belt is 50 m wide (25 m on each side of the transect line), and the supplementary belt represents the area beyond the main belt as far as birds can be detected. Every observation is assigned to either the main or the supplementary belt. Birds crossing the main belt belong to the supplementary belt even if first observed above the main belt. Species‐specific annual proportions of displaying birds and birds in the main belt remained stable between 1987–2010, indicating that there have been no major changes in species detectability36. The data are curated by the Finnish Museum of Natural History. We used records between 1978 and 2017, including a total of 189 species sampled in 1,105 transects after applying our selection criteria (‘Study design and data preparation’ below).MammalsA systematic monitoring programme of counts of mammal snow tracks was established in 1989 by the Natural Resources Institute Finland (Luke; Game triangle data37,38). The wildlife triangle scheme is based on a network of 4 + 4 + 4 km triangle-shaped transects (totalling 12 km per triangle) with fixed locations, covering the entire country. The triangles are located in forested areas covering the main forest types and are usually situated in hunting areas with the observations carried out by volunteers (mainly hunters). Around 2,000 triangles have been established and about half of these are counted annually. In the winter count, the transect is walked or skied during one day and all snow tracks of 24 mammal species crossing the count line are recorded, usually from mid-January to mid-March (when snow cover conditions are good). The number of crossings are typically related to the transect length and number of days since last snowfall, when snow tracks have been accumulating. Snowfall can be replaced with a pre-count, where the existing tracks are marked or erased, to be disregarded during the actual count. We used records between 1989 and 2017, including a total of 18 species sampled in 1,958 sites after applying our selection criteria (‘Study design and data preparation’ below).Small rodentsSince the 1960s, the Natural Resources Institute Finland (Luke) carries out an inventory of vole species to support forest management planning (small rodents data39). Data were collected biannually in spring (mid‐March to mid‐June) and autumn (mid‐August to mid‐October) in forested and field habitats in 34 locations throughout Finland. Individual trapping sites within locations were often not constant over the study period, primarily due to changes in land use. New sites were selected to be as close and as similar as possible to earlier sites regarding habitat characteristics. Trapping was conducted in two habitat types in almost all locations: spruce (Picea abies) or mountain birch (Betula pubescens) forests and in open grassland habitats, primarily old agricultural land no longer in use. We used records between 1978 and 2017, including a total of 15 species sampled in 19 sites after applying our selection criteria (‘Study design and data preparation’ below).ButterfliesWe combined two similar surveys of butterflies conducted in agricultural landscapes in Finland. The first is a butterfly monitoring network based on volunteer transects initiated by the Finnish Environment Institute (SYKE)40. Between 1999 and 2017, the network included 101 sites, with an average of 47 sites recorded annually (range 30–59). At each site, the walking route (transect) is kept constant from year to year and walked repeatedly during the summer. Along each transect, the number of individuals for each species is recorded from a 5 × 5 × 5 m3 cube ahead of the observer41. The transects are monitored by volunteer butterfly enthusiasts with high species identification skills, who are asked to conduct a minimum of seven annual visits per transect, approximately once a fortnight from late May to late August. Weekly counts are recommended and are usually carried out on nearly half of the transects. The sampling period is typically no longer than 16 weeks and less than ten weeks in the northernmost transects. The second survey type spans between 2001 and 2014 and consists of 68 standardized transects of 1 km length in southern Finland and the Åland islands. These transects were sampled by researchers with a constant sampling frequency of seven counts per summer42. The median transect length of the combined data is 1.95 km (mean = 2.41 km). We used records between 1999 and 2017, including a total of 68 species sampled in 98 transects after applying our selection criteria (‘Study design and data preparation’ below).MothsData on moths have been collected under the National Moth Monitoring scheme (Nocturna) between 1993 and 201743,44, which is coordinated by the Finnish Environment Institute (SYKE). Nocturnal moths were sampled using ‘Jalas’ light traps45 equipped with 125 W Hg vapour or 160 W mixed-light bulbs, located mainly in forested areas across Finland. Traps are located in the same location from year to year and are usually emptied weekly. Sampling occurred every night from early spring to late autumn (usually between April and October). Sampling effort (that is, trapping period) was constant across years for each trap, but given that sampling aimed to cover the entire moth activity period at each location, the trapping period was longer in more southern traps. Volunteers empty the traps and identify the specimens43, with a variable number of traps being sampled per year. The taxonomic skills of the volunteer lepidopterists were typically excellent, and data quality control and cross checking was carried out by the monitoring coordinators43,44,46. The data used here consist of species records collected from 65 traps with at least eight years of sampling. We used records between 1993 and 2017, including a total of 615 species after applying our selection criteria (‘Study design and data preparation’ below).Understory vascular plants of forestsUnderstory vegetation was surveyed on a systematic network of 1,700 sites established on mineral soil in forested land between 1985 and 1986 (as part of the 8th Finnish National Forest Inventory47). This network consists of clusters, each including four sites with 400 m intervals. The clusters were located 16 km apart from each other in southern Finland, and 24 km and 32 km apart in northern Finland along east–west and north–south axes, respectively. All 1,700 sites were resurveyed in 1995, and a subset of 443 sites were resurveyed in 2006. The spatial extent of sampling was comparable across surveys covering the whole country. In all three surveys, vascular plant species (dwarf shrubs, herbs, ferns and graminoids, including also tree and shrub seedlings and saplings up to 50 cm tall) were identified and species’ cover (0.1–100%) was visually estimated; this was based on three to six permanent square‐shaped sampling plots of 2 m2, located 5 m apart from each other within each site. The data are curated by the Natural Resources Institute Finland (Luke). For this analysis, we selected all sites with four sampling plots. The average of species cover across these four sampling plots is used as an estimate of species abundance at each site. This included occurrence records from 1,518, 1,673 and 443 sites in years 1985, 1995 and 2006, respectively. After applying the selection criteria (‘Study design and data preparation’ below), the data included a total of 109 species sampled in 1,712 sites.PhytoplanktonThe National Finnish Phytoplankton Monitoring Database maintained by the Finnish Environment Institute (SYKE; open data portal http://www.syke.fi/en-US/Open_information) comprises nationwide phytoplankton community data of lake surface water samples. We used data collected in the late summer months with samples taken during early July to late August, reflecting the peak productivity season of lake phytoplankton communities. To ensure consistent sampling methodology, we included only data between the years 1977 and 2017. All phytoplankton samples were preserved with acid Lugol’s solution and analysed using the standard Utermöhl technique48. We used records between 1978 and 2017, including a total of 464 species sampled in 1,544 sites after applying our selection criteria (‘Study design and data preparation’ below).Study design and data preparationBefore running the joint species distribution models (below), we converted abundance data into presence records. Each site was assigned to one of the four bioclimatic zones in Finland49—from south to north: hemiboreal (HB), southern boreal (SB), middle boreal (MB) and northern boreal (NB). We combined the two southernmost regions by pooling the occurrence records to obtain a better distribution and number of samples given the much smaller extent of the HB zone. Each occurrence record was also assigned to a different decade based on the year of sampling: decade 1 (1978–1987), decade 2 (1988–1997), decade 3 (1998–2007) and decade 4 (2008–2017) (Fig. 1). Scarce records before 1978 were excluded. Splitting the data into discrete zone and decade subsets allowed us to use independent (and computationally manageable) data to jointly model species responses within each taxonomic group and to disentangle any contrasting imprints of climatic changes between regions and periods. Each subset therefore covered a wide range of climatic conditions for each taxon, with the majority including ten years of data. While the number of sites varied over decades, zones and taxa, the change was not systematic, neither over time nor across taxonomic groups—that is, there was no consistent pattern of more sampling in later decades, and for each group, there could be more sampling sites in a given zone in a later or in an earlier decade (Supplementary Table S2). In addition, the frequency distribution of the pairwise distances between all sites remained similar across decades in each zone (Supplementary Fig. S4), suggesting no changes in the spatial aggregation of sites over time. Furthermore, our analyses are model based and thus explicitly account for the number and distribution of sampling sites, while making inference on both environmental covariates and spatial random effects50. Thus, the variation in the number and distribution of sampling sites affects the uncertainty on trend estimates, rather than affecting the estimates themselves.Species were included if they had a minimum of ten occurrences in each zone × decade combination. Finally, due to the difficulty of obtaining reliable model estimates from very sparse data, we set two additional criteria, including only data subsets with at least 20 samples and at least six species; this meant that despite data being available, some subsets were not included in the analyses (for instance, small rodents in NB).Environmental dataFor each site across the different taxonomic datasets and for each year sampled, we extracted values of daily mean temperature, daily precipitation sum and daily snow depth from the Finnish Meteorological Institute (https://etsin.fairdata.fi/datasets/fmi?keys=Finnish%20Meteorological%20Insitute&terms=organization_name_en.keyword&p=1&sort=best; first accessed in April 2019 and updated in May 2020). These datasets are part of the Finnish Meteorological Institute ClimGrid, which is a gridded daily climatology dataset of Finland, with a spatial resolution of 10 × 10 km 51. From these, we calculated values of annual mean temperature, total precipitation and number of days with snow cover. We extracted annual NAO values from the Climate Analysis Section, National Center for Atmospheric Research (NCAR), Boulder, Colorado, United States52. The NAO index is calculated based on surface sea level pressure difference between the Subtropical (Azores) High and the Subpolar (Iceland) Low, with a high index (NAO +) indicating cool summers and mild and wet winters, whereas low values (NAO −) indicate cold dry winters53. We explored whether other climatic variables might also be relevant for our analysis. Specifically, we additionally calculated January temperature, July temperature, temperature range, mean temperature standard deviation, temperature seasonality, growing degree days (over 5 °C), summer precipitation, precipitation range and mean snow depth. For calculating these additional variables, we extracted daily maximum and minimum temperature data from the ClimGrid data. We evaluated the correlation patterns among these variables and found they were highly correlated, particularly with annual values (Supplementary Fig. S5). As such, we included annual mean and summed values in our models because the relevance of the more detailed variables is likely to vary among taxa. Using annual values also facilitates comparisons with other studies and climate scenarios and allows overcoming issues regarding the overlap between some variables and seasonal sampling of the different species surveys.Quantifying climatic change patternsWe used two approaches to quantify changes in the different climatic variables. First, we used linear regressions with an interaction term between decade and bioclimatic zone to test whether changes over time differed between the different zones in our analysis framework. Second, for a more spatio-temporally resolved assessment of changing patterns we used the k-means clustering method to characterize regions of common climatic profiles for each variable considering all available data over space and time. We set k = 4, resulting in four groups of respectively similar variable conditions. Subsequently we calculated the mean and standard deviation of each resulting cluster for all variables and highlighted the average climate regimes over the whole latitudinal gradient and decades.Joint species distribution modellingWe used a joint species distribution modelling framework to (1) determine how the different climatic variables affected species occurrence patterns and (2) assess whether their relative importance in structuring assemblages has changed over time or across latitude. We fitted separate spatially explicit models for each combination of taxonomic group × bioclimatic zone × decade, yielding a total of 63 models. We modelled the probability of species presence in response to temperature, precipitation and snow cover with quadratic and to NAO with a linear function. Because we model presence data, we used probit regression models. Towards (1), we used the species responses to the climatic variables to quantify the proportion of species at the lower end of their niche (that is, occurrences increasing along the climatic gradient), at the upper end of their niche (that is, occurrences decreasing along the gradient) or at the optimum of their niche (that is, occurrences peaking within the gradient; Fig. 2b; ‘Scoring species’ position within niche domains’ below) within each bioclimatic zone and decade. Towards (2), we compared the proportion of explained variance attributed to each variable and examined whether their relative contribution shifted through time and/or space.For each taxon × bioclimatic zone × decade combination, we fitted latent-variable joint species distribution models using the Hierarchical Modelling of Species Communities (HMSC) framework. HMSC is a multi-variate Bayesian generalized linear mixed-effect model framework, which allows joint modelling of the responses of entire species assemblages and explicit modelling of spatial and temporal autocorrelation18,54,55. We used spatially structured latent variables which were originally proposed by Ovaskainen et al. 56. and later expanded to big spatial data by Tikhonov et al. 57. We fitted the models with the ‘Hmsc’ v 3.0.9 package55 in R58 with a probit link function and assuming the default prior distributions. As fixed effects, we included the climatic variables described above, estimating a second-order polynomial term for all covariates except for NAO, for which we estimated a linear term only. To account for variation in other (unmeasured) environmental variables and potential year-to-year variation not captured by the climatic covariates, we included the random effects of site and year, respectively. All models had the same structure for all the taxon × zone × decade subsets, except for the understory vegetation data for which we did not include the covariate NAO nor the random effect of year because these data were collected only in three individual years, corresponding to a single year per decade in our analytical framework (Fig. 1). Finally, due to computational bottlenecks for large data subsets, some model runs failed to complete with available resources; when this was the case, we randomly subsetted 1,000 records before re-fitting the models (specifically, bird and phytoplankton subsets for the last decade in SB and six mammal subsets for MB and SB in the second, third and fourth decades). We performed posterior sampling using four Markov Chain Monte Carlo chains, each collecting 250 samples, yielding a total of 1,000 samples. We used a thinning interval of 100 and excluded the first 12,500 iterations as burn-in, only sampling the subsequent 25,000 iterations per chain. For phytoplankton in the southernmost region in decade 4, we used a thinning of 10 due to computational constraints due to large site and species numbers. To evaluate Markov Chain Monte Carlo convergence, we examined the distribution of the potential scale reduction factor over the parameters related to the fixed effects and the random effects (equivalent to the Gelman-Rubin statistic59). We assessed model fit via the Area Under the Curve (AUC) statistic60 and model discriminatory power was quantified by Tjur’s R2, which is recommended as a standard measure of discriminatory power for binary outcomes61.To quantify shifts in the explanatory power associated with each covariate, we assessed variance component estimates, that is, the relative explanatory power of each environmental covariate in the HMSC models18,54. We estimated how the relative importance of the covariates in explaining species occurrences varied over time by fitting linear regression models to the species variance component estimate values as a function of decade (using the function ‘lm’ in R) and then compared these changes across zones for the different taxonomic groups (Fig. 4). These model comparisons were carried out after weighing the variance component values by each model’s ability to explain species occurrence patterns (that is, discriminatory power quantified using Tjur’s R2 values).Scoring species’ position within niche domainsTo analyse whether a species occurred at the lower end, at the optimum or at the upper end of its climatic niche within a particular bioclimatic zone and decade, we assessed the species’ responses to each of the climatic variables as follows. First, we classified a species as non-responsive to a specific climate variable within the measured range of that variable if the posterior distribution of the corresponding beta parameter estimates included zero with a probability of more than 10% (corresponding to having less than 90% posterior probability for the response). The non-zero responses were then classified as positive, negative or ‘bell-shaped’ based on the sign of the derivative of the response over the observed environmental gradient. A positive response corresponds to a species being at the lower end of its niche, ‘bell-shaped’ response corresponds to a species being at the optimum of its niche, and a negative response corresponds to a species being at the upper end of its niche. In cases where the derivative is positive/negative at both ends of the environmental gradient, responses were classified as either positive or negative, respectively. Cases where the derivative changed from positive to negative required subsequent evaluation. More specifically, if the derivative was positive or negative over less than 20% of the gradient, we classified the response as negative or positive, respectively. If the derivative was positive or negative over more than 80% of the gradient, we classified the response as positive or negative, respectively. Finally, if the derivative was positive or negative over at most 60% of the environmental gradient, we classified the response as bell-shaped. We evaluated whether this threshold affected the overall results by implementing the same classification using two other criteria for the derivatives being positive or negative: over less than 10% and more than 90% of the environmental gradient, and over less than 30% and more than 70% of the gradient. This showed that our classification of species’ responses was robust to these choices (Supplementary Fig. S6). This classification procedure did not apply to the beta parameters for NAO because we did not include a polynomial term for this covariate, as explained above. Thus, we obtained the number of species for each taxon × bioclimatic zone × decade model that showed responses to the different covariates and calculated the proportion of these species relative to the total number of species in each model (Fig. 3 and Extended Data Fig. 2).Comparing overall species composition between decadesWe compiled the species list present in each decade and each zone for the different taxonomic groups and compared these lists between consecutive decades, that is, comparing decades 1 and 2, 2 and 3, and 3 and 4. For each comparison, we noted how many species were present in both decades (‘shared’ = A), were present only in the first decade in the comparison (‘unique to earlier decade’ = C) or were present only in the last decade in the comparison (‘unique to later decade’ = B). We did this exercise for all taxa in all zones, plotting the sum of ‘shared’ and ‘unique species in each decade’ (Supplementary Fig. S2) and all the consecutive decade comparisons for each taxon (Supplementary Fig. S1). To quantify these patterns over a larger temporal extent, we implemented the same procedure but only comparing the first and last decades sampled for each taxon (Supplementary Figs. S1 and S2; note that for butterflies, these analyses are identical, because this taxon was sampled only in decades 3 and 4).Quantifying community dissimilarityTo assess how community composition changed over space and time, we calculated overall dissimilarity among all the sampling units within a given zone and decade—that is, we quantified variation in composition among sites within a given spatio-temporal extent, regardless of their location. Dissimilarity indices range between 0 and 1, representing cases where all or no species are shared between sites, respectively. We used the same occurrence matrices that were analysed with the HMSC models, that is, the raw species data matrices. We used the function ‘beta.sample’ in the ‘betapart’ package v 1.5.262,63 to calculate total dissimilarity (Sørensen index), which can be additively decomposed into the turnover (Simpson index) and nestedness components64. ‘beta.sample’ randomly selects a specified number of sites to generate distributions of the multiple‐site dissimilarity measures. This is important because the number of sites affects the estimated compositional change values. For each taxonomic group, we first determined the minimum number of sites among the different zone × decade combinations, which was used to define the number of sites to be randomly sampled from the original occurrence matrix, performing this subsampling 1,000 times. We then plotted the mean and standard deviation of these distributions to compare compositional change for each taxonomic group across the different zones and decades. We focus on the turnover metric (that is, species replacement among sites independent of changes in species richness; Extended Data Fig. 3), as it was systematically the main component of dissimilarity except for the small rodent data where the nestedness component had a relatively higher contribution to total dissimilarity. We show the results for the three metrics in Supplementary Fig. S3.Reporting SummaryFurther information on research design is available in the Nature Research Reporting Summary linked to this article. More