Ecology

1.Chesson, P. General theory of competitive coexistence in spatially-varying environments. Theor. Popul. Biol. 58, 211–237 (2000).2.Hubbell, S. P. The Unified Neutral Theory of Biodiversity and Biogeography (Princeton Univ. Press, 2001).3.Ellner, S. P., Snyder, R. E., Adler, P. B. & Hooker, G. An expanded modern coexistence theory for empirical applications. Ecol. Lett. 22, 3–18 (2019).4.Rosenzweig, M. L. Paradox of enrichment: destabilization of exploitation ecosystems in ecological time. Science 171, 385–387 (1971).5.Costantino, R. F., Cushing, J. M., Dennis, B. & Desharnais, R. A. Experimentally induced transitions in the dynamic behaviour of insect populations. Nature 375, 227–230 (1995).6.Fussmann, G. F., Ellner, S. P., Shertzer, K. W. & Hairston, N. G. Jr. Crossing the Hopf bifurcation in a live predator-prey system. Science 290, 1358–1360 (2000).7.Dalziel, B. D. et al. Persistent chaos of measles epidemics in the prevaccination United States caused by a small change in seasonal transmission patterns. PLoS Comput. Biol. 12, e1004655 (2016).8.Darwin, C. On the Origin of Species by Means of Natural Selection, or The Preservation of Favoured Races in the Struggle for Life (John Murray, 1859).9.Gause, G. F. Experimental analysis of Vito Volterra’s mathematical theory of the struggle for existence. Science 79, 16–17 (1934).10.Hutchinson, G. E. The paradox of the plankton. Am. Nat. 95, 137–145 (1961).Article 

Google Scholar 
11.Chesson, P. Multispecies competition in variable environments. Theor. Popul. Biol. 45, 227–276 (1994).Article 

Google Scholar 
12.McCann, K., Hastings, A. & Huxel, G. R. Weak trophic interactions and the balance of nature. Nature 395, 794–798 (1998).13.Rooney, N., McCann, K., Gellner, G. & Moore, J. C. Structural asymmetry and the stability of diverse food webs. Nature 442, 265–269 (2006).14.Coyte, K. Z., Schluter, J. & Foster, K. R. The ecology of the microbiome: networks, competition, and stability. Science 350, 663–666 (2015).15.May, R. M. Host-parasitoid systems in patchy environments: a phenomenological model. J. Anim. Ecol. 47, 833–844 (1978).Article 

Google Scholar 
16.Briggs, C. J. & Hoopes, M. F. Stabilizing effects in spatial parasitoid–host and predator–prey models: a review. Theor. Popul. Biol. 65, 299–315 (2004).17.Vicsek, T. & Zafeiris, A. Collective motion. Phys. Rep. 517, 71–140 (2012).Article 

Google Scholar 
18.Berdahl, A., Torney, C. J., Ioannou, C. C., Faria, J. J. & Couzin, I. D. Emergent sensing of complex environments by mobile animal groups. Science 339, 574–576 (2013).19.Nagy, M., Akos, Z., Biro, D. & Vicsek, T. Hierarchical group dynamics in pigeon flocks. Nature 464, 890–893 (2010).20.Dalziel, B. D., Corre, M. L., Côté, S. D. & Ellner, S. P. Detecting collective behaviour in animal relocation data, with application to migrating caribou. Methods Ecol. Evol. 7, 30–41 (2015).Article 

Google Scholar 
21.Torney, C. J. et al. Inferring the rules of social interaction in migrating caribou. Phil. Trans. R. Soc. B 373, 20170385 (2018).22.Fryxell, J. M., Mosser, A., Sinclair, A. R. E. & Packer, C. Group formation stabilizes predator–prey dynamics. Nature 449, 1041–1043 (2007).23.Vicsek, T., Czirók, A., Ben-Jacob, E., Cohen, I. & Shochet, O. Novel type of phase transition in a system of self-driven particles. Phys. Rev. Lett. 75, 1226 (1995).CAS 
Article 

Google Scholar 
24.Buhl, J. et al. From disorder to order in marching locusts. Science 312, 1402–1406 (2006).25.King, A. J., Fehlmann, G., Biro, D., Ward, A. J. & Fürtbauer, I. Re-wilding collective behaviour: an ecological perspective. Trends Ecol. Evol. 33, 347–357 (2018).26.Sumpter, D. J. T. Collective Animal Behavior (Princeton Univ. Press, 2010).27.Guttal, V. & Couzin, I. D. Social interactions, information use, and the evolution of collective migration. Proc. Natl Acad. Sci. USA 107, 16172–16177 (2010).CAS 
Article 

Google Scholar 
28.Barbier, M. & Watson, J. R. The spatial dynamics of predators and the benefits and costs of sharing information. PLoS Comput. Biol. 12, e1005147 (2016).29.Lotka, A. J. Analytical note on certain rhythmic relations in organic systems. Proc. Natl Acad. Sci. USA 6, 410–415 (1920).CAS 
Article 

Google Scholar 
30.Rosenzweig, M. L. & MacArthur, R. H. Graphical representation and stability conditions of predator-prey interactions. Am. Nat. 97, 209–223 (1963).Article 

Google Scholar 
31.Couzin, I. D., Krause, J., James, R., Ruxton, G. D. & Franks, N. R. Collective memory and spatial sorting in animal groups. J. Theor. Biol. 218, 1–11 (2002).32.Couzin, I. D., Krause, J., Franks, N. R. & Levin, S. A. Effective leadership and decision-making in animal groups on the move. Nature 433, 513–516 (2005).33.MacArthur, R. H. Population ecology of some warblers of northeastern coniferous forests. Ecology 39, 599–619 (1958).Article 

Google Scholar 
34.Dalziel, B. D., Thomann, E., Medlock, J. & De Leenheer, P. Global analysis of a predator-prey model with variable predator search rate. J. Math. Biol. 81, 159–183 (2020).35.Lukas, D. & Clutton-Brock, T. Social complexity and kinship in animal societies. Ecol. Lett. 21, 1129–1134 (2018).36.Purves, D. W., Lichstein, J. W., Strigul, N. & Pacala, S. W. Predicting and understanding forest dynamics using a simple tractable model. Proc. Natl Acad. Sci. USA 105, 17018–17022 (2008).CAS 
Article 

Google Scholar 
37.Dalziel, B. D. et al. Urbanization and humidity shape the intensity of influenza epidemics in U.S. cities. Science 362, 75–79 (2018).38.Monk, C. T. et al. How ecology shapes exploitation: a framework to predict the behavioural response of human and animal foragers along exploration-exploitation trade-offs. Ecol. Lett. 21, 779–793 (2018).39.Hutchins, D. A. & Fu, F. Microorganisms and ocean global change. Nat. Microbiol. 2, 17058 (2017).40.Zakem, E. J. et al. Ecological control of nitrite in the upper ocean. Nat. Commun. 9, 1206 (2018).41.Axtell, R. L. Zipf distribution of U.S. firm sizes. Science 293, 1818–1820 (2001).42.Turchin, P. et al. Quantitative historical analysis uncovers a single dimension of complexity that structures global variation in human social organization. Proc. Natl Acad. Sci. USA 115, E144–E151 (2018).CAS 
Article 

Google Scholar 
43.Press, W. H. Numerical Recipes in C (Cambridge Univ. Press, 1986). More

1.Lewis, S. L. & Maslin, M. A. Defining the anthropocene. Nature 519, 171–180 (2015).ADS 
CAS 
PubMed 
Article 
PubMed Central 

Google Scholar 
2.Tilman, D. et al. Future threats to biodiversity and pathways to their prevention. Nature 546, 73–81 (2017).ADS 
CAS 
PubMed 
Article 

Google Scholar 
3.Pinto-Ledezma, J. N. & Rivero Mamani, M. L. Temporal patterns of deforestation and fragmentation in lowland Bolivia: Implications for climate change. Clim. Change 127, 43–54 (2014).ADS 
Article 

Google Scholar 
4.Allen, J. M., Folk, R. A., Soltis, P. S., Soltis, D. E. & Guralnick, R. P. Biodiversity synthesis across the green branches of the tree of life. Nat. Plants 5, 11–13 (2019).PubMed 
Article 

Google Scholar 
5.Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services, IPBES. Summary for policymakers of the global assessment report on biodiversity and ecosystem services. Accessed 15 Feb 2021. https://zenodo.org/record/3553579. https://doi.org/10.5281/ZENODO.3553579 (2019). 6.Cavender-Bares, J., Balvanera, P., King, E. & Polasky, S. Ecosystem service trade-offs across global contexts and scales. Ecol. Soc. 20, art22 (2015).Article 

Google Scholar 
7.Chaplin-Kramer, R. et al. Global modeling of nature’s contributions to people. Science 366, 255–258 (2019).ADS 
CAS 
Article 
PubMed 

Google Scholar 
8.Díaz, S. et al. Pervasive human-driven decline of life on Earth points to the need for transformative change. Science 366, eaax3100 (2019).Article 
CAS 
PubMed 

Google Scholar 
9.Watson, J. E. M. et al. Set a global target for ecosystems. Nature 578, 360–362 (2020).ADS 
CAS 
PubMed 
Article 
PubMed Central 

Google Scholar 
10.Jetz, W. et al. Essential biodiversity variables for mapping and monitoring species populations. Nat. Ecol. Evol. 3, 539–551 (2019).PubMed 
Article 
PubMed Central 

Google Scholar 
11.Mateo, R. G., Mokany, K. & Guisan, A. Biodiversity models: What if unsaturation is the rule?. Trends Ecol. Evol. 32, 556–566 (2017).PubMed 
PubMed Central 
Article 

Google Scholar 
12.Pereira, H. M. et al. Essential biodiversity variables. Science 339, 277–278 (2013).ADS 
CAS 
PubMed 
Article 

Google Scholar 
13.Ferrier, S. & Guisan, A. Spatial modelling of biodiversity at the community level. J. Appl. Ecol. 43, 393–404 (2006).Article 

Google Scholar 
14.Guisan, A. & Rahbek, C. SESAM—A new framework integrating macroecological and species distribution models for predicting spatio-temporal patterns of species assemblages: Predicting spatio-temporal patterns of species assemblages. J. Biogeogr. 38, 1433–1444 (2011).Article 

Google Scholar 
15.Cavender-Bares, J., Schweiger, A. K., Pinto-Ledezma, J. N. & Meireles, J. E. Applying remote sensing to biodiversity science. in Remote Sensing of Plant Biodiversity (eds. Cavender-Bares, J. et al.) 13–42 (Springer, 2020). https://doi.org/10.1007/978-3-030-33157-3_2.Chapter 

Google Scholar 
16.Fawcett, D. et al. Advancing retrievals of surface reflectance and vegetation indices over forest ecosystems by combining imaging spectroscopy, digital object models, and 3D canopy modelling. Remote Sens. Environ. 204, 583–595 (2018).ADS 
Article 

Google Scholar 
17.Randin, C. F. et al. Monitoring biodiversity in the Anthropocene using remote sensing in species distribution models. Remote Sens. Environ. 239, 111626 (2020).ADS 
Article 

Google Scholar 
18.Turner, W. Sensing biodiversity. Science 346, 301–302 (2014).ADS 
CAS 
PubMed 
Article 

Google Scholar 
19.D’Amen, M., Pradervand, J.-N. & Guisan, A. Predicting richness and composition in mountain insect communities at high resolution: A new test of the SESAM framework: Community-level models of insects. Glob. Ecol. Biogeogr. 24, 1443–1453 (2015).Article 

Google Scholar 
20.Pottier, J. et al. The accuracy of plant assemblage prediction from species distribution models varies along environmental gradients: Climate and species assembly predictions. Glob. Ecol. Biogeogr. 22, 52–63 (2013).Article 

Google Scholar 
21.Zurell, D. et al. Testing species assemblage predictions from stacked and joint species distribution models. J. Biogeogr. 47, 101–113 (2020).Article 

Google Scholar 
22.D’Amen, M. et al. Improving spatial predictions of taxonomic, functional and phylogenetic diversity. J. Ecol. 106, 76–86 (2018).Article 

Google Scholar 
23.Dobrowski, S. Z. et al. Modeling plant ranges over 75 years of climate change in California, USA: Temporal transferability and species traits. Ecol. Monogr. 81, 241–257 (2011).Article 

Google Scholar 
24.Soria-Auza, R. W. et al. Impact of the quality of climate models for modelling species occurrences in countries with poor climatic documentation: A case study from Bolivia. Ecol. Model. 221, 1221–1229 (2010).Article 

Google Scholar 
25.Rocchini, D. et al. Satellite remote sensing to monitor species diversity: Potential and pitfalls. Remote Sens. Ecol. Conserv. 2, 25–36 (2016).Article 

Google Scholar 
26.Schulte to Bühne, H. & Pettorelli, N. Better together: Integrating and fusing multispectral and radar satellite imagery to inform biodiversity monitoring, ecological research and conservation science. Methods Ecol. Evol. 9, 849–865 (2018).Article 

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

Google Scholar 
28.Hobi, M. L. et al. A comparison of Dynamic Habitat Indices derived from different MODIS products as predictors of avian species richness. Remote Sens. Environ. 195, 142–152 (2017).ADS 
Article 

Google Scholar 
29.Radeloff, V. C. et al. The Dynamic Habitat Indices (DHIs) from MODIS and global biodiversity. Remote Sens. Environ. 222, 204–214 (2019).ADS 
Article 

Google Scholar 
30.Pinto-Ledezma, J. N. & Cavender-Bares, J. Using remote sensing for modeling and monitoring species distributions. In Remote Sensing of Plant Biodiversity (eds. Cavender-Bares, J., Gamon, J. A. & Townsend, P. A.) 199–223 (Springer International Publishing, 2020). https://doi.org/10.1007/978-3-030-33157-3_9.31.Fernández, N., Ferrier, S., Navarro, L. M. & Pereira, H. M. Essential biodiversity variables: Integrating in-situ observations and remote sensing through modeling. In Remote Sensing of Plant Biodiversity (eds. Cavender-Bares, J., Gamon, J. A. & Townsend, P. A.) 485–501 (Springer International Publishing, 2020). https://doi.org/10.1007/978-3-030-33157-3_18.32.Skidmore, A. K. et al. Priority list of biodiversity metrics to observe from space. Nat. Ecol. Evol. https://doi.org/10.1038/s41559-021-01451-x (2021).Article 
PubMed 
PubMed Central 

Google Scholar 
33.Saatchi, S., Buermann, W., ter Steege, H., Mori, S. & Smith, T. B. Modeling distribution of Amazonian tree species and diversity using remote sensing measurements. Remote Sens. Environ. 112, 2000–2017 (2008).ADS 
Article 

Google Scholar 
34.He, K. S. et al. Will remote sensing shape the next generation of species distribution models?. Remote Sens. Ecol. Conserv. 1, 4–18 (2015).Article 

Google Scholar 
35.Cord, A. F., Meentemeyer, R. K., Leitão, P. J. & Václavík, T. Modelling species distributions with remote sensing data: Bridging disciplinary perspectives. J. Biogeogr. 40, 2226–2227 (2013).Article 

Google Scholar 
36.Jetz, W. et al. Monitoring plant functional diversity from space. Nat. Plants 2, 16024 (2016).PubMed 
Article 

Google Scholar 
37.Scherrer, D., D’Amen, M., Fernandes, R. F., Mateo, R. G. & Guisan, A. How to best threshold and validate stacked species assemblages? Community optimisation might hold the answer. Methods Ecol. Evol. 9, 2155–2166 (2018).Article 

Google Scholar 
38.Cavender-Bares, J. Diversification, adaptation, and community assembly of the American oaks (Quercus), a model clade for integrating ecology and evolution. New Phytol. 221, 669–692 (2019).PubMed 
Article 

Google Scholar 
39.Cavender-Bares, J., Ackerly, D. D., Baum, D. A. & Bazzaz, F. A. Phylogenetic overdispersion in Floridian oak communities. Am. Nat. 163, 823–843 (2004).CAS 
PubMed 
Article 
PubMed Central 

Google Scholar 
40.Cavender-Bares, J. et al. The role of diversification in community assembly of the oaks (Quercus L.) across the continental U.S. Am. J. Bot. 105, 565–586 (2018).PubMed 
Article 

Google Scholar 
41.Colwell, R. K. & Rangel, T. F. Hutchinson’s duality: The once and future niche. Proc. Natl. Acad. Sci. 106, 19651–19658 (2009).ADS 
CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
42.Townsend Peterson, A. et al. Ecological Niches and Geographic Distributions. (Princeton University Press, 2011). Book 

Google Scholar 
43.Cavender-Bares, J., Fontes, G. C. & Pinto-Ledezma, J. Open questions in understanding the adaptive significance of plant functional trait variation within a single lineage. New Phytol. https://doi.org/10.1111/nph.16652 (2020).Article 
PubMed 
PubMed Central 

Google Scholar 
44.Cavender-Bares, J., Kitajima, K. & Bazzaz, F. A. Multiple trait associations in relation to habitat differentiation among 17 Floridian oak species. Ecol. Monogr. 74, 635–662 (2004).Article 

Google Scholar 
45.Menges, E. S. & Hawkes, C. V. Interactive effects of fire and microhabitat on plants of Florida scrub. Ecol. Appl. 8, 935–946 (1998).Article 

Google Scholar 
46.Calabrese, J. M., Certain, G., Kraan, C. & Dormann, C. F. Stacking species distribution models and adjusting bias by linking them to macroecological models: Stacking species distribution models. Glob. Ecol. Biogeogr. 23, 99–112 (2014).Article 

Google Scholar 
47.Hurlbert, A. H. & Jetz, W. Species richness, hotspots, and the scale dependence of range maps in ecology and conservation. Proc. Natl. Acad. Sci. 104, 13384–13389 (2007).ADS 
CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
48.Pinto-Ledezma, J. N., Jahn, A. E., Cueto, V. R., Diniz-Filho, J. A. F. & Villalobos, F. Drivers of phylogenetic assemblage structure of the Furnariides, a widespread clade of lowland neotropical birds. Am. Nat. 193, E41–E56 (2019).PubMed 
Article 

Google Scholar 
49.Gamon, J. A. et al. Consideration of scale in remote sensing of biodiversity. In Remote Sensing of Plant Biodiversity (eds. Cavender-Bares, J., Gamon, J. A. & Townsend, P. A.) 425–447 (Springer International Publishing, 2020). https://doi.org/10.1007/978-3-030-33157-3_16.50.Pollock, L. J. et al. Understanding co-occurrence by modelling species simultaneously with a Joint Species Distribution Model (JSDM). Methods Ecol. Evol. 5, 397–406 (2014).Article 

Google Scholar 
51.Ovaskainen, O. et al. How to make more out of community data? A conceptual framework and its implementation as models and software. Ecol. Lett. 20, 561–576 (2017).PubMed 
Article 
PubMed Central 

Google Scholar 
52.Ovaskainen, O. Joint Species Distribution Modelling: with Applications in R (Cambridge University Press, 2020).Book 

Google Scholar 
53.Poggiato, G. et al. On the interpretations of joint modeling in community ecology. Trends Ecol. Evol. https://doi.org/10.1016/j.tree.2021.01.002 (2021).Article 
PubMed 

Google Scholar 
54.Wilkinson, D. P., Golding, N., Guillera-Arroita, G., Tingley, R. & McCarthy, M. A. Defining and evaluating predictions of joint species distribution models. Methods Ecol. Evol. 12, 394–404 (2021).Article 

Google Scholar 
55.Bystrova, D. et al. Clustering species with residual covariance matrix in joint species distribution models. Front. Ecol. Evol. 9, 601384 (2021).Article 

Google Scholar 
56.Mateo, R. G. et al. Hierarchical species distribution models in support of vegetation conservation at the landscape scale. J. Veg. Sci. 30, 386–396 (2019).ADS 
Article 

Google Scholar 
57.Petitpierre, B. et al. Will climate change increase the risk of plant invasions into mountains?. Ecol. Appl. 26, 530–544 (2016).PubMed 
Article 

Google Scholar 
58.Cavender-Bares, J. et al. Harnessing plant spectra to integrate the biodiversity sciences across biological and spatial scales. Am. J. Bot. 104, 966–969 (2017).PubMed 
Article 

Google Scholar 
59.Schweiger, A. K. et al. Spectral Niches Reveal Taxonomic Identity and Complementarity in Plant Communities. (2020) https://doi.org/10.1101/2020.04.24.060483. 60.Cavender-Bares, J. et al. Associations of leaf spectra with genetic and phylogenetic variation in oaks: Prospects for remote detection of biodiversity. Remote Sens. 8, 221 (2016).ADS 
Article 

Google Scholar 
61.Meireles, J. E. et al. Leaf reflectance spectra capture the evolutionary history of seed plants. New Phytol. 228, 485–493 (2020).PubMed 
PubMed Central 
Article 

Google Scholar 
62.Williams, L. J. et al. Remote spectral detection of biodiversity effects on forest biomass. Nat. Ecol. Evol. 5, 46–54 (2021).PubMed 
Article 
PubMed Central 

Google Scholar 
63.Alonso, K. et al. Data products, quality and validation of the DLR earth sensing imaging spectrometer (DESIS). Sensors 19, 4471 (2019).ADS 
CAS 
PubMed Central 
Article 

Google Scholar 
64.Stavros, E. N. et al. ISS observations offer insights into plant function. Nat. Ecol. Evol. 1, 0194 (2017).Article 

Google Scholar 
65.Féret, J.-B. & Asner, G. P. Mapping tropical forest canopy diversity using high-fidelity imaging spectroscopy. Ecol. Appl. 24, 1289–1296 (2014).PubMed 
Article 
PubMed Central 

Google Scholar 
66.Cavender-Bares, J. et al. BII-Implementation: The causes and consequences of plant biodiversity across scales in a rapidly changing world. Res. Ideas Outcomes 7, e63850 (2021).Article 

Google Scholar 
67.Hipp, A. L. et al. Sympatric parallel diversification of major oak clades in the Americas and the origins of Mexican species diversity. New Phytol. 217, 439–452 (2018).CAS 
PubMed 
Article 
PubMed Central 

Google Scholar 
68.Cavender-Bares, J. et al. Phylogeny and biogeography of the American live oaks (Quercus subsection Virentes): A genomic and population genetics approach. Mol. Ecol. 24, 3668–3687 (2015).PubMed 
Article 
PubMed Central 

Google Scholar 
69.Aiello-Lammens, M. E., Boria, R. A., Radosavljevic, A., Vilela, B. & Anderson, R. P. spThin: An R package for spatial thinning of species occurrence records for use in ecological niche models. Ecography 38, 541–545 (2015).Article 

Google Scholar 
70.Lobo, J. M., Jiménez-Valverde, A. & Hortal, J. The uncertain nature of absences and their importance in species distribution modelling. Ecography 33, 103–114 (2010).Article 

Google Scholar 
71.Barnett, D. T. et al. The plant diversity sampling design for The National Ecological Observatory Network. Ecosphere 10, e02603 (2019).
Google Scholar 
72.Deblauwe, V. et al. Remotely sensed temperature and precipitation data improve species distribution modelling in the tropics: Remotely sensed climate data for tropical species distribution models. Glob. Ecol. Biogeogr. 25, 443–454 (2016).Article 

Google Scholar 
73.Hijmans, R. J., Phillips, S., Leathwick, J. & Elith, J. dismo: Species Distribution Modeling. R package version 1.3. https://CRAN.R-project.org/package=dismo (2020).74.Myneni, R. B. et al. Global products of vegetation leaf area and fraction absorbed PAR from year one of MODIS data. Remote Sens. Environ. 83, 214–231 (2002).ADS 
Article 

Google Scholar 
75.Gower, S. T., Kucharik, C. J. & Norman, J. M. Direct and indirect estimation of leaf area index, fAPAR, and net primary production of terrestrial ecosystems. Remote Sens. Environ. 70, 29–51 (1999).ADS 
Article 

Google Scholar 
76.Reich, P. B. Key canopy traits drive forest productivity. Proc. R. Soc. B Biol. Sci. 279, 2128–2134 (2012).Article 

Google Scholar 
77.Xiao, Z. et al. Use of general regression neural networks for generating the GLASS leaf area index product from time-series MODIS surface reflectance. IEEE Trans. Geosci. Remote Sens. 52, 209–223 (2014).ADS 
Article 

Google Scholar 
78.Soberón, J. Grinnellian and Eltonian niches and geographic distributions of species. Ecol. Lett. 10, 1115–1123 (2007).PubMed 
Article 

Google Scholar 
79.Farr, T. G. et al. The shuttle radar topography mission. Rev. Geophys. 45, RG2004 (2007).ADS 
Article 

Google Scholar 
80.Dubuis, A. et al. Predicting spatial patterns of plant species richness: A comparison of direct macroecological and species stacking modelling approaches: Predicting plant species richness. Divers. Distrib. 17, 1122–1131 (2011).Article 

Google Scholar 
81.Schoener, T. W. Anolis lizards of Bimini: Resource partition in a complex fauna. Ecology 49, 704–726 (1968).Article 

Google Scholar 
82.Barve, N. et al. The crucial role of the accessible area in ecological niche modeling and species distribution modeling. Ecol. Model. 222, 1810–1819 (2011).Article 

Google Scholar 
83.Cooper, J. C. & Soberón, J. Creating individual accessible area hypotheses improves stacked species distribution model performance. Glob. Ecol. Biogeogr. 27, 156–165 (2018).Article 

Google Scholar 
84.Barbet-Massin, M., Jiguet, F., Albert, C. H. & Thuiller, W. Selecting pseudo-absences for species distribution models: How, where and how many? How to use pseudo-absences in niche modelling?. Methods Ecol. Evol. 3, 327–338 (2012).Article 

Google Scholar 
85.Carlson, C. J. et al. The global distribution of Bacillus anthracis and associated anthrax risk to humans, livestock and wildlife. Nat. Microbiol. 4, 1337–1343 (2019).CAS 
PubMed 
Article 
PubMed Central 

Google Scholar 
86.Chipman, H. A., George, E. I. & McCulloch, R. E. BART: Bayesian additive regression trees. Ann. Appl. Stat. 4, 266–298 (2010).MathSciNet 
MATH 
Article 

Google Scholar 
87.Yen, J. D. L., Thomson, J. R., Vesk, P. A. & Mac Nally, R. To what are woodland birds responding? Inference on relative importance of in-site habitat variables using several ensemble habitat modelling techniques. Ecography 34, 946–954 (2011).Article 

Google Scholar 
88.Carlson, C. J. embarcadero: Species distribution modelling with Bayesian additive regression trees in r. Methods Ecol. Evol. 11, 850–858 (2020).Article 

Google Scholar 
89.Dorie, V. dbarts: Discrete Bayesian Additive Regression Trees Sampler. (2020).90.Hastie, T. & Tibshirani, R. Bayesian backfitting. Stat. Sci. 15(3), 196–223 (2000). MathSciNet 
MATH 
Article 

Google Scholar 
91.Allouche, O., Tsoar, A. & Kadmon, R. Assessing the accuracy of species distribution models: Prevalence, kappa and the true skill statistic (TSS): Assessing the accuracy of distribution models. J. Appl. Ecol. 43, 1223–1232 (2006).Article 

Google Scholar 
92.Di Cola, V. et al. ecospat: An R package to support spatial analyses and modeling of species niches and distributions. Ecography 40, 774–787 (2017).Article 

Google Scholar 
93.Hipp, A. L. et al. Genomic landscape of the global oak phylogeny. New Phytol. 226, 1198–1212 (2020).CAS 
PubMed 
Article 
PubMed Central 

Google Scholar 
94.Webb, C. O., Ackerly, D. D., McPeek, M. A. & Donoghue, M. J. Phylogenies and community ecology. Annu. Rev. Ecol. Syst. 33, 475–505 (2002).Article 

Google Scholar 
95.Kembel, S. W. et al. Picante: R tools for integrating phylogenies and ecology. Bioinformatics 26, 1463–1464 (2010).CAS 
PubMed 
Article 

Google Scholar 
96.Kruschke, J. Doing Bayesian Data Analysis, 2nd Ed. (2014).97.Mills, J. A. & Parent, O. Bayesian MCMC estimation. In Handbook of Regional Science (eds Fischer, M. M. & Nijkamp, P.) 1571–1595 (Springer, Berlin, 2014). https://doi.org/10.1007/978-3-642-23430-9_89.Chapter 

Google Scholar 
98.Carpenter, B. et al. Stan : A probabilistic programming language. J. Stat. Softw. 76, 1–32 (2017).Article 

Google Scholar 
99.Goodrich, B., Gabry, J., Ali, I. & Brilleman, S. rstanarm: Bayesian applied regression modeling via Stan. (2020).100.Bürkner, P.-C. brms: An R package for Bayesian multilevel models using Stan. J. Stat. Softw. 80, 1–28 (2017).Article 

Google Scholar 
101.Makowski, D., Ben-Shachar, M. & Lüdecke, D. bayestestR: Describing effects and their uncertainty, existence and significance within the Bayesian framework. J. Open Source Softw. 4, 1541 (2019).ADS 
Article 

Google Scholar  More

1.Palme, R. Monitoring stress hormone metabolites as a useful, non-invasive tool for welfare assessment in farm animals. Anim. Welf. 21, 331–337 (2012).CAS 
Article 

Google Scholar 
2.Jankord, R. & Herman, J. P. Limbic regulation of hypothalamo-pituitary-adrenocortical function during acute and chronic stress. Ann. N. Y. Acad. Sci. 1148, 64–73 (2008).ADS 
PubMed 
PubMed Central 
Article 

Google Scholar 
3.Romero, L. M. Physiological stress in ecology: Lessons from biomedical research. Trends Ecol. Evol. 19, 249–255 (2004).PubMed 
Article 

Google Scholar 
4.Haase, C. G., Long, A. K. & Gillooly, J. F. Energetics of stress: Linking plasma cortisol levels to metabolic rate in mammals. Biol. Lett. 12, 20150867 (2016).PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
5.Selye, H. A syndrome produced by diverse nocuous agents. Nature 1936, 32 (1936).ADS 
Article 

Google Scholar 
6.Fink, G. Stress Science: Neuroendocrinology (Academic Press, Elsevier Science, 2010).
Google Scholar 
7.Hing, S., Narayan, E. J., Thompson, R. C. A. & Godfrey, S. S. The relationship between physiological stress and wildlife disease: Consequences for health and conservation. Wildl. Res. 43, 51 (2016).Article 

Google Scholar 
8.Tsigos, C. & Chrousos, G. P. Hypothalamic-pituitary-adrenal axis, neuroendocrine factors and stress. J. Psychosom. Res. 53, 865–871 (2002).PubMed 
Article 

Google Scholar 
9.Tryphonopoulos, P. D., Letourneau, N. & Azar, R. Approaches to salivary cortisol collection and analysis in infants. Biol. Res. Nurs. 16, 398–408 (2014).PubMed 
Article 

Google Scholar 
10.Palme, R. Non-invasive measurement of glucocorticoids: Advances and problems. Physiol. Behav. 199, 229–243 (2019).CAS 
PubMed 
Article 

Google Scholar 
11.Russell, E., Koren, G., Rieder, M. & Van Uum, S. Hair cortisol as a biological marker of chronic stress: Current status, future directions and unanswered questions. Psychoneuroendocrinology 37, 589–601 (2012).CAS 
PubMed 
Article 

Google Scholar 
12.McEwen, B. S. Central effects of stress hormones in health and disease: Understanding the protective and damaging effects of stress and stress mediators. Eur. J. Pharmacol. 583, 174–185 (2008).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
13.O’Connor, E. A. et al. The impact of chronic environmental stressors on growing pigs, Sus scrofa (Part 1): Stress physiology, production and play behaviour. Animal 4, 1899–1909 (2010).PubMed 
Article 

Google Scholar 
14.Kadarmideen, H. N. & Janss, L. L. G. Population and systems genetics analyses of cortisol in pigs divergently selected for stress. Physiol. Genomics 29, 57–65 (2007).CAS 
PubMed 
Article 

Google Scholar 
15.Romano, M. C. et al. Stress in wildlife species: Noninvasive monitoring of glucocorticoids. NeuroImmunoModulation 17, 209–212 (2010).CAS 
PubMed 
Article 
PubMed Central 

Google Scholar 
16.Sales, L. P. et al. Niche conservatism and the invasive potential of the wild boar. J. Anim. Ecol. 86, 1214–1223 (2017).PubMed 
Article 
PubMed Central 

Google Scholar 
17.Briedermann, L. Schwarzwild (Franckh-Kosmos Verlags-GmnH & Co. KG, 2009).
Google Scholar 
18.Keuling, O. et al. Eurasian wild boar Sus scrofa (Linnaeus, 1758). In Ecology, Conservation and Management of Wild Pigs and Peccaries (eds Melletti, M. & Meijaard, E.) 202–233 (Cambridge University Press, 2018).
Google Scholar 
19.Niedersächsisches Ministerium für Ernährung, Landwirtschaft und Verbraucherschutz. Aktuelle Jagdzeiten in Niedersachsen (konsolidierte Fassung) Stand: 25. Januar 2021 inkl. Verordnung zur Durchführung des Nieders. Jagdgesetzes (DVO-NJagdG) vom 23. Mai 2008 (Nds. GVBl. S. 194), zuletzt geändert durch Verordnung vom 18. Januar 2021 (Nds. GVBl. S. 24). (2021). https://www.ml.niedersachsen.de/download/163729/Aktuelle_Jagdzeiten_in_Niedersachsen_Stand_25.01.2021_nicht_vollstaendig_barrierefrei_.pdf. Accessed 01 June 2021.20.Casas-Díaz, E. et al. Hematologic and biochemical reference intervals for Wild Boar (Sus scrofa) captured by cage trap. Vet. Clin. Pathol. 44, 215–222 (2015).PubMed 
Article 

Google Scholar 
21.Gentsch, R. P., Kjellander, P. & Röken, B. O. Cortisol response of wild ungulates to trauma situations: Hunting is not necessarily the worst stressor. Eur. J. Wildl. Res. 64, 11 (2018).Article 

Google Scholar 
22.Adcock, S. J. J., Martin, G. M. & Walsh, C. J. The stress response and exploratory behaviour in Yucatan minipigs (Sus scrofa): Relations to sex and social rank. Physiol. Behav. 152, 194–202 (2015).CAS 
PubMed 
Article 

Google Scholar 
23.Bratton, S. P. The effect of the European wild boar (Sus scrofa) on gray beech forest in the great smokey mountains. Ecology 56, 1356–1366 (1975).Article 

Google Scholar 
24.Singer, F. J., Swank, W. T. & Clebsh, E. E. C. The effects of wild pig rooting in a deciduous forest. J. Wildl. Manage. 48, 464–473 (1984).CAS 
Article 

Google Scholar 
25.Wlazelko, M. & Labudzki, L. Über Nahrungskomponenten und trophische Stellung des Schwarzwildes im Forschungsgebiet Zielonka. Z. Jagdwiss. 38, 81–87 (1992).
Google Scholar 
26.Killian, G., Miller, L., Rhyan, J. & Doten, H. Immunocontraception of Florida feral swine with a single-dose GnRH vaccine. Am. J. Reprod. Immunol. 55, 378–384 (2006).CAS 
PubMed 
Article 

Google Scholar 
27.Gortázar, C., Ferroglio, E., Höfle, U., Frölich, K. & Vicente, J. Diseases shared between wildlife and livestock: A European perspective. Eur. J. Wildl. Res. 53, 241–256 (2007).Article 

Google Scholar 
28.Gräber, R., Strauß, E. & Johanshon, S. Wild und Jagd—Landesjagdbericht 2017/2018 (Niedersächsisches Ministerium für Ernährung, Landwirtschaft und Verbraucherschutz, Hannover, 2018).29.Wölfel, H. Bewegungsjagden (Leopold Stocker Verlag, 2003).30.Eisenbarth, E. & Ophoven, E. Bewegungsjagd auf Schalenwild (Franckh-Kosmos Verlags-GmbH & Co., 2002).
Google Scholar 
31.Böhm, E. Drückjagd auf Sauen (Neumann-Neudamm, 2004).
Google Scholar 
32.Bradshaw, E. L. & Bateson, P. Welfare implications of culling red deer (Cervus elaphus). Anim. Welf. 9, 3–24 (2000).
Google Scholar 
33.Sheriff, M. J., Dantzer, B., Delehanty, B., Palme, R. & Boonstra, R. Measuring stress in wildlife: Techniques for quantifying glucocorticoids. Oecologia 166, 869–887 (2011).ADS 
PubMed 
Article 

Google Scholar 
34.Hellhammer, D. H., Wüst, S. & Kudielka, B. M. Salivary cortisol as a biomarker in stress research. Psychoneuroendocrinology 34, 163–171 (2009).CAS 
PubMed 
Article 

Google Scholar 
35.Palme, R., Rettenbacher, S., Touma, C., El-Bahr, S. M. & Möstl, E. Stress hormones in mammals and birds: Comparative aspects regarding metabolism, excretion, and noninvasive measurement in fecal samples. Ann. N. Y. Acad. Sci. 1040, 162–171 (2005).ADS 
CAS 
PubMed 
Article 

Google Scholar 
36.Kanitz, E., Otten, W., Tuchscherer, M. & Manteuffel, G. Effects of prenatal stress on corticosteroid receptors and monoamine concentrations in limbic areas of suckling piglets (Sus scrofa) at different ages. J. Vet. Med. Ser. A 50, 132–139 (2003).CAS 
Article 

Google Scholar 
37.Campbell, E. A. et al. Plasma corticotropin-releasing hormone concentrations during pregnancy and parturition. J. Clin. Endocrinol. Metab. 64, 1054–1059 (1987).CAS 
PubMed 
Article 

Google Scholar 
38.Seth, S., Lewis, A. J. & Galbally, M. Perinatal maternal depression and cortisol function in pregnancy and the postpartum period: A systematic literature review. BMC Pregn. Childbirth 16, 124 (2016).Article 
CAS 

Google Scholar 
39.Gethöffer, F. Reproduktionsparameter und Saisonalität der Fortpflanzung des Wildschweins (Sus scrofa) in drei Untersuchungsgebieten Deutschlands (University of Veterinary Medicine Hannover, 2005).
Google Scholar 
40.Frauendorf, M., Gethöffer, F., Siebert, U. & Keuling, O. The influence of environmental and physiological factors on the litter size of wild boar (Sus scrofa) in an agriculture dominated area in Germany. Sci. Total Environ. 541, 877–882 (2016).ADS 
CAS 
PubMed 
Article 

Google Scholar 
41.Gethöffer, F., Sodeikat, G. & Pohlmeyer, K. Reproductive parameters of wild boar (Sus scrofa) in three different parts of Germany. Eur. J. Wildl. Res. 53, 287–297 (2007).Article 

Google Scholar 
42.DWD. Deutscher Wetterdienst—Wetter und Klima—Klimadaten (2019). https://www.dwd.de. Accessed 01 Oct 2019.43.Keuling, O., Stier, N. & Roth, M. Annual and seasonal space use of different age classes of female wild boar Sus scrofa L.. Eur. J. Wildl. Res. 54, 403–412 (2008).Article 

Google Scholar 
44.Malmsten, A., Jansson, G., Lundeheim, N. & Dalin, A.-M. The reproductive pattern and potential of free ranging female wild boars (Sus scrofa) in Sweden. Acta Vet. Scand. 59, 52 (2017).PubMed 
PubMed Central 
Article 

Google Scholar 
45.R Core Team. R: A Language and Environment for Statistical Computing Version R3.5.2.  R Foundation for Statistical Computing, Vienna, Austria.  http://www.R-project.org/ (2018).46.Dunn, O. J. Multiple comparisons using rank sums. Technometrics 6, 241–252 (1964).Article 

Google Scholar 
47.Ogle, D. H., Wheeler, P. & Dinno, A. FSA: Fisheries Stock Analysis. R Package Version 0.8.25. https://github.com/droglenc/FSA (2019).48.Kassambara, A. ggpubr: ‘ggplot2’ Based Publication Ready Plots. R Package Version 0.2.3. http://www.sthda.com/english/rpkgs/ggpubr (2019).49.Palme, R. Measuring fecal steroids: Guidelines for practical application. Ann. N. Y. Acad. Sci. 1046, 75–80 (2005).ADS 
CAS 
PubMed 
Article 

Google Scholar 
50.Cockrem, J. F. Individual variation in glucocorticoid stress responses in animals. Gen. Comp. Endocrinol. 181, 45–58 (2013).CAS 
PubMed 
Article 

Google Scholar 
51.Mormède, P. et al. Exploration of the hypothalamic-pituitary-adrenal function as a tool to evaluate animal welfare. Physiol. Behav. 92, 317–339 (2007).PubMed 
Article 
CAS 

Google Scholar 
52.Goymann, W. Noninvasive monitoring of hormones in bird droppings: Physiological validation, sampling, extraction, sex differences, and the influence of diet on hormone metabolite levels. Ann. N. Y. Acad. Sci. 1046, 35–53 (2005).ADS 
CAS 
PubMed 
Article 

Google Scholar 
53.Guilliams, T. G. & Edwards, L. Chronic stress and the HPA axis: Clinical assessment and therapeutic considerations. Stand. 9, 1–12 (2010).
Google Scholar 
54.Merta, D., Mocala, P., Pomykacz, M. & Frackowiak, W. Autumn-winter diet and fat reserves of wild boars (Sus scrofa) inhabiting forest and forest-farmland environment in south-western Poland. Folia Zool. 63, 95–102 (2014).Article 

Google Scholar 
55.Poteaux, C. et al. Socio-genetic structure and mating system of a wild boar population. J. Zool. 278, 116–125 (2009).Article 

Google Scholar 
56.Kaminski, G., Brandt, S., Baubet, E. & Baudoin, C. Life-history patterns in female wild boars (Sus scrofa): Mother–daughter postweaning associations. Can. J. Zool. 83, 474–480 (2005).Article 

Google Scholar 
57.Krause, J. & Ruxton, G. D. Living in Groups. Oxford Series in Ecology and Evolution (Oxford University Press, 2002).
Google Scholar 
58.Kudielka, B. M. & Kirschbaum, C. Sex differences in HPA axis responses to stress: A review. Biol. Psychol. 69, 113–132 (2005).PubMed 
Article 

Google Scholar 
59.Balhara, Y. S., Verma, R. & Gupta, C. Gender differences in stress response: Role of developmental and biological determinants. Ind. Psychiatry J. 20, 4 (2012).Article 

Google Scholar 
60.Sutherland, M. A., Rodriguez-Zas, S. L., Ellis, M. & Salak-Johnson, J. L. Breed and age affect baseline immune traits, cortisol, and performance in growing pigs. J. Anim. Sci. 83, 2087–2095 (2005).CAS 
PubMed 
Article 

Google Scholar 
61.Foury, A. et al. Stress hormones, carcass composition and meat quality in Large White × Duroc pigs. Meat Sci. 69, 703–707 (2005).CAS 
PubMed 
Article 

Google Scholar 
62.Ruis, M. A. W. et al. The circadian rhythm of salivary cortisol in growing pigs: Effects of age, gender, and stress. Physiol. Behav. 62, 623–630 (1997).CAS 
PubMed 
Article 

Google Scholar  More

Our results suggest that certain plant functional traits can be retrieved from simple RGB photographs. The key for this trait retrieval are deep learning-based Convolutional Neural Networks in concert with Big Data derived from open and citizen science projects. Although these models are subject to some noise, there is a wealth of applications for this approach, such as global trait maps, monitoring of trait shifts in time and the identification of large-scale ecological gradients. This way, the problem of limited data that still impedes us to picture global gradients7 could be alleviated by harnessing the exponentially growing iNaturalist database16. The performance of the CNN models across traits varied strongly, but revealed a clear trend: As expected, the more a trait referred to morphological features, the more accurate the predictions were. The models of the Baseline setup explained a substantial amount of the variance for LA and GH, whereas traits that are partly related to morphological features, SLA and SM, show moderate (R^2) values. The predictions of LNC and SSD explain almost none of the variance, suggesting that tissue constituents are not directly expressed or related to visible features. It also indicates that the strong covariance among these traits13 does not suffice in supporting their predictions from photographs. If the RGB images do not contain relevant information, the model will minimise the prediction errors by the regression-to-the-mean bias seen in Fig. 3 (especially lower panels).The value of informing the model on the known trait variability through an augmentation of the target values (Plasticity setup) depended on the results of the Baseline setup of the trait. That is, the better the predictive performance of the Baseline setup, the more the trait seemed to profit from the Plasticity setup, rendering it ineffective for SSD, LNC and SM (Fig. 3). Refraining to cling to species mean values by considering within-species trait variation has been applied before using conventional methods26, but to our knowledge has never been tried in CNN models, yet. We expected that providing a distribution of trait values rather than a single mean for each species can convey to the CNN that different trait realisations can be expected from the same species. Obviously, this idea can only work if a distribution rather than a single value is available for each species. The SM dataset, for instance, contained only one image per species (Table 1). In this case, the Plasticity setup reduced the predictive performance compared to the Baseline setup, possibly by increasing the discrepancy to the true trait value. Since the traits with more accurate predictions profited most from the Plasticity setup, we assume that it supports the model in learning to predict the trait expressions themselves rather than extracting them indirectly through taxa-specific morphological features. Given that we restricted the number of images per species to a maximum of 8, while successful deep learning-based plant species identification usually requires thousands of images12,22, it seems very unlikely that the models inferred traits from species-specific plant features visible in the imagery. The latter was underpinned by our finding that the predictions of most traits are void of phylogenetic autocorrelation (Supplementary Information 1 and Supplementary Table 5), indicating that taxonomic relationships were insignificant for the trait predictions. The absence of phylogenetic autocorrelation of the prediction errors underlines that the models did not learn species-specific features for most traits, as this would imply similar trait predictions for related species.On the contrary, the SSD model predictions express a phylogenetic signal (Supplementary Information 1 and Supplementary Table 5). Trait expressions are generally clustered under similar climatic conditions29,30,31. Simultaneously, climatic conditions constrain the geographic distribution of species and growth forms29,30,31,32. The SSD dataset is biased towards woody species (Table 1), which confines it to a smaller taxonomic range. Hence, the phylogenetic signal of SSD might result from its phylogenetic clustering and predominant dependence on bioclimatic information rather than on RGB imagery (Fig. 2).Nevertheless, the benefit of including climate information on temperature, precipitation and their seasonality8,20,21,26 on predicting trait expressions was confirmed for all traits in this study, which underlines the value of contextual constraints in CNN models10 (see below for a discussion of the relevance of climate vs. image data). This also highlights the general flexibility of deep learning frameworks in adapting to variable input data from different scales and sensors10, which makes them a promising tool for ecological research. Our results particularly revealed this effect for SSD, SLA and LA, whereas it was smaller for GH, LNC and SM (Fig. 2). For the latter traits, other physical constraints such as disturbance33,34, seasonal variation35,36 and soil conditions6,26,28 come into consideration. As the focus of the Worldclim setup was to show that contextual cues can improve the trait retrieval from photographs rather than identifying the best set of auxiliary data, we confined the analysis to the most promising20,26 data source (WorldClim37).In the Worldclim setup, a single model accumulated knowledge about the trait learning task. Combined predictions of different CNN models, however, have shown to surpass the predictive performance of single CNN, e.g. in plant species identification tasks22. Each of the CNN models is prone to literally ‘look’ at different aspects of the learning task by focusing on different image features. Previous research also showed increased model performance in a trait prediction task in case of ensembles of regression and machine learning models26. Accordingly, and as demonstrated by our results, an ensemble approach seems promising to further enhance predictive performance of CNN models concerning trait prediction.The predictive performance of these Ensembles has shown to be reproducible with different sets of training images (cp. Figs. 2, 3, Supplementary Fig. 2). In our heterogeneous dataset, model performance was not affected by different growth forms, image qualities and image-target distances (Fig. 4). Different growth forms and plant functional types show their own characteristic trait spectrum13. Possibly, contextual cues within the image might have supported the CNN in inferring the plant functional type of a species, e.g. by a long-distance image being indicative of a tree species. Yet, since the majority of the images only shows single plant organs on close-up photographs (Fig. 4), we assume that the trait predictions are not confounded by the identification of growth forms. Furthermore, the absence of a phylogenetic signal in the prediction errors for most traits highlights the model’s ability to generalise by extracting trait information independent of taxonomic relationships, meaning that the models (except for SSD) did not learn species-specific mean trait expressions (see Supplementary Information 1 and Supplementary Table 5).Additionally, we disclosed the high generalisability of these results by investigating the datasets’ underlying distributions both spatially (Supplementary Fig. 3) and across biomes (Supplementary Fig. 4). Although some regions such as Central Europe and North America show higher data coverage, the datasets used for this study contain data across all biomes and regions on Earth. Therefore and despite this clustering, we expect the models to be applicable for all biomes around the globe. This was highlighted by an additional analysis showing that the predictive performance of the models is reasonably constant across biomes (Supplementary Fig. 5). As suggested by refs.38,39, we tolerated a certain spatial bias in favor of larger datasets. Although the SSD dataset predominantly contained woody species, neither of the six datasets expressed an exclusion of either growth form (Table 1, Fig. 4).The application of our models to global gradients of traits revealed that our GTDM indeed cover macroecological patterns and trends known from other publications: The latitudinal distributions could roughly be confirmed for GH26, LNC26,27 and SM6,8 (Fig. 5). Predicted trends for maximum leaf size hint at the applicability of our GTDM of LA40. The trait gradients for North America were confirmed for SLA6,8,26,27,28, SM6,8, LNC27 and SSD6 alike. Although based on different input data and modelling methods, the major global latitudinal gradients found in previous studies could be reproduced by our GTDM, which indicates the plausibility of the latter6,8,26,27.We further validated the GTDM quantitatively by means of correlations with other GTDM. Regarding SSD, the detected high correlations might be due to method similarity, as our GTDM product of SSD primarily builds upon climate data (see above), just as ref.6,26. For GH, SLA and SM, however, the high correlations are unlikely to result from climate data exclusively, as the explained variances of the RGB imagery ((R^2) of Plasticity setup) are higher than the additional contributions of the Worldclim setup (approx. 94%, 70% and 79% share of imagery on total explained variance, respectively; Supplementary Table 2). We decoupled the GTDM products from bioclimatic information in an additional analysis (Supplementary Fig. 6). Remarkably, the macroecological patterns could roughly be reproduced when the GTDM were based exclusively on RGB imagery, which shows that the bioclimatic information merely serves to smooth the macroecological trait patterns for most of the traits.Despite of all GTDM being at least partly build upon climate data and using trait data from the same source (TRY database), some GTDM of SLA and all GTDM of LNC vary strongly in their correlations (Supplementary Fig. 1). On the one hand, this might indicate that LNC varies at a different scale, e.g. on account of its seasonal and within-species variation35,36. On the other hand, other GTDM products are based on mean trait values weighted by abundances of plant functional types27,28 rather than single trait predictions, which might explain negative or non-significant correlations as well.Hence, a potential pitfall of the presented approach is that it is prone to express an observation bias, e.g. by citizen scientists only taking pictures of the most striking species. The sampling design underlying the GTDM does not account for plant community composition, meaning that we cannot tell if plant photographs at a certain location represent the actual community structure. Since many images contain more than one individual plant and different species, the CNN model predictions, however, might be based on more than one species, thereby partly resembling trait expressions of the community. The representativeness of trait data for plant communities, though, remains an ubiquitous problem of global trait maps, including those fully based on trait data from the TRY database7, since every available dataset is far from representing the actual plant community composition7. Hence, at present our GTDM have to be considered a plausibility check of the model predictions rather than an application-ready trait map product, not least because the sampling of images might not be representative of the respective plant community.Nevertheless, our results indicate that exploiting a Big Data approach is viable to reveal macroecological trait patterns, maybe because the most striking species of an ecosystem are likely to suffice in describing its functional footprint5. Since the strong growth of the iNaturalist database leads to a steadily increasing geographic coverage, the representativeness of these data is likely to grow as well. A recent study investigating the records of FloraIncognita12, a citizen-science and deep learning-based application for identifying plant species from photographs, suggested that such crowd-sourced data can reproduce primary dimensions in plant community composition41. This underlines the future potential of harnessing citizen science databases for identifying these patterns. Here, we demonstrated the practical value and applicability of the CNN models by producing GTDM that were able to reproduce known macroecological trait patterns while displaying one anticipated application of this method. Additionally, in these GTDM we bypass the issue of spatial error analysis that is challenging for most GTDM products26 by obtaining a potentially arbitrary number of observations in light of the strongly increasing number of observations in iNaturalist, almost rendering an extrapolation obsolete. Our GTDM are based on individual trait measurements rather than estimated on behalf of a small set of covariates, which is typical for climate-based GTDM26. Since plant traits vary strongly within species17,18,19, these measurements express a high practical relevance. As the iNaturalist plant photograph database is witnessing an exponential growth of data inputs, the potential of exploiting this data source for plant trait predictions is growing rapidly. It is worth mentioning that this approach also led to the first publication of a GTDM of mean LA (available for download under https://doi.org/10.6084/m9.figshare.13312040), since former publications were limited to modelling upper limits of LA based on climatic constraints40.Future studies building on our work, which benefit from the ever-growing data accumulation of both the iNaturalist and TRY database, might not face restrictions of dataset size as we did in our study. This might allow for more representative samples in future studies, e.g. enabling to stratify training data by species while simultaneously balancing the trait distribution. This might support a reduction of the regression-to-the-mean bias seen in all of the results (Fig. 3) by avoiding to overrepresent common trait expressions. Another possible approach would be to select only species with particularly low variability for model training, since it decreases the chance of incorporating images showing plants with an extreme trait expression that differ strongly from the chosen mean trait values from TRY. By that, we might be able to derive more reliable and accurate predictions in the context of weakly supervised learning by reducing noise in the training data.Although weakly supervised learning approaches generally have shown to be an effective way of compensating a shortage of individually labeled data42,43, an image dataset including in-field trait measurements under natural conditions representing the global trait spectrum would be necessary for a conclusive validation. In our study, it even remains unclear to what extent the trait values actually refer to the individual plant shown in an image, particularly as the images sometimes show more than one individual plant and more than one species (Supplementary Fig. 7). This may hinder the model from predicting a trait value corresponding to the dominant species in the image (but might also partly resemble the community composition, see above). Although we attempted to compensate the lack of a dataset that enables a conclusive validation by eliminating possible biases concerning image settings (Fig. 4), growth forms (Fig. 4), phylogenetic autocorrelation (Supplementary Information 1, Supplementary Table 5), predictions based only on climate data (Supplementary Fig. 6), predictive performance across biomes (Supplementary Fig. 5), a training dataset subject to limited geographic or climatic coverage (Supplementary Figs. 3, 4) and effects of a specific set of training data (Supplementary Fig. 2), we cannot conclusively prove that the model predictions are based on causal relationships. Our model results suggest that the trait predictions reflect the feature space of natural trait expressions (Fig. 3), but an in-depth analysis of the image features the models learned for inferring the respective traits will be necessary to rule out any possible biases in future studies. An explicit analysis might involve investigating which plant organs are relevant for the trait predictions by means of feature attribution techniques and could ultimately provide clear evidence. This may not only enable to build trust in such artificial intelligence (AI) models, but also to generate new knowledge from them in order to deepen our understanding of plant morphology and trait covariance.Nevertheless, this study can only be considered a pioneering work testing the feasibility of the approach, as application-ready models require a conclusive and explicit validation. A dataset enabling this has to incorporate image-trait pairs measured and photographed on the same individuals. One possible solution would be to generate a database of plant traits including respective photographs, which then can serve as a benchmark for future studies. More

Log normal concentrations in the environmentEmpirical concentration distributions in the environment exhibit a wide variability, ranging from familiar parametric forms, e.g., normal and log-normal distributions, to more complex shapes9,10,11,12. Overall, the concentration distribution reflects the different processes that govern environmental fate, e.g., emissions/formation, transport processes, source mixing, sinks and dependences on other variables and processes. In principle, we may therefore extract mechanistic insights into the lifecycle of an environmental component from analysis of the concentration distribution. Although such deconvolution is intangible in the general case, there are specific cases where there are known mechanistic origins for certain distributions. For instance, source mixing in the environment will tend the distribution towards normal.In this paper we present a model for how log-normal distributions may emerge in the environment (Fig. 1A). This topic has been addressed in a few earlier studies6,7,8. A joint argument in these is the implication of the ‘Multiplicative Central Limit Theorem’ (or ‘Gibrat’s law’), which applies to processes that include the product of many random variables. If we take the logarithm of the product of many random variables, we have the sum of many random variables, by which the Central Limit Theorem applies, which predict a normal distribution. Back-transforming the logarithm, yields a log-normal distribution. Although this heuristic argumentation does not provide specific physico/chemical mechanisms as to how the multiplication of multiple random variables may commonly appear in the environment, it does provide an intuitive framework.The present contribution is based on a physico/chemical first order exponential kinetics model (Fig. 1B). Starting with an ordinary differential equation (Eq. 1), we introduce a stochastically variable rate, resulting in a stochastic differential equation (Eq. 2). By solving the corresponding Fokker–Planck equation (FPE) (see Theory section and SI for details) we derive the log-normal distribution (Eq. 3), under certain assumptions about stochastic variability. Since exponential kinetics are commonly observed for many processes in the environment, this model provides a potential explanation for the relative ubiquity of lognormal distributions in the environment.Observations of lognormal-like concentration distributions in the environment include a wide array of components (e.g., aerosols; bulk organic carbon; heavy metals; organic pollutants; minerals; trace gases; inorganic ions; humic matter; biomarkers; radionuclides; microplastics; pharmaceuticals and pesticides) in various environmental sub-compartments (e.g., groundwater; watersheds; urban air; precipitation; rocks; stormwater; indoor air; soils; wastewater; marine sediments; landfills; lake sediments; peat; glaciers; background air; biota and humans)2,3,12,13,14,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38. However, even though log normality is relatively common in the environment, we emphasize that it by no means is generally observed10,11,12: see discussion in the next section.Ascribing a data set to a specific parametric form, e.g., log-normal, may provide mechanistic insights, but also typically simplifies statistical treatment11. Among multiple approaches, a straight-forward approach to test for log-normality is by log-transforming the data and then use one of the many well-established tests for normality, e.g., the Shapiro-Wilks test9,39,40. Specific aspects of analysis of log-normal data sets, e.g., the impact of non-detects or data averaging, or data sampling strategies, have been discussed in some detail in the literature12,41,42,43. However, even though we may attribute a certain data set to a parametric distribution, there are many empirical situations where the data appears to follow a complex/un-known distribution. Can we identify some general principles that may help disentangle observed variability, even beyond common parametric forms?Three mechanisms influencing concentration distributionsAny environmental system is governed by a multitude of different processes, occurring simultaneously. Fortunately, some mechanisms appears to be more prevalent than others, and we can make simplifications, allowing modeling, e.g., the currently presented model for log-normal dependence. To attempt a more general description for concentration distributions, it may prove useful to consider a few different principal mechanisms. Here we explore three such mechanisms:KineticsThe model presented here is a common first order approximation of kinetics, e.g., of source and sink processes (Fig. 2A–B). In reality, dynamical systems may be highly complex, beyond the present formulation with a constant deterministic part (μk, Eq. 2), with multiple coupled components, non-linearities, feedbacks and heterogenous interactions. How such more complex kinetics influence concentration distributions is a topic for further scientific discovery.Figure 2Numerical simulations (random sampling) of time-series concentration data and corresponding distributions, illustrating how different processes may influence concentration distributions. Panels (A)–(B): Log normally-distributed data (Eq. 3), e.g., influenced by exponential kinetics. Panels (C)–(D): Data reflecting the random fluctuation between three different log-normally distributed states (green, blue and red), yielding a multi-modal (mixture) distribution. Panels (E)–(F): Random mixing of the three distributions from panels (C)–(D), yielding a convolution distribution, tending towards a normal distribution. Panels (G)–(H): The log-normal distribution of panels (A)–(B), modulated by an oscillatory function (here sine function), illustrating the impact of, e.g. diurnal or seasonal cycles on observed distributions. The simulations were conducted using Matlab (ver. R2019b).Full size imageMixingThe concentration of a component in the environment is typically the sum of emissions from many sources. If these combine randomly, the concentration distribution approach normality (Fig. 2E–F). However, this type represents one limit of mixing—where the resulting distribution at each time point is the sum distribution (convolution) of many variables. Another limit is where the contributions from different sources are separated, such that the concentration at any given measurement point (e.g., time point) reflects one individual source (Fig. 2C–D). An example could be an atmospheric measurement site located between two cities: by wind direction each time point is mainly dominated by either city. In this limit the concentration data will tend to a multi-modal (or ‘mixture’) distribution.External variablesIn the model described in the Theory section the non-stochastic part of Eq. (2) is a constant (μk). But this parameter may also vary with external parameters, e.g., state variables. For instance, diurnal or seasonal variations of temperature, pressure or sunlight, including phase transitions, may strongly influence concentration distributions (Fig. 2G–H).These three classes of mechanisms influence the lifecycle of a component in the environment, but their relative importance regarding how concentration distributions are affected may be highly variable. For certain instances it may be possible to ascribe a certain parametric function to the data, e.g., a normal distribution for well-mixed sources or a log-normal suggesting a kinetic domain. However, for situations with more complex, multi-modal or broad distributions both statistical treatment and interpretations are more challenging. For instance, quite similarly looking distributions may emerge from rather different mechanisms, e.g., dependence on an externally oscillating parameter or a stochastically jumping mixture distribution (e.g., compare Fig. 2C–D with Fig. 2G–H).OutlookThe shape of empirical concentration distributions is determined by the lifecycle dynamics and thereby contain information of environmental fate. In this paper we show that log-normality suggests influence by first order exponential kinetics (Eqs. 1 and 5; Fig. 1A–B). Reflecting the overall non-equilibrium state of environmental systems, there are a number of different processes/mechanisms that may exhibit exponential kinetics and thereby drive concentrations towards log-normality, e.g., emissions/chemical formation (e.g., of primary or secondary pollutants), degradation/decomposition (e.g., chemical reactions or deposition), kinetic transfer between different pools/layers/reservoirs (e.g., between the troposphere and the stratosphere) or kinetic partitioning (e.g., between gas and liquid phases). Some of these processes, e.g., sink kinetics, are active throughout the lifecycle of a component, and thereby continuously push concentrations towards log-normality.Overall, the implications of log-normal concentration distributions span a broad spectrum of potential applications, ranging from data analysis methodologies, sampling strategies, emissions estimation, source apportionment calculations, modelling of chemical fate, estimation of toxic exposure to future climate scenarios9,25,29,41,42,43,44,45,46,47,48,49,50,51,52. Given the general mathematical formulation of the model (Eq. 2), we note that the applicability extends to log-normal concentration distributions also beyond Environmental systems53,54.A specific example, where mechanistic insights may be derived from analysis of concentration distributions is the estimation of lifetimes (τ = 1/k, Eq. 1)55,56. Our present findings suggest that observation of log-normally distributed concentration data, perhaps especially at remote or receptor sites, may indicate sink kinetics, and may therefore provide insights into the sink rate (Eqs. 4a–b and 5). Another specific situation where concentration distribution analysis is of importance is the common challenge of trend detection in environmental data11. For certain time-series data, log-transformation is commonly used prior to statistical analysis51,57. The present analysis may provide a physico-chemical motivation for such transformations, potentially contributing to interpretations. A third specific example regards the analysis of ratios of diagnostic markers, which are common tools to assess sources and processes across Earth and Environmental sciences. Examples include ratios of chemical markers and isotope signatures, where the latter commonly are reported as concentration ratios. The ratio of two log-normal random variables is another log-normal variable58. We may then predict that if the overall concentrations are log-normal, then, e.g., isotope signatures should also be log-normal, with potential implications for data analysis methodology and interpretation.Finally, we note that rates in environmental systems often are proportional to concentrations, e.g., reactants. As an example, the sink rate for CO is proportional to the OH concentration (see Theory). Consequently, if the concentration is log-normal, so is the rate. Furthermore, fluxes (e.g., emission or sink fluxes) are often defined as the product of a rate and a concentration (or at least the amount, by some proportional measure). Taken together, this suggests that log-normal concentration distributions may imply log-normal distributions also for rates and/or fluxes for certain systems, with implications for, e.g., emission estimation or box models47,49,59.All-in-all, this paper presents a mechanistic model for how log-normal concentrations may emerge in the environment, which in turn suggests an explanation for the relative abundance. More

1.Wilcove, D. S., McLellan, C. H. & Dobson, A. P. Habitat fragmentation in the temperate zone. Conserv. Biol. 6, 237–256 (1986).
Google Scholar 
2.Crooks, K. R. et al. Quantification of habitat fragmentation reveals extinction risk in terrestrial mammals. Proc. Natl. Acad. Sci. USA 114, 7635–7640 (2017).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
3.Fahrig, L. Effects of habitat fragmentation on biodiversity. Annu. Rev. Ecol. Evol. Syst. 34, 487–515 (2011).Article 

Google Scholar 
4.Okie, J. G. & Brown, J. H. Niches, body sizes, and the disassembly of mammal communities on the Sunda Shelf islands. Proc. Natl. Acad. Sci. USA 106, 19679–19684 (2009).ADS 
CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
5.Viveiros De Castro, E. B. & Fernandez, F. A. S. Determinants of differential extinction vulnerabilities of small mammals in Atlantic forest fragments in Brazil. Biol. Conserv. 119, 73–80 (2004).Article 

Google Scholar 
6.Feeley, K. J. & Terborgh, J. W. Direct versus indirect effects of habitat reduction on the loss of avian species from tropical forest fragments. Anim. Conserv. 11, 353–360 (2008).Article 

Google Scholar 
7.Prugh, L. R., Hodges, K. E., Sinclair, A. R. E. & Brashares, J. S. Effect of habitat area and isolation on fragmented animal populations. Proc. Natl. Acad. Sci. USA 105, 20770–20775 (2008).ADS 
CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
8.Crooks, K. R., Burdett, C. L., Theobald, D. M., Rondinini, C. & Boitani, L. Global patterns of fragmentation and connectivity of mammalian carnivore habitat. Philos. Trans. R. Soc. B Biol. Sci. 366, 2642–2651 (2011).Article 

Google Scholar 
9.Janecka, J. E. et al. Genetic differences in the response to landscape fragmentation by a habitat generalist, the bobcat, and a habitat specialist, the ocelot. Conserv. Genet. 17, 1093–1108 (2016).Article 

Google Scholar 
10.Creel, S. Four factors modifying the effect of competition on Carnivore population dynamics as illustrated by African wild dogs. Conserv. Biol. 15, 271–274 (2001).Article 

Google Scholar 
11.Crooks, K. R. Relative sensitivities of mammalian carnivores to habitat fragmentation. Conserv. Biol. 16, 488–502 (2002).Article 

Google Scholar 
12.Ripple, W. J. et al. Status and ecological effects of the world’s largest carnivores. Science 343 (2014).13.Sanderson, C. E., Jobbins, S. E. & Alexander, K. A. With Allee effects, life for the social carnivore is complicated. Popul. Ecol. 56, 417–425 (2014).Article 

Google Scholar 
14.Kamler, J. F. et al. Cuon alpinus. The IUCN Red List of Threatened Species 2015: e.T5953A72477893. https://doi.org/10.2305/IUCN.UK.2015-4.RLTS.T5953A72477893.en (2015).15.Bashir, T., Bhattacharya, T., Poudyal, K., Roy, M. & Sathyakumar, S. Precarious status of the endangered dhole cuon alpinus in the high elevation eastern himalayan habitats of khangchendzonga biosphere reserve, Sikkim, India. Oryx 48, 125–132 (2014).Article 

Google Scholar 
16.Pal, R., Thakur, S., Arya, S., Bhattacharya, T. & Sathyakumar, S. Recent records of dhole (Cuon alpinus, Pallas 1811) in Uttarakhand, Western Himalaya, India. Mammalia 82, 614–617 (2018).Article 

Google Scholar 
17.Karanth, K. K., Nichols, J. D., UllasKaranth, K., Hines, J. E. & Christensen, N. L. The shrinking ark: Patterns of large mammal extinctions in India. Proc. R. Soc. B Biol. Sci. 277, 1971–1979 (2010).Article 

Google Scholar 
18.Keyghobadi, N. The genetic implications of habitat fragmentation for animals. Can. J. Zool. 85, 1049–1064 (2007).Article 

Google Scholar 
19.Lourenço, A., Álvarez, D., Wang, I. J. & Velo-Antón, G. Trapped within the city: Integrating demography, time since isolation and population-specific traits to assess the genetic effects of urbanization. Mol. Ecol. 26, 1498–1514 (2017).PubMed 
Article 
CAS 
PubMed Central 

Google Scholar 
20.Ghaskadbi, P., Habib, B. & Qureshi, Q. A whistle in the woods: An ethogram and activity budget for the dhole in central India. J. Mammal. 97, 1745–1752 (2016).Article 

Google Scholar 
21.Karanth, K. U. & Sunquist, M. E. Behavioural correlates of predation by tiger (Panthera tigiris), leopard (Panthera pardus) and dhole (Cuon alpinus) in Nagarahole, India. J. Zool. Lond. 250, 255–265 (2000).Article 

Google Scholar 
22.Johnsingh, A. J. T. Reproduction and social behaviour of the dhole, Cuon alpinus (Canidae). J. Zool. 198, 443–463 (1982).Article 

Google Scholar 
23.Ngoprasert, D. & Gale, G. A. Tiger density, dhole occupancy, and prey occupancy in the human disturbed Dong Phayayen—Khao Yai Forest Complex, Thailand. Mammal. Biol. 95, 51–58 (2019).Article 

Google Scholar 
24.Selvan, K. M., Lyngdoh, S., Habib, B. & Gopi, G. V. Population density and abundance of sympatric large carnivores in the lowland tropical evergreen forest of Indian Eastern Himalayas. Mammal. Biol. 79, 254–258 (2014).Article 

Google Scholar 
25.Jenks, K. E. et al. Comparative movement analysis for a sympatric dhole and golden jackal in a human-dominated landscape. Raffles Bull. Zool. 63, 546–554 (2015).
Google Scholar 
26.Modi, S., Habib, B., Ghaskadbi, P., Nigam, P. & Mondol, S. Standardization and validation of a panel of cross-species microsatellites to individually identify the Asiatic wild dog (Cuon alpinus). PeerJ 7, e7453 (2019).PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
27.Modi, S. et al. Noninvasive DNA-based species and sex identification of Asiatic wild dog (Cuonalpinus). J. Genet. 97, 1457–1461 (2018).CAS 
PubMed 
Article 
PubMed Central 

Google Scholar 
28.Iyengar, A. et al. Phylogeography, genetic structure, and diversity in the dhole (Cuon alpinus). Mol. Ecol. 14, 2281–2297 (2005).CAS 
PubMed 
Article 
PubMed Central 

Google Scholar 
29.Durbin, L., Venkataraman, A. & Hedges, S. D. J. Dhole (Cuon alpinus). In Status Survery and Conservation Action Plan. Canids: Foxes, Wolves, Jackals and Dogs (eds. Sillero-Zubiri, C., Hoffman, M. & Macdonald, D. W.) 210–219 (2004).30.Smith, O. & Wang, J. When can noninvasive samples provide sufficient information in conservation genetics studies?. Mol. Ecol. Resour. 14, 1011–1023 (2014).CAS 
PubMed 

Google Scholar 
31.Godinho, R. et al. Real-time assessment of hybridization between wolves and dogs: Combining noninvasive samples with ancestry informative markers. Mol. Ecol. Resour. 15, 317–328 (2015).CAS 
PubMed 
Article 

Google Scholar 
32.Venkataraman, A. B., Arumugam, R. & Sukumar, R. The foraging ecology of dhole (Cuon alpinus) in Mudumalai Sanctuary, southern India. J. Zool. 237, 543–561 (1995).Article 

Google Scholar 
33.Srivathsa, A., Karanth, K. U., Kumar, N. S. & Oli, M. K. Insights from distribution dynamics inform strategies to conserve a dhole Cuon alpinus metapopulation in India. Sci. Rep. 9, 1–12 (2019).ADS 
CAS 
Article 

Google Scholar 
34.Reddy, C. S., Sreelekshmi, S., Jha, C. S. & Dadhwal, V. K. National assessment of forest fragmentation in India: Landscape indices as measures of the effects of fragmentation and forest cover change. Ecol. Eng. 60, 453–464 (2013).Article 

Google Scholar 
35.Dutta, T., Sharma, S. & DeFries, R. Targeting restoration sites to improve connectivity in a tiger conservation landscape in India. PeerJ 6, e5587 (2018).PubMed 
PubMed Central 
Article 

Google Scholar 
36.Mondal, I., Habib, B., Talukdar, G. & Nigam, P. Triage of means: Options for conserving tiger corridors beyond designated protected lands in India. Front. Ecol. Evol. 4, 2–7 (2016).ADS 
Article 

Google Scholar 
37.Lowther, A. D., Harcourt, R. G., Goldsworthy, S. D. & Stow, A. Population structure of adult female Australian sea lions is driven by fine-scale foraging site fidelity. Anim. Behav. 83, 691–701 (2012).Article 

Google Scholar 
38.Marsden, C. D. et al. Spatial and temporal patterns of neutral and adaptive genetic variation in the endangered African wild dog (Lycaon pictus). Mol. Ecol. 21, 1379–1393 (2012).PubMed 
Article 
PubMed Central 

Google Scholar 
39.Yumnam, B. et al. Prioritizing tiger conservation through landscape genetics and habitat linkages. PLoS ONE 9 (2014).40.Dutta, T. et al. Fine-scale population genetic structure in a wide-ranging carnivore, the leopard (Panthera pardus fusca) in central India. Divers. Distrib. 19, 760–771 (2013).Article 

Google Scholar 
41.Thatte, P. et al. Human footprint differentially impacts genetic connectivity of four wide-ranging mammals in a fragmented landscape. Divers. Distrib. 26, 299–314 (2020).Article 

Google Scholar 
42.Slatkin M. Gene flow and population structure. Ecol. Genet. 3–17 (1994).43.Bhandari, A., Ghaskadbi, P., Nigam, P. & Habib, B. Dhole pack size variation: Assessing effect of Prey availability and Apex predator. Ecol. Evol. 00, 1–12 (2021).
Google Scholar 
44.Davies, K. F., Margules, C. R. & Lawrence, J. F. Which traits of species predict population declines in experimental forest fragments?. Ecology 81, 1450–1461 (2000).Article 

Google Scholar 
45.Bhatt, S., Biswas, S., Karanth, K., Pandav, B. & Mondol, S. Genetic analyses reveal population structure and recent decline in leopards (Panthera pardus fusca) across the Indian subcontinent. PeerJ 8, e8482 (2020).PubMed 
PubMed Central 
Article 

Google Scholar 
46.Mondol, S., Karanth, K. U. & Ramakrishnan, U. Why the Indian subcontinent holds the key to global tiger recovery. PLoS Genet. 5 (2009).47.Nijman, V. et al. Illegal wildlife trade–surveying open animal markets and online platforms to understand the poaching of wild cats. Biodiversity 20, 58–61 (2019).Article 

Google Scholar 
48.Srivathsa, A., Sharma, S., Singh, P., Punjabi, G. A. & Oli, M. K. A strategic road map for conserving the Endangered dhole Cuon alpinus in India. Mammal. Rev. 50, 399–412 (2020).Article 

Google Scholar 
49.Richards, J. F. & Elizabeth, P. F. A century of land-use change in South and Southeast Asia. In Effects of land-use change on atmospheric CO2 concentrations 15–66 (1994).50.Goldewijk, K. K. & Ramankutty, N. Land use changes during the past 300 years (EOLSS Publisher Co., 2009).
Google Scholar 
51.Sharma, S. et al. Forest corridors maintain historical gene flow in a tiger metapopulation in the highlands of central India. Proc. R. Soc. B Biol. Sci. 280, 14 (2013).
Google Scholar 
52.Rangarajan, M. Fencing the forest: Conservation and ecological change in India’s central provinces 1860–1914 (1999).53.Gadgil, M. Towards an ecological history of India. Econ. Pol. Wkly. 20, 1909–1911 (2011).
Google Scholar 
54.Bebarta, K. C. Teak; ecology, silviculture, management and profitability (International Book Distributors, 1999).
Google Scholar 
55.Waples, R. S. & England, P. R. Estimating contemporary effective population size on the basis of linkage disequilibrium in the face of migration. Genetics 189, 633–644 (2011).PubMed 
PubMed Central 
Article 

Google Scholar 
56.Frankham, R., Bradshaw, C. J. A. & Brook, B. W. Genetics in conservation management: Revised recommendations for the 50/500 rules, Red List criteria and population viability analyses. Biol. Conserv. 170, 56–63 (2014).Article 

Google Scholar 
57.de Manuel, M. et al. The evolutionary history of extinct and living lions. Proc. Natl. Acad. Sci. USA 117, 10927–10934 (2020).PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
58.Creel, S. Social organization and effective population size in carnivores. Behav. Ecol. Conserv. Biol. 264–265 (1998).59.Lande, R. & Barrowclough, G. Effective population size, genetic variation, and their use in population. Viable Popul. Conserv. 87–123 (1987).60.Neel, M. C. et al. Estimation of effective population size in continuously distributed populations: There goes the neighborhood. Heredity 111, 189–199 (2013).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
61.Girman, D. J. et al. Patterns of population subdivision, gene flow and genetic variability in the African wild dog (Lycaon pictus). Mol. Ecol. 10, 1703–1723 (2001).CAS 
PubMed 
Article 
PubMed Central 

Google Scholar 
62.Sacks, B. N., Mitchell, B. R., Williams, C. L. & Ernest, H. B. Coyote movements and social structure along a cryptic population genetic subdivision. Mol. Ecol. 14, 1241–1249 (2005).CAS 
PubMed 
Article 
PubMed Central 

Google Scholar 
63.Stronen, A. V. et al. Population genetic structure of gray wolves (Canis lupus) in a marine archipelago suggests island-mainland differentiation consistent with dietary niche. BMC Ecol. 14, 1–9 (2014).Article 

Google Scholar 
64.Wolf, C. & Ripple, W. J. Range contractions of the world’s large carnivores. R. Soc. Open Sci. 4 (2017).65.Walston, J. et al. Bringing the tiger back from the brink-the six percent solution. PLoS Biol. 8, 6–9 (2010).Article 
CAS 

Google Scholar 
66.Champion, H. G. & Seth, S. K. A revised survey of the forest types of India. (Manager of Publications, 1968).67.Biswas, S. et al. A practive faeces collection protocol for multidisciplinary research in wildlife science. Curr. Sci. 116, 1878 (2019).CAS 
Article 

Google Scholar 
68.Hallsworth, J. E., Nomura, Y. & Iwahara, M. Ethanol-induced water stress and fungal growth. J. Ferment. Bioeng. 86, 451–456 (1998).CAS 
Article 

Google Scholar 
69.van Oosterhout, C., Hutchinson, W. F., Wills, D. P. M. & Shipley, P. MICRO-CHECKER: Software for identifying and correcting genotyping errors in microsatellite data. Mol. Ecol. Notes 4, 535–538 (2004).Article 
CAS 

Google Scholar 
70.Broquet, T. & Petit, E. Quantifying genotyping errors in noninvasive population genetics. Mol. Ecol. 13, 3601–3608 (2004).CAS 
PubMed 
Article 

Google Scholar 
71.Kalinowski, S. T., Taper, M. L. & Marshall, T. C. Revising how the computer program CERVUS accommodates genotyping error increases success in paternity assignment. Mol. Ecol. 16, 1099–1106 (2007).PubMed 
Article 

Google Scholar 
72.Waits, L., Taberlet, P. & Luikart, G. Estimating the probability of identity among genotypesin natural populations: Cautions and guidelines. Mol. Ecol. 10, 249–256 (2001).CAS 
PubMed 
Article 

Google Scholar 
73.Valière, N. GIMLET: A computer program for analysing genetic individual identification data. Mol. Ecol. Notes 2, 377–379 (2002).
Google Scholar 
74.Excoffier, L., Laval, G. & Schneider, S. Arlequin (version 3.0): An integrated software package for population genetics data analysis. Evol. Bioinf. 1, 117693430500100 (2005).75.Pritchard, J. K. & Stephens, M. D. M. Inference of population structure using multilocus genotype data. Genetics 155, 945–959 (2000).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
76.Evanno, G., Regnaut, S. & Goudet, J. Detecting the number of clusters of individuals using the software STRUCTURE: A simulation study. Mol. Ecol. 14, 2611–2620 (2005).CAS 
PubMed 
Article 

Google Scholar 
77.Earl, D. A. & vonHoldt, B. M. STRUCTURE HARVESTER: A website and program for visualizing STRUCTURE output and implementing the Evanno method. Conserv. Genet. Resour. 4, 359–361 (2012).78.Kopelman, N. M., Mayzel, J., Jakobsson, M., Rosenberg, N. A. & Mayrose, I. Clumpak: a program for identifying clustering modes and packaging population structure inferences across K. Mol. Ecol. Resour. 15, 1179–1191 (2015).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
79.Caye, K., Deist, T. M., Martins, H., Michel, O. & François, O. TESS3: Fast inference of spatial population structure and genome scans for selection. Mol. Ecol. Resour. 16, 540–548 (2016).CAS 
PubMed 
Article 

Google Scholar 
80.Jombart, T. et al. Discriminant analysis of principal components: a new method for the analysis of genetically structured populations. BMC Genet. 11, 94 (2010).PubMed 
PubMed Central 
Article 

Google Scholar 
81.Jombart, T. Adegenet: A R package for the multivariate analysis of genetic markers. Bioinformatics 24, 1403–1405 (2008).CAS 
PubMed 
Article 
PubMed Central 

Google Scholar 
82.Jombart, T., Devillard, S., Dufour, A. B. & Pontier, D. Revealing cryptic spatial patterns in genetic variability by a new multivariate method. Heredity 101, 92–103 (2008).CAS 
PubMed 
Article 
PubMed Central 

Google Scholar 
83.Thioulouse, J., Chessel, D. & Champely, S. Multivariate analysis of spatial patterns: a unified approach to local and global structures. Environ. Ecol. Stat. 2, 1–14 (1995).Article 

Google Scholar 
84.Moran, P. The interpretation of statistical maps. J. R. Stat. Soc. Ser. B Stat. Methodol. 10, 243–251 (1948).85.Hedrick, P. W. A standardized genetic differentiation measure. Evolution 59, 1633–1638 (2005).CAS 
PubMed 
Article 
PubMed Central 

Google Scholar 
86.Jost, L. GST and its relatives do not measure differentiation. Mol. Ecol. 17, 4015–4026 (2008).PubMed 
Article 
PubMed Central 

Google Scholar 
87.Keenan, K., Mcginnity, P., Cross, T. F., Crozier, W. W. & Prodöhl, P. A. DiveRsity: An R package for the estimation and exploration of population genetics parameters and their associated errors. Methods Ecol. Evol. 4, 782–788 (2013).Article 

Google Scholar 
88.Sundqvist, L., Keenan, K., Zackrisson, M., Prodöhl, P. & Kleinhans, D. Directional genetic differentiation and relative migration. Ecol. Evol. 6, 3461–3475 (2016).PubMed 
PubMed Central 
Article 

Google Scholar 
89.Ryman, N. & Leimar, O. GST is still a useful measure of genetic differentiation—A comment on Jost’s D. Mol. Ecol. 18, 2084–2087 (2009).PubMed 
Article 
PubMed Central 

Google Scholar 
90.Meirmans, P. G. & Hedrick, P. W. Assessing population structure: FST and related measures. Mol. Ecol. Resour. 11, 5–18 (2011).PubMed 
Article 
PubMed Central 

Google Scholar 
91.Wilson, G. A. & Rannala, B. Bayesian inference of recent migration rates using multilocus genotypes. Genetics 163, 1177–1191 (2003).PubMed 
PubMed Central 
Article 

Google Scholar 
92.Faubet, P., Waples, R. S. & Gaggiotti, O. E. Evaluating the performance of a multilocus Bayesian method for the estimation of migration rates. Mol. Ecol. 16, 1149–1166 (2007).PubMed 
Article 
PubMed Central 

Google Scholar 
93.Do, C. et al. NeEstimator v2: Re-implementation of software for the estimation of contemporary effective population size (Ne) from genetic data. Mol. Ecol. Resour. 14, 209–214 (2014).CAS 
PubMed 
Article 
PubMed Central 

Google Scholar 
94.Waples, R. S. & Do, C. LDNE: A program for estimating effective population size from data on linkage disequilibrium. Mol. Ecol. Resour. 8, 753–756 (2008).PubMed 
Article 
PubMed Central 

Google Scholar 
95.Piry, S., Luikart, G. & Cornuet, J. M. BOTTLENECK: A computer program for detecting recent reductions in the effective population size using allele frequency data. J. Hered. 90, 502–503 (1999).Article 

Google Scholar 
96.Nikolic, N. & Chevalet, C. Detecting past changes of effective population size. Evol. Appl. 7, 663–681 (2014).PubMed 
PubMed Central 
Article 

Google Scholar 
97.Kimura, M. & Ohta, T. Stepwise mutation model and distribution of allelic frequencies in a finite population. Proc. Natl. Acad. Sci. USA 75, 2868–2872 (1978).ADS 
CAS 
PubMed 
PubMed Central 
MATH 
Article 

Google Scholar 
98.Ruiz-Garcia, M. et al. Determination of microsatellite DNA mutation rates, mutation models and mutation bias in four main Felidae lineages (European wild cat, F. silvestris; ocelot, Leopardus pardalis; puma, Puma concolor; jaguar, Panthera onca). In Molecular Population Genetics, Evolutionary Biology & Biological Conservation of Neotropical Carnivores. (Nova Science Publishers Inc., New York, 2013).99.Xu, X., Peng, M., Fang, Z. & Xu, X. The direction of microsatellite mutations is dependent upon allele length. Nat. Genet. 24, 396–399 (2000).CAS 
PubMed 
Article 
PubMed Central 

Google Scholar  More

Understanding the spatial patterns and drivers of animal movement is a crucial first step to controlling disease spread4. Our study provides novel information about where, how and when cattle move in a region beset by endemic pathogens2,39,40. Because contacts occur heterogeneously through time and space, interventions targeting areas and times of high contact risk could effectively break the chain of transmission across wide areas. We found that cattle herds had the highest probability of contact at dipping sites, far from their bomas, in small herds and during periods of low rainfall, indicating that transmission of all pathogens may be particularly elevated under these conditions (Figs. 5, 6). Nonetheless, cattle spent most of their time in other areas (i.e. near bomas or in grazing areas) where the direction and magnitude of effect of spatiotemporal scale on contact rates varies. This suggests that interventions for different pathogens in these systems will likely require a consideration of scale of transmission and be tailored to particular pathogens. Overall, our study provides a framework for risk-based livestock disease control approaches for the most dominant management systems in sub-Saharan Africa.Daily movement patterns of cattle in pastoral and agropastoral settings in sub-Saharan Africa largely reflect the distribution of shared resources, which determines the distance animals move each day and the probability of contacting each other. Our results are similar to those reported in other regions of Africa, suggesting broadly comparable patterns of daily displacement. For instance, cattle in our agropastoral study area travel to grazing, watering and dipping locations that are ~ 4 km from their bomas and primarily during daylight hours (Fig. 2). Similarly, in Kenya, cattle in the pastoral Mara and Ol Pajeta regions move less than 6 km from their bomas and movements peak around 12:00–14:00 h each day9,41. Despite the predominance of short-distance daily movements, we observed occasional long-distance movements (i.e. up to 12 km), particularly by larger herds. Transhumant cattle in Cameroon also moved up to 23 km/day for short periods, while relocating to seasonal grazing areas on the edge of the Sahel, though in most observations (86%) they moved less than 5 km/day8. Although we observed no contacts among cattle from bomas  > 17 km apart (Supplementary Fig. S5), regardless of how contact was defined, infrequent long-distance movements by large herds may provide a conduit for disease transmission between villages42. Indeed, larger herds actually had a lower relative probability of contact across spatiotemporal scales (Fig. 5), which may reflect the fact that large herds were more likely to move to areas away from other collared cattle, either because they were moving outside the study area, or because they had exclusive use of particular areas, whereas smaller herds that were mostly moved around bomas mixed more frequently. While interventions (e.g. vaccination or quarantine) targeting small herds would address local disease events, particularly within villages, halting larger-scale transmission requires an understanding of livestock pathways enabling inter-village connectivity and strategies tailored to herds driving these processes.A key difference between the movement of cattle in agropastoral and pastoral systems lies in the seasonal variation of daily movement. In our study, agropastoralists move their herds farther in the wet compared to the dry season, while the opposite has been reported for pastoralists8,9,41. During the wet season, agropastoralists cultivate crops near their homesteads, which increases competition for space and displaces cattle to reserved grazing areas far from cultivated land11. During the dry season, particularly in the early period, cattle graze harvested fields around the homestead and tend to move short distances each day. In our study, although individual herds travelled more (marginally) in the wet compared to the dry season, there were more contacts following low rainfall periods when resources were typically scarce (Fig. 5). Similarly, a previous study has shown that more villages were connected at shared resource areas during dry spells, which resulted in higher contacts11. This suggests a higher disease risk in the dry compared to wet seasons in agropastoral management systems.Translating movements into contact between individuals is challenging because the definition of a “contact” depends on the distance at which pathogens can travel in space, and the time period that pathogens survive, or mature to an infectious state, in the environment. Most studies that attempt to measure contact, however, focus only on a single scale. Here, we show that pairwise contact rates between cattle herds generally increase with broader spatiotemporal definitions of contact. Yet, there was no difference at spatial scales between 50 m, 100 m and 200 m for a temporal scale of one hour, suggesting these scales are functionally equivalent definitions of contact. Thus, we define “close contact” as proximity of livestock herds within 200 m in any given hour, which would be applicable to multiple disease systems and vital for understanding infectious disease spread in traditionally managed herds. However, given that herds tracked in our study ranged in size from 30 to 500 cattle, for households with herds of  More

Soil physicochemical propertiesThe test results of physicochemical factors of the soil are shown in Table 2. In the soil with 15-year vines, the average contents of TK and SK were highest and the contents of SOM and TN were lowest. In the soil with 5-year vines, the contents of XN and SK were relatively higher, and the soil pH was between 7.86 and 7.98, thus it is alkaline soil.Table 2 Determined results of soil physicochemical properties.Full size tableThe analysis of variance showed that grape planting year had significant effect on TK and SP (P  More