Ecology

Lawrence, H. In City Trees: A Historical Geography from the Renaissance through to the Nineteenth Century (Charlottesville and London: University of Virginia Press, 2006, Lewis Mumford. The City in History: Its Origins, Its Transformations and Its Prospects (San Diego: Harvest Book Harcourt, 1961).Frawley, J. Campaigning for street trees, Sydney botanic gardens, 1890s–1920s. Environ. Hist. 15(3), 303–322. https://doi.org/10.3197/096734009X12474738199953 (2009).Article 

Google Scholar 
Seburanga, J. L., Kaplin, B. A., Zhang, Q.-X. & Gatesire, T. Amenity trees and green space structure in urban settlements of Kigali, Rwanda. Urban. For. Urban Green. 13(84–9313), 84–93. https://doi.org/10.1016/j.ufug.2013.08.001 (2014).Article 

Google Scholar 
Wilson, E. H. Northern trees in southern lands. J. Arnold Arbor. 4(2), 61–90 (1923).Article 

Google Scholar 
Gwedla, N. & Shackleton, C. M. Population size and development history determine street tree distribution and composition within and between Eastern Cape towns, South Africa. Urban. For. Urban. Gree. 25, 11–18. https://doi.org/10.1016/j.ufug.2017.04.014 (2017).Article 

Google Scholar 
Jacobs, A. B., Macdonald, E. & Rofé, Y. In The Boulevard Book: History, Evolution, Design of Multiway Boulevards (MIT Press, Cambridge, MA 2002), Robinson, W. The Parks and Gardens of Paris Considered in Relation to the Wants of Other Cities and of Private and Public Gardens (McMillan and Co., London , 1878).Akbari, A. H., Pomerantz, M. & Taha, H. Cool surfaces and shade trees to reduce energy use and improve air quality in urban. Sol. Energy. 70(3), 295–310 (2001).ADS 
Article 

Google Scholar 
Roy, S., Byrne, J. & Pickering, C. A systematic quantitative review of urban tree benefits, costs, and assessment methods across cities in different climatic zones. Urban For. Urban Green. 11, 351–363. https://doi.org/10.1016/j.ufug.2012.06.006 (2012).Article 

Google Scholar 
Schäffler, A. & Swilling, M. Valuing green infrastructure in an urban environment under pressure—The Johannesburg case. Ecol. Econ. 86, 246–257. https://doi.org/10.1016/j.ecolecon.2012.05.008 (2013).Article 

Google Scholar 
Santamour, F. S. Trees for urban planting: Diversity, uniformity and common sense. In Proceedings of the 7th Conference of the Metropolitan Tree Improvement Alliance (METRIA), vol. 7, 57–65 (1990).Shams, Z. I. Changes in diversity and composition of flora along a corridor of different land uses in Karachi over 20 years: caUses and implications. Urban. For. Urban Green. 17, 71–79. https://doi.org/10.1016/j.ufug.2016.03.002 (2016).Article 

Google Scholar 
Kambites, C. & Owen, S. Renewed prospects for green infrastructure planning in the UK. Plan. Prac. Res. 21(94), 483–496. https://doi.org/10.1080/02697450601173413 (2006).Article 

Google Scholar 
Cho, M. A., Malahlelac, O. & Ramoeloa, A. Assessing the utility WorldView-2 imagery for tree species mapping in South African subtropical humid forest and the conservation implications: Dukuduku forest patch as case study. Int. J. Appl. Earth. Obs. 38, 349–357. https://doi.org/10.1016/j.jag.2015.01.015 (2015).Article 

Google Scholar 
Niculescu, S., Lardeux, C., Grigoras, I., Hanganu, J. & David, L. Synergy between LiDAR, RADARSAT-2, and spot-5 images for the detection and mapping of wetland vegetation in the Danube Delta. IEEE J Sel. Top. Appl. Earth. Obs. Remote Sens. 9, 3651–3666 (2016).ADS 
Article 

Google Scholar 
Lefebvre, A., Picand, P.-A. & Sannier, C. Mapping tree cover in European cities: Comparison of classification algorithms for an operational production framework. In 2015 Joint Urban Remote Sensing Event (JURSE), IEEE, 1–4 (2015) https://doi.org/10.1109/JURSE.2015.7120511.Wyndham, C. H., Strydom, N. B., Van Rensburg, A. J. & Rogers, G. G. Effects on maximal oxygen intake of acute changes in altitude in a deep mine. J. Appl. Physiol. 29(5), 552–555 (1970).CAS 
Article 

Google Scholar 
Hegnauer, R. Chemotaxonomie der Pflanzen, vol. 3, 268–281 (Birkhäuser Verlag, Basel, 1964).Mabberley, D. J. The Plant-Book, 2nd edn. 87, 368–369 (Cambridge University Press, Cambridge, 1997).Gachet, M. S. & Schühly, W. Jacaranda—An ethnopharmacological and phytochemical review. J. Ethnopharmacol. 121, 14–27. https://doi.org/10.1016/j.jep.2008.10.015 (2009).CAS 
Article 
PubMed 

Google Scholar 
Gilman, E. F. & Watson, D. G. Jacaranda mimosifolia. Fact Sheet ST-317, Environmental Horticulture Department, Florida Cooperative Extension Service, University of Florida, Gainesville, http://www.ci.milpitas.ca.gov/_pdfs/council/2016/021616/item_04.pdf Accessed 6 June 2020 (1993).Dineva, S. B. Comparative studies of the leaf morphology and structure of white ash Fraxinus americana L. and London plane tree Platanus acerifolia Willd growing in polluted area. Dendrobiology 52, 3–8 (2004).
Google Scholar 
Liu, G., Li, Z. & Bao, M. Colchicine-induced chromosome doubling in Platanus acerifolia and its effect on plant morphology. Euphytica 157, 145–154. https://doi.org/10.1007/s10681-007-9406-6 (2007).Article 

Google Scholar 
Henry, A. & Flood, M. G. The history of the London plane, Platanus acerifolia, with notes on the Genus Platanus. Proc. R. Irish Acad Sect. B Biol. Geol. Chem. Sci. 35, 9–28 (1919).
Google Scholar 
Chavez, P. S. Image-based atmospheric corrections revisited and improved. Photogram. Eng. Rem. S. 62, 1025–1036 (1996).
Google Scholar 
Riano, D., Chuvieco, E., Salas, J. & Aguado, I. Assessment of different topographic corrections in Landsat-T. M. data for mapping vegetation types. IEEE Trans. Geosci. Remote Sens. 41, 1056–1061. https://doi.org/10.1109/TGRS.2003.811693 (2003).ADS 
Article 

Google Scholar 
Rouse J. W., Haas, R. H., Schell, J. A. & Deering, D. W. Proceedings of the Third Earth Resources Technology Satellite-1 Symposium, Greenbelt, USA: NASASP-351; 1974. Monitoring vegetation system in the great plains with ERTS, 3010–3017 (1974).R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/ (2021).Du, Y. et al. New hyperspectral discrimination measure for spectral characterization. Opt. Eng. 43(8), 1777–1786 (2004).ADS 
Article 

Google Scholar 
Bhattacharyya, A. On a measure of divergence between two statistical populations defined by their probability distributions’. Bull. Calcutta Math. Soc. 35, 99–109 (1943).MathSciNet 
MATH 

Google Scholar 
Bruzzone, L., Roli, F. & Serpico, S. B. An extension to multiclass cases of the Jefferys-Matusita distance. IEEE Trans. Pattern. Anal. Mach. Intell. 33, 1318–1321 (1995).
Google Scholar 
Kaufman, Y. & Remer, L. Detection of forests using mid-IR reflectance: An application for aerosol studies. IEEE Trans. Geosci. Remote Sens. 32(3), 672–683 (1994).ADS 
Article 

Google Scholar 
Padma, S. & Sanjeevi, S. Jeffries Matusita based mixed-measure for improved spectral matching in hyperspectral image analysis. Int. J. Appl. Earth. Obs. 32, 138–151. https://doi.org/10.1016/j.jag.2014.04.001 (2014).Article 

Google Scholar 
Kavzoglu, T. & Mather, P. M.. The use of feature selection techniques in the context of artificial neural networks. In Proceedings of the 26th Annual Conference of the Remote Sensing Society (CD-ROM), 12–14 September (Leicester, UK, 2000).Gunal, S. & Edizkan, R. Subspace based feature selection for pattern recognition. Info. Sci. 178, 3716–3726. https://doi.org/10.1016/j.ins.2008.06.001 (2008).Article 

Google Scholar 
Tolpekin, V. A. & Stein, A. Quantification of the effects of land-cover-class spectral separability on the accuracy of markov-random-field-based superresolution mapping. IEEE Trans. Geosci. Remote Sens. 47(9), 3283–3297. https://doi.org/10.1109/TGRS.2009.2019126 (2009).ADS 
Article 

Google Scholar 
Paterson, M., Lucas, R. M. & Chisholm, L. Differentiation of selected Australian woodland species using CASI data. In Proceedings IEEE International Geoscience and Remote Sensing Symposium, 643–645 (University of New South Wales, Australia, 2001).Richards, J. A. & Jai, X. Remote Sensing Digital Analysis: An Introduction, 4th edition (Springer, Berlin, 1999).Veraverbeke, S., Harris, S. & Hook, S. Evaluating spectral indices for burned area discrimination using MODIS/ASTER (MASTER) airborne simulator data. Remote Sens. Environ. 115, 2702–2709. https://doi.org/10.1016/j.rse.2011.06.010 (2011).ADS 
Article 

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

Google Scholar 
Georganos, S. et al. Geographical random forests: a spatial extension of the random forest algorithm to address spatial heterogeneity in remote sensing and population modelling. Geocarto Int. https://doi.org/10.1080/10106049.2019.1595177 (2019).Article 

Google Scholar 
Mellor, A., Haywood, A., Stone, C. & Jones, S. The performance of random forests in an operational setting for large area sclerophyll forest classification. Remote Sens. 5, 2838–2856. https://doi.org/10.3390/rs5062838 (2013).ADS 
Article 

Google Scholar 
Congalton, R. G. Accuracy assessment and validation of remotely sensed and other spatial information. Int. J. Wildland. Fire. 10, 321–328 (2001).Article 

Google Scholar 
Thomas, I. L., Ching, N. P., Benning, V. M. & D’aguanno, J. A. Review Article A review of multi-channel indices of class separability. Int. J. Remote Sens. 8(3), 331–350. https://doi.org/10.1080/01431168708948645 (1987).Article 

Google Scholar 
Mausel, P. W., Kramber, W. J. & Lee, J. K. Optimum band selection for supervised classification of multispectral data. Photogramm. Eng. Remote. Sens. 56(1), 55–60 (1990).
Google Scholar 
Singh, A. Some clarifications about the pairwise divergence measure in remote sensing. Int. J. Remote Sens. 5(3), 623–627. https://doi.org/10.1080/01431168408948845 (1984).Article 

Google Scholar 
Kumar, P. et al. A statistical significance of differences in classification accuracy of crop types using different classification algorithms. Geocarto Int. 32(2), 206–224. https://doi.org/10.1080/10106049.2015.1132483 (2017).Article 

Google Scholar 
McPherson, E. G., Simpson, J. R., Peper, P. J., Xiao, Q. & Wu, C. Los Angeles 1-Million Tree Canopy Cover Assessment. General Technical Report PSW-GTR-207. U.S. Department of Agriculture Forest Service Pacific Southwest Research Station. Albany, CA, 1–64 (2008).Rahimizadeh, N., Kafaky, S. B., Sahebi, M. R. & Mataji, A. Forest structure parameter extraction using SPOT-7 satellite data by object- and pixel-based classification methods. Environ. Monit. Assess. 192, 43. https://doi.org/10.1007/s10661-019-8015-x (2020).Article 

Google Scholar 
McRoberts, R. E. Satellite image-based maps: Scientific inference or pretty pictures?. Remote. Sens. Environ. 115, 715–724. https://doi.org/10.1016/j.rse.2010.10.013 (2011).ADS 
Article 

Google Scholar 
McRoberts, R. E. Probability- and model-based approaches to inference for proportion forest using satellite imagery as ancillary data. Remote. Sens. Environ. 114, 1017–1025. https://doi.org/10.1016/j.rse.2009.12.013 (2010).ADS 
Article 

Google Scholar 
Kokubu, Y., Hara, S. & Tani, A. Mapping seasonal tree canopy cover and leaf area using worldview-2/3 satellite imagery: A megacity-scale case study in Tokyo urban area. Remote. Sens. 12(9), 1505. https://doi.org/10.3390/rs12091505 (2020).Article 

Google Scholar 
Johannesburg City Parks and Zoo. 2018. The city that’s a rain forest. http://www.jhbcityparks.com/index.php/street-trees-contents-29. Accessed 14 June 2020.Tesfamichael, S. G., Newete, S. W., Adam, E. & Dubula, B. Field spectroradiometer and simulated multispectral bands for discriminating invasive species from morphologically similar cohabitant plants. GIsci. Remote Sens. 55(3), 417–436. https://doi.org/10.1080/15481603.2017.1396658 (2018).Article 

Google Scholar 
McPherson, E. G., Simpsona, J. R., Xiao, Q. & Wu, C. Million trees Los Angeles canopy cover and benefit assessment. Landsc. Urban. Plan. 99, 40–50 (2011).Article 

Google Scholar 
Baines, O., Wilkes, P. & Disney, M. Quantifying urban forest structure with open-access remote sensing data sets. Urban For. Urban Green. 50, 126653. https://doi.org/10.1016/j.ufug.2020.126653 (2020).Article 

Google Scholar 
Nowak, D. J. et al. Measuring and analyzing urban tree cover. Landsc. Urban Plan. 36, 49–57 (1996).Article 

Google Scholar 
Estoque, R. C. et al. Remotely sensed tree canopy cover-based indicators for monitoring global sustainability and environmental initiatives. Environ. Res. Lett. 16, 044047. https://doi.org/10.1088/1748-9326/abe5d9 (2021).ADS 
CAS 
Article 

Google Scholar 
Paap, T., de Beer, W., Migliorini, D., Nel, W. J. & Wingfield, M. J. The polyphagous shot hole borer (PSHB) and its fungal symbiont Fusarium euwallaceae: A new invasion in South Africa Trudy. Aust. Plant. Pathol. 47, 231–237. https://doi.org/10.1007/s13313-018-0545-0 (2018).Article 

Google Scholar  More

Cox, P. A., Banack, S. A. & Murch, S. J. Biomagnification of cyanobacterial neurotoxins and neurodegenerative disease among the Chamorro people of Guam. Proc. Natl. Acad. Sci. U.S.A. 100, 13380–13383 (2003).ADS 
CAS 
Article 

Google Scholar 
Brand, L. E., Pablo, J., Compton, A., Hammerschlag, N. & Mash, D. C. Cyanobacterial blooms and the occurrence of the neurotoxin, beta-N-methylamino-L-alanine (BMAA), in south Florida aquatic food webs. Harmful Algae 9, 620–635 (2010).CAS 
Article 

Google Scholar 
Metcalf, J. S., Banack, S. A., Richer, R. & Cox, P. A. Neurotoxic amino acids and their isomers in desert environments. J. Arid Environ. 112, 140–144 (2015).ADS 
Article 

Google Scholar 
Violi, J. P., Mitrovic, S. M., Colville, A., Main, B. J. & Rodgers, K. J. Prevalence of (beta)-methylamino-L-alanine (BMAA) and its isomers in freshwater cyanobacteria isolated from eastern Australia. Ecotoxicol. Environ. Saf. 172, 72–81 (2019).CAS 
Article 

Google Scholar 
Jonasson, S. et al. Transfer of a cyanobacterial neurotoxin within a temperate aquatic ecosystem suggests pathways for human exposure. Proc. Natl. Acad. Sci. 107, 9252–9257 (2010).ADS 
CAS 
Article 

Google Scholar 
Metcalf, J. et al. Toxin analysis of freshwater cyanobacterial and marine harmful algal blooms on the west coast of Florida and implications for estuarine environments. Neurotox. Res. 39, 27–35 (2021).CAS 
Article 

Google Scholar 
Cox, P. A. et al. Cyanobacteria and BMAA exposure from desert dust: a possible link to sporadic ALS among Gulf War veterans. Amyotroph. Lateral Scler. 10, 109–117 (2009).CAS 
Article 

Google Scholar 
Charlton, T. S., Marini, A. M., Markey, S. P., Norstog, K. & Duncan, M. W. Quantification of the neurotoxin 2-amino-3-(methylamino)-propanoic acid (BMAA) in Cycadales. Phytochemistry 31, 3429–3432 (1992).CAS 
Article 

Google Scholar 
Whiting, M. G. Toxicity of cycads. Econ. Bot. 17, 270–302 (1963).Article 

Google Scholar 
Cox, P. A., Davis, D. A., Mash, D. C., Metcalf, J. S. & Banack, S. A. Dietary exposure to an environmental toxin triggers neurofibrillary tangles and amyloid deposits in the brain. Proc. R. Soc. B: Biol. Sci. 283, 20152397 (2016).Article 

Google Scholar 
Scott, L. L. & Downing, T. G. A single neonatal exposure to BMAA in a rat model produces neuropathology consistent with neurodegenerative diseases. Toxins 10, 22 (2018).Article 

Google Scholar 
Roy, U. et al. Metabolic profiling of zebrafish (Danio rerio) embryos by NMR spectroscopy reveals multifaceted toxicity of (beta)-methylamino-L-alanine (BMAA). Sci. Rep. 7, 1–12 (2017).ADS 
Article 

Google Scholar 
Purdie, E. L., Metcalf, J. S., Kashmiri, S. & Codd, G. A. Toxicity of the cyanobacterial neurotoxin (beta)-N-methylamino-L-alanine to three aquatic animal species. Amyotroph. Lateral Scler. 10, 67–70 (2009).CAS 
Article 

Google Scholar 
Brenner, E. D. et al. Arabidopsis mutants resistant to s (+)-(beta)-methyl-(alpha), (beta)-diaminopropionic acid, a cycad-derived glutamate receptor agonist. Plant Physiol. 124, 1615–1624 (2000).CAS 
Article 

Google Scholar 
Schneider, D., Wink, M., Sporer, F. & Lounibos, P. Cycads: Their evolution, toxins, herbivores and insect pollinators. Naturwissenschaften 89, 281–294 (2002).ADS 
CAS 
Article 

Google Scholar 
Koi, S. & Daniels, J. Life history variations and seasonal polyphenism in Eumaeus atala (Lepidoptera: Lycaenidae). Florida Entomol. 100, 219–229 (2017).Article 

Google Scholar 
Koi, S. A butterfly picks its poison: Cycads (Cycadaceae), integrated pest management and Eumaeus atala Poey (Lepidoptera: Lycaenidae). Entomol. Ornithol. Herpetol. 6 (2017).Brenner, E. D., Stevenson, D. W. & Twigg, R. W. Cycads: Evolutionary innovations and the role of plant-derived neurotoxins. Trends Plant Sci. 8, 446–452 (2003).CAS 
Article 

Google Scholar 
Prado, A. The cycad herbivores. Bull. Soc. D’entomol. Quebec 18, 3–6 (2011).
Google Scholar 
Popova, A. & Koksharova, O. Neurotoxic non-proteinogenic amino acid (beta)-N-methylamino-L-alanine and its role in biological systems. Biochem. Mosc. 81, 794–805 (2016).CAS 
Article 

Google Scholar 
Salzman, S., Whitaker, M. R. L. & Pierce, N. E. Cycad-feeding insects share a core gut microbiome. Biol. J. Lin. Soc. 123, 728–738 (2018).Article 

Google Scholar 
Whitaker, M. R. & Salzman, S. Ecology and evolution of cycad-feeding Lepidoptera. Ecol. Lett. 23, 1862–1877 (2020).Article 

Google Scholar 
Zhou, X., Escala, W., Papapetropoulos, S., Bradley, W. G. & Zhai, R. G. BMAA neurotoxicity in Drosophila. Amyotroph. Lateral Scler. 10, 61–66 (2009).CAS 
Article 

Google Scholar 
Zhou, X., Escala, W., Papapetropoulos, S. & Zhai, R. G. (beta)-N-methylamino-L-alanine induces neurological deficits and shortened life span in Drosophila. Toxins 2, 2663–2679 (2010).CAS 
Article 

Google Scholar 
Mekdara, N. T. et al. A novel lenticular arena to quantify locomotor competence in walking fruit flies. J. Exp. Zool. A Ecol. Genet. Physiol. 317, 382–394 (2012).Article 

Google Scholar 
Goto, J. J., Koenig, J. H. & Ikeda, K. The physiological effect of ingested (beta)-N-methylamino-L-alanine on a glutamatergic synapse in an in vivo preparation. Comp. Biochem. Physiol. Part C: Toxicol. Pharmacol. 156, 171–177 (2012).CAS 

Google Scholar 
Okle, O., Rath, L., Galizia, C. G. & Dietrich, D. R. The cyanobacterial neurotoxin (beta)-N-methylamino-L-alanine (BMAA) induces neuronal and behavioral changes in honeybees. Toxicol. Appl. Pharmacol. 270, 9–15 (2013).CAS 
Article 

Google Scholar 
Spencer, P. S. et al. Guam amyotrophis lateral sclerosis-parkinsonism-dementia linked to a plant excitant neurotoxin. Science 237, 517–522 (1987).ADS 
CAS 
Article 

Google Scholar 
Bernays, E. A. & Chapman, R. F. Host-plant selection by phytophagous insects. In Host-Plant Selection by Phytophagous Insects. Contemporary Topics in Entomology, vol. 2, 201–213 (Springer, Boston, MA, 1994).Zandt, P. A. V. Plant defense, growth, and habitat: A comparative assessment of constitutive and induced resistance. Ecology 88, 1984–1993 (2007).Article 

Google Scholar 
Duncan, M. W. Role of the cycad neurotoxin BMAA in the amyotrophic lateral sclerosi-parkisonism dementia complex of the Western Pacific. Adv. Neurol. 56, 301–310 (1991).CAS 
PubMed 

Google Scholar 
Banack, S. A. & Cox, P. A. Distribution of the neurotoxic nonprotein amino acid BMAA in Cycas micronesica. Bot. J. Linn. Soc. 143, 165–168 (2003).Article 

Google Scholar 
R Core Team. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria (2021).Therneau, T. M. A Package for Survival Analysis in R. R package version 3.2-11 (2021).Kassambara, A., Kosinski, M. & Biecek, P. survminer: Drawing Survival Curves using ’ggplot2’. R package version 0.4.9 (2021).Pennington, Z. T. et al. eztrack: An open-source video analysis pipeline for the investigation of animal behavior. Sci. Rep. 9, 1–11 (2019).Article 

Google Scholar 
Pérez, F. & Granger, B. E. IPython: A system for interactive scientific computing. Comput. Sci. Eng. 9, 21–29 (2007).Article 

Google Scholar 
Hammer, T. J., Janzen, D. H., Hallwachs, W., Jaffe, S. P. & Fierer, N. Caterpillars lack a resident gut microbiome. Proc. Natl. Acad. Sci. 114, 9641–9646 (2017).CAS 
Article 

Google Scholar 
Karlsson, O., Roman, E. & Brittebo, E. B. Long-term cognitive impairments in adult rats treated neonatally with (beta)-N-methylamino-L-alanine. Toxicol. Sci. 112, 185–195 (2009).CAS 
Article 

Google Scholar 
Whitaker, M. R. L., Salzman, S., Gratacos, X. & Tucker Lima, J. Localized overabundance of an otherwise rare butterfly threatens endangered cycads. Florida Entomol. 103, 519–522 (2021).Article 

Google Scholar 
Backmann, P. et al. Delayed chemical defense: Timely expulsion of herbivores can reduce competition with neighboring plants. Am. Nat. 193, 125–139 (2019).Article 

Google Scholar 
Yáñez-Espinosa, L. & Sosa-Sosa, F. Population structure of Dioon purpusii rose in Oaxaca, Mexico. Neotrop. Biol. Conserv. 2, 46–54 (2007).
Google Scholar 
Robbins, R. K. et al. A switch to feeding on cycads generates parallel accelerated evolution of toxin tolerance in two clades of Eumaeus caterpillars (Lepidoptera: Lycaenidae). Proc. Natl. Acad. Sci.118 (2021).Grunseich, J. M., Thompson, M. N., Aguirre, N. M. & Helms, A. M. The role of plant-associated microbes in mediating host-plant selection by insect herbivores. Plants 9, 6 (2020).CAS 
Article 

Google Scholar 
Zhang, Y. & Whalen, J. K. Production of the neurotoxin beta-N-methylamino-L-alanine may be triggered by agricultural nutrients: An emerging public health issue. Water Res. 170, 115335 (2020).CAS 
Article 

Google Scholar  More

McFall-Ngai M, Hadfield MG, Bosch TCG, Carey HV, Domazet-Lošo T, Douglas AE, et al. Animals in a bacterial world, a new imperative for the life sciences. PNAS. 2013;s110:3229–36. https://doi.org/10.1073/pnas.1218525110.Article 

Google Scholar 
Davenport ER, Sanders JG, Song SJ, Amato KR, Clark AG, Knight R. The human microbiome in evolution. BMC Biol. 2017;15:127 https://doi.org/10.1186/s12915-017-0454-7.Article 
PubMed 
PubMed Central 

Google Scholar 
Mazel F, Davis KM, Loudon A, Kwong WK, Groussin M, Parfrey LW. Is host filtering the main driver of phylosymbiosis across the Tree of Life? mSystems. 2018;3:e00097–18. https://doi.org/10.1128/mSystems.00097-18.Article 
PubMed 
PubMed Central 

Google Scholar 
Kohl KD. Ecological and evolutionary mechanisms underlying patterns of phylosymbiosis in host-associated microbial communities. Phil Trans R Soc B. 2020;375:20190251 https://doi.org/10.1098/rstb.2019.0251.CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Greyson-Gaito CJ, Bartley TJ, Cottenie K, Jarvis WMC, Newman AEM, Stothart MR. Into the wild: microbiome transplant studies need broader ecological reality. Proc Royal Soc B. 2020;287:20192834 https://doi.org/10.1098/rspb.2019.2834.Article 

Google Scholar 
Brooks AW, Kohl KD, Brucker RM, van Opstal EJ, Bordenstein SR. Phylosymbiosis: relationships and functional effects of microbial communities across host evolutionary history. PLOS Biol. 2016. https://doi.org/10.1371/journal.pbio.2000225.Delsuc F, Metcalf JL, Parfrey LW, Song SJ, González A, Knight R. Convergence of gut microbiomes in myrmecophagus mammals. Mol Ecol. 2014;23:1301–17. https://doi.org/10.1111/mec.12501.CAS 
Article 
PubMed 

Google Scholar 
Song SJ, Sanders JG, Delsuc F, Metcalf J, Amato K, Taylor MW, et al. Comparative analyses of vertebrate gut microbiomes reveal convergence between birds and bats. mBio. 2020;11:e02901–19. https://doi.org/10.1128/mBio.02901-19.CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Bodawatta KH, Hird SM, Grond K, Poulsen M, Jønsson KA. Avian gut microbiomes taking flight. Trends Microbiol. 2021;30:268–0. https://doi.org/10.1016/j.tim.2021.07.003.CAS 
Article 
PubMed 

Google Scholar 
Egerton S, Culloty S, Whooley J, Stanton C, Ross RP. The gut microbiota of marine fish. Front Microbiol. 2018; 9. https://doi.org/10.3389/fmicb.2018.00873.Mallott EK, Amato KR. Host specificity of the gut microbiome. Nat Rev Microbiol. 2021;19:639–653. https://doi.org/10.1038/s41579-021-00562-3.CAS 
Article 
PubMed 

Google Scholar 
Ley RE, Hamady M, Lozupone C, Turnbaugh P, Ramey RR, Bircher S, et al. Evolution of mammals and their gut microbes. Science. 2008;320:1647–51. https://doi.org/10.1126/science.1155725.CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Gogarten JF, Davies TJ, Benjamino J, Gogarten JP, Graf J, Mielke A, et al. Factors influencing bacterial microbiome composition in a wild non-human primate community in Tai National Park, Côte d’Ivoire. ISME J. 2018;12:2559–74. https://doi.org/10.1038/s41396-018-0166-1.CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Perofsky AC, Lewis RJ, Meyers LA. Terrestriality and bacterial transfer: a comparative study of GMs in sympatric Malagasy mammals. ISME. 2019;13:50–63. https://doi.org/10.1038/s41396-018025-5.Article 

Google Scholar 
Amato KR, Mallott EK, McDonald D, Dominy NJ, Goldberg T, Lambert LE, et al. Convergence of human and old world monkey gut microbiomes demonstrate the importance of human ecology over phylogeny. Genome Biol. 2019;20:201 https://doi.org/10.1186/s13059-019-1807-z.CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Gomez A, Sharma AK, Mallott EK, Petrzelkova KJ, Robinson CAJ, Yeoman CJ, et al. Plasticity in the human gut microbiome defies evolutionary constraints. mSphere. 2019;4:e00271–19. https://doi.org/10.1128/mSphere.00271-19.Article 
PubMed 
PubMed Central 

Google Scholar 
Hale VL, Tan CL, Niu K, Yang Y, Knight R, Zhang Q, et al. Diet versus phylogeny: a comparison of gut microbiota in captive colobine monkey species. Microb Ecol. 2018;75:515–27. https://doi.org/10.1007/s00248-017-1041-8.Article 
PubMed 

Google Scholar 
Ochman H, Worobey M, Kuo CH, Ndjando JBN, Peeters M, Hahn BH, et al. Evolutionary relationships of wild hominids recapitulated by gut microbial communities. PLOS Biol. 2010. https://doi.org/10.1371/journal.pbio.1000546.McKenney EA, Maslanka M, Rodrigo A, Yoder AD. Bamboo specialists from two mammalian orders (Primates, Carnivora) share a high number of low-abundance gut microbes. Microb Ecol. 2018. https://doi.org/10.1007/s00248-017-1114-8.Amato KR, Sanders J, Song SJ, Nute M, Metcalf JL, Thompson LR, et al. Evolutionary trends in host physiology outweigh dietary niche in structuring primate gut microbiomes. ISME J. 2019;13:576–87. https://doi.org/10.1038/s41396-018-0175-0.CAS 
Article 
PubMed 

Google Scholar 
Bornbusch SL, Greene LK, McKenney EA, Volkoff SJ, Midani FS, Joseph G, et al. A comparative study of gut microbiomes in captive nocturnal strepsirrhines. Am J Primatol. 2019;81:e22986 https://doi.org/10.1002/ajp.22986.Article 
PubMed 

Google Scholar 
Greene LK, Bornbusch SL, McKenney EA, Harris RL, Gorvetzian SR, Yoder AD, et al. The importance of scale in comparative microbiome research: new insights from the gut and glands of captive and wild lemurs. Am J Primatol. 2019. https://doi.org/10.1002/ajp.22974.Yoder AD, Nowak MD. Has vicariance or dispersal been the predominant biogeographic force in Madagascar? Only time will tell. Annu Rev Ecol Evol Syst. 2006;37:405–31. https://doi.org/10.1146/annurev.ecolsys.37.091305.110239.Article 

Google Scholar 
Horvath JE, Weisrock DW, Embry SL, Fiorentino I, Balhodd JP, Kappeler P, et al. Development and application of a phylogenomic toolkit: resolving the evolutionary history of Madagascar’s lemurs. Genome Res. 2008;18:489–99. https://doi.org/10.1101/gr.7265208.CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Herrera JP, Dávalos LM. Phylogeny and divergence times of lemurs inferred with recent and ancient fossils in the tree. Syst Biol. 2016;65:772–91. https://doi.org/10.1093/sysbio/syw035.Article 
PubMed 

Google Scholar 
Herrera JP. Testing the adaptive radiation hypothesis for the lemurs of Madagascar. R Soc Open Sci. 2017;4:161014 https://doi.org/10.1098/rsos.161014.Article 
PubMed 
PubMed Central 

Google Scholar 
Herrera JP. Convergent evolution in lemur environmental niches. J Biogeogr. 2019;47:795–806. https://doi.org/10.1111/jbi.13741.Article 

Google Scholar 
Wright PC. Lemur traits and Madagascar ecology: coping with an island environment. Am J Phys Anthropol. 1999;110:31–72. 10.1002/(SICI)1096-8644(1999)110:29+3.0.CO;2-0.Article 

Google Scholar 
Greene LK, McKenney EA, O’Connell TM, Drea CM. The critical role of dietary foliage in maintaining the gut microbiome and metabolome of folivorous sifakas. Sci Rep. 2018;8:1–13. https://doi.org/10.1038/s41598-018-32759-7.CAS 
Article 

Google Scholar 
Greene LK, Clayton JB, Rothman RS, Semel BP, Semel M, Gillespie TR, et al. Local habitat, not phylogenetic relatedness, predicts gut microbiota better within folivorous than frugivorous lemur lineages. Biol Lett. 2019;15:20190028 https://doi.org/10.1098/rsbl.2019.0028.Article 
PubMed 
PubMed Central 

Google Scholar 
McKenney EA, O’Connell TM, Rodrigo A, Yoder AD. Feeding strategy shapes gut metagenomic enrichment and functional specialization in captive lemurs. Gut Microbes. 2018;9:202–17. https://doi.org/10.1080/19490976.2017.1408762.CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Bennett G, Malone M, Sauther ML, Cuozzo FP, White B, Nelson KE, et al. Host age, social group, and habitat type influence the gut microbiota of wild ring-tailed lemurs (Lemur catta). Am J Primatol. 2016;78:883–92. https://doi.org/10.1002/ajp.22555.CAS 
Article 
PubMed 

Google Scholar 
de Winter I, Umanets A, Ijdema F, Ramiro-Garcia J, van Hooft P, Heitkönig IMA, et al. Occupancy strongly influences faecal microbial composition of wild lemurs. FEMS Microbiol Ecol. 2018;94:fiy017 https://doi.org/10.1093/femsec/fiy017.CAS 
Article 

Google Scholar 
Donohue ME, Absanga AE, Ralainirina J, Weisrock DW, Stumpf RM, Wright PC. Extensive variability in the gut microbiome of a highly-specialized and critically endangered lemur species across sites. Am J Primatol. 2019;81:e23046 https://doi.org/10.1002/ajp.23046.Article 
PubMed 

Google Scholar 
Fogel AT. The gut microbiome of wild lemurs: a comparison of sympatric Lemur catta and Propithecus verreauxi. Folia Primatol. 2015;86:85–95. https://doi.org/10.1159/000369971.Article 

Google Scholar 
de Winter, Umanets A, Gort G, Nieuwland H, van Hooft P, Heitkönig IMA, et al. Effects of seasonality and previous logging on faecal helminth-microbiota associations in wild lemurs. Sci Rep. 2020;10:16818 https://doi.org/10.1038/s41598-020-73827-1.CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Rowe AK, Donohue ME, Clare EL, Drinkwater R, Koenig A, Ridgway ZM, et al. Exploratory analysis reveals arthropod consumption in 10 lemur species using DNA metabarcoding. Am J Primatol. 2021;83:e23256 https://doi.org/10.1002/ajp.23256.CAS 
Article 
PubMed 

Google Scholar 
Kozich JJ, Westcott SL, Baxter NT, Highlander SK, Schloss PD. Development of a dual-index sequencing strategy and curation pipeline for analyzing amplicon sequence data on the MiSeq Illumina sequencing platform. AEM. 2013. https://doi.org/10.1128/AEM.01043-13.Callahan BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AJA, Holmes SP. DADA2: high resolution sample inference from Illumina amplicon data. Nat Methods. 2016;13:581–3. https://doi.org/10.1038/nmeth.3869.CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Bolyen E, Rideout JR, Dillon MR, Bokulich NA, Abnet C, Al-Ghalith GA, et al. QIIME2: reproducible, interactive, scalable, and extensive microbiome data science. Nat Biotechnol. 2018;6:e27295v2 https://doi.org/10.7287/peerj.preprints.27295v2.Article 

Google Scholar 
Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucl Acids Res. 2013;41:D590–6. https://doi.org/10.1093/nar/gks1219.CAS 
Article 
PubMed 

Google Scholar 
Karstens L, Asquith M, Davin S, Fair D, Gregory WT, Wolfe AJ, et al. Controlling for contaminants in low-biomass 16S rRNA gene sequencing experiments. mSystems. 2019;4:e00290–19. https://doi.org/10.1128/mSystems.00290-19.Article 
PubMed 
PubMed Central 

Google Scholar 
Lozupone CA, Knight R. Species divergence and the measurement of microbial diversity. FEMS Microbiol Rev. 2008;32:557–78. https://doi.org/10.1111/j.1574-6976.2008.00111.x.CAS 
Article 
PubMed 

Google Scholar 
Sanders JG, Powell S, Kronauer DJC, Vasconcelos HL, Frederickson ME, Pierce NE. Stability and phylogenetic correlation in gut microbiota: lessons from ants and apes. Mol Ecol. 2014;23:1268–83. 10.111/mec.12611.Article 

Google Scholar 
McMurdie PJ, Holmes S. phyloseq: an R package for reproducible interactive analysis and graphics of microbiome census data. PLOS One. 2013. https://doi.org/10.1371/journal.pone.0061217.Paliy O, Shankar V. Application of multivariate statistical techniques in microbial ecology. Mol Ecol. 2016;25:1032–57. https://doi.org/10.1111/mec.13536.CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Wickham H. ggplot2: elegant graphics for data analysis. Springer-Verlag New York. 2016.Paradis E, Schliep K. ape 5.0: an environment for modern phylogenetics and evolutionary analyses in R. Bioinformatics. 2019;35:526–8. https://doi.org/10.1093/bioinformatics/bty633.CAS 
Article 
PubMed 

Google Scholar 
Revell LJ. phytools: an R package for phylogenetic comparative biology (and other things). Methods Ecol Evol. 2012;3:217–23. https://doi.org/10.1111/j.2041-210X.2011.00169.x.Article 

Google Scholar 
Robinson DF, Foulds LR. Comparison of phylogenetic trees. Math Biosci. 1981;53:131–47. https://doi.org/10.1016/0025-5564(81)90043-2.Article 

Google Scholar 
Schliep KP. phangorn: phylogenetic analysis in R. Bioinformatics. 2011;27:592–3. https://doi.org/10.1093/bioinformatics/btq706.CAS 
Article 
PubMed 

Google Scholar 
Jombart T, Balloux F, Dray S. adephylo: new tools for investigating the phylogenetic signal in biological traits. Bioinformatics. 2010;26:1907–9. https://doi.org/10.1093/bioinformatics/btq292.CAS 
Article 
PubMed 

Google Scholar 
Oksanen J, Blanchet FG, Kindt R, Legendre P, Minchin PR, O’Hara R, et al. Package “Vegan”. Community Ecology Package. 2016. http://CRAN.R-project.org/package=vegan.Easson CG, Thacker RW. Phylogenetic signal in the community structure of host-specific microbiomes of tropical marine sponges. Front Microbiol. 2014;5:532 https://doi.org/10.3389/fmicb.2014.00532.Article 
PubMed 
PubMed Central 

Google Scholar 
Blomberg SP, Garland T, Ives AR. Testing for phylogenetic signal in comparative data: behavioral traits are more labile. Evolution. 2007;57:717–45. https://doi.org/10.1111/j.00143820.2003.tb00285.x.Article 

Google Scholar 
Pagel M. Detecting correlated evolution on phylogenies: a general method for the comparative analysis of discrete characters. Proc Royal Soc B. 1994;255:37–45. https://doi.org/10.1098/rspb.1994.0006.Article 

Google Scholar 
Harmon LJ, Weird JT, Brock CD, Glor RE, Challenger W. GEIGER: investigating evolutionary radiations. Bioinformatics. 2007;24:129–31. https://doi.org/10.1093/bioinformatics/btm538.CAS 
Article 
PubMed 

Google Scholar 
Phillips SJ, Anderson RP, Dudík M, Schapire RE, Blair ME. Opening the black box: an open-source release of Maxent. Ecography. 2017;40:887–93. https://doi.org/10.1111/ecog.03049.Article 

Google Scholar 
Hijmans RJ, Cameron SE, Parra JL, Jones PG, Jarvis A. Very high resolution interpolated climate surfaces for global land areas. Int J Climatol. 2005;25:1965–78. https://doi.org/10.1002/joc.1276.Article 

Google Scholar 
Orme D, Freckleton R, Thomas G, Petzoldt T, Fritz S, Isaac N, et al. CAPER: comparative analyses of phylogenetics and evolution in R. 2012. http://cran.r-project.org/package=caper.Groussin M, Mazel F, Sanders JG, Smillie CS, Lavergne S, Thuiller W, et al. Unraveling the processes shaping mammalian GMs over evolutionary time. Nat Commun. 2017;8:14319 https://doi.org/10.1038/ncomms14319.CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Knowles SCL, Eccles RM, Baltrūnaitė L. Species identity dominates over environment in shaping the microbiota of small mammals. Ecol Lett. 2019;22:826–37. https://doi.org/10.1111/ele.13240.CAS 
Article 
PubMed 

Google Scholar 
Kohl KD, Varner J, Wilkening JL, Dearing MD. Gut microbial communities of American pikas (Ochotona princeps): evidence for phylosymbiosis and adaptations to novel diets. J Anim Ecol. 2018a;87:323–330. https://doi.org/10.1111/1365-2656.12692.Article 
PubMed 

Google Scholar 
Kohl KD, Dearing MD, Bordenstein SR. Microbial communities exhibit host species distinguishability and phylosymbiosis along the length of the gastrointestinal tract. Mol Ecol. 2018;27:1874–83. https://doi.org/10.1111/mec.14460.Article 
PubMed 

Google Scholar 
Weinstein SB, Martínez-Mota R, Stapleton TE, Klure DM, Greeenhalgh R, Orr TJ, et al. Microbiome stability and structure is governed by host phylogeny over diet and geography in woodrats (Neotoma spp.). PNAS. 118:e2108787118. https://doi.org/10.1073/pnas.2108787118.Grond K, Bell KC, Demboski JR, Santos M, Sullivan JM, Hird SM. No evidence for phylosymbiosis in western chipmunk species. FEMS Microbiol Ecol. 2020;96:fiz182 https://doi.org/10.1093/femsec/fiz182.CAS 
Article 
PubMed 

Google Scholar 
Perofsky AC, Lewis RJ, Adondano LA, Di Fiore A, Meyers LA. Hierarchical social networks shape gut microbial composition in wild Verreaux’s sifaka. Proc R Soc B. 2017;284:20172274 https://doi.org/10.1098/rspb.2017.2274.Article 
PubMed 
PubMed Central 

Google Scholar 
Raulo A, Ruokolainen L, Lane A, Amato K, Knight R, Leigh S, et al. Social behavior and gut microbiota in red-bellied lemurs (Eulemur rubriventer): in search of the role of immunity in the evolution of sociality. J Anim Ecol. 2017;87:388–99. https://doi.org/10.1111/1365-2656.12781.Article 
PubMed 

Google Scholar 
Overdorff DJ. Similarities, differences, and seasonal patterns in the diets of Eulemur rubriventer and Eulemur fulvus rufus in the Ranomafana National Park, Madagascar. Int J Primatol. 1993;14:721–53. https://doi.org/10.1007/BF02192188.Article 

Google Scholar 
Johnson SE. Ecology and speciation in brown lemurs: white-collared lemur (Eulemur albocollaris) and hybrids (Eulemur albocollaris x Eulemur fulvus rufus) in southeastern Madagascar. 2002. PhD Dissertation. The University of Texas at Austin.Wyner YM, Johnson SE, Stumpf RM, DeSalle R. Genetic assessment of a white-collared x red-fronted lemur hybrid zone at Andringitra, Madagascar. Am J Primatol. 2002;67:51–66. https://doi.org/10.1002/ajp.10033.Article 

Google Scholar 
Hladik CM, Charles-Dominique P, Petter JJ. Feeding strategies of five nocturnal prosimians in the dry forest of the west coast of Madagascar. In: Charles-Dominique P, Cooper HM, Hladik A, Pages E, Pariente GF, Petter-Rousseaux A, Schilling A (eds). Nocturnal Malagasy primates: ecology, physiology, and behavior. Academic Press, New York, NY. 1980. pp 41-73. https://doi.org/10.1016/B978-0-12-169350-3.50007-1.Perelman P, Johnson WE, Roos C, Seuánez HN, Horvath JE, Moreira MAM, et al. A molecular phylogeny of living primates. PLOS Genet. 2011. https://doi.org/10.1371/journal.pgen.1001342. More

The coefficient of variation of phytoplankton ((CV_P)) varies with the changes in environmental factors, namely, light, temperature and salinity and many more. The focus of our discussion will be on the variation of (CV_P) of phytoplankton.Case 1: (CV_P < 1) Measured (CV_P) values are 0.32, 0.37, 0.78 at the depth of 10 m, 50 m, 50 m of Region 3, Region 4 and Region 2 respectively. From Fig. 1c, we observe that for Region 3, concentrated mean of phytoplankton has escalated over a larger domain along the horizontal axis, while spread of phytoplankton is comparatively very low and constant for all times, whereas for Region 2 and Region 4 (Fig. 1,b,e), spread of phytoplankton is comparatively high, but, quantity of concentrated biomass is higher at Region 4 than Region 2, which is also supported by higher phytoplankton productivity at Region 4 than Region 2.Nature of spread of phytoplankton is obtained from the dynamics of normalized variance x of phytoplankton, which depends on (beta). At a fixed depth, x increases with increasing (beta) (Fig. 5b). For all regions where (CV_P1). Therefore, spread x remains comparatively low (Fig. 7b), whereas (p_0) is close to 1 (Fig. 7a), which causes (CV_P) to be less than 1 (Fig. 7c) at this zone.From above discussion we observe that when (varepsilon) belongs to (0.035, 0.1) and due to this range of (varepsilon), domain of (beta) reduces for a location, then (CV_P) remains less than 1 at that zone. These domains of (varepsilon , beta) are determined from nature of phytoplankton productivity at a location during the period of observation and nature of the spread of dominating class. It has been observed that in case of Region 3, during early summer season (May), the existing phytoplankton communities are Skeletonema Costatum, Navicula species and Pyraminonas Grossii36, for Region 4, the existing phytoplankton communities in Sep are diatom Skeletonema Costatum, Dinoflagellates, Raphidophytes and others35, whereas for Region 2, the existing classes in May are diatom Skeletonema Costatum, Raphidophytes and others35. But, for all three regions during corresponding time periods, most of the phytoplankton biomass is dominated by the diatom class, Skeletonema Costatum35,36. Spread of this phytoplankton class has a peculiar nature, which is influenced by its measure of stickiness (alpha), where (alpha in (0,,0.98))43. Now, during the period of observation, since the dominating class Skeletonema Costatum coexists with some other phytoplankton classes at all three regions, therefore range of its measure of stickiness (alpha) should belong to (0.02, 0.25) for these regions and depending on (alpha), scatteredness of Skeletonema Costatum has varied for these zones, that is, when (alpha) is high, scatteredness of Skeletonema Costatum reduces and when (alpha) is low, this scatteredness increases. In field observation, we have seen that, at Region 3, scatteredness of Skeletonema Costatum is very low in May 2011, whereas for Region 4 and Region 2, it is slightly higher in Sep 2007 and May 2011. For all three zones, (alpha) belongs to ((0.02,,0.25)) but its value has varied differently for each zone. If we consider (alpha) to be high for Region 3 in May 2011, then Skeletonema Costatum will be more sticky for that zone during that time period which will hinder the scatteredness. If we assume (alpha) to be slightly high for Region 2, Region 4 for corresponding time periods, then Skeletonema Costatum will be less sticky than Region 3 and scatteredness will be slightly higher for these zones by that time.In the model, spread due to scatteredness is controlled by low (beta) value. Therefore, ecologically it might be considered that during early summer at Region 3, (alpha) value was close to 0.25, which has caused Skeletonema Costatum to remain more sticky at that zone, as a result, spread was very low which represents low (beta) value. Similar ecological assumptions can be drawn in case of Region 2, Region 4, but the only difference is probably, for these two zones in summer and early spring season respectively, (alpha) was slightly low than Region 3. As a result, the dominating class Skeletonema Costatum was less sticky than Region 3 and spread due to scatteredness was slightly higher than Region 3 (Fig. S4b). Hence, differences in the nature of total biomass of a system, nature of productivity and finally nature of stickiness of dominating phytoplankton species cause high irregularity in phytoplankton distribution and produce low (CV_P) values for Region 2, Region 3 (Fig. 7c, Fig. S4c) and Region 4 (Fig. 8c, Fig. S4c). Case 2: (CV_P > 1)
In case of Region 4, at the depth of 50 m, (CV_P) remains 1.61 and 1.36 in Dec 2006 and Feb 2008 respectively. In Dec 2006, Feb 2008, due to very low productivity, range of (varepsilon) remains (0.35, 1.0) at Region 4, which generates larger domain of (beta) (considering total biomass and half saturation constant remain the same at Region 4 during both time periods Dec 2006 and Feb 2008). Since total biomass A is conserved, large value of (beta) indicates larger value of B, which ecologically implies spread of all fluctuating components of nutrient and phytoplankton remains higher. Therefore, in Dec 2006 and Feb 2008, spread of phytoplankton remains higher, whereas due to very low productivity, most of the total biomass A is dominated by nutrient biomass (n_0) and phytoplankton biomass (p_0) remains very low, that is, (p_0 More

Experimental design and crop managementThe study was conducted during 2019 in a field experiment on an arable soil (classified as Luvisols) in the deep root experimental facility at the University of Copenhagen, Denmark (Supplementary Table S4). The experiment was conducted with two diverse perennial deep-rooted species: the tap-rooted forage legume lucerne (Medicago sativa L. (cv. Creno); Family: Fabaceae) with the capacity to fix N2 and the intermediate wheatgrass (Thinopyrum intermedium; Family: Poaceae) kernza developed by the Land Institute (Salina, Kansas, USA). Kernza was initially sown on April 11th, 2015 and lucerne on September 9th, 2016 with a seeding density of 20 kg seeds ha−1. Every year, kernza was fertilized with NPK fertilizer (21:7:3; NH4:NO3 = 1.28) as a single dose in early spring (before the onset of plant growth). Kernza was harvested every year in August using a combine harvester and lucerne three times per year in June, August, and October. Plants were rainfed with a subsurface drain installed at both 1 and 2 m depth running between the plots.For each species, fixed frames of 0.75 m2 were inserted in the soil (ca. 5 cm) within each field plot. Specifically, three field plots of lucerne (with observable root nodulation) and kernza were used where each of the three kernza field plots contained two subplots of N fertilized kernza at 100 kg N ha−1 (K100) (i.e., the standard fertilization within this field) and N fertilized kernza at 200 kg N ha−1 (K200) (i.e., within the range of standard fertilization practices for kernza). Before the onset of plant growth, all plots received 15N (as 15NH4Cl; 98 atom%) in trace amounts (corresponding to 1 kg N ha−1) to trace N allocation from the surface to deeper layers.
13C/14C-CO2-labelingWithin each fixed frame, the 13C/14C-CO2-labeling was conducted using an atmospheric labeling chamber41. Labeling with C-tracers was done with multiple-pulse labeling (three times per week) over two months until first harvest (May 2nd to June 20th 2019). Glass beakers containing 13C labeled bicarbonate (0.1 g mL−1 labeling solution; 99 atom%), and 14C labeled bicarbonate (11 kBq mL−1 labeling solution) within a solution of NaOH (1 M) were added within each of the labeling chambers. Once chambers were sealed, hydrochloric acid (HCl; 2 M) was added to the labeling solution (in equivalent amounts) via a syringe promoting 14CO2/13CO2 evolution. Chambers remained sealed for one to three hours (between 9 am and 12 pm) depending on weather conditions (i.e., the duration and intensity of sunshine). The amount of added labeling solution sequentially increased with increasing plant growth (i.e., 5 mL per 20 cm increase in plant height) reaching a plant height of 100–120 cm at the termination of the labeling.Shoot, root, and soil samplingThe labeling plots (0.75 m2) were harvested on June 20th, 2019 to obtain the aboveground biomass of lucerne and kernza (K100 and K200). The aboveground biomass in addition to samples obtained from unlabeled parts of the field was directly stored at − 20 °C until drying at 105 °C for two days. For each plot and unlabeled samples, the plant biomass was homogenized and ball-milled for subsequent isotopic analyses.Soil cores to 1.5 m depth were taken inside all labeling plots, and cores were subdivided into four depth intervals: 0–25, 25–50, 50–100, and 100–150 cm. The soil coring was conducted in 25 cm intervals using a soil auger (6 cm inner diameter). Specifically, per depth three soil samples were taken and stored at 4–5 °C (ca. two days) and then immediately processed and stored at -20 °C until analyses. Roots, bulk soil and rhizosphere soil (adhering to the roots), were separated by sequential sieving of the soil with finer mesh sizes to 1 mm as described by Peixoto, et al.26. A subsample of the bulk soil (ca. 150 g) from each depth in all labeling plots was washed on a 250 µm sieve to recover root fragments for subsequent isotopic determination in unrecovered root fragments. Soil samples (and associated roots) from unlabeled parts of the larger field plots were used to determine natural abundance of 13C/14C/15N with depth. The collection of plant material complied with relevant institutional guidelines and seeds were gifted by University of Copenhagen.Determination of 13C/14C/15N enrichment, and C and N quantityFor each defined depth, samples of roots and soil were homogenized, freeze-dried (except PLFA samples that were stored at − 20 °C), and ground in a ball-mill for the determination of total C and N, 13C, 15N, and 14C activity. Total C, N, 13C, and 15N were measured with a FLASH 2000 CHNS/O Elemental Analyzer (Thermo Fisher Scientific, Cambridge, UK) combined to a Delta V Advantage isotope ratio mass spectrometer via a ConFlo III interface (Thermo Fisher Scientific, Bremen, Germany) at the Centre for Stable Isotope Research and Analysis (Georg August University Göttingen, Göttingen, Germany).All δ13C values are standardized to the Vienna PeeDee Belemnite international isotope standard and δ15N values standardized to the δ15N values of atmospheric N2. 13C and 15N enrichment is expressed as atom% excess as calculated by the atom% difference between the respective labeled and unlabeled samples. The 14C activity was determined by combustion in a Hidex 600 OX Oxidizer (Hidex, Turku, Finland) and counted on a liquid scintillation counter (Tri-Carb 3180TR/SL, PerkinElmer, Waltham, MA, USA). 14C enrichment is determined by the difference in the 14C activity (Bq g−1) between the respective labeled and unlabeled samples.Calculation of root C and net rhizodepositionThe amount of root C (mg C kg−1 soil) was calculated based on the root dry matter and C concentration divided by the quantity of soil sampled38. For the determination of net rhizodeposition, 14C was used due to lower detection limits in deeper soil layers42. A modified tracer mass balance approach described by Rasmussen, et al.43 with adjusted unrecovered root fragments41 was used to determine the net rhizodeposition based on the following equations where the %ClvR is the relative proportion of rhizodeposition expressed as the percent C lost via rhizodeposition:$${text{%ClvR}} = frac{{^{{{14}}} {text{C Soil (rhizosphere + adjusted bulk)}}}}{{^{{{14}}} {text{C bulk soil }} + ,^{{{14}}} {text{C rhizosphere soil}} + ,^{{{14}}} {text{C Root}}}} times 100.$$$${text{Net rhizodeposition}} = frac{{{text{%ClvR }} times {text{ root C content}}}}{{left( {100 – % {text{ClvR}}} right)}}$$The 14C soil content was the sum of the adjusted bulk soil 14C and rhizosphere 14C content for each soil sample. The 14C rhizosphere and bulk soil content for each soil sample were determined by multiplying the total quantity of C by the 14C enrichment of the soil. The adjusted bulk soil 14C content was calculated as the difference between the bulk 14C soil content by the 14C root washed content as determined by the multiplication of 14C enrichment in root fragments recovered from a subsample of soil by the total C content within the entire soil volume sampled. The 14C root content was determined by multiplying the total quantify of C in roots by the 14C enrichment. Similar equations were used to calculate the net rhizodeposition of N based on 15N enrichment within the soil and roots.Biomarker analysesPhospholipid fatty acid (PLFA)The analysis of PLFAs was done according to a modified protocol by Frostegård, et al.44 with a detailed description of the modifications provided by Gunina, et al.45. In brief, 25 μL of 1,2-Dinonadecanoyl-sn-Glycero-3-Phosphatidylcholine (C19:0) (1 mg mL–1) were added to each of the samples and used in the quantification of recovery of the phospholipids. The lipid fraction from 5–6 g of rhizosphere soil was extracted twice using a one-phase Bligh-Dyer extractant46 of chloroform, methanol (MeOH), and citrate buffer (pH 4) (1:2:0.8, v/v/v). To isolate the phospholipid fraction, a solid-phase extraction with activated silica gel and methanol elution was conducted. The derivatization into fatty acid methyl esters occurred via a sequential hydrolyzation with 0.5 mL sodium hydroxide (NaOH) (0.5 M) in MeOH for 10 min at 100 °C and methylation with 0.75 mL of boron trifluoride (BF3) (1.3 M) in MeOH for 15 min at 80 °C. An external standard stock solution containing 28 individual fatty acids (ca. 1 mg mL–1 per fatty acid) used in the quantification of PLFA content was simultaneously derivatized with the samples. The residues were dissolved in 185 μL of toluene, and 15 μL of the internal standard 2, tridecanoic acid methyl ester (C13:0) (1 mg mL–1) were added to each sample prior to measurement using an Agilent 7820A GC coupled to an Agilent 5977 quadrupole mass spectrometer (Agilent, Waldbronn, Germany). The sum of all PLFAs was used as a proxy of the living microbial biomass based on the direct relation between PLFAs and microbial biomass.Amino sugars (AS)Amino sugars were extracted according to a modified protocol by Zhang and Amelung47 with a detailed description of the procedure by Peixoto, et al.26. In brief, 0.8–1.5 g of freeze-dried rhizosphere soil were hydrolyzed with the addition of 11 mL of 6 M HCl for 8 h at 105 °C. Following hydrolysis, soil samples were filtered and HCl was removed via rotary evaporation at 45 °C to dry the filtrate. Prior to derivatization both iron precipitates and salts were removed from the filtrate and 25 μL of the internal standard 1, methylglucamine (MeGlcN) (1 mg mL–1) was added and used for quantification of recovery. The derivatization into aldononitrile acetates was conducted as described by Zhang and Amelung47. For the quantification of AS, an external standard stock solution containing the AS: N-acetylglucosamine (GlcN) (2 mg mL–1), N-acetylgalactosamine (GalN) (2 mg mL–1), N-acetylmuramic acid (MurN) (1 mg mL–1), mannosamine (ManN) (2 mg mL–1), and MeGlcN (1 mg mL–1) was derivatized and analyzed with the samples. The residues were dissolved in 185 μL of ethyl acetate-hexane (1:1, v/v), and 15 μL of the internal standard 2, tridecanoic acid methyl ester (1 mg mL–1), were added to the samples for measurement using an Agilent 7890A GC coupled to Agilent 7000A triple quadrupole mass spectrometer (Agilent, Waldbronn, Germany). Total amino sugars content was calculated as the summation of the four detected amino sugars: GlcN, MurN, GalN, and ManN.Amino acids (AA)Amino acids were extracted from both freeze-dried rhizosphere soil and root samples according to the protocol by Enggrob, et al.48. In brief, 0.8–3 g of rhizosphere soil and 0.02 g of root were hydrolyzed with the addition of 2 mL of 6 M HCl for 20 h at 110 °C to break the peptide bonds. Samples were subsequently purified via the removal of lipophilic and solid compounds by the addition of 4 mL n-hexane/dichloromethane (6:5, v/v) to the soil and root samples. Following centrifugation, the aqueous phase was filtered through glass wool and rinsed with 2 × 0.5 mL 0.1 M HCl into new glass tubes with the addition of 300 μL of the internal standard, norleucine (2.5 mM). The samples were freeze-dried and the residues dissolved in 1 mL 0.01 M HCl prior to the separation of amino acids and amino sugars (i.e., N containing compounds) on a polypropylene column with a cation exchange resin. The amino acids were eluted with a 2.5 M ammonium hydroxide solution and freeze-dried prior to derivatization of the amino acids as described by Enggrob, et al.48. For the quantification of AA, an external standard stock solution containing 14 AA was derivatized and analyzed with the samples. The amino acids were measured using a trace GC Ultra mounted with a TriPlus autosampler (Thermo Scientific, Hvidovre, Denmark) coupled via a combustion reactor (GC IsoLink, Thermo Scientific) to an isotope ratio mass spectrometer (Delta V Plus IRMS, Thermo Scientific). The total AA content of the rhizosphere soil and roots was based on the summation of the AA: alanine, Asx (asparagine and aspartate), Glx (glutamine and glutamate), glycine, isoleucine, lysine, phenylalanine, Pro/Thr (proline and threonine), serine, tyrosine, and valine.Compound-specific stable isotope probingTo determine the 13C enrichment of biomarkers, all raw δ13C were measured individually for AS and PLFA using a Delta V Advantage isotope ratio mass spectrometer via a ConFlo III interface (Thermo Fisher Scientific, Bremen, Germany). For AA, all raw δ13C were measured using a trace GC Ultra mounted with a TriPlus autosampler (Thermo Scientific, Hvidovre, Denmark) coupled via a combustion reactor (GC IsoLink, Thermo Scientific) to an isotope ratio mass spectrometer (Delta V Plus IRMS, Thermo Scientific). For each sample, chromatogram peaks identified based on retention times specific for the measured amino sugars, PLFA, and AA were integrated using Isodat v. 3.0 (Thermo Fisher Scientific). All raw δ13C values were corrected for dilution by additional C atoms added during the derivatization, amount dependence, offset, and drift (for PLFA samples)49,50,51. To determine the 13C incorporation into each biomarker, the 13C excess for each biomarker as determined by the difference between the 13C of the labeled and unlabeled biomarker was multiplied by the C content of the specific biomarker.Relative microbial stabilization (RMS)The relative microbial stabilization is based on the relation of rhizodeposited 13C in the PLFA and amino sugar pools as described in detail by Peixoto, et al.26. The underlying assumption is that 13C incorporation into the amino sugar pool indicates the transformation of rhizodeposited C into necromass52,53, and the 13C incorporation into the PLFA pool (i.e., the living microbial biomass) represents a temporary C pool as PLFAs are immediately exposed to degradation following cell lysis54. The relative microbial stabilization (RMS) is calculated as follows:$${text{Relative microbial stabilization}} = {text{log}}frac{{{text{Average weighted atom% }},^{{{13}}} {text{C excess AS}}}}{{{text{Average weighted atom% }},^{{{13}}} {text{C excess PLFA}}}}$$where the average weighted atom% 13C excess is determined by the total 13C incorporation divided by the total C content of the respective PLFA or amino sugar pools. Accordingly RMS  0 is indicative of higher stabilization of C based on the dominant entry of C into the microbial necromass. However, the RMS indicator does not imply the absolute stability of rhizodeposited C, but rather signifies the potential for microbial stabilization among contrasting experimental variables (i.e., depth and plant species).Molecular analysisDNA extractionFrom each sample, 0.5 g of freeze-dried rhizosphere soil was used for DNA extraction using the Fast DNA Spin kit for Soil (MP Biomedicals, Solon, OH, USA) according to the manufacturer’s protocol with a single modification. Following, the addition of Binding Matrix, the suspension was washed with 5.5 M Guanidine Thiocyanate (protocol from MP Biomedicals) to remove humic acids that could inhibit preceding polymerase chain reaction (PCR) steps. The DNA was eluted in DNase free water and purified using the NucleoSpin gDNA Clean-up kit following the manufacturer’s protocol (Macherey–Nagel, Düren, Germany). The purity and concentration of DNA were checked on Nanodrop and Qubit, respectively.Amplicon sequencingExtracted DNA was sent to Novogene Europe (Cambridge, United Kingdom) for library preparation and amplicon sequencing. For 16S rRNA gene amplicon sequencing of the V3-V4 regions, the primer pair 341 F and 806 R were used (Supplementary Table S5). To identify the fungal communities, we targeted the Internal Transcribed Spacer (ITS) Region 1, using the primer pair ITS1 and ITS2 (Supplementary Table S5). The constructed libraries were sequenced using a Novaseq 6000 platform producing 2 × 250 bp paired-end reads. Raw sequences were deposited in the NCBI Sequence Read Archive (Bioproject number PRJNA736561).Quantitative PCRCopy numbers of the 16S rRNA gene were determined by quantitative PCR (qPCR) using the primers 341F and 805R (Supplementary Table S5) on an AriaMX Real-Time PCR System (Agilent Technologies, Santa Clara, CA, USA). An external plasmid standard curve was made based on the pCR 2.1 TOPO vector (Thermo Fisher Scientific, Waltham, MA, USA) with a 16S rRNA gene insert amplified from bulk soil. The PCR reaction was performed in 20 µl reactions containing: 1 × Brilliant III Ultra-Fast SYBR green low ROX qPCR Master Mix (Agilent Technologies, Santa Clara, CA, USA), 0.05 µg/µl BSA (New England Biolabs Inc., Ipswich, MA, USA), 0.4 µM of each primer and 2 μl of template DNA. The thermal cycling conditions were 3 min at 95 °C followed by 40 cycles of 20 s at 95 °C and 30 s at 58 °C, and a final extension for 1 min at 95 °C. A melting curve was included according to the default settings of the AriaMx qPCR software (Agilent Technologies). The reaction efficiencies were between 97 and 102%. Fungal quantification was done by qPCR amplification of the Internal Transcribed Spacer 1 (ITS1) using the primers ITS1-F and ITS2 (Supplementary Table S5). A plasmid standard curve was made using the pCR 2.1 TOPO vector containing an ITS1 region from Penicillium aculeatum. Reaction mixture and cycling conditions were as described above for the 16S rRNA gene (Supplementary Table S5). The reaction efficiency was 84%.Quantification of functional genes involved in N cyclingThe five bacterial genes amoA, nirK, nirS, nosZ, and nifH coding for enzymes involved in N-cycling were quantified by qPCR on an AriaMx Real-Time PCR System (Agilent Technologies). Reaction mixtures and cycling conditions were as described above for the 16S rRNA gene (Supplementary Table S5). The standard curves were prepared as described in Garcia-Lemos, et al.55. The reaction efficiencies were in the range 87%-105%.Sequence processingRaw reads were treated using DADA2 version 1.14.156. In brief, reads were quality checked and primers were removed using Cutadapt v. 1.1557. We followed the protocol DADA2 using default parameters, with a few modifications. For 16S rRNA sequences, the forward and reverse reads were trimmed to 222 and 219 bp, respectively, while the maxEE was set to 2 and 5 for forward and reverse reads, respectively. Detection of amplicon sequence variants (ASVs) was done using the pseudo-pool option and forward and reverse reads were merged with a minimum overlap of 10 bp. Merged reads in the range of 395–439 bp were kept, as reads outside this range are considered too long or too short for the sequenced region. Taxonomy was assigned using the Ribosomal Database Project (RDP) classifier58 with the Silva database v.13859. For ITS region 1, quality filtered reads shorter than 50 bp were removed prior to merging the forward and the reverse reads, with maxEE set to two for both forward and reverse reads. During merging, the minimum overlap was set to 20 (default). Taxonomy was assigned with the RDP classifier using the Unite v. 8.2 database60 after removal of chimeras.As ITS region 1 has a variable length, reads can be lost during merging. Hence, to validate our dataset we ran only the forward reads through the DADA2 pipeline and compared the overall community structure with the dataset from the merging using a Mantel test. No significant changes were observed in the community structures between the two datasets (r = 0.99; p = 0.0001). To obtain the highest taxonomic resolution, the dataset based on the merged reads was used. Further analysis was done using the phyloseq v. 1.30.0 R package61.Statistical analysisAnalyses of variance (ANOVA) were conducted to examine the effects of N fertilized kernza at 100 kg N ha−1 (K100) and kernza at 200 kg N ha−1 (K200) as well as to test the effect of the deep-rooted plant species: kernza and lucerne, and soil depth on each of the dependent variables. An average across the two subplots within each of the three kernza field plots was used when measured variables did not significantly differ between K100 and K200. Subsequent pairwise comparisons of the means was conducted using the TukeyHSD post-hoc test. Homogeneity of variance and normality were confirmed (data log-transformed when required) for all comparisons using the Fligner-Killeen test of homogeneity of variances62 and the Shapiro–Wilk test of normality63. A permutational multivariate analysis of variance (PERMANOVA) using the Bray–Curtis dissimilarity matrix with the adonis function in the vegan R package was used to test the effect of K100 and K200, lucerne across both K100 and K200, and depth on the bacterial and fungal communities. The multivariate homogeneity of group dispersions or variances were confirmed for all comparisons using the function betadisper in vegan. The bacterial and fungal communities in response to the ascribed variables were visually represented as ordination plots with a Principle Coordinates Analysis (PCoA). Unique ASVs were defined for each depth and between K100, K200, and lucerne as ASVs only present in those samples belonging to a specific depth and treatment. Significance testing was conducted at p  More

Shackleton, N. J., Sánchez-Goñi, M. F., Pailler, D. & Lancelot, Y. Marine isotope substage 5e and the eemian interglacial. Glob. Planet. Change 36, 151–155 (2003).ADS 

Google Scholar 
Shackleton, N. J., Chapman, M., Sánchez-Goñi, M. F., Pailler, D. & Lancelot, Y. The classic marine isotope substage 5e. Quat. Res. 58, 14–16 (2002).CAS 

Google Scholar 
Hofreiter, M. & Stewart, J. Ecological change, range fluctuations and population dynamics during the pleistocene. Curr. Biol. 19, R584–R594 (2009).CAS 
PubMed 

Google Scholar 
Hewitt, G. M. Post-glacial re-colonization of European biota. Biol. J. Linn. Soc. 68, 87–112 (1999).
Google Scholar 
Petit, R. J. et al. Glacial refugia: hotspots but not melting pots of genetic diversity. Science 300, 1563–1565 (2003).ADS 
CAS 
PubMed 

Google Scholar 
Magri, D., Di Rita, F., Aranbarri, J., Fletcher, W. & González-Sampériz, P. Quaternary disappearance of tree taxa from Southern Europe: timing and trends. Quat. Sci. Rev. 163, 23–55 (2017).ADS 

Google Scholar 
Calatayud, J. et al. Pleistocene climate change and the formation of regional species pools. Proc. R. Soc. B Biol. Sci. 286, 20190291 (2019).
Google Scholar 
Ebdon, S. et al. The Pleistocene species pump past its prime: evidence from European butterfly sister species. Mol. Ecol. 30, 3575–3589 (2021).Záveská, E. et al. Multiple auto- and allopolyploidisations marked the Pleistocene history of the widespread Eurasian steppe plant Astragalus onobrychis (Fabaceae). Mol. Phylogenet. Evol. 139, 106572 (2019).Wesche, K. et al. The Palaearctic steppe biome: a new synthesis. Biodivers. Conserv. 25, 2197–2231 (2016).
Google Scholar 
Walter, H. & Breckle, S. Ökologie der Erde, Band 1. (Spektrum Akademischer Verlag, 1991).Braun-Blanquet, J. Die inneralpine Trockenvegetation: von der Provence bis zur Steiermark. (Gustav Fischer, 1961).Hurka, H. et al. The Eurasian steppe belt: Status quo, origin and evolutionary history. Turczaninowia 22, 5–71 (2019).
Google Scholar 
Jännicke, W. Die Sandflora von Mainz, ein Relict aus der Steppenzeit. (Gebrueder Knauer, 1892).Allen, J. R. M. et al. Rapid environmental changes in southern Europe during the last glacial period. Nature 400, 740–743 (1999).ADS 
CAS 

Google Scholar 
Reille, M. & de Beaulieu, J. L. Pollen analysis of a long upper Pleistocene continental sequence in a Velay maar (Massif Central, France). Palaeogeogr. Palaeoclimatol. Palaeoecol. 80, 35–48 (1990).
Google Scholar 
Sadori, L. et al. Pollen-based paleoenvironmental and paleoclimatic change at Lake Ohrid (south-eastern Europe) during the past 500 ka. Biogeosciences 13, 1423–1437 (2016).ADS 
CAS 

Google Scholar 
Ellenberg, H. & Leuschner, C. Vegetation Mitteleuropas mit den Alpen: in ökologischer, dynamischer und historischer Sicht. (Stuttgart: Verlag Eugen Ulmer, 2010).Kirschner, P. et al. Long-term isolation of European steppe outposts boosts the biomes conservation value. Nat. Commun. 11, 1968 (2020).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Fonseca, E. M., Colli, G. R., Werneck, F. P. & Carstens, B. C. Phylogeographic model selection using convolutional neural networks. Mol. Ecol. Resour. 21, 2661–2675 (2021).Beaumont, M. A., Zhang, W. & Balding, D. J. Approximate Bayesian computation in population genetics. Genetics 162, 2025–2035 (2002).PubMed 
PubMed Central 

Google Scholar 
Csilléry, K., Blum, M. G. B., Gaggiotti, O. E. & François, O. Approximate Bayesian computation (ABC) in practice. Trends Ecol. Evol. 25, 410–418 (2010).PubMed 

Google Scholar 
Flagel, L., Brandvain, Y. & Schrider, D. R. The unreasonable effectiveness of convolutional neural networks in population genetic inference. Mol. Biol. Evol. 36, 220–238 (2019).CAS 
PubMed 

Google Scholar 
Robert, C. P., Cornuet, J.-M., Marin, J.-M. & Pillai, N. S. Lack of confidence in approximate Bayesian computation model choice. Proc. Natl Acad. Sci. USA 108, 15112–15117 (2011).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Sanchez, T., Cury, J., Charpiat, G. & Jay, F. Deep learning for population size history inference: design, comparison and combination with approximate Bayesian computation. Mol. Ecol. Resour. 21, 2645–2660 (2021).Liu, X. & Fu, Y.-X. Stairway Plot 2: demographic history inference with folded SNP frequency spectra. Genome Biol. 21, 280 (2020).PubMed 
PubMed Central 

Google Scholar 
Liu, X. & Fu, Y.-X. Exploring population size changes using SNP frequency spectra. Nat. Genet. 47, 555–559 (2015).CAS 
PubMed 
PubMed Central 

Google Scholar 
Magri, D. et al. A new scenario for the quaternary history of European beech populations: palaeobotanical evidence and genetic consequences. New Phytol. 171, 199–221 (2006).CAS 
PubMed 

Google Scholar 
Pironon, S. et al. Geographic variation in genetic and demographic performance: new insights from an old biogeographical paradigm. Biol. Rev. 92, 1877–1909 (2017).PubMed 

Google Scholar 
Arenas, M., Ray, N., Currat, M. & Excoffier, L. Consequences of range contractions and range shifts on molecular diversity. Mol. Biol. Evol. 29, 207–218 (2012).CAS 
PubMed 

Google Scholar 
Excoffier, L., Foll, M. & Petit, R. J. Genetic consequences of range expansions. Annu. Rev. Ecol. Evol. Syst. 40, 481–501 (2008).
Google Scholar 
Mona, S., Ray, N., Arenas, M. & Excoffier, L. Genetic consequences of habitat fragmentation during a range expansion. Heredity 112, 291–299 (2014).CAS 
PubMed 

Google Scholar 
Szűcs, M., Melbourne, B. A., Tuff, T. & Hufbauer, R. A. The roles of demography and genetics in the early stages of colonization. Proc. R. Soc. B Biol. Sci. 281, 20141073 (2014).
Google Scholar 
Loog, L. Sometimes hidden but always there: the assumptions underlying genetic inference of demographic histories. Philos. Trans. R. Soc. B Biol. Sci. 376, 20190719 (2021).
Google Scholar 
Narbona, E., Arista, M. & Ortiz, P. L. Explosive seed dispersal in two perennial Mediterranean Euphorbia species (Euphorbiaceae). Am. J. Bot. 92, 510–516 (2005).PubMed 

Google Scholar 
Stevens, V. M. et al. A comparative analysis of dispersal syndromes in terrestrial and semi-terrestrial animals. Ecol. Lett. 17, 1039–1052 (2014).PubMed 

Google Scholar 
Flouri, T., Jiao, X., Rannala, B. & Yang, Z. Species tree inference with BPP using genomic sequences and the multispecies coalescent. Mol. Biol. Evol. 35, 2585–2593 (2018).CAS 
PubMed 
PubMed Central 

Google Scholar 
Willeit, M., Ganopolski, A., Calov, R. & Brovkin, V. Mid-Pleistocene transition in glacial cycles explained by declining CO2 and regolith removal. Sci. Adv. 5, eaav7337 (2019).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Hansen, J., Sato, M., Russell, G. & Kharecha, P. Climate sensitivity, sea level and atmospheric carbon dioxide. Philos. Trans. R. Soc. Math. Phys. Eng. Sci. 371, 20120294 (2013).ADS 

Google Scholar 
Martinson, D. G. et al. Age dating and the orbital theory of the ice ages: Development of a high-resolution 0 to 300,000-year chronostratigraphy. Quat. Res. 27, 1–29 (1987).CAS 

Google Scholar 
OConnell, K. A. et al. Impacts of the Toba eruption and montane forest expansion on diversification in Sumatran parachuting frogs (Rhacophorus). Mol. Ecol. 29, 2994–3009 (2020).
Google Scholar 
Theodoridis, S. et al. How do cold-adapted plants respond to climatic cycles? Interglacial expansion explains current distribution and genomic diversity in Primula farinosa L. Syst. Biol. 66, 715–736 (2017).PubMed 

Google Scholar 
Williams, M. The More

Good, S. P., Noone, D. & Bowen, G. Hydrologic connectivity constrains partitioning of global terrestrial water fluxes. Science 349, 175–177 (2015).ADS 
CAS 
PubMed 

Google Scholar 
Masson-Delmotte, V. et al. Climate change 2021: The physical science basis. contribution of working group I to the sixth assessment report of the intergovernmental panel of climate change. Global warming of 1.5 C. An IPCC Special Report (2021).Milly, P. C. D., Dunne, K. A. & Vecchia, A. V. Global pattern of trends in streamflow and water availability in a changing climate. Nature 438, 347–350 (2005).ADS 
CAS 
PubMed 

Google Scholar 
Konapala, G., Mishra, A. K., Wada, Y. & Mann, M. E. Climate change will affect global water availability through compounding changes in seasonal precipitation and evaporation. Nat. Commun. 11, 3044 (2020).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Miralles, D. G., Gentine, P., Seneviratne, S. I. & Teuling, A. J. Land-atmospheric feedbacks during droughts and heatwaves: state of the science and current challenges. Ann. N. Y. Acad. Sci. 1436, 19–35 (2019).ADS 
PubMed 

Google Scholar 
Schwalm, C. R. et al. Global patterns of drought recovery. Nature 548, 202–205 (2017).ADS 
CAS 
PubMed 

Google Scholar 
Sippel, S. et al. Drought, heat, and the carbon cycle: a review. Curr. Clim. Change Rep. 4, 266–286 (2018).
Google Scholar 
Peterson, T. J., Saft, M., Peel, M. C. & John, A. Watersheds may not recover from drought. Science 372, 745–749 (2021).ADS 
CAS 
PubMed 

Google Scholar 
Vicente-Serrano, S. M., Beguería, S. & L`ópez-Moreno, J. I. A multiscalar drought index sensitive to global warming: the standardized precipitation evapotranspiration index. J. Clim. 23, 1696–1718 (2010).
Google Scholar 
Anderson, M. C. et al. The evaporative stress index as an indicator of agricultural drought in brazil: an assessment based on crop yield impacts. Remote Sens. Environ. 174, 82–99 (2016).ADS 

Google Scholar 
Fisher, J. B. et al. The future of evapotranspiration: global requirements for ecosystem functioning, carbon and climate feedbacks, agricultural management, and water resources. Water Resour. Res. 53, 2618–2626 (2017).ADS 

Google Scholar 
Kalma, J. D., McVicar, T. R. & McCabe, M. F. Estimating land surface evaporation: a review of methods using remotely sensed surface temperature data. Surv. Geophys. 29, 421–469 (2008).ADS 

Google Scholar 
Melton, F. S. et al. Openet: Filling a critical data gap in water management for the western united states. JAWRA Journal of the American Water Resources Association (2021). https://onlinelibrary.wiley.com/doi/abs/10.1111/1752-1688.12956. https://onlinelibrary.wiley.com/doi/pdf/10.1111/1752-1688.12956.Lawrence, D. M. et al. The community land model version 5: Description of new features, benchmarking, and impact of forcing uncertainty. J. Adv. Modeling Earth Syst. 11, 4245–4287 (2019).ADS 

Google Scholar 
Niu, G.-Y. et al. The community noah land surface model with multiparameterization options (noah-mp): 1. model description and evaluation with local-scale measurements. J. Geophys. Res.: Atmosph. 116 (2011). https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1029/2010JD015139.Miralles, D. G. et al. Global land-surface evaporation estimated from satellite-based observations. Hydrol. Earth Syst. Sci. 15, 453–469 (2011).ADS 

Google Scholar 
Fisher, J. B., Tu, K. P. & Baldocchi, D. D. Global estimates of the land-atmosphere water flux based on monthly avhrr and islscp-ii data, validated at 16 fluxnet sites. Remote Sens. Environ. 112, 901–919 (2008).ADS 

Google Scholar 
Mu, Q., Zhao, M. & Running, S. W. Improvements to a modis global terrestrial evapotranspiration algorithm. Remote Sens. Environ. 115, 1781–1800 (2011).ADS 

Google Scholar 
Mueller, B. & Seneviratne, S. I. Systematic land climate and evapotranspiration biases in cmip5 simulations. Geophys. Res. Lett. 41, 128–134 (2014).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Koppa, A., Alam, S., Miralles, D. G. & Gebremichael, M. Budyko-based long-term water and energy balance closure in global watersheds from earth observations. Water Resour. Res. 57, e2020WR028658 (2021).ADS 
PubMed 
PubMed Central 

Google Scholar 
Fisher, J. B. et al. The future of evapotranspiration: global requirements for ecosystem functioning, carbon and climate feedbacks, agricultural management, and water resources. Water Resour. Res. 53, 2618–2626 (2017).ADS 

Google Scholar 
Penman, H. L. & Keen, B. A. Natural evaporation from open water, bare soil and grass. Proc. R. Soc. Lond. Ser. A. Math. Phys. Sci. 193, 120–145 (1948).ADS 
CAS 

Google Scholar 
Priestley, C. H. B. & Taylor, R. J. On the assessment of surface heat flux and evaporation using large-scale parameters. Monthly Weather Rev. 100, 81–92 (1972).
Google Scholar 
Maes, W. H., Gentine, P., Verhoest, N. E. C. & Miralles, D. G. Potential evaporation at eddy-covariance sites across the globe. Hydrol. Earth Syst. Sci. 23, 925–948 (2019).ADS 

Google Scholar 
Zhao, W. L. et al. Physics-constrained machine learning of evapotranspiration. Geophys. Res. Lett. 46, 14496–14507 (2019).ADS 

Google Scholar 
Miralles, D. G. et al. The wacmos-et project – part 2: Evaluation of global terrestrial evaporation data sets. Hydrol. Earth Syst. Sci. 20, 823–842 (2016).ADS 

Google Scholar 
Green, J. K., Berry, J., Ciais, P., Zhang, Y. & Gentine, P. Amazon rainforest photosynthesis increases in response to atmospheric dryness. Sci. Adv. 6 (2020). https://advances.sciencemag.org/content/6/47/eabb7232. https://advances.sciencemag.org/content/6/47/eabb7232.full.pdf.Verhoef, A. & Egea, G. Modeling plant transpiration under limited soil water: Comparison of different plant and soil hydraulic parameterizations and preliminary implications for their use in land surface models. Agric. For. Meteorol. 191, 22–32 (2014).ADS 

Google Scholar 
Powell, T. L. et al. Confronting model predictions of carbon fluxes with measurements of amazon forests subjected to experimental drought. N. Phytologist 200, 350–365 (2013).
Google Scholar 
Wu, X. et al. Parameterization of the water stress reduction function based on soil–plant water relations. Irrig. Sci. 39, 101–122 (2021).
Google Scholar 
Zhang, J., Liu, P., Zhang, F. & Song, Q. Cloudnet: Ground-based cloud classification with deep convolutional neural network. Geophys. Res. Lett. 45, 8665–8672 (2018).ADS 

Google Scholar 
Hengl, T. et al. Soilgrids250m: global gridded soil information based on machine learning. PLoS ONE 12, 1–40 (2017).
Google Scholar 
Hansen, M. C. et al. High-resolution global maps of 21st-century forest cover change. Science 342, 850–853 (2013).ADS 
CAS 
PubMed 

Google Scholar 
Jung, M. et al. The fluxcom ensemble of global land-atmosphere energy fluxes. Sci. Data 6, 74 (2019).PubMed 
PubMed Central 

Google Scholar 
McGovern, A. et al. Using artificial intelligence to improve real-time decision-making for high-impact weather. Bull. Am. Meteorological Soc. 98, 2073–2090 (2017).
Google Scholar 
Kratzert, F. et al. Toward improved predictions in ungauged basins: exploiting the power of machine learning. Water Resour. Res. 55, 11344–11354 (2019).ADS 

Google Scholar 
Reichstein, M. et al. Deep learning and process understanding for data-driven earth system science. Nature 566, 195–204 (2019).ADS 
CAS 
PubMed 

Google Scholar 
Rasp, S., Pritchard, M. S. & Gentine, P. Deep learning to represent subgrid processes in climate models. Proc. Natl Acad. Sci. USA 115, 9684–9689 (2018).CAS 
PubMed 
PubMed Central 

Google Scholar 
de Bézenac, E., Pajot, A. & Gallinari, P. Deep learning for physical processes: incorporating prior scientific knowledge. J. Stat. Mech.: Theory Exp. 2019, 124009 (2019).MathSciNet 
MATH 

Google Scholar 
Kraft, B., Jung, M., Körner, M. & Reichstein, M. Hybrid modeling: Fusion of a deep learning approach and a physics-based model for global hydrological modeling. Int. Arch. Photogramm., Remote Sens. Spat. Inf. Sci. XLIII-B2-2020, 1537–1544 (2020).
Google Scholar 
Chen, H., Huang, J. J., Dash, S. S., Wei, Y. & Li, H. A hybrid deep learning framework with physical process description for simulation of evapotranspiration. J. Hydrol. 606, 127422 (2022).
Google Scholar 
Martens, B. et al. Gleam v3: satellite-based land evaporation and root-zone soil moisture. Geoscientific Model Dev. 10, 1903–1925 (2017).ADS 

Google Scholar 
Gash, J. H. C. An analytical model of rainfall interception by forests. Q. J. R. Meteorological Soc. 105, 43–55 (1979).ADS 

Google Scholar 
Grossiord, C. et al. Plant responses to rising vapor pressure deficit. N. Phytologist 226, 1550–1566 (2020).
Google Scholar 
Urban, J., Ingwers, M., McGuire, M. A. & Teskey, R. O. Stomatal conductance increases with rising temperature. Plant Signal. Behav. 12, e1356534 (2017). PMID: 28786730.PubMed 
PubMed Central 

Google Scholar 
Matthews, J. S. A., Vialet-Chabrand, S. & Lawson, T. Role of blue and red light in stomatal dynamic behaviour. J. Exp. Bot. 71, 2253–2269 (2019).PubMed Central 

Google Scholar 
Xu, Z., Jiang, Y., Jia, B. & Zhou, G. Elevated-co2 response of stomata and its dependence on environmental factors. Front. Plant Sci. 7, 657 (2016).PubMed 
PubMed Central 

Google Scholar 
Jung, M. et al. Recent decline in the global land evapotranspiration trend due to limited moisture supply. Nature 467, 951–954 (2010).ADS 
CAS 
PubMed 

Google Scholar 
Peng, Y., Bloomfield, K. J., Cernusak, L. A., Domingues, T. F. & Colin Prentice, I. Global climate and nutrient controls of photosynthetic capacity. Commun. Biol. 4, 462 (2021).CAS 
PubMed 
PubMed Central 

Google Scholar 
Knoben, W. J. M., Freer, J. E. & Woods, R. A. Technical note: Inherent benchmark or not? comparing nash–sutcliffe and kling–gupta efficiency scores. Hydrol. Earth Syst. Sci. 23, 4323–4331 (2019).ADS 

Google Scholar 
Pagán, B. R., Maes, W. H., Gentine, P., Martens, B. & Miralles, D. G. Exploring the potential of satellite solar-induced fluorescence to constrain global transpiration estimates. Remote Sens. 11 (2019). https://www.mdpi.com/2072-4292/11/4/413.Jonard, F. et al. Value of sun-induced chlorophyll fluorescence for quantifying hydrological states and fluxes: current status and challenges. Agric. For. Meteorol. 291, 108088 (2020).ADS 

Google Scholar 
Bauer, P. et al. The digital revolution of earth-system science. Nat. Comput. Sci. 1, 104–113 (2021).
Google Scholar 
Bonan, G. B. Forests and climate change: forcings, feedbacks, and the climate benefits of forests. Science 320, 1444–1449 (2008).ADS 
CAS 
PubMed 

Google Scholar 
Pastorello, G. et al. The fluxnet2015 dataset and the oneflux processing pipeline for eddy covariance data. Sci. Data 7, 225 (2020).PubMed 
PubMed Central 

Google Scholar 
Wei, Z. et al. Revisiting the contribution of transpiration to global terrestrial evapotranspiration. Geophys. Res. Lett. 44, 2792–2801 (2017).ADS 

Google Scholar 
Stoy, P. C. et al. Reviews and syntheses: turning the challenges of partitioning ecosystem evaporation and transpiration into opportunities. Biogeosciences 16, 3747–3775 (2019).ADS 
CAS 

Google Scholar 
Poyatos, R. et al. Global transpiration data from sap flow measurements: the sapfluxnet database. Earth Syst. Sci. Data 13, 2607–2649 (2021).ADS 

Google Scholar 
Falster, D. S. et al. Baad: a biomass and allometry database for woody plants. Ecology 96, 1445–1445 (2015).
Google Scholar 
Granier, A. & Loustau, D. Measuring and modelling the transpiration of a maritime pine canopy from sap-flow data. Agric. For. Meteorol. 71, 61–81 (1994).ADS 

Google Scholar 
Aumann, H. et al. Airs/amsu/hsb on the aqua mission: design, science objectives, data products, and processing systems. IEEE Trans. Geosci. Remote Sens. 41, 253–264 (2003).ADS 

Google Scholar 
Wielicki, B. A. et al. Clouds and the earth’s radiant energy system (ceres): an earth observing system experiment. Bull. Am. Meteorological Soc. 77, 853–868 (1996).ADS 

Google Scholar 
Moesinger, L. et al. The global long-term microwave vegetation optical depth climate archive (vodca). Earth Syst. Sci. Data 12, 177–196 (2020).ADS 

Google Scholar 
Abadi, M. et al. TensorFlow: Large-scale machine learning on heterogeneous systems (2015). https://www.tensorflow.org/. Software available from tensorflow.org.Gupta, H. V., Kling, H., Yilmaz, K. K. & Martinez, G. F. Decomposition of the mean squared error and NSE performance criteria: implications for improving hydrological modelling. J. Hydrol. 377, 80–91 (2009).ADS 

Google Scholar 
Yu, L., Wen, J., Chang, C. Y., Frankenberg, C. & Sun, Y. High-resolution global contiguous sif of oco-2. Geophys. Res. Lett. 46, 1449–1458 (2019).ADS 

Google Scholar 
Koppa, A., Rains, D., Hulsman, P., Poyatos, R. & Miralles, D. G. A Deep learning-based hybrid model of global terrestrial evaporation (2022). https://doi.org/10.5281/zenodo.5886608.Koppa, A., Rains, D., Hulsman, P., Poyatos, R. & Miralles, D. G. A Deep learning-based hybrid model of global terrestrial evaporation (2022). https://doi.org/10.5281/zenodo.6343005. More

In this study we analysed how tree and arborescent palm species richness was related to aboveground carbon stock, commercially relevant timber stock, and commercially relevant NTFP abundance in tropical forests, and how these relationships were influenced by environmental stratification at different spatial scales. We found that species richness showed significant relationships with all three ecosystem services stock components, but its relationships were strongly influenced by variation across forest types and biogeographical strata. This is further explained below.Across the Guiana Shield, species richness showed a positive relationship with carbon stock and timber, but not with NTFP abundance. Although relationships only differed in significance among the biogeographical subregions, they differed in direction between terra firme forests and white sand forests. Species richness was positively related to carbon stock and timber stock in terra firme forests, whereas it was negatively related to NTFP abundance in white sand forests. The positive species-carbon relationship across forests of the Guiana Shield is in line with the effects described by hypotheses such as the ‘niche complementarity’ and ‘selection effect’10 and is in line with previous findings at regional spatial scales6,21. To our knowledge, the relationship between species richness and timber stock has not been previously analysed for tropical forests. Interestingly, the observed positive species-timber relationship in terra firme forests of the Guiana Shield contrasts with the negative species-timber relationship found for subtropical forests in both the U.S.A. and Spain20, although this may be explained by the difference in ecosystems. The non-significant species-NTFP abundance relationship across the Guiana Shield and the negative relationship within white sand forests seems to contradict previous findings. Steur et al.24 found a negative species-NTFP abundance relationship for tropical forests in Suriname. However, this negative relationship was found across multiple forest types, including flooded forests that had low species richness and high NTFP abundance. These flooded forests most likely influenced the species-NTFP abundance relationship across all forest types.In contrast to the relationship between species richness and carbon stock, no mechanism has been proposed for how species richness would influence commercial timber stock and NTFP abundance. Although our results suggest that species richness had a positive relationship with timber, the relationship was not found within multiple biogeographical subregions. For NTFP abundance, species richness did not contribute to explaining variation when variation across biogeographical subregions was accounted for (i.e. was included as an explanatory variable). We here tentatively propose that both commercial relevant timber stock and NTFP abundance are driven by variation in species floristic composition, rather than by species richness. For services such as commercial timber and NTFP provisioning, only a subset of all species is relevant (in this study, 9.4% of all morphospecies for timber and 3.8% for NTFPs), and such subsets are likely not random selections. For example, for Suriname, it was found that variation in commercially relevant NTFP abundance was driven by a particularly small selection of NTFP producing species with high abundances (referred to as ‘NTFP oligarchs’)24, and for commercial relevant timber stock, it is commonly known that selections tend to include more abundant than rare species. Additionally, as the relative abundance of species tends to vary across floristic regions in Amazonia, where, for example, certain species are dominant in particular forest types and biogeographical regions31,32, it can be expected that commercial timber stock and NTFP abundance are determined by floristic composition. In support, for NTFP abundance in Suriname tropical forests, Steur et al.24 found that floristic composition was a stronger predictor of NTFP abundance than species richness.Across all of Amazonia, species richness had a positive relationship with carbon stock, but only when variation among biogeographical regions was accounted for. The positive species-carbon relationship across Amazonia partly contrasts with previous findings at continental spatial scales11,13. When variation across climatic and/or edaphic variables was accounted for, Sullivan et al.13 found no significant species-carbon relationship across South-America, while Poorter et al.33 did find a positive relationship across Meso- and South-America. Here, we propose that accounting for differences among biogeographical regions can explain the previously found contrasts at continental spatial scales. In our dataset, for individual regions, we found either a positive relationship or a non-significant, but weakly positive, relationship between carbon stock and species richness (Fig. 2). However, when the data were aggregated across all regions, this resulted in a non-significant, and weakly negative, relationship. This reflects a known statistical phenomenon referred to as a ‘Simpson’s paradox’34, in which a relationship appears in multiple distinct groups but disappears or reverses when the groups are combined. Additional post-hoc tests of leaving one region out at a time showed that this pattern was not dependent of any particular biogeographical region. This is the first time that an analysis based on empirical data provides evidence for a Simpson’s paradox in species-ecosystem service relationships.It is likely that the observed differences in carbon stock across the biogeographical regions of Amazonia are influenced by multiple factors. For example, the biogeographical regions used in our analyses were recognised according to differences in substrate history, geological age and floristic composition, which could all contribute to variation in carbon stock. The substrate history and geological age of the biogeographical regions have been related to differences in soil fertility35, while multiple spatial gradients in floristic composition identified across the Amazon coincide with a spatial gradient in wood density28. However, further analysis is needed to obtain better insight into the relative contributions of these and other variables to explain the observed variation in carbon stock across the biogeographical regions. This requires data on multiple environmental variables, including floristic composition, climatic variables such as the length of the dry period, soil conditions, and intensity of disturbance.In our analyses, terra firme forests determined the relationship of species richness with the carbon stock, timber stock, and NTFP abundance across the datasets. Although this is most likely the effect of unequal sample sizes, with terra firme forests being the dominant forest type in terms of sample size (n = 130 vs. n = 21 for the Guiana Shield dataset; n = 257 vs. n = 26 for the Amazonia dataset), we expect that the observed relationships reflect the general pattern. Terra firme forests are the most dominant forest type in terms of geographical area32 and were representatively sampled. Regardless, the analyses per forest type had added value. The significant relationship between species richness and NTFP abundance in white sand forests across the Guiana Shield would otherwise have been overlooked.Due to the known scarcity of reliable and adequate information on which timber and NTFP species are being commercially traded36,37,38,39, we used a fixed set of timber and NTFP species to apply across the Guiana Shield plots. However, in reality, timber and NTFP species can be expected to vary according to socio-economic factors, such as culture, access, and harvest costs, which may change over space and time. Therefore, estimates of timber stock and NTFP abundance can be expected to differ across spatial gradients, and thus, their possible relationships with species richness cannot be easily generalised. To circumvent this, timber stock and NTFP abundance would have to be estimated on the basis of ‘flexible’ species selections that can change according to local socio-economic contexts. To this end, detailed information on both commercially relevant timber and NTFP species is urgently needed. Yet, for our study area, we did not observe major differences in selected species, and we included broad selections of species, which should make timber stock and NTFP abundance robust against small deviations in species selection. It must be noted that our approach of quantifying commercial relevant timber stock and NTFP abundance does not consider the value of timber and NTFPs for subsistence use. In addition, NTFPs can also be derived from other growth forms, such as lianas, shrubs and herbs. Last, because NTFP production data was not available we used NTFP abundance as a proxy for NTFP stock, following similar assessments of NTFP stock 24,40. A limitation of this approach is that each NTFP species individual has an equal contribution to NTFP stock, whereas it can be expected that large individuals may have a larger contribution than smaller individuals and that production volumes can differ for different types of NTFPs, for example barks vs. seeds.Our findings illustrate the importance of considering environmental stratification and spatial scale when analysing relationships between biodiversity and ecosystem services. First, environmental stratification can help detect relationships that are otherwise obscured by environmental heterogeneity. For example, although the association between species richness and carbon stock across Amazonia was relatively weak (explaining ~ 3% of total variation vs. ~ 15% in the Guiana Shield) and was obscured by variation in carbon stock across biogeographical strata, by using environmental stratification the positive relationship remained detectable. Second, environmental heterogeneity tends to vary with spatial scale; therefore, its importance needs to be checked according to spatial scale. For example, at the regional scale of the Guiana Shield, biogeographical subregions explained a moderate amount of variation in carbon stock (~ 20%), while at the spatial scale of Amazonia, biogeographical regions explained more than half of total variation in carbon stock (~ 55%). Such an increase and ultimate importance of variation across biogeographical strata might also explain the absence of a significant relationship between species richness and carbon stock across African and/or Asian tropical forests as reported by Sullivan et al.13.In our analyses, we found evidence of a positive relationship between species richness and carbon stock across and within Amazonia. This supports the notion that win–win scenarios are possible in conservation approaches, where, for example, REDD+ can be expected to help conserve tropical forests that contain large amounts of carbon stock and high concentrations of species9. However, we conclude that species richness is not always a strong predictor of biomass-based ecosystem services. In our analyses, NTFP abundance was not driven by species richness, and we ultimately expect the same for timber stock. We expect that differences in floristic composition, linked to differences across forest types and biogeographical strata, will be more relevant than species richness in explaining variation in timber stock and NTFP abundance. This would mean that conserving timber and NTFP related ecosystem services requires the development of additional region-specific strategies that account for differences in floristic composition. For example, areas with high concentrations of timber or NTFPs could be considered in the designation of multiple use protected areas41, such as the extractive reserves in Brazil, or be included as ‘high conservation value areas’ (HCVAs) in sustainable forest management certification42. More