www.nature.com/scientificreports OPEN Unraveling Amazon tree community assembly using Maximum Information Entropy: a quantitative analysis of tropical forest ecology Edwin Pos 1,2*, Luiz de Souza Coelho 3, Diogenes de Andrade Lima Filho 3, Rafael P. Salomão 4,5, Iêda Leão Amaral 3, Francisca Dionízia de Almeida Matos 3, Carolina V. Castilho 6, Oliver L. Phillips 7, Juan Ernesto Guevara 8,9, Marcelo de Jesus Veiga Carim 10, Dairon Cárdenas López 11, William E. Magnusson 12, Florian Wittmann 13,14, Mariana Victória Irume 3, Maria Pires Martins 3, Daniel Sabatier 15, José Renan da Silva Guimarães 10, Jean‑François Molino 15, Olaf S. Bánki 2, Maria Teresa Fernandez Piedade 16, Nigel C. A. Pitman 17, Abel Monteagudo Mendoza 18, José Ferreira Ramos 3, Joseph E. Hawes 19, Everton José Almeida 20, Luciane Ferreira Barbosa 20, Larissa Cavalheiro 20, Márcia Cléia Vilela dos Santos 20, Bruno Garcia Luize 21, Evlyn Márcia Moraes de Leão Novo 22, Percy Núñez Vargas 23, Thiago Sanna Freire Silva 24, Eduardo Martins Venticinque 25, Angelo Gilberto Manzatto 26, Neidiane Farias Costa Reis 27, John Terborgh 28,29, Katia Regina Casula 27, Euridice N. Honorio Coronado 7,30, Juan Carlos Montero 3,31, Beatriz S. Marimon 32, Ben Hur Marimon‑Junior 32, Ted R. Feldpausch 7,33, Alvaro Duque 34, Chris Baraloto 35, Nicolás Castaño Arboleda 11, Julien Engel 15,35, Pascal Petronelli 36, Charles Eugene Zartman 3, Timothy J. Killeen 37, Rodolfo Vasquez 18, Bonifacio Mostacedo 38, Rafael L. Assis 39, Jochen Schöngart 16, Hernán Castellanos 40, Marcelo Brilhante de Medeiros 41, Marcelo Fragomeni Simon 41, Ana Andrade 42, José Luís Camargo 42, Layon O. Demarchi 16, William F. Laurance 29, Susan G. W. Laurance 29, Emanuelle de Sousa Farias 43,44, Maria Aparecida Lopes 45, José Leonardo Lima Magalhães 46,47, Henrique Eduardo Mendonça Nascimento 3, Helder Lima de Queiroz 48, Gerardo A. C. Aymard 49, Roel Brienen 7, Juan David Cardenas Revilla 3, Flávia R. C. Costa 3, Adriano Quaresma 16, Ima Célia Guimarães Vieira 5, Bruno Barçante Ladvocat Cintra 50, Pablo R. Stevenson 51, Yuri Oliveira Feitosa 52, Joost F. Duivenvoorden 53, Hugo F. Mogollón 54, Leandro Valle Ferreira 5, James A. Comiskey 55,56, Freddie Draper 35,57, José Julio de Toledo 58, Gabriel Damasco 59, Nállarett Dávila 60, Roosevelt García‑Villacorta 61,62, Aline Lopes 16,63, Alberto Vicentini 12, Janaína Costa Noronha 64, Flávia Rodrigues Barbosa 64, Rainiellen de Sá Carpanedo 64, Thaise Emilio 12,65, Carolina Levis 66,67, Domingos de Jesus Rodrigues 64, Juliana Schietti 3, Priscila Souza 3, Alfonso Alonso 56, Francisco Dallmeier 56, Vitor H. F. Gomes 68,69, Jon Lloyd 70, David Neill 71, Daniel Praia Portela de Aguiar 16, Alejandro Araujo‑Murakami 72, Luzmila Arroyo 72, Fernanda Antunes Carvalho 12,73, Fernanda Coelho de Souza 7,12, Dário Dantas do Amaral 5, Kenneth J. Feeley 74,75, Rogerio Gribel 3, Marcelo Petratti Pansonato 3,76, Jos Barlow 77, Erika Berenguer 77,78, Joice Ferreira 47, Paul V. A. Fine 59, Marcelino Carneiro Guedes 79, Eliana M. Jimenez 80, Juan Carlos Licona 31, Maria Cristina Peñuela Mora 81, Carlos A. Peres 82, Boris Eduardo Villa Zegarra 83, Carlos Cerón 84, Terry W. Henkel 85, Paul Maas 2, Marcos Silveira 86, Juliana Stropp 87, Raquel Thomas‑Caesar 88, Tim R. Baker 7, Doug Daly 89, Kyle G. Dexter 90,91, John Ethan Householder 13, Isau Huamantupa‑Chuquimaco 23, Toby Pennington 33,91, Marcos Ríos Paredes 92, Alfredo Fuentes 93,94, José Luis Marcelo Pena 95, Miles R. Silman 96, J. Sebastián Tello 94, Jerome Chave 97, Fernando Cornejo Valverde 98, Scientific Reports | (2023) 13:2859 | https://doi.org/10.1038/s41598-023-28132-y 1 Vol.:(0123456789) www.nature.com/scientificreports/ Anthony Di Fiore 99, Renato Richard Hilário 58, Juan Fernando Phillips 100, Gonzalo Rivas‑Torres 101,102, Tinde R. van Andel 2,103, Patricio von Hildebrand 104, Edelcilio Marques Barbosa 3, Luiz Carlos de Matos Bonates 3, Hilda Paulette Dávila Doza 92, Émile Fonty 15,105, Ricardo Zárate Gómez 106, Therany Gonzales 107, George Pepe Gallardo Gonzales 92, Jean‑Louis Guillaumet 108,145, Bruce Hoffman 109, André Braga Junqueira 110, Yadvinder Malhi 111, Ires Paula de Andrade Miranda 3, Linder Felipe Mozombite Pinto 92, Adriana Prieto 112, Agustín Rudas 112, Ademir R. Ruschel 47, Natalino Silva 113, César I. A. Vela 114, Vincent Antoine Vos 115, Egleé L. Zent 116, Stanford Zent 116, Bianca Weiss Albuquerque 16, Angela Cano 51,117, Diego F. Correa 51,118, Janaina Barbosa Pedrosa Costa 79, Bernardo Monteiro Flores 119, Milena Holmgren 120, Marcelo Trindade Nascimento 121, Alexandre A. Oliveira 76, Hirma Ramirez‑Angulo 122, Maira Rocha 16, Veridiana Vizoni Scudeller 123, Rodrigo Sierra 124, Milton Tirado 124, Maria Natalia Umaña 125, Geertje van der Heijden 126, Emilio Vilanova Torre 122,127, Corine Vriesendorp 17, Ophelia Wang 128, Kenneth R. Young 129, Manuel Augusto Ahuite Reategui 130, Cláudia Baider 76,131, Henrik Balslev 132, Sasha Cárdenas 51, Luisa Fernanda Casas 51, William Farfan‑Rios 23,94,133, Cid Ferreira 3, Reynaldo Linares‑Palomino 56, Casimiro Mendoza 134,135, Italo Mesones 59, Armando Torres‑Lezama 122, Ligia Estela Urrego Giraldo 34, Daniel Villarroel 72, Roderick Zagt 136, Miguel N. Alexiades 137, Karina Garcia‑Cabrera 96, Lionel Hernandez 40, William Milliken 65, Walter Palacios Cuenca 138, Susamar Pansini 27, Daniela Pauletto 139, Freddy Ramirez Arevalo 140, Adeilza Felipe Sampaio 27, Elvis H. Valderrama Sandoval 140,141, Luis Valenzuela Gamarra 18, Gerhard Boenisch 142, Jens Kattge 143, Nathan Kraft 144, Aurora Levesley 7, Karina Melgaço 7, Georgia Pickavance 7, Lourens Poorter 67 & Hans ter Steege 2 In a time of rapid global change, the question of what determines patterns in species abundance distribution remains a priority for understanding the complex dynamics of ecosystems. The constrained maximization of information entropy provides a framework for the understanding of such complex systems dynamics by a quantitative analysis of important constraints via predictions using least biased probability distributions. We apply it to over two thousand hectares of Amazonian tree inventories across seven forest types and thirteen functional traits, representing major global axes of plant strategies. Results show that constraints formed by regional relative abundances of genera explain eight times more of local relative abundances than constraints based on directional selection for specific functional traits, although the latter does show clear signals of environmental dependency. These results provide a quantitative insight by inference from large-scale data using cross-disciplinary methods, furthering our understanding of ecological dynamics. 1 Quantitative Biodiversity Dynamics, Ecology and Biodiversity, Utrecht University Botanic Gardens, Utrecht University, Padualaan 8, Utrecht 3584 CH, The Netherlands. 2Naturalis Biodiversity Center, PO Box 9517, Leiden 2300 RA, The Netherlands. 3Coordenação de Biodiversidade, Instituto Nacional de Pesquisas da Amazônia - INPA, Av. André Araújo, 2936, Petrópolis, Manaus, AM 69067‑375, Brazil. 4Programa Professor Visitante Nacional Sênior Na Amazônia ‑ CAPES, Universidade Federal Rural da Amazônia, Av. Perimetral, s/n, Belém, PA, Brazil. 5Coordenação de Botânica, Museu Paraense Emílio Goeldi, Av. Magalhães Barata 376, C.P. 399, Belém, PA 66040‑170, Brazil. 6EMBRAPA – Centro de Pesquisa Agroflorestal de Roraima, BR 174, km 8 – Distrito Industrial, Boa Vista, RR 69301‑970, Brazil. 7School of Geography, University of Leeds, Woodhouse Lane, Leeds LS2 9JT, UK. 8Grupo de Investigación en Biodiversidad, Medio Ambiente y Salud-BIOMAS, Universidad de Las Américas, Campus Queri, Quito, Ecuador. 9Keller Science Action Center, The Field Museum, 1400 S. Lake Shore Drive, Chicago, IL 60605‑2496, USA. 10Departamento de Botânica, Instituto de Pesquisas Científicas e Tecnológicas do Amapá - IEPA, Rodovia JK, Km 10, Campus Do IEPA da Fazendinha, Amapá 68901‑025, Brazil. 11Herbario Amazónico Colombiano, Instituto SINCHI, Calle 20 No 5‑44, Bogotá, DC, Colombia. 12Coordenação de Pesquisas em Ecologia, Instituto Nacional de Pesquisas da Amazônia - INPA, Av. André Araújo, 2936, Petrópolis, Manaus, AM 69067‑375, Brazil. 13Department of Wetland Ecology, Institute of Geography and Geoecology, Karlsruhe Institute of Technology - KIT, Josefstr.1, 76437 Rastatt, Germany. 14Biogeochemistry, Max Planck Institute for Chemistry, Hahn‑Meitner Weg 1, 55128 Mainz, Germany. 15AMAP, IRD, Cirad, CNRS, INRA, Université de Montpellier, 34398 Montpellier, France. 16Coordenação de Dinâmica Ambiental, Instituto Nacional de Pesquisas da Amazônia - INPA, Av. André Araújo, 2936, Petrópolis, Manaus, AM 69067‑375, Brazil. 17Science and Education, The Field Museum, 1400 S. Lake Shore Drive, Chicago, IL 60605‑2496, USA. 18Jardín Botánico de Missouri, Oxapampa, Pasco, Peru. 19Applied Ecology Research Group, School of Life Sciences, Anglia Ruskin University, East Road, Cambridge CB1 1PT, UK. 20ICNHS, Universidade Federal de Mato Grosso, Av. Alexandre Ferronato, 1200, Sinop, MT 78557‑267, Brazil. 21Departamento de Ecologia, Universidade Estadual Paulista - UNESP – Instituto de Biociências – IB, Av. 24 A, 1515, Bela Vista, Rio Claro, SP 13506‑900, Brazil. 22Divisao de Sensoriamento Remoto – DSR, Instituto Nacional de Pesquisas Espaciais – INPE, Av. Dos Astronautas, 1758, Jardim da Granja, São José Dos Scientific Reports | (2023) 13:2859 | https://doi.org/10.1038/s41598-023-28132-y 2 Vol:.(1234567890) www.nature.com/scientificreports/ Campos, SP 12227‑010, Brazil. 23Herbario Vargas, Universidad Nacional de San Antonio Abad del Cusco, Avenida de La Cultura, Nro 733, Cusco, Cuzco, Peru. 24Biological and Environmental Sciences, University of Stirling, Stirling FK9 4LA, UK. 25Centro de Biociências, Departamento de Ecologia, Universidade Federal do Rio Grande do Norte, Av. Senador Salgado Filho, 3000, Natal, RN 59072‑970, Brazil. 26Departamento de Biologia, Universidade Federal de Rondônia, Rodovia BR 364 s/n Km 9, 5 ‑ Sentido Acre, Unir, Porto Velho, RO 76.824‑027, Brazil. 27Programa de Pós‑ Graduação em Biodiversidade e Biotecnologia PPG‑ Bionorte, Universidade Federal de Rondônia, Campus Porto Velho Km 9, 5 Bairro Rural, Porto Velho, RO 76.824‑027, Brazil. 28Department of Biology and Florida Museum of Natural History, University of Florida, Gainesville, FL 32611, USA. 29Centre for Tropical Environmental and Sustainability Science and College of Science and Engineering, James Cook University, Cairns, QLD 4870, Australia. 30Instituto de Investigaciones de la Amazonía Peruana (IIAP), Av. A. Quiñones Km 2,5, Iquitos, Loreto 784, Peru. 31Instituto Boliviano de Investigacion Forestal, Av. 6 de Agosto #28, Km. 14, Doble via La Guardia, 6204 Santa Cruz, Santa Cruz, Casilla, Bolivia. 32Programa de Pós-Graduação em Ecologia e Conservação, Universidade do Estado de Mato Grosso, Nova Xavantina, MT, Brazil. 33Geography, College of Life and Environmental Sciences, University of Exeter, Rennes Drive, Exeter EX4 4RJ, UK. 34Departamento de Ciencias Forestales, Universidad Nacional de Colombia, Calle 64 X Cra 65, 1027 Medellín, Antioquia, Colombia. 35International Center for Tropical Botany (ICTB) Department of Biological Sciences, Florida International University, 11200 SW 8Th Street, OE 243, Miami, FL 33199, USA. 36Cirad UMR Ecofog, AgrosParisTech, CNRS, INRA, Univ Guyane, Campus Agronomique, 97379 Kourou Cedex, France. 37Agteca-Amazonica, Santa Cruz, Bolivia. 38Facultad de Ciencias Agrícolas, Universidad Autónoma Gabriel René Moreno, Santa Cruz, Santa Cruz, Bolivia. 39Natural History Museum, University of Oslo, Postboks 1172, 0318 Oslo, Norway. 40Centro de Investigaciones Ecológicas de Guayana, Universidad Nacional Experimental de Guayana, Calle Chile, Urbaniz Chilemex, Puerto Ordaz, Bolivar, Venezuela. 41Prédio da Botânica e Ecologia, Embrapa Recursos Genéticos e Biotecnologia, Parque Estação Biológica, Av. W5 Norte, Brasilia, DF 70770‑917, Brazil. 42Projeto Dinâmica Biológica de Fragmentos Florestais, Instituto Nacional de Pesquisas da Amazônia - INPA, Av. André Araújo 2936, Petrópolis, Manaus, AM 69067‑375, Brazil. 43Laboratório de Ecologia de Doenças Transmissíveis da Amazônia (EDTA), Instituto Leônidas e Maria Deane, Fiocruz, Rua Terezina, 476, Adrianópolis, Manaus, AM 69060‑001, Brazil. 44Programa de Pós-Graduação em Biodiversidade e Saúde, Instituto Oswaldo Cruz - IOC/FIOCRUZ, Pav. Arthur Neiva – Térreo, Av. Brasil, 4365 – Manguinhos, Rio de Janeiro, RJ 21040‑360, Brazil. 45Instituto de Ciências Biológicas, Universidade Federal do Pará, Av. Augusto Corrêa 01, Belém, PA 66075‑110, Brazil. 46Programa de Pós‑Graduação em Ecologia, Universidade Federal do Pará, Av. Augusto Corrêa 01, Belém, PA 66075‑110, Brazil. 47Embrapa Amazônia Oriental, Trav. Dr. Enéas Pinheiro S/nº, Belém, PA 66095‑100, Brazil. 48Diretoria Técnico-Científica, Instituto de Desenvolvimento Sustentável Mamirauá, Estrada do Bexiga, 2584, Tefé, AM 69470‑000, Brazil. 49Programa de Ciencias del Agro y el Mar, Herbario Universitario (PORT), UNELLEZ-Guanare, Guanare, Portuguesa 3350, Venezuela. 50Instituto de Biociências – Department of Botanica, Universidade de Sao Paulo - USP, Rua do Matão 277, Cidade Universitária, São Paulo, SP 05508‑090, Brazil. 51Laboratorio de Ecología de Bosques Tropicales y Primatología, Universidad de los Andes, Carrera 1 # 18a‑ 10, 111711 Bogotá, DC, Colombia. 52Programa de Pós‑Graduação Em Biologia (Botânica), Instituto Nacional de Pesquisas da Amazônia - INPA, Av. André Araújo, 2936, Petrópolis, Manaus, AM 69067‑375, Brazil. 53Institute of Biodiversity and Ecosystem Dynamics, University of Amsterdam, Sciencepark 904, Amsterdam 1098 XH, The Netherlands. 54Endangered Species Coalition, 8530 Geren Rd., Silver Spring, MD 20901, USA. 55Inventory and Monitoring Program, National Park Service, 120 Chatham Lane, Fredericksburg, VA 22405, USA. 56Center for Conservation and Sustainability, Smithsonian Conservation Biology Institute, 1100 Jefferson Dr. SW, Suite 3123, Washington, DC 20560‑0705, USA. 57Department of Global Ecology, Carnegie Institution for Science, 260 Panama St., Stanford, CA 94305, USA. 58Universidade Federal do Amapá, Ciências Ambientais, Rod. Juscelino Kubitschek km2, Macapá, AP 68902‑280, Brazil. 59Department of Integrative Biology, University of California, Berkeley, CA 94720‑3140, USA. 60Biologia Vegetal, Universidade Estadual de Campinas, Caixa Postal 6109, Campinas, SP 13.083‑970, Brazil. 61Department of Ecology and Evolutionary Biology, Cornell University, Corson Hall, 215 Tower Road, Ithaca, NY 14850, USA. 62Peruvian Center for Biodiversity and Conservation (PCBC), Iquitos, Peru. 63Department of Ecology, University of Brasilia, Brasilia, DF 70904‑970, Brazil. 64ICNHS, Federal University of Mato Grosso, Av. Alexandre Ferronato 1200, Setor Industrial, Sinop, MT 78.557‑267, Brazil. 65Natural Capital and Plant Health, Royal Botanic Gardens, Kew, Richmond TW9 3AB, Surrey, UK. 66Programa de Pós‑Graduação em Ecologia, Instituto Nacional de Pesquisas da Amazônia - INPA, Av. André Araújo, 2936, Petrópolis, Manaus, AM 69067‑375, Brazil. 67Forest Ecology and Forest Management Group, Wageningen University and Research, Droevendaalsesteeg 3, P.O. Box 47, Wageningen 6700 AA, The Netherlands. 68Escola de Negócios Tecnologia e Inovação, Centro Universitário do Pará, Belém, PA, Brazil. 69Universidade Federal do Pará, Rua Augusto Corrêa 01, Belém, PA 66075‑110, Brazil. 70Faculty of Natural Sciences, Department of Life Sciences, Imperial College London, South Kensington Campus, Silwood ParkLondon SW7 2AZ, UK. 71Ecosistemas, Biodiversidad y Conservación de Especies, Universidad Estatal Amazónica, Km. 2 1/2 Vía a Tena (Paso Lateral), Puyo, Pastaza, Ecuador. 72Museo de Historia Natural Noel Kempff Mercado, Universidad Autónoma Gabriel Rene Moreno, Avenida Irala 565 Casilla Post Al 2489, Santa Cruz, Santa Cruz, Bolivia. 73Departamento de Genética, Ecologia e Evolução, Universidade Federal de Minas Gerais, Instituto de Ciências Biológicas, Av. Antônio Carlos, 6627 Pampulha, Belo Horizonte, MG 31270‑901, Brazil. 74Department of Biology, University of Miami, Coral Gables, FL 33146, USA. 75Fairchild Tropical Botanic Garden, Coral Gables, FL 33156, USA. 76Instituto de Biociências ‑ Dept. Ecologia, Universidade de Sao Paulo - USP, Rua do Matão, Trav. 14, No. 321, Cidade Universitária, São Paulo, SP 05508‑090, Brazil. 77Lancaster Environment Centre, Lancaster University, Lancaster LA1 4YQ, Lancashire, UK. 78Environmental Change Institute, University of Oxford, Oxford OX1 3QY, Oxfordshire, UK. 79Empresa Brasileira de Pesquisa Agropecuária, Embrapa Amapá, Rod. Juscelino Kubitschek Km 5, Macapá, Amapá 68903‑419, Brazil. 80Grupo de Ecología y Conservación de Fauna y Flora Silvestre, Instituto Amazónico de Investigaciones Imani, Universidad Nacional de Colombia Sede Amazonia, Leticia, Amazonas, Colombia. 81Universidad Regional Amazónica IKIAM, Km 7 Via Muyuna, Tena, Napo, Scientific Reports | (2023) 13:2859 | https://doi.org/10.1038/s41598-023-28132-y 3 Vol.:(0123456789) www.nature.com/scientificreports/ Ecuador. 82School of Environmental Sciences, University of East Anglia, Norwich NR4 7TJ, UK. 83Direccíon de Evaluación Forestal y de Fauna Silvestre, Av. Javier Praod Oeste 693, Magdalena del Mar, Peru. 84Escuela de Biología Herbario Alfredo Paredes, Universidad Central, Ap. Postal 17.01.2177, Quito, Pichincha, Ecuador. 85Department of Biological Sciences, Humboldt State University, 1 Harpst Street, Arcata, CA 95521, USA. 86Museu Universitário / Centro de Ciências Biológicas e da Natureza / Laboratório de Botânica e Ecologia Vegetal, Universidade Federal do Acre, Rio Branco, AC 69915‑559, Brazil. 87Institute of Biological and Health Sciences, Federal University of Alagoas, Av. Lourival Melo Mota, s/n, Tabuleiro do Martins, Maceio, AL 57072‑970, Brazil. 88Iwokrama International Centre for Rain Forest Conservation and Development, Georgetown, Guyana. 89New York Botanical Garden, 2900 Southern Blvd, Bronx, New York, NY 10458‑5126, USA. 90School of Geosciences, University of Edinburgh, 201 Crew Building, King’s Buildings, Edinburgh EH9 3JN, UK. 91Tropical Diversity Section, Royal Botanic Garden Edinburgh, 20a Inverleith Row, Edinburgh EH3 5LR, Scotland, UK. 92Servicios de Biodiversidad EIRL, Jr. Independencia 405, Iquitos, Loreto 784, Peru. 93Herbario Nacional de Bolivia, Universitario UMSA, Casilla 10077 Correo Central, La Paz, La Paz, Bolivia. 94Center for Conservation and Sustainable Development, Missouri Botanical Garden, P.O. Box 299, St. Louis, MO 63166‑0299, USA. 95Universidad Nacional de Jaén, Carretera Jaén San Ignacio Km 23, Jaén, Cajamarca 06801, Peru. 96Biology Department and Center for Energy, Environment and Sustainability, Wake Forest University, 1834 Wake Forest Rd, Winston Salem, NC 27106, USA. 97Laboratoire Evolution et Diversité Biologique, CNRS and Université Paul Sabatier, UMR 5174 EDB, 31000 Toulouse, France. 98Andes to Amazon Biodiversity Program, Madre de Dios, Madre de Dios, Peru. 99Department of Anthropology, University of Texas at Austin, SAC 5.150, 2201 Speedway Stop C3200, Austin, TX 78712, USA. 100Fundación Puerto Rastrojo, Cra 10 No. 24‑76 Oficina 1201, Bogotá, DC, Colombia. 101Colegio de Ciencias Biológicas y Ambientales‑COCIBA and Galapagos Institute for the Arts and Sciences‑GAIAS, Universidad San Francisco de Quito-USFQ, Quito, Pichincha, Ecuador. 102Department of Wildlife Ecology and Conservation, University of Florida, 110 Newins‑Ziegler Hall, Gainesville, FL 32611, USA. 103Biosystematics Group, Wageningen University, Droevendaalsesteeg 1, Wageningen 6708 PB, The Netherlands. 104Fundación Estación de Biología, Cra 10 No. 24‑76 Oficina, 1201 Bogotá, DC, Colombia. 105Direction Régionale de la Guyane, ONF, Cayenne 97300, French Guiana. 106PROTERRA, Instituto de Investigaciones de la Amazonía Peruana (IIAP), Av. A. Quiñones Km 2,5, Iquitos, Loreto 784, Peru. 107ACEER Foundation, Jirón Cusco N° 370, Puerto Maldonado, Madre de Dios, Peru. 108Departement EV, Muséum National d’histoire Naturelle de Paris, 16 Rue Buffon, Paris 75005, France. 109Amazon Conservation Team, Doekhieweg Oost #24, Paramaribo, Suriname. 110Institut de Ciència i Tecnologia Ambientals, Universitat Autònoma de Barcelona, 08193 Bellaterra, Barcelona, Spain. 111Environmental Change Institute, Dyson Perrins Building, Oxford University Centre for the Environment, South Parks Road, Oxford OX1 3QY, England, UK. 112Instituto de Ciencias Naturales, Universidad Nacional de Colombia, 7945 Apartado, Bogotá, DC, Colombia. 113Instituto de Ciência Agrárias, Universidade Federal Rural da Amazônia, Av. Presidente Tancredo Neves 2501, Belém, PA 66.077‑830, Brazil. 114Escuela Profesional de Ingeniería Forestal, Universidad Nacional de San Antonio Abad del Cusco, Jirón San Martín 451, Puerto Maldonado, Madre de Dios, Peru. 115Universidad Autónoma del Beni José Ballivián, Campus Universitario Final, Av. Ejercito, Riberalta, Beni, Bolivia. 116Laboratory of Human Ecology, Instituto Venezolano de Investigaciones Científicas - IVIC, Ado 20632, Caracas 1020A, DC, Venezuela. 117Cambridge University Botanic Garden, 1 Brookside., Cambridge CB2 1JE, UK. 118School of Agriculture and Food Sciences ‑ ARC Centre of Excellence for Environmental Decisions CEED, The University of Queensland, St. Lucia, QLD 4072, Australia. 119Plant Biology Department, Rua Monteiro Lobato, University of Campinas, 255, Cidade Universitária Zeferino Vaz, Barão Geraldo, Campinas, São Paulo CEP 13083‑862, Brazil. 120Resource Ecology Group, Wageningen University and Research, Droevendaalsesteeg 3a, Lumen, Building Number 100, Wageningen, Gelderland 6708 PB, The Netherlands. 121Laboratório de Ciências Ambientais, Universidade Estadual do Norte Fluminense, Av. Alberto Lamego 2000, Campos dos, Goyatacazes, RJ 28013‑620, Brazil. 122Instituto de Investigaciones Para el Desarrollo Forestal (INDEFOR), Universidad de los Andes, Conjunto Forestal, Mérida, Mérida 5101, Venezuela. 123Departamento de Biologia, Universidade Federal do Amazonas - UFAM – Instituto de Ciências Biológicas – ICB1, Av General Rodrigo Octavio 6200, Manaus, AM 69080‑900, Brazil. 124GeoIS, el Día 369 y el Telégrafo, 3° Piso, Quito, Pichincha, Ecuador. 125Department of Ecology and Evolutionary Biology, University of Michigan, Ann Arbor, MI 48109, USA. 126University of Nottingham, University Park, Nottingham NG7 2RD, UK. 127School of Environmental and Forest Sciences, University of Washington, Seattle, WA 98195‑2100, USA. 128Environmental Science and Policy, Northern Arizona University, Flagstaff, AZ 86011, USA. 129Geography and the Environment, University of Texas at Austin, 305 E. 23Rd Street, CLA Building, Austin, TX 78712, USA. 130Medio Ambiente, PLUSPRETOL, Iquitos, Loreto, Peru. 131The Mauritius Herbarium, Agricultural Services, Ministry of Agro-Industry and Food Security, Reduit 80835, Mauritius. 132Department of Bioscience, Aarhus University, Building 1540 Ny Munkegade, 8000 Aarhus C, Aarhus, Denmark. 133Living Earth Collaborative, Washington University in Saint Louis, St. Louis, MO 63130, USA. 134Escuela de Ciencias Forestales (ESFOR), Universidad Mayor de San Simon (UMSS), Sacta, Cochabamba, Bolivia. 135FOMABO, Manejo Forestal en Las Tierras Tropicales de Bolivia, Sacta, Cochabamba, Bolivia. 136Tropenbos International, Lawickse Allee 11, PO Box 232, Wageningen 6700 AE, The Netherlands. 137School of Anthropology and Conservation, University of Kent, Marlowe Building, Canterbury, Kent CT2 7NR, UK. 138Herbario Nacional del Ecuador, Universidad Técnica del Norte, Quito, Pichincha, Ecuador. 139Instituto de Biodiversidade e Floresta, Universidade Federal do Oeste do Pará, Rua Vera Paz, Campus Tapajós, Santarém, PA 68015‑110, Brazil. 140Facultad de Biologia, Universidad Nacional de la Amazonia Peruana, Pevas 5Ta Cdra, Iquitos, Loreto, Peru. 141Department of Biology, University of Missouri, St. Louis, MO 63121, USA. 142Department of Biogeochemical Integration, Max-Planck-Institute for Biogeochemistry, P.O. Box 10 01 64, 07701 Jena, Germany. 143Functional Biogeography, Max-Planck-Institute for Biogeochemistry, P.O. Box 10 01 64, 07701 Jena, Germany. 144Department of Ecology and Evolutionary Biology, UCLA, 621 Charles E. Young Drive South, Box 951606, Los Angeles, CA 90095, USA. 145Jean-Louis Guillaumet is deceased. *email: e.t.pos@uu.nl Scientific Reports | (2023) 13:2859 | https://doi.org/10.1038/s41598-023-28132-y 4 Vol:.(1234567890) www.nature.com/scientificreports/ Drivers of species distributions and their predictions have been a long-standing search in ecology, with approaches varying from deterministic to neutral (i.e. stochastic) and almost everything in between (e.g. near- neutral, continuum or emergent-neutral1,2). Most models are based on prior assumptions of processes that drive community dynamics. The Maximum Entropy Formalism (hereafter called MEF) makes no such, potentially unjustified, a-priori assumptions in generating predictions of species abundance distributions, as such it is a use- ful construct to infer processes driving community dynamics given the constraints imposed by prior knowledge (e.g. functional traits or summed regional abundances)3. Quantifying the relative importance of these distinct constraints can thus provide additional answers to understand the complexity of community dynamics (see Supporting Materials SM: boxes S1–S3). This is especially so because, although many different tests are available that link variation in taxon abundances to (1) trait variation, (2) taxon turnover between habitats or environ- ments and (3) the distance decay of similarities between samples, none quantify the importance of these relative to each other. The MEF as applied here, however, is capable of and designed to do exactly this by decomposing variation to separate information explained by each of these aspects in a four-step model (Fig. 1 and Box S2). Its application to an unprecedented large tree inventory database on genus level taxonomy consisting of > 2,000 1-ha plots distributed over ­Amazonia4 and a genus trait database of 13 key functional traits representing global axes of plant ­strategies5 allows us to advance the study of Amazonian tree community dynamics from a new cross-disciplinary perspective. Results Principles from information t­ heory6–8 can be used in an ecological setting to predict the most likely abundance state for each taxon while simultaneously maximizing entropy based on constraints. Maximization of entropy allows quantifying the information yield for each constraint and therefor identifies which constraints reduce entropy the most. Here we specifically use Shipley’s mathematical framework (CATS) for the MEF calculations, similar to earlier s­ tudies9–11. Predictive power of the four‑step model. Using a uniform prior and CWM values (Community Weighted Means) as constraints accounted for 23% on average of total deviance between observed and pre- dicted relative abundances (measured by ­R2KL values, see Box S2 Eq. 5). Filtered by forest type this was 34% for podzol forests, várzea 25%, igapó 23%, swamp forests 34%, 21% and 24% for Guyana Shield and Pebas terra firme respectively and 20% for Brazilian Shield terra firme forests (see Table S1 for detailed decomposition). Using observed metacommunity relative abundances as prior regardless of functional traits accounted for on average 56% for the combined dataset with for all forest types between 51 and 55%, except for the Guyana Shield terra firme with 62%. The hybrid model (including both traits as constraints and the metacommunity prior) performed slightly better for the combined dataset (average 60%) with a minimum of 57% for swamp and várzea forests and a maximum of 66% for Guyana Shield terra firme forests. To compensate for spurious relation- ships between regional abundances and local trait constraints, regardless of selection, explanatory power was regarded relative to model bias yielding the pure trait and metacommunity effects (Box S3, Fig. 2 and Table S1). This lowered the proportion of information accounted for and yielded average pure metacommunity effects of 40% for the overall dataset ranging between 26 and 45% for each forest type separately with pure trait effects explaining only 5% of information for the combined dataset on average with for each forest type between 3 and 9%. Although the latter was lowered substantially, the explanatory power did appear to be strongly dependent on forest type. The online supplementary material provides additional results relating to the predictive power of each model as well as the spatial gradient of the pure trait and metacommunity effects (Figs. S2–S3). Direction and strength of selection of trait‑based constraints. Each trait showed significant dif- ferences in lambda when compared between forest types (Fig. S1, see methods for a definition of lambda). Scat- terplots of CWM trait values versus lambda show that, in general, higher lambda values correspond with higher CWM trait values (Fig. S7). A number of functional traits associated with low nutrient conditions (e.g. ectomy- corrhiza) and life history strategies suited for protection against herbivores (e.g. latex, resin and high leaf C con- tent) were clearly positively associated with abundance in nutrient poor environments (podzols), indicated by the positive lambda values. In contrast, having fleshy fruits and high leaf N and P content were clearly negatively associated with abundance on these soils. Nodulation was also negatively associated with abundance on poor soils. The ability to accumulate aluminium was positively associated with abundance on those soils commonly associated with higher aluminium content such as igapó (strong positive effects). In contrast, it was strongly negatively associated with abundance on other soils, with negative lambda values for podzol and várzea forests. Traits such as SLA or winged fruits also showed patterns dependent on forest type, although less pronounced. Effect of regional metacommunity prior. There was a remarkable similar mean 22% decrease of the information explained purely by the metacommunity prior for each forest type (Fig. S3). For the separate forest types, although the initial pure metacommunity effect varied, the decline appeared remarkably similar with a mean 25% decrease in pure metacommunity effect for podzol, 23% for várzea, 26% for igapó and 27% for swamp forests with terra firme forests having a smaller decline of approximately 21%, averaged over the three subre- gions. It should be noted there is an obvious risk that when sampling size is increased, this also includes more environmental heterogeneity as samples are coming from a variety of localities potentially leading to changing composition. If this were the case, however, the regional prior (­ qi from Fig. 1 and Box S2) would also change, as taxa might be abundant in some places but rare or absent in others. As the metacommunity effect is the explained information that remains relative to any trait effects (i.e. information unique to the neutral prior) and the pure trait effects are the explained information remaining after correcting for pure metacommunity effects Scientific Reports | (2023) 13:2859 | https://doi.org/10.1038/s41598-023-28132-y 5 Vol.:(0123456789) www.nature.com/scientificreports/ Figure 1.  Schematic depiction of the MEF procedure. Left panel shows a genus abundances per site and a functional trait matrix per genus, bottom half outlines calculations. Middle and right panel show different scenarios of neutral and deterministic dynamics under infinite or limited migration. Figure was custom made using Adobe Illustrator (Adobe Inc., 2019. Adobe Illustrator). (Box S3) this effect should then be accompanied by an increase in pure trait effect for each sample. This was not observed, not even within the different forest types. Instead, the trait effect gradually went up and then remained constant (Fig. S4). Discussion The MEF emerges from a well-founded theoretical and empirical body of ecology and evolutionary biology, regarding natural selection, migration and population dynamics. From an ecological point of view, it can be used to quantify the relative association between directional or stabilizing selection for functional traits versus the importance of relative regional abundance regardless of these traits by imposing these as constraints. Our results show that pure trait effects, on average, explained only 5% of the information when all forest types were taken together whereas the pure metacommunity effect, however, explained eight times more taken all forest types together (40%). Greater trait dissimilarity was positively associated with higher pure trait effects, indicating trait-based selection, although the assumed influence of dispersal regardless of these traits appeared to confer more information explaining tree genus composition of the Amazon rainforest. The strength and direction of selection indicated clear directional selective pressure for life history strategies of either growth or protection, depending on forest type (see supplementary online material S-A for a more detailed exploration of ecological interpretation). Including community weighted variance as reflective of potential stabilizing selection did not provide additional information. Although this could be interpreted as indicative of weak or absent stabilizing selection, it is more likely to be an artefact of many genera not being shared among localities due to the sheer geographical scale resulting in a strong mismatch between observed, predicted and uniform relative abundances resulting in a model bias higher than information yielded by including these constraints (see also box S2). Despite showing clear patterns in environmental selection and dispersal effects, there was a large proportion of information left unexplained (44% on average). Potentially, local demographic stochasticity could weaken any link between functional traits measured and regional abundances of genera. This would, however, mean that almost half of the information contained in relative abundances are the result of random population dynamics and are not structurally governed. Alternatively, this could be due to functional traits reflective of processes not taken into account in this study, such as traits reflective of interactions between trophic levels (e.g. traits influenc- ing specific plant-pathogen interactions). Another and at least equally likely hypothesis for (local) unexplained information is that when scaling up, the ratio of genus richness to total abundance decreases rapidly initially but stabilizes again as relatively non-overlapping habitats are included in the regional abundance distributions Scientific Reports | (2023) 13:2859 | https://doi.org/10.1038/s41598-023-28132-y 6 Vol:.(1234567890) www.nature.com/scientificreports/ Figure 2.  Visual representation of pure trait, pure metacommunity, hybrid model and the remaining unexplained information for each separate forest type. Abbreviations indicate different types: igapó (IG), podzol (PZ), swamp (SW), Brazilian shield terra firme (TFBS), Guiana Shield terra firme (TFGS), Pebas terra firme (TFPB) and várzea (VA). Boxplots show median value of pure effects over all samples, with lower and upper hinges corresponding to 25th and 75th percentiles. Whiskers extends from hinge to largest or smallest value no further than 1.5 * IQR from hinge. Points beyond this range are plotted individually and only positive values were plotted. and more genera are included again due to the different habitats. This would result in a change of the regional abundance distribution (i.e. the prior) to which each local community is compared, resulting in higher local unexplained information. Further study into these aspects could provide additional insight, though the data necessary for these scales is lacking for Amazonian trees. Metacommunity importance. Although the initial explanatory power of the metacommunity prior dif- fered between forest types, the decay pattern was very similar. As the effects of either traits or the metacom- munity are measured in the goodness-of-fit predictions on local relative abundances, this implies that at small spatial scales the surrounding regional abundances provide better estimators than functional traits, while at larger spatial scales this shifts to the traits. The ecological translation would be that on small spatial scales, local communities share similar environmental conditions leaving dispersal and drift acting predominantly in changing community composition, at least for genus level taxonomy. As the potential regional pool is increased, more and more environmental heterogeneity and non-overlapping regions are likely to be introduced. The more gradual decline of terra firme forests can then arguably be attributed to these forests having the largest relative surface area of Amazonia (even for the separate subregions), potentially giving these forests an almost continu- ous metacommunity without gaps, resulting in a more gradual transition from metacommunity to trait relative importance. The fact that metacommunity effects do not change anymore after certain distances would indicate the effect of dispersal potentially occurs over very large distances. It should be noted that as these calculations are done at community and genus level, they do not measure single dispersal events but rather the effect of dispersal on community composition much deeper in time. In other words, this effect suggests more than a dispersal event every now and then. Instead, it argues for prolonged mixing of forests on large geographical and temporal scales, supported by recent findings demonstrating a lack of geographical phylogenetic structure of lineages for Amazonian tree ­genera12. Scientific Reports | (2023) 13:2859 | https://doi.org/10.1038/s41598-023-28132-y 7 Vol.:(0123456789) www.nature.com/scientificreports/ Conclusion. Using an unprecedented scale of data and applying the Maximum Entropy Formalism from information theory we show that constraints formed by regional relative abundances of genera explain eight times more of local relative abundances then constraints based on directional selection for specific functional traits, although the latter does show clear signals of environmental dependency. There is, however, still much to be explored due to the large unexplained effects and analyses on finer taxonomic (i.e. species level) and envi- ronmental (e.g. microhabitat) scales could resolve these issues. The relatively large effects of the regional pool of genera over great distances does suggest an important role for long term dispersal and mixing of Amazonian trees, especially for the Amazonian interior. Methods Empirical data. The ­ATDN4,13,14 consists of over 2000 tree inventory plots distributed over the Amazon basin and the Guiana Shield, collectively referred to as Amazonia (a map of all current plots can be found at https://​atdn.​myspe​cies.​info/). Only those plots with trees ≥ 10 cm diameter at breast height were used, leaving 2011 plots with a mean of 558 individuals per plot identified to at least genus level. Most plots used are 1 ha in size (1414) with 492 being smaller (minimum size of 0.1 ha) and 105 larger (maximum size of 80 ha). Genera have been standardized to the W3 Tropicos ­database15 using the Taxonomic Name Resolution Service (TNRS, ­see16). After filtering based on above criteria and solving nomenclature issues, 1,121,935 individuals belonging to over 828 genera remained. Plots were distributed over seven abiotically different forest types: Podzol forests (PZ), Igapó (IG, black water flood forests), Várzea (VA white water flood forests), Swamp (SW) and Terra firme forests (TF) with subregions BS (Brazilian Shield), GS (Guyana Shield) and PB (Pebas) (see ­also17 for details regarding these forest types). Trait data were extracted from several sources. Wood density was mostly derived ­from18. Traits related to leaf characteristics mostly came from four large d ­ atasets19–24, including additional data from other ­sources25–27 as well as unpublished data (J. Lloyd, A.A. de Oliveira, L. Poorter, M. van de Sande & Mazzei, M. van de Sande & L. Poorter). Data on seed mass came ­from28–30 as well as different flora’s and tree guides. As this particular trait can vary over several orders of magnitude, this was included on a log-scale29,31. Ectomycorrhizal aspects were derived from l­iterature32, the same applies to n ­ odulation33,34. Traits involved in aluminum accumulation were based ­on35,36 and references therein. For binary traits (yes/no), a genus was con- sidered having a certain trait only when > 50% of the genus was positive for that specific trait. Functional traits and trait imputation. Constraints were formed by Community Weighted genus Means (CWM) of functional traits (Table 1), related to key ecological life history aspects. According to prin- ciples of natural selection, CWM values will likely be biased towards favourable trait values for that particular environment in the case of directional selection, as taxa with these traits will be more abundant due to envi- ronmental selection. Previous studies included community weighted variance (CWV) as well as indicative of potential stabilizing ­selection11,37. In our case, however, including CWV as constraints resulted in a model bias that was consistently higher than information including trait or metacommunity aspects, CWV was therefore not included as constraints in the final analysis. As for many traits it has been shown earlier that the interspecific variability was larger than the intraspecific variability, this allowed the use of data from different sources to at least calculate a mean species trait value. Genus trait values were subsequently computed as genus-level means of species values if known within the genus and considered constant for each genus. Genus level of taxonomy was used as the available trait database had the most information on this taxonomic level (see Table 1). Unknown val- ues for traits were estimated by Multiple Imputation with Chained Equations (MICE, s­ ee15) by delta adjustment, subtracting a fixed amount (delta), with sensitivity of this adjustment to the imputations of the observed versus imputed data analysed using density plots (Fig. S8) and a linear regression model. This procedure was done using the mice ­package38, available on the R repository, under predictive mean matching (pmm setting, 50 iterations). Results showed imputations were stable and showed near identical patterns with each imputation scenario (see Figs. S5–S6 and Table S2). After imputation, all trait valueswere transformed to Community Weighted Means (CWM) of each trait (J) for each plot (K) ( T JK ) as T JK = Si=1 tij raik with ra the relative abundance of the ith genus in the kth plot following earlier ­uses37. MEF procedure predictions and ecological inference. Figure 1 provides a schematic procedure over- view, box S1 provides an overview of important terms and Boxes S2–S3 further mathematical details. Initially, a maximally uninformative prior is specified, where ­qi (Box S1 Eq. 1) equals 1/S, indicative of each species having equal abundances, and trait constraints are randomly permuted multiple times (n = 50) among genera to test whether inclusion of specified constraints significantly changes derived probability distributions (see a­ lso39). Subsequently, the same prior is used but now observed trait CWM values belonging to specific genera are used as constraints (following earlier applications using simulated ­communities11). Third, observed regional abundances are used as prior with permutated trait constraints and finally both observed regional abundances and observed trait CWM are used as prior and constraints. Maxent2, an updated version of the maxent function currently in the FD library of R provided the computational platform. Proportions of uncertainty explained by each model are given by the Kullback–Leibler divergence ­R2KL, a generalization of the classic ­R2 goodness of fit. In contrast with standard linear regression models having squared goodness-of-fits measurements, the ­R2KL is much more related to the concept of relative entropy, quantifying the information lost when one distribution is compared to another by means of quantifying the statistical distance between two d ­ istributions40. Pure trait, pure metacom- munity, joint metacommunity-trait and unexplained effects are calculated as proportions of total biologically relevant information (Box S1 and Box S2). Data was rarefied to smallest sample size (swamp forests; 28) and calculations bootstrapped 25 times. Results indicated no significant change compared to using all data, hence the total dataset was used for all analyses. Scientific Reports | (2023) 13:2859 | https://doi.org/10.1038/s41598-023-28132-y 8 Vol:.(1234567890) www.nature.com/scientificreports/ Functional trait Units Mean SD Est % Associated challenge Wood density (WD) g/cm3 0.63 0.17 30 Longevity43 Seed mass class (SMC) categorical (1–8) 4.3 1.4 31 Dispersal, Fecundity, ­Establishment43 2 Specific leaf area (SLA) mm /mg 15 5.9 41 Establishment, Plasticity, ­Disturbance43 Leaf nitrogen content (N) mg/g 22.3 7.30 41 Photosynthetic ­capacity43 Leaf phosphorus content (P) mg/g 1 0.77 50 Limited available P for ­metabolism44 Leaf carbon content (C) mg/g 468 38.1 54 Herbivore resistance (C:N)45 Latex 1 = no, 2 = yes 1.2 0.43 46 Herbivore ­resistance46 Resin 1 = no, 2 = yes 1.1 0.35 58 Herbivore ­resistance46 Root nodules (Nodules) 1 = no, 2 = yes 1.1 0.28 0 Nitrogen ­fixation47 Ectomycorrhiza (EctoMyco) 1 = no, 2 = yes 1.01 0.11 0 Organic N fixation , heavy metal ­pollution48 Aluminum accumulation (AlAcc) 1 = no, 2 = yes 1.1 0.21 3 Heavy metal ­pollution49 Fleshy fruits (Fleshy) 1 = no, 2 = yes 1.6 0.50 7 Dispersal (specificity)50 Winged seeds (Wings) 1 = no, 2 = yes 1.2 0.42 39 Dispersal (limitation)50 Table 1.  Overview of used functional traits. Mean and standard deviation (SD) are calculated after predictive mean matching (percentage of estimated values is given by Est (%)). Associated challenge indicates different aspects of life history and selective environment related to specific functional traits, sources are given in the footnote. For specific methodology of measurement protocols and calculation for each trait we refer to the original sources of the data (see main text). Strength and direction of selection. Predictions of genus relative abundances are computed as a func- tion of traits reflected in the CWM values and a series of constants (λjk: the Lagrange Multipliers). Each multi- plier quantifies the association between a unit of change for a particular trait j and a proportional change in predicted relative abundance  pik (the ith genus in the kth community) considering all other traits are constant, formally described as: ∂pik ∂tij = jk pik 1 − pik (see appendix 1 f­rom41). Positive values indicate larger trait values associated with higher abundances (positive selection), negative values indicate the opposite (negative selection) with changes proportional to lambda. Values approximating zero indicate no association between specific traits and relative abundances of species. Decomposing λjk and comparing by means of a One-Way Analysis of Variance for each trait separately between forest types allows studying both the strength and direc- tion of selection in different habitats. Note that this is done for the same constraint between forest types, as lambda values for each constraint do not scale linearly between different constraints. Estimation of metacommunity size. Iteratively increasing the regional species pool considered as prior in concentric circles of a fixed radius of 50 km allows estimating the spatial effect of metacommunity size. Due to computational limits, the number of permutations for the MEF calculations (see above) was reduced to two, shuffling the combinations of genera and traits at least once. Comparison of results from the analyses using all plots indicated small effects of a smaller perturbations (average of 5% difference for metacommunity effect between 5 and 50 permutations). The relationship between pure metacommunity effect and radius of meta- community size was fitted using a smoothing loess regression (function loess and predict; R-package stats with span set at 0.1). Fits subsequently were used to predict values of metacommunity effect based on geographical distance to visualize general patterns for each forest type. Exponential decay of pure metacommunity effect was described using a self-start asymptotic regression function (SSasymp) of the form y(t)∼yf + (y0 − yf)e−exp(log(α))t (nls from stats42. A list of all packages used in R in addition to those preloaded can be found in the supplementary online material (SA2). Data availability R scripts are available on the github repository of E.T. Pos (EdwinTPos). The data that support the findings of this study are available from The Amazon Tree Diversity Network (ATDN) upon reasonable request. Received: 11 March 2021; Accepted: 13 January 2023 References 1. Chisholm, R. A. & Pacala, S. W. Niche and neutral models predict asymptotically equivalent species abundance distributions in high-diversity ecological communities. Proc. Natl. Acad. Sci. U. S. A. 107, 15821 (2010). 2. Tilman, D. Niche tradeoffs, neutrality, and community structure: A stochastic theory of resource competition, invasion, and com- munity assembly. Proc. Natl. Acad. Sci. U. S. A. 101, 10854 (2004). 3. Harte, J. Maximum Entropy and Ecology: A Theory of Abundance, Distribution, and Energetics (OUP Oxford, 2011). 4. The Amazon Tree Diversity Network, http://​atdn.​myspe​cies.​info/. 5. Díaz, S. et al. The global spectrum of plant form and function. Nature 529, 167 (2016). 6. Jaynes, E. T. Information theory and statistical mechanics I. Phys. Rev. 106, 620 (1957). 7. Jaynes, E. T. Information theory and statistical mechanics. II. Phys. Rev. 108, 171 (1957). 8. Shannon, C. E. A mathematical theory of communication. Bell Syst. Tech. J. 27, 379 (1948). Scientific Reports | (2023) 13:2859 | https://doi.org/10.1038/s41598-023-28132-y 9 Vol.:(0123456789) www.nature.com/scientificreports/ 9. Shipley, B. Measuring and interpreting trait-based selection versus meta-community effects during local community assembly. J. Veg. Sci. 25, 55 (2014). 10. Shipley, B., Paine, C. E. T. & Baraloto, C. Quantifying the importance of local niche-based and stochastic processes to tropical tree community assembly. Ecology 93, 760 (2012). 11. Loranger, J., Munoz, F., Shipley, B. & Violle, C. What makes trait-abundance relationships when both environmental filtering and stochastic neutral dynamics are at play?. Oikos 127, 1735 (2018). 12. Dexter, K. G. et al. Dispersal assembly of rain forest tree communities across the amazon basin. Proc. Natl. Acad. Sci. 114, 2645 (2017). 13. ter Steege, H. et al. Hyperdominance in the amazonian tree flora. Science 342, 1243092 (2013). 14. ter Steege, H. et al. Biased-corrected richness estimates for the amazonian tree Flora, Sci. Rep. 10, 1 (2020). 15. Tropicos, Tropicos.Org. Missouri Botanical Garden. Accessed 2018, Version Number Unknown Due to Lack of Release-History. 16. Boyle, B. et al. The taxonomic name resolution service: An online tool for automated standardization of plant names. BMC Bio- inform. 14, 16 (2013). 17. Haugaasen, T. & Peres, C. Floristic, edaphic and structural characteristics of flooded and unflooded forests in the lower rio purús region of Central Amazonia, Brazil. Acta Amaz 36, 25 (2006). 18. Chave, J., Coomes, D. & Jansen, S. Towards a worldwide wood economics spectrum. Ecol. Lett. 351, 855 (2009). 19. Kraft, N. J. B., Valencia, R. & Ackerly, D. D. Functional traits and niche-based tree community assembly in an amazonian forest. Science 322, 580 (2008). 20. Fyllas, N. M. et al. Basin-wide variations in foliar properties of amazonian forest: Phylogeny, soils and climate. Biogeosciences 6, 2677 (2009). 21. Baraloto, C. et al. Decoupled leaf and stem economics in rain forest trees. Ecol. Lett. 13, 1338 (2010). 22. Fortunel, C., Fine, P. V. A. & Baraloto, C. Leaf, stem and root tissue strategies across 758 neotropical tree species. Funct. Ecol. 26, 1153 (2012). 23. Paine, C. E. T., Baraloto, C., Chave, J. & Hérault, B. Functional traits of individual trees reveal ecological constraints on community assembly in tropical rain forests. Oikos 120, 720 (2011). 24. Patiño, S. et al. Coordination of physiological and structural traits in amazon forest trees. Biogeosciences 9, 775 (2012). 25. F. M. J. Ohler, Phytomass and Mineral Content in Untouched Forest, (1980). 26. Thompson, J. et al. Ecological studies on a lowland evergreen rain forest on Maraca Island, Roraima, Brazil. I. Physical environ- ment, forest structure and leaf chemistry. J. Ecol. 689, 89 (1992). 27. Pons, T. L., Perreijn, K., Van Kessel, C. & Werger, M. J. A. Symbiotic nitrogen fixation in a tropical rainforest: 15N natural abundance measurements supported by experimental isotopic enrichment. New Phytol. 173, 154 (2006). 28. Foster, S. & Janson, C. H. The relationship between seed size and establishment conditions in tropical woody plants. Ecology 66, 773 (1985). 29. Hammond, D. S. & Brown, V. K. Hammond and brown 1995 seed size. Ecology.Pdf. Ecology 76, 2544 (1995). 30. Royal Botanic Gardens Kew, 2020, Seed Information Database (SID). 31. ter Steege, H. & Hammond, D. Character convergence, diversity, and disturbance in tropical rain forest in Guyana. Ecology 82, 3197 (2001). 32. Tedersoo, L. & Brundrett, M. C. Evolution of ectomycorrhizal symbiosis in plants. Biogeogr. Mycorrhizal Symbiosis. 407, 89 (2017). 33. Sprent, J. I. Nodulation in Legumes, Royal Botanic Gardens, Kew. 34. Soltis, P. S., Soltis, D. E. & Chase, M. W. Angiosperm phylogeny inferred from multiple genes as a tool for comparative biology. Nature 402, 402 (1999). 35. Jansen, S., Watanabe, T., Dessein, S., Smets, E. & Robbrecht, E. A comparative study of metal levels in leaves of some Al-accumu- lating Rubiaceae. Ann. Bot. 91, 657 (2003). 36. Jansen, S., Watanabe, T. & Smets, E. Aluminium accumulation in leaves of 127 species in Melastomataceae, with comments on the order myrtales. Ann. Bot. 90, 53 (2002). 37. Sonnier, G., Shipley, B. & Navas, M. L. Quantifying relationships between traits and explicitly measured gradients of stress and disturbance in early successional plant communities. J. Veg. Sci. 21, 1014 (2010). 38. van Buuren, S. & Groothuis-Oudshoorn, K. Mice: Multivariate imputation by chained equations in R. J. Stat. Softw. 45, 1 (2011). 39. Roxburgh, S. H. & Mokany, K. On testing predictions of species relative abundance from maximum entropy optimisation. Oikos 119, 583 (2010). 40. Cameron, A. C. & Windmeijer, F. A. G. An R-squared measure of goodness of fit for some common nonlinear regression models. J. Econ. 77, 329 (1997). 41. Sonnier, G., Navas, M. L., Fayolle, A. & Shipley, B. Quantifying trait selection driving community assembly: A test in herbaceous plant communities under contrasted land use regimes. Oikos 121, 1103 (2012). 42. R Core Team, R: A Language and Environment for Statistical Computing. 43. Weiher, E. et al. Challenging theophrastus: A common core list of plant traits for functional ecology. J. Veg. Sci. 10, 609 (1999). 44. Raghothama, K. G. Phosphate acquisition. Annu. Rev. Plant Physiol. Plant Mol. Biol. 50, 665 (1999). 45. Poorter, L., Van De Plassche, M., Willems, S. & Boot, R. G. A. Leaf traits and herbivory rates of tropical tree species differing in successional status. Plant Biol. 6, 746 (2004). 46. Agrawal, A. A. Macroevolution of plant defense strategies. Trends Ecol. Evol. 22, 103 (2007). 47. Sprent, J. I. & Parsons, R. Nitrogen fixation in legume and non-legume trees. F. Crop. Res. 65, 183 (2000). 48. Colpaert, J. V., Wevers, J. H. L., Krznaric, E. & Adriaensen, K. How metal-tolerant ecotypes of ectomycorrhizal fungi protect plants from heavy metal pollution. Ann. For. Sci. 68, 17 (2011). 49. Poschenrieder, C., Gunsé, B., Corrales, I. & Barceló, J. A Glance into aluminum toxicity and resistance in plants. Sci. Total Environ. 400, 356 (2008). 50. Howe, H. F. & Smallwood, J. Ecology of Seed Dispersal. Ecology 13, 201 (2013). Acknowledgements This paper is the result of the work of hundreds of different scientists and research institutions in the Amazon over the past 80 years. Without their hard work this analysis would have been impossible. The 25-ha Long-Term Ecological Research Project of Amacayacu is a collaborative project of the Instituto Amazónico de Investiga- ciones Científicas Sinchi and the Universidad Nacional de Colombia Sede Medellín, in partnership with the Unidad de Manejo Especial de Parques Naturales Nacionales and the Center for Tropical Forest Science of the Smithsonian Tropical Research Institute (CTFS). The Amacayacu Forest Dynamics Plot is part of the Center for Tropical Forest Science, a global network of large-scale demographic tree plots. We acknowledge the Director and staff of the Amacayacu National Park for supporting and maintaining the project in this National Park. We would also like to thank Prof. John Harte, Prof. Bill Shipley and the anonymous reviewers for their constructive feedback during the review process. Scientific Reports | (2023) 13:2859 | https://doi.org/10.1038/s41598-023-28132-y 10 Vol:.(1234567890) www.nature.com/scientificreports/ Author contributions E.T.P. and H.T.S. designed the study. E.T.P. performed analyses and took the lead in writing the manuscript, H.T.S. supervised the writing and provided regular feedback both for the manuscript and the interpretation of the results. All other authors provided feedback on the manuscript and provided their data from the Amazon Tree Diversity Network or trait data. Authors E.T.P. to L.V.G. provided tree inventory data, authors G.B., J.K., N.K., A.L., K.M., G.P., L.P. provided data on functional traits, C.B., J.L., A.A.O. and H.T.S. provided both tree inventory and functional trait data. Funding HtS and RS were supported by grant 407232/2013-3 – PVE - MEC/MCTI/CAPES/CNPq/FAPs; CB was sup- ported by grant FAPESP 95/3058-0 - CRS 068/96 WWF Brasil - The Body Shop; DS, JFM, JE, PP and JC ben- efited from an “Investissement d’Avenir” grant managed by the Agence Nationale de la Recherche (CEBA: ANR-10-LABX-25-01); HLQ/MAP/JLLM received financial supported by MCT/CNPq/CT-INFRA/GEOMA #550373/2010-1 and # 457515/2012-0, and JLLM were supported by grant CAPES/PDSE # 88881.135761/2016-01 and CAPES/Fapespa #1530801; The Brazilian National Research Council (CNPq) provided a productivity grant to EMV (Grant 308040/2017-1); Floristic identification in plots in the RAINFOR forest monitoring network have been supported by the Natural Environment Research Council (grants NE/B503384/1, NE/ D01025X/1, NE/I02982X/1, NE/F005806/1, NE/D005590/1 and NE/I028122/1) and the Gordon and Betty Moore Founda- tion; B.M.F. is funded by FAPESP grant 2016/25086-3. BSM, BHMJ and OLP were supported by grants CNPq/ CAPES/FAPS/BC-Newton Fund #441244/2016-5 and FAPEMAT/0589267/2016; TWH was funded by National Science Foundation grant DEB-1556338. Competing interests The authors declare no competing interests. Additional information Supplementary Information The online version contains supplementary material available at https://​doi.​org/​ 10.​1038/​s41598-​023-​28132-y. Correspondence and requests for materials should be addressed to E.P. Reprints and permissions information is available at www.nature.com/reprints. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. © The Author(s) 2023 Scientific Reports | (2023) 13:2859 | https://doi.org/10.1038/s41598-023-28132-y 11 Vol.:(0123456789)