Philippa Kaur

Sapolsky, R. M. The influence of social hierarchy on primate health. Science 308, 648–652 (2005).Article 
ADS 
CAS 

Google Scholar 
Snyder-Mackler, N. et al. Social status alters immune regulation and response to infection in macaques. Science 354, 1041–1045 (2016).Article 
ADS 
CAS 

Google Scholar 
Levy, E. J. et al. Higher dominance rank is associated with lower glucocorticoids in wild female baboons: A rank metric comparison. Horm. Behav. 125, 104826 (2020).Article 
CAS 

Google Scholar 
Sapolsky, R. M. Social status and health in humans and other animals. Annu. Rev. Anthropol. 33, 393–418 (2004).Article 

Google Scholar 
Goymann, W. & Wingfield, J. C. Allostatic load, social status and stress hormones: the costs of social status matter. Anim. Behav. 67, 591–602 (2004).Article 

Google Scholar 
Cavigelli, S. A. & Chaudhry, H. S. Social status, glucocorticoids, immune function, and health: Can animal studies help us understand human socioeconomic-status-related health disparities?. Horm. Behav. 62, 295–313 (2012).Article 
CAS 

Google Scholar 
Meyer, J. S. & Hamel, A. F. Models of stress in nonhuman primates and their relevance for human psychopathology and endocrine dysfunction. ILAR J. 55, 347–360 (2014).Article 
CAS 

Google Scholar 
Saltzman, W., Schultz-Darken, N. J., Scheffler, G., Wegner, F. H. & Abbott, D. H. Social and reproductive influences on plasma cortisol in female marmoset monkeys. Physiol. Behav. 56, 801–810 (1994).Article 
CAS 

Google Scholar 
Abbott, D. H. et al. Are subordinates always stressed? A comparative analysis of rank differences in cortisol levels among primates. Horm. Behav. 43, 67–82 (2003).Article 
CAS 

Google Scholar 
Sadoughi, B., Lacroix, L., Berbesque, C., Meunier, H. & Lehmann, J. Effects of social tolerance on stress: Hair cortisol concentrations in the tolerant Tonkean macaques (Macaca tonkeana) and the despotic long-tailed macaques (Macaca fascicularis). Stress 1, 1–9 (2021).
Google Scholar 
Kawachi, I. & Berkman, L. Social cohesion, social capital, and health. Social Epidemiol. 174, 290–314 (2000).
Google Scholar 
Dong, M. et al. Insights into causal pathways for ischemic heart disease: adverse childhood experiences study. Circulation 110, 1761–1766 (2004).Article 

Google Scholar 
Galobardes, B., Lynch, J. W. & Davey Smith, G. Childhood socioeconomic circumstances and cause-specific mortality in adulthood: Systematic review and interpretation. Epidemiol. Rev. 26, 7–21 (2004).Article 

Google Scholar 
Lockwood, K. G., John-Henderson, N. A. & Marsland, A. L. Early life socioeconomic status associates with interleukin-6 responses to acute laboratory stress in adulthood. Physiol. Behav. 188, 212–220 (2018).Article 
CAS 

Google Scholar 
Taylor, S. E. Mechanisms linking early life stress to adult health outcomes. Proc. Natl. Acad. Sci. 107, 8507–8512 (2010).Article 
ADS 
CAS 

Google Scholar 
Uchino, B. N. Social Support and Physical Health: Understanding the Health Consequences of Relationships (Yale University Press, 2004).Book 

Google Scholar 
Holt-Lunstad, J. & Uchino, B. N. Social support and health. Health Behav. Theory Res. Pract. 1, 183–204 (2015).
Google Scholar 
Gurven, M., Allen-Arave, W., Hill, K. & Hurtado, A. M. Reservation food sharing among the Ache of Paraguay. Hum. Nat. 12, 273–297 (2001).Article 
CAS 

Google Scholar 
Hill, K. & Hurtado, A. M. Ache Life History: The Ecology and Demography of a Foraging People (Routledge, 2017).Book 

Google Scholar 
Kraft, T. S., Venkataraman, V. V., Tacey, I., Dominy, N. J. & Endicott, K. M. Foraging performance, prosociality, and kin presence do not predict lifetime reproductive success in Batek hunter-gatherers. Hum. Nat. 30, 71–97 (2019).Article 

Google Scholar 
Venkataraman, V. V., Kraft, T. S., Dominy, N. J. & Endicott, K. M. Hunter-gatherer residential mobility and the marginal value of rainforest patches. Proc. Natl. Acad. Sci. 114, 3097–3102 (2017).Article 
ADS 
CAS 

Google Scholar 
Woodburn, J. Egalitarian societies. Man 1, 431–451 (1982).Article 

Google Scholar 
Marlowe, F. The Hadza: Hunter-Gatherers of Tanzania Vol. 3 (University of California Press, 2010).
Google Scholar 
Fedurek, P. et al. Status does not predict stress: Women in an egalitarian hunter–gatherer society. Evol. Hum. Sci. 2, 1–10 (2020).
Google Scholar 
Kornienko, O. & Santos, C. E. The effects of friendship network popularity on depressive symptoms during early adolescence: Moderation by fear of negative evaluation and gender. J. Youth Adolesc. 43, 541–553 (2014).Article 

Google Scholar 
Smelser, N. J. & Baltes, P. B. International Encyclopedia of the Social & Behavioral Sciences Vol. 11 (Elsevier, 2001).
Google Scholar 
Kim, D. A., Benjamin, E. J., Fowler, J. H. & Christakis, N. A. Social connectedness is associated with fibrinogen level in a human social network. Proc. R. Soc. B Biol. Sci. 283, 20160958 (2016).Article 

Google Scholar 
Kindermann, T. A. & Gest, S. D. Assessment of the peer group: Identifying naturally occurring social networks and capturing their effects. In Handbook of peer interactions, relationships, and groups, 100–117 (2009).Kornienko, O., Clemans, K. H., Out, D. & Granger, D. A. Friendship network position and salivary cortisol levels. Soc. Neurosci. 8, 385–396 (2013).Article 

Google Scholar 
La Greca, A. M. & Lopez, N. Social anxiety among adolescents: Linkages with peer relations and friendships. J. Abnorm. Child Psychol. 26, 83–94 (1998).Article 

Google Scholar 
Okamoto, J. et al. Social network status and depression among adolescents: An examination of social network influences and depressive symptoms in a Chinese sample. Res. Hum. Dev. 8, 67–88 (2011).Article 

Google Scholar 
Ulset, V. S. et al. Are unpopular children more likely to get sick? Longitudinal links between popularity and infectious diseases in early childhood. PLoS ONE 14, e0222222 (2019).Article 
CAS 

Google Scholar 
Hawkes, K. Showing off: Tests of an hypothesis about men’s foraging goals. Ethol. Sociobiol. 12, 29–54 (1991).Article 

Google Scholar 
Smith, E. A. Why do good hunters have higher reproductive success?. Hum. Nat. 15, 343–364 (2004).Article 

Google Scholar 
Apicella, C. L., Feinberg, D. R. & Marlowe, F. W. Voice pitch predicts reproductive success in male hunter-gatherers. Biol. Lett. 3, 682–684 (2007).Article 
CAS 

Google Scholar 
Apicella, C. L. Upper-body strength predicts hunting reputation and reproductive success in Hadza hunter–gatherers. Evol. Hum. Behav. 35, 508–518 (2014).Article 

Google Scholar 
Smith, K. M., Olkhov, Y. M., Puts, D. A. & Apicella, C. L. Hadza men with lower voice pitch have a better hunting reputation. Evol. Psychol. 15, 1474704917740466 (2017).Article 

Google Scholar 
MacDougall-Shackleton, S. A., Bonier, F., Romero, L. M. & Moore, I. T. Glucocorticoids and “stress” are not synonymous. Integr. Organ. Biol. 1, 017 (2019).
Google Scholar 
Ouellette, S. J. et al. Hair cortisol concentrations in higher-and lower-stress mother–daughter dyads: A pilot study of associations and moderators. Dev. Psychobiol. 57, 519–534 (2015).Article 
CAS 

Google Scholar 
Stalder, T. et al. Stress-related and basic determinants of hair cortisol in humans: A meta-analysis. Psychoneuroendocrinology 77, 261–274 (2017).Article 
CAS 

Google Scholar 
Heimbürge, S., Kanitz, E. & Otten, W. The use of hair cortisol for the assessment of stress in animals. Gen. Comp. Endocrinol. 270, 10–17 (2019).Article 

Google Scholar 
Fedurek, P. et al. Relationship between proximity and physiological stress levels in hunter-gatherers: The Hadza. Horm. Behav. 147, 105294 (2023).Article 

Google Scholar 
Bowers, K. et al. Maternal distress and hair cortisol in pregnancy among women with elevated adverse childhood experiences. Psychoneuroendocrinology 95, 145–148 (2018).Article 
CAS 

Google Scholar 
Wells, S. et al. Associations of hair cortisol concentration with self-reported measures of stress and mental health-related factors in a pooled database of diverse community samples. Stress 17, 334–342 (2014).Article 
CAS 

Google Scholar 
Faresjö, T. et al. Elevated levels of cortisol in hair precede acute myocardial infarction. Sci. Rep. 10, 1–8 (2020).Article 

Google Scholar 
Fuchs, A. et al. Link between children’s hair cortisol and psychopathology or quality of life moderated by childhood adversity risk. Psychoneuroendocrinology 90, 52–60 (2018).Article 
CAS 

Google Scholar 
Staufenbiel, S. M., Penninx, B. W., Spijker, A. T., Elzinga, B. M. & van Rossum, E. F. Hair cortisol, stress exposure, and mental health in humans: A systematic review. Psychoneuroendocrinology 38, 1220–1235 (2013).Article 
CAS 

Google Scholar 
Davison, B., Singh, G. R. & McFarlane, J. Hair cortisol and cortisone as markers of stress in Indigenous and non-Indigenous young adults. Stress 22, 210–220 (2019).Article 
CAS 

Google Scholar 
Kim, E., Bolkan, C., Crespi, E. & Madigan, J. The relationship between hair cortisol, chronic stress, and well-being among older adults with dementia. Innov. Aging 3, S468 (2019).Article 

Google Scholar 
Woodburn, J. Egalitarian societies revisited. Proper. Equal. 1, 18–31 (2005).
Google Scholar 
Berbesque, J. C., Wood, B. M., Crittenden, A. N., Mabulla, A. & Marlowe, F. W. Eat first, share later: Hadza hunter–gatherer men consume more while foraging than in central places. Evol. Hum. Behav. 37, 281–286 (2016).Article 

Google Scholar 
Marlowe, F. W. & Berbesque, J. C. Tubers as fallback foods and their impact on Hadza hunter-gatherers. Am. J. Phys. Anthropol. 140, 751–758 (2009).Article 

Google Scholar 
Berbesque, J. C. & Marlowe, F. W. Sex differences in food preferences of Hadza hunter-gatherers. Evol. Psychol. 7, 147470490900700400 (2009).Article 

Google Scholar 
Hawkes, K., O’Connell, J. F. & Blurton Jones, N. G. Hunting income patterns among the Hadza: Big game, common goods, foraging goals and the evolution of the human diet. Philos. Trans. R. Soc. Lond. B 334, 243–251 (1991).Article 
ADS 
CAS 

Google Scholar 
Hawkes, K. Hunting and the evolution of egalitarian societies: Lessons from the Hadza. Hierarch. Action Cui Bono 27, 1–10 (2000).
Google Scholar 
Stibbard-Hawkes, D. N., Attenborough, R. D. & Marlowe, F. W. A noisy signal: To what extent are Hadza hunting reputations predictive of actual hunting skills?. Evol. Hum. Behav. 39, 639–651 (2018).Article 

Google Scholar 
Smith, K. M. & Apicella, C. Partner choice in human evolution: The role of character, hunting ability, and reciprocity in Hadza campmate selection. (2019).Smith, K. M. & Apicella, C. L. Hadza hunter-gatherers disagree on perceptions of moral character. Soc. Psychol. Pers. Sci. 11, 616–625 (2020).Article 

Google Scholar 
Gurven, M., Allen-Arave, W., Hill, K. & Hurtado, M. “It’s a wonderful life”: Signaling generosity among the Ache of Paraguay. Evol. Hum. Behav. 21, 263–282 (2000).Article 
CAS 

Google Scholar 
Aktipis, A. et al. Cooperation in an uncertain world: For the Maasai of East Africa, need-based transfers outperform account-keeping in volatile environments. Hum. Ecol. 44, 353–364 (2016).Article 

Google Scholar 
Cronk, L. et al. Managing risk through cooperation: Need-based transfers and risk pooling among the societies of the Human Generosity Project. in Global Perspectives on Long Term Community Resource Management, 41–75 (Springer, 2019).Cronk, L. & Aktipis, A. Design principles for risk-pooling systems. Nat. Hum. Behav. 1, 1–9 (2021).
Google Scholar 
Jones, N. B. Demography and Evolutionary Ecology of Hadza Hunter-Gatherers Vol. 71 (Cambridge University Press, 2016).
Google Scholar 
Crittenden, A. N. et al. Oral health in transition: The Hadza foragers of Tanzania. PLoS ONE 12, e0172197 (2017).Article 

Google Scholar 
Bennett, F. J., Barnicot, N. A., Woodburn, J. C., Pereira, M. S. & Henderson, B. E. Studies on viral, bacterial, rickettsial and treponemal diseases in the Hadza of Tanzania and a note on injuries. Hum. Biol. 1, 243–272 (1973).
Google Scholar 
Ibar, C. et al. Evaluation of stress, burnout and hair cortisol levels in health workers at a University Hospital during COVID-19 pandemic. Psychoneuroendocrinology 128, 105213 (2021).Article 
CAS 

Google Scholar 
Rajcani, J., Vytykacova, S., Solarikova, P. & Brezina, I. Stress and hair cortisol concentrations in nurses during the first wave of the COVID-19 pandemic. Psychoneuroendocrinology 129, 105245 (2021).Article 
CAS 

Google Scholar 
Hill, K. R., Wood, B. M., Baggio, J., Hurtado, A. M. & Boyd, R. T. Hunter-gatherer inter-band interaction rates: Implications for cumulative culture. PLoS ONE 9, e102806 (2014).Article 
ADS 

Google Scholar 
Bird, D. W., Bird, R. B., Codding, B. F. & Zeanah, D. W. Variability in the organization and size of hunter-gatherer groups: Foragers do not live in small-scale societies. J. Hum. Evol. 131, 96–108 (2019).Article 

Google Scholar 
Fedurek, P. et al. Social status does not predict in-camp integration among egalitarian hunter-gatherer men. Behav. Ecol. 33, 65–76 (2022).Article 

Google Scholar 
Ponzi, D., Muehlenbein, M. P., Geary, D. C. & Flinn, M. V. Cortisol, salivary alpha-amylase and children’s perceptions of their social networks. Soc. Neurosci. 11, 164–174 (2016).Article 

Google Scholar 
Marlowe, F. W. Mate preferences among Hadza hunter-gatherers. Hum. Nat. 15, 365–376 (2004).Article 

Google Scholar 
Von Rueden, C. R. & Jaeggi, A. V. Men’s status and reproductive success in 33 nonindustrial societies: Effects of subsistence, marriage system, and reproductive strategy. Proc. Natl. Acad. Sci. 113, 10824–10829 (2016).Article 

Google Scholar 
Townsend, C. Egalitarianism, Evolution Of (Wiley, 2018).Book 

Google Scholar 
Winterhalder, B. Diet choice, risk, and food sharing in a stochastic environment. J. Anthropol. Archaeol. 5, 369–392 (1986).Article 

Google Scholar 
Cornell, T. & Allen, T. B. War and Games Vol. 3 (Boydell Press, 2002).
Google Scholar 
Smáradóttir, S. Health and Wellbeing in the Arctic: The Critical Issues of Food Insecurity and Suicide Among Indigenous People.Finkler, H. W. Violence and the administration of justice: A focus on inuit communities in Northern Canada. BC Third World LJ 4, 137 (1983).
Google Scholar 
Bowles, S. Did warfare among ancestral hunter-gatherers affect the evolution of human social behaviors?. Science 324, 1293–1298 (2009).Article 
ADS 
CAS 

Google Scholar 
Fry, D. P. & Söderberg, P. Lethal aggression in mobile forager bands and implications for the origins of war. Science 341, 270–273 (2013).Article 
ADS 
CAS 

Google Scholar 
Gat, A. Proving communal warfare among hunter-gatherers: The quasi-rousseauan error. Evol. Anthropol. 24, 111–126 (2015).Article 

Google Scholar 
Kreyszig, E. Bernstein polynomials and numerical integration. Int. J. Numer. Meth. Eng. 14, 292–295 (1979).Article 
MATH 

Google Scholar 
Meyer, D. et al. Misc functions of the department of statistics, probability theory group (formerly: E1071). Package e1071. TU Wien (2015).R Development Core. A Language ans Environment for Statistical Computing. (R Found Stat Comput Vienna, 2018).Wennig, R. Potential problems with the interpretation of hair analysis results. Forensic Sci. Int. 107, 5–12 (2000).Article 
CAS 

Google Scholar 
Kumari, M., Shipley, M., Stafford, M. & Kivimaki, M. Association of diurnal patterns in salivary cortisol with all-cause and cardiovascular mortality: Findings from the Whitehall II study. J. Clin. Endocrinol. Metab. 96, 1478–1485 (2011).Article 
CAS 

Google Scholar 
Marmot, M. G. & Sapolsky, R. Of baboons and men: Social circumstances, biology, and the social gradient in health. in Sociality, hierarchy, health: Comparative biodemography: Papers from a workshop (2014).Hoffman, M. C., Karban, L. V., Benitez, P., Goodteacher, A. & Laudenslager, M. L. Chemical processing and shampooing impact cortisol measured in human hair. Clin. Investig. Med. 37, E252 (2014).Article 
CAS 

Google Scholar 
Sauvé, B., Koren, G., Walsh, G., Tokmakejian, S. & Van Uum, S. H. Measurement of cortisol in human hair as a biomarker of systemic exposure. Clin. Investig. Med. 30, E183–E191 (2007).Article 

Google Scholar 
Slominski, R., Rovnaghi, C. R. & Anand, K. J. Methodological considerations for hair cortisol measurements in children. Ther. Drug Monit. 37, 812 (2015).Article 
CAS 

Google Scholar 
Xiang, L., Sunesara, I., Rehm, K. E. & Marshall, G. D. Jr. A modified and cost-effective method for hair cortisol analysis. Biomarkers 21, 200–203 (2016).Article 
CAS 

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

Google Scholar 
Mangiafico, S. & Mangiafico, M. S. Package ‘rcompanion’. Cran Repos 1–71 (2017).Kuznetsova, A., Brockhoff, P. B. & Christensen, R. H. lmerTest package: Tests in linear mixed effects models. J. Stat. Softw. 82, 1–26 (2017).Article 

Google Scholar 
Bates, D. M. lme4: Mixed-Effects Modeling with R. (2010).Lüdecke, D. ggeffects: Tidy data frames of marginal effects from regression models. J. Stat. Softw. 3(26), 772. https://doi.org/10.21105/joss.00772 (2018).Article 

Google Scholar 
Nowok, B., Raab, G. M. & Dibben, C. synthpop: Bespoke creation of synthetic data in R. J. Stat. Softw. 74, 1–26 (2016).Article 

Google Scholar  More

Foley, J. A. et al. Global consequences of land use. Science 309, 570–574 (2005).Article 
CAS 

Google Scholar 
Rockström, J. et al. A safe operating space for humanity. Nature 461, 472–475 (2009).Article 

Google Scholar 
Geiger, F. et al. Persistent negative effects of pesticides on biodiversity and biological control potential on European farmland. Basic Appl. Ecol. 11, 97–105 (2010).Article 
CAS 

Google Scholar 
Zhang, W., Ricketts, T. H., Kremen, C., Carney, K. & Swinton, S. M. Ecosystem services and dis-services to agriculture. Ecol. Econ. 64, 253–260 (2007).Article 

Google Scholar 
Tilman, D., Cassman, K. G., Matson, P. A., Naylor, R. & Polasky, S. Agricultural sustainability and intensive production practices. Nature 418, 671–677 (2002).Article 
CAS 

Google Scholar 
Helming, K. et al. Managing soil functions for a sustainable bioeconomy-assessment framework and state of the art. Land Degrad. Dev. 29, 3112–3126 (2018).Article 

Google Scholar 
Rockström, J. et al. Sustainable intensification of agriculture for human prosperity and global sustainability. Ambio 46, 4–17 (2017).Article 

Google Scholar 
Lehmann, J., Bossio, D. A., Kögel-Knabner, I. & Rillig, M. C. The concept and future prospects of soil health. Nat. Rev. Earth Environ. 1, 544–553 (2020).Article 

Google Scholar 
Smith, J., Pearce, B. D. & Wolfe, M. S. Reconciling productivity with protection of the environment: Is temperate agroforestry the answer? Renew. Agric. Food Syst. 28, 80–92 (2013).Article 

Google Scholar 
European Commission. A Greener and Fairer CAP (EC, 2021).Grass, I. et al. Trade-offs between multifunctionality and profit in tropical smallholder landscapes. Nat. Commun. 11, 1186 (2020).Article 
CAS 

Google Scholar 
Mayer, S. et al. Soil organic carbon sequestration in temperate agroforestry systems – a meta-analysis. Agric. Ecosyst. Environ. 323, 107689 (2022).Article 
CAS 

Google Scholar 
Pardon, P. et al. Juglans regia (walnut) in temperate arable agroforestry systems: effects on soil characteristics, arthropod diversity and crop yield. Renew. Agric. Food Syst. 35, 533–549 (2020).Article 

Google Scholar 
Schmidt, M. et al. Nutrient saturation of crop monocultures and agroforestry indicated by nutrient response efficiency. Nutr. Cycl. Agroecosyst. 119, 69–82 (2021).Article 
CAS 

Google Scholar 
Beule, L. & Karlovsky, P. Tree rows in temperate agroforestry croplands alter the composition of soil bacterial communities. PLoS ONE 16, e0246919 (2021).Article 
CAS 

Google Scholar 
Palma, J. H. N. et al. Modeling environmental benefits of silvoarable agroforestry in Europe. Agric. Ecosyst. Environ. 119, 320–334 (2007).Article 

Google Scholar 
Kay, S. et al. Spatial similarities between European agroforestry systems and ecosystem services at the landscape scale. Agroforest Syst. 92, 1075–1089 (2018).Article 

Google Scholar 
Swieter, A., Langhof, M., Lamerre, J. & Greef, J. M. Long-term yields of oilseed rape and winter wheat in a short rotation alley cropping agroforestry system. Agroforest Syst. 93, 1853–1864 (2019).Article 

Google Scholar 
Ivezić, V., Yu, Y. & van der Werf, W. Crop yields in European agroforestry systems: a meta-analysis. Front. Sustain. Food Syst. 5, 606631 (2021).Article 

Google Scholar 
Cardinael, R. et al. High organic inputs explain shallow and deep SOC storage in a long-term agroforestry system – combining experimental and modeling approaches. Biogeosciences 15, 297–317 (2018).Article 
CAS 

Google Scholar 
Smith, P. Carbon sequestration in croplands: the potential in Europe and the global context. Eur. J. Agron. 20, 229–236 (2004).Article 
CAS 

Google Scholar 
Kay, S. et al. Agroforestry creates carbon sinks whilst enhancing the environment in agricultural landscapes in Europe. Land Use Policy 83, 581–593 (2019).Article 

Google Scholar 
Cardinael, R. et al. Impact of alley cropping agroforestry on stocks, forms and spatial distribution of soil organic carbon — a case study in a Mediterranean context. Geoderma 259–260, 288–299 (2015).Article 

Google Scholar 
Cardinael, R. et al. Spatial variation of earthworm communities and soil organic carbon in temperate agroforestry. Biol. Fertil. Soils 55, 171–183 (2019).Article 
CAS 

Google Scholar 
Boinot, S. et al. Alley cropping agroforestry systems: reservoirs for weeds or refugia for plant diversity? Agric. Ecosyst. Environ. 284, 106584 (2019).Article 

Google Scholar 
Barnes, A. D. et al. Direct and cascading impacts of tropical land-use change on multi-trophic biodiversity. Nat. Ecol. Evol. 1, 1511–1519 (2017).Article 

Google Scholar 
Kehoe, L. et al. Biodiversity at risk under future cropland expansion and intensification. Nat. Ecol. Evol. 1, 1129–1135 (2017).Article 

Google Scholar 
DuPont, S. T., Culman, S. W., Ferris, H., Buckley, D. H. & Glover, J. D. No-tillage conversion of harvested perennial grassland to annual cropland reduces root biomass, decreases active carbon stocks, and impacts soil biota. Agric. Ecosyst. Environ. 137, 25–32 (2010).Article 
CAS 

Google Scholar 
Bengtsson, J. et al. Grasslands-more important for ecosystem services than you might think. Ecosphere 10, e02582 (2019).Article 

Google Scholar 
Beule, L. et al. Conversion of monoculture cropland and open grassland to agroforestry alters the abundance of soil bacteria, fungi and soil-N-cycling genes. PLoS ONE 14, e0218779 (2019).Article 
CAS 

Google Scholar 
Borrelli, P., Ballabio, C., Panagos, P. & Montanarella, L. Wind erosion susceptibility of European soils. Geoderma 232–234, 471–478 (2014).Article 

Google Scholar 
Amundson, R. et al. Soil and human security in the 21st century. Science 348, 12610711–12610716 (2015).Article 

Google Scholar 
Olson, K. R., Al-Kaisi, M., Lal, R. & Cihacek, L. Impact of soil erosion on soil organic carbon stocks. J. Soil Water Conserv. 71, 61A–67A (2016).Article 

Google Scholar 
Larney, F. J., Bullock, M. S., Janzen, H. H., Ellert, B. H. & Olson, E. C. S. Wind erosion effects on nutrient redistribution and soil productivity. J. Soil Water Conserv. 53, 133–140 (1998).
Google Scholar 
de Jong, E. & Kowalchuk, T. E. The effect of shelterbelts on erosion and soil properties. Soil Sci. 159, 337–345 (1995).Article 

Google Scholar 
Deutsch, M. & Otter, V. Nachhaltigkeit und förderung? Akzeptanzfaktoren im Entscheidungsprozess deutscher Landwirte zur Anlage von Agroforstsystemen. Berichte über Landwirtschaft – Zeitschrift für Agrarpolitik und Landwirtschaft Aktuelle Beiträge (2021).Tsonkova, P., Böhm, C., Quinkenstein, A. & Freese, D. Ecological benefits provided by alley cropping systems for production of woody biomass in the temperate region: a review. Agroforest Syst. 85, 133–152 (2012).Article 

Google Scholar 
Lehmann, J., Weigl, D., Droppelmann, K., Huwe, B. & Zech, W. Nutrient cycling in an agroforestry system with runoff irrigation in Northern Kenya. Agroforestry Syst. 43, 49–70 (1998).Article 

Google Scholar 
Shao, G. et al. Impacts of monoculture cropland to alley cropping agroforestry conversion on soil N2O emissions. GCB Bioenergy https://doi.org/10.1111/gcbb.13007 (2022).Isaac, M. E. & Borden, K. A. Nutrient acquisition strategies in agroforestry systems. Plant Soil 444, 1–19 (2019).Article 
CAS 

Google Scholar 
Cannell, M. G. R., van Noordwijk, M. & Ong, C. K. The central agroforestry hypothesis: the trees must acquire resources that the crop would not otherwise acquire. Agroforestry Syst. 34, 27–31 (1996).Article 

Google Scholar 
Beule, L., Vaupel, A. & Moran-Rodas, V. E. Abundance, diversity, and function of soil microorganisms in temperate alley-cropping agroforestry systems: a review. Microorganisms 10, 616 (2022).Article 
CAS 

Google Scholar 
Thevathasan, N. V. & Gordon, A. M. in New Vistas in Agroforestry, Vol. 1 (eds Nair, P. K. R., Rao, M. R. & Buck, L. E.) 257–268 (Springer Netherlands, 2004).Veldkamp, E. & Keller, M. Fertilizer-induced nitric oxide emissions from agricultural soils. Nutr. Cycling Agroecosyst. 48, 69–77 (1997).Article 
CAS 

Google Scholar 
Luo, J., Beule, L., Shao, G., Veldkamp, E. & Corre, M. D. Reduced soil gross N2O emission driven by substrates rather than denitrification gene abundance in cropland agroforestry and monoculture. JGR Biogeosciences 127, e2021JG006629 (2022).Article 
CAS 

Google Scholar 
Langenberg, J., Feldmann, M. & Theuvsen, L. Alley cropping agroforestry systems: using Monte-Carlo simulation for a risk analysis in comparison with arable farming systems. German J. Agric. Econ. 67, 95–112 (2018).
Google Scholar 
Otter, V. & Langenberg, J. Willingness to pay for environmental effects of agroforestry systems: a PLS-model of the contingent evaluation from German taxpayers’ perspective. Agroforest Syst. 94, 811–829 (2020).Article 

Google Scholar 
Zhang, X. et al. Quantification of global and national nitrogen budgets for crop production. Nat. Food 2, 529–540 (2021).Article 

Google Scholar 
Markwitz, C., Knohl, A. & Siebicke, L. Evapotranspiration over agroforestry sites in Germany. Biogeosciences 17, 5183–5208 (2020).Article 

Google Scholar 
Pardon, P. et al. Trees increase soil organic carbon and nutrient availability in temperate agroforestry systems. Agric. Ecosyst. Environ. 247, 98–111 (2017).Article 
CAS 

Google Scholar 
European Commission. Commission regulation (EC) No 1120/2009 (EC, 2009).Piñeiro, V. et al. A scoping review on incentives for adoption of sustainable agricultural practices and their outcomes. Nat. Sustain. 3, 809–820 (2020).Article 

Google Scholar 
Kay, S. et al. Agroforestry is paying off – economic evaluation of ecosystem services in European landscapes with and without agroforestry systems. Ecosyst. Serv. 36, 100896 (2019).Article 

Google Scholar 
European Council. Council agrees its position on the next EU common agricultural policy. Press release. https://www.consilium.europa.eu/en/press/press-releases/2020/10/21/council-agrees-its-position-on-the-next-eu-common-agricultural-policy/ (2020).IUSS Working Group WRB. World Reference Base for Soil Resources 2014. International Soil Classification System for Naming Soils and Creating Legends for Soil Maps (FAO, 2014).Garland, G. et al. A closer look at the functions behind ecosystem multifunctionality: a review. J. Ecol. 109, 600–613 (2021).Article 

Google Scholar 
Naumann, C. & Bassler, R. Die Chemische Untersuchung von Futtermitteln 3. Auflage (Chemical Analysis of Feedstuff 3rd Edition) (VDLUFA-Verlag, 1976).Beule, L., Lehtsaar, E., Rathgeb, A. & Karlovsky, P. Crop diseases and mycotoxin accumulation in temperate agroforestry systems. Sustainability 11, 2925 (2019).Article 
CAS 

Google Scholar 
Verwijst, T. & Telenius, B. Biomass estimation procedures in short rotation forestry. For. Ecol. Manag. 121, 137–146 (1999).Article 

Google Scholar 
Harris, D., Horwáth, W. R. & van Kessel, C. Acid fumigation of soils to remove carbonates prior to total organic carbon or carbon-13 isotopic analysis. Soil Sci. Soc. Am. J. 65, 1853–1856 (2001).Article 
CAS 

Google Scholar 
Blake, G. & Hartge, K. in Methods of Soil Analysis: Part 1 – Physical and Mineralogical Methods 363–375 (Americal Society of Agronomy, Inc., 1995).Davidson, E. A., Hart, S. C., Shanks, C. A. & Firestone, M. K. Measuring gross nitrogen mineralization, and nitrification by 15N isotopic pool dilution in intact soil cores. J. Soil Sci. 42, 335–349 (1991).Article 
CAS 

Google Scholar 
Tiessen, H. & Moir, J. O. in Soil Sampling and Methods of Analysis Ch. 25 (CRC Press, 1993).Beule, L. et al. Poplar rows in temperate agroforestry croplands promote bacteria, fungi, and denitrification genes in soils. Front. Microbiol. 10, 3108 (2020).Article 

Google Scholar 
Ando, S. et al. Detection of nifH sequences in sugarcane (Saccharum officinarum L.) and pineapple (Ananas comosus [L.] Merr.). Soil Sci. Plant Nutr. 51, 303–308 (2005).Article 
CAS 

Google Scholar 
Singh, J., Singh, S. & Vig, A. P. Extraction of earthworm from soil by different sampling methods: a review. Environ. Dev. Sustain. 18, 1521–1539 (2016).Article 

Google Scholar 
Brookes, P. C., Landman, A., Pruden, G. & Jenkinson, D. S. Chloroform fumigation and the release of soil nitrogen: a rapid direct extraction method to measure microbial biomass nitrogen in soil. Soil Biol. Biochem. 17, 837–842 (1985).Article 
CAS 

Google Scholar 
Shen, S. M., Pruden, G. & Jenkinson, D. S. Mineralization and immobilization of nitrogen in fumigated soil and the measurement of microbial biomass nitrogen. Soil Biol. Biochem. 16, 437–444 (1984).Article 
CAS 

Google Scholar 
Marx, M.-C., Wood, M. & Jarvis, S. C. A microplate fluorimetric assay for the study of enzyme diversity in soils. Soil Biol. Biochem. 33, 1633–1640 (2001).Article 
CAS 

Google Scholar 
Matson, A. L., Corre, M. D., Langs, K. & Veldkamp, E. Soil trace gas fluxes along orthogonal precipitation and soil fertility gradients in tropical lowland forests of Panama. Biogeosciences 14, 3509–3524 (2017).Article 
CAS 

Google Scholar 
Wen, Y., Corre, M. D., Schrell, W. & Veldkamp, E. Gross N2O emission and gross N2O uptake in soils under temperate spruce and beech forests. Soil Biol. Biochem. 112, 228–236 (2017).Article 
CAS 

Google Scholar 
McKenzie, N. J., Green, T. W. & Jacquier, D. W. in Soil Physical Measurement and Interpretation for Land Evaluation 150–162 (Csiro Publishing, 2002).Priesack, E. Expert-N model library documentation. https://expert-n.uni-hohenheim.de/en/documentation (2005).Formaglio, G., Veldkamp, E., Duan, X., Tjoa, A. & Corre, M. D. Herbicide weed control increases nutrient leaching compared to mechanical weeding in a large-scale oil palm plantation. Biogeosciences 17, 5243–5262 (2020).Article 
CAS 

Google Scholar 
Kroetsch, D. & Wang, C. in Soil sampling and methods of analysis (eds Angers, D. A. & Larney, F. J.) 713–725 (CRC Press, 2008).Kurniawan, S. et al. Conversion of tropical forests to smallholder rubber and oil palm plantations impacts nutrient leaching losses and nutrient retention efficiency in highly weathered soils. Biogeosciences 15, 5131–5154 (2018).Article 
CAS 

Google Scholar 
Markwitz, C. Micrometeorological Measurements and Numerical Simulations of Turbulence and Evapotranspiration over Agroforestry (University of Göttingen, 2021).Jarrah, M., Mayel, S., Tatarko, J., Funk, R. & Kuka, K. A review of wind erosion models: data requirements, processes, and validity. Catena 187, 104388 (2020).Article 

Google Scholar 
van Ramshorst, J. G. V. et al. Reducing wind erosion through agroforestry: a case study using large eddy simulations. Sustainability 14, 13372 (2022).Article 

Google Scholar 
Kanzler, M., Böhm, C., Mirck, J., Schmitt, D. & Veste, M. Microclimate effects on evaporation and winter wheat (Triticum aestivum L.) yield within a temperate agroforestry system. Agroforest Syst. 93, 1821–1841 (2019).Article 

Google Scholar 
Clough, Y. et al. Land-use choices follow profitability at the expense of ecological functions in Indonesian smallholder landscapes. Nat. Commun. 7, 13137 (2016).Article 
CAS 

Google Scholar  More

Our study is focussed on population trends of large carnivores; a culturally important group32, essential for regulating ecosystem function33. Large carnivores represent an important study group as their population status is unclear, with reports of devastating declines33 contrasted with remarkable recoveries23. Further, as a well-studied taxa with abundant trend and trait datasets, large carnivores present a good system to evaluate important drivers of trends without being impacted by poor inference from missing data34. Finally, as large carnivores are considered indicator species of the overall status of biodiversity within an area35, our inference may provide insight beyond our focal taxa.Population trendsWe sourced population (defined by the authors of the original studies, who reported on population trends for one or more studied groups of individuals) trend information for species in the families Canidae, Felidae, Hyaenidae, and Ursidae of the order Carnivora from two large trend datasets: CaPTrends12 and the Living Planet Database13. The CaPTrends database is the product of a semi-systematic literature search for population trends of large carnivore species (from the families listed above); the dataset possesses trend information for 50 species from locations around the world, and trends are reported in a variety of ways. The Living Planet Database contains population abundance time-series for vertebrates from thousands of sites around the world and is one of the larger population trend datasets. Combined, these datasets produce a cumulative 1123 trends (after removing duplicates and records we deemed unreliable or unsuitable), derived from >10,000 individual population estimates. In the Living Planet Database, and for most records in CaPTrends, trends are reported as a time-series of abundance (or density) estimates. We modelled these time-series with log-linear regressions, where abundance (the response) was loge transformed, and year of abundance estimates was selected as the predictor. We included a continuous Ornstein-Uhlenbeck (OU) autoregressive process to control for temporal autocorrelation in these models. The OU process estimates covariance between abundance values, under the assumption that abundances in time point 1 will be more similar to abundances in time point 2, than time-point 3, 4, 5, etc. Accounting for covariance resolves non-independence within time-series. We extracted the slope coefficient which represents the annual instantaneous rate of change, sometimes called the population growth rate (rt). Alongside the abundance time-series, CaPTrends also has three other quantitative datatypes, all of which we converted into an annual instantaneous rate of change (rt): (1) a mean finite rate of change; (2) estimates of percentage abundance change between two points in time; and (3) time-series’ of population change estimates (e.g. in year 1 the population doubled and in year 2 it halved). We converted all annual instantaneous rates of change into an annual rate of change percentage to improve interpretability. These annual rates of change ranged from −75 to 68%, but the majority of values fell within −10 to 10% (Supplementary Fig. 1a).Alongside the quantitative records, 138 populations in the CaPTrends dataset were only described qualitatively with categories: Increase, Stable, and Decrease. These qualitative records were more common for populations located in traditionally poorer-sampled countries (e.g. with lower human development), so whilst they are less informative (only describing the direction and not the magnitude), we deem them important to reduce bias (Fig. 1). As a result, we used a combination of percentage annual rates of change (N = 985) and qualitative categories (N = 138) as our response in our model (see below), representing 50 large carnivore species.CovariatesFor each population, we extracted sixteen covariates (each z-transformed) that fell into four categories: land-use, climate, governance, and traits. Our covariates were designed to cover a diverse array of factors that could influence population trends in large carnivores (Supplementary Table 1). Each covariate is described briefly in Fig. 2 with full descriptions of how variables were derived in the Supplementary material: Covariates.One of the challenges in identifying how covariates—which can vary in space and time—impact population trends is matching the spatial and temporal scale of the covariate with the population i.e. how much of the population is affected by the covariate at a given point in time. To tackle the spatial element of this problem, we used data on the area of extent of each population (e.g. how large is the spatial extent of the population or monitoring zone) to generate a circular distribution zone around the population’s coordinate centroid. We refer to this as the ‘population area’ hereafter. We then sampled covariate values within each population area, with more sampling points in larger areas (range: 13–295 sampling points, Supplementary Fig. 2b). For covariates which varied over time, we extracted the covariates across the ‘population monitoring period’, which refers to the period (from start to end year) the population was monitored for. However, as evidence suggests there can be a lag period between impact or change and any detectable changes in population abundance3, we tested how 0-, 5-, and 10-year lags in covariates changed model fits and effect sizes. We implemented these lags by extending the start of the population monitoring period backwards for each given lag e.g. for a 10-year lag, a normal population monitoring period of 1990–2000, would then capture covariates between 1980–2000. Sensitivity analysis showed a 10-year lag had the greatest balance of improved model fit, with high taxonomic and spatial coverage (see Supplementary: Sensitivity analysis).ModellingAt its core, our model is a linear mixed effects model, regressing annual rates of change against a combined 23 covariates and interactions, using random intercepts to account for phylogenetic and spatial nesting. The model was written in BUGS language and implemented in JAGS 4.3.036 via R 4.0.337. The model structure is summarised below, with a full description in Supplementary: Modelling.ResponseWe modelled our annual rate of change response with a normal error prior. However, to allow the two different types of population trend data (quantitative rates of change and qualitative descriptions of change) to be included in the same model, we treated the qualitative records as partially known. Specifically, we censored the qualitative records to indicate that the true value is unknown, but it occurs within a specified range, with annual rate changes ranging from −50 to 0%, −5 to 5%, and 0 to 50% within the decrease, stable and increase categories, respectively. The overlapping nature of these thresholds is by design, as we wanted to acknowledge that there is likely a grey area between the different categories. For instance, in one study, a 2% trend could be called stable, whilst a different study would consider this as increasing, our overlapping thresholds address this grey area. Admittedly, our category thresholds were arbitrarily selected—this is as a consequence of there being no strict rules on what population change is needed to be assigned a given category. However, despite being arbitrary, they were still carefully selected. For instance, our censoring range thresholds are similar to the range of the observed change (−75 to 68%). Further, whilst we don’t have a clear definition for what an increasing or decreasing population looks like (is it 1% or 10%), we can be confident that increasing and decreasing populations will fall above and below 0%, respectively. The stable category is most vulnerable to subjectivity, and so without clear definitions, we set a large range e.g. the maximum and minimum value we considered could be plausibly called stable was 5% and −5%, respectively.Many of the qualitative and short-term (brief monitoring period) quantitative records address known data biases as they occur in less-well represented regions, species, and time-periods (Fig. 1). However, these lower quality records can be more prone to error. As a result, we developed a weighting term within the model to inflate uncertainty around trends derived over a short timeframe, with few abundance observations, and less robust methods—see Supplementary: Modelling—Weighted error.CovariatesPrior to modelling, we identified missing values in some covariates (e.g. some species were missing Maximum longevity values), which can be problematic for inference if ignored34. We used imputation approaches38,39 to predict these missing values and recorded the associated imputation uncertainty alongside these predictions. Within our model, we accounted for uncertainty in the imputed estimates by treating imputed values of the covariates as distributions instead of point estimates. Specifically, for each imputed value we assigned a normal distribution defined by the mean and standard deviation of the imputed estimates. This approach allowed us to capture imputation uncertainty and improve inference robustness.With 16 covariates and a further seven interactive effects (23 effects in total), we were conscious of overparameterizing the model. As a result, we split these parameters into three groups: (1) core parameters—which included main effects that were considered likely drivers of population change; (2) optional parameters—which included main effects we considered interesting but with little evidence to-date of any influence on trends; and (3) interaction parameters—which includes the seven proposed interaction terms. We included our core parameters (Change in human density, Primary land loss, Population area, Body mass, Change in extreme heat, Governance, and Protected area coverage) in every model, but used Kuo and Mallick variable selection40 to identify parameters from the optional and interaction groups that improve model fit whilst balancing the risk of overfitting.Random interceptsWe used a hierarchical model structure to account for phylogenetic and spatial non-independence in the data, including species as a random intercept nested with genus, and country as a random intercept nested within sub-regions, as defined by the United Nations (https://www.un.org/about-us/member-states).Model runningWe ran the full model through three chains, each with 150,000 iterations. The first 50,000 iterations in each chain were discarded, and we only stored every 10th iteration along the chain (thinning factor of 10). We opted for a large chain and burn-in due to the model complexity, and to allow a broad selection of parameter combinations to be tested under variable selection. We assessed convergence of the full model on all parameters monitored in the sensitivity analysis, as well as the model intercept, and all 23 main and interactive effect slope coefficients. We checked the standard assumptions of a mixed effect linear model (normal residuals and heterogeneity of variance), and tested the residuals to ensure there was no spatial (Moran’s test) or phylogenetic (Pagel’s lambda) autocorrelation. We also conducted posterior predictive checks to ensure independently simulated values were broadly reminiscent of model predicted values.We report the median slope coefficient and associated credible intervals for each of the main and interactive effects, and produce marginal effect plots for a selection of important parameters. These marginal effects hold all other covariates at zero (which is the equivalent of the mean, as covariates were z-transformed).LimitationsDeveloping macro-scale models of population change is challenging as response data are biased41 and hard to summarise42, and response-covariate relationships are likely complex and numerous2. Within our workflow, we attempted to address these challenges, and overall, this allowed us to achieve a moderate model fit (conditional R2 ~ 0.4). We minimised biases in the trend data by integrating qualitative trends with quantitative estimates, which allowed us to increase the taxonomic and spatial scale of the work. However, biases are likely still present to some extent. For instance, whilst we have population trend data covering the full parameter space of our most influential variable (change in human development), we have more population trends in high human development countries (Supplementary Fig. 20)—given these biases, caution should be used when interpreting results. While we could not avoid some biases, we found inference was similar across different fragments of the data and model structures (Supplementary results: Sensitivity analysis). We also attempted to capture complexity by covering a more comprehensive array of covariates than many previous analyses, but we still lack data on likely important aspects that are cryptic and difficult to measure (e.g. poaching, persecution, culling, and the conservation benefits of being flagship species). Further, there are temporal lags between disturbance-events and observable changes in the population10 and we tested several to incorporate the lag that maximised model fit. However, it is possible that responses to different types of disturbance (e.g. habitat loss and climate change) have different lags, although this has not been quantified. Long lags (the maximum lag we explored was 10-years) may also occur and be associated with slow recoveries, but an absence of longer temporal extents in the response and covariate data largely prohibits this analysis at global scales (long temporal extent data is less available outside of the global north).Counterfactual scenariosTo explore how observed changes in land-use, climate and human development have influenced population trends, we developed three counterfactual scenarios, where we compared observed population change to predicted population change if habitat, climate, and human development remained static. For instance, in the climate change counterfactual scenario, we predicted each population trend using the global model (all covariate parameters) with available covariate data (e.g. land-use, governance and trait covariates), as well as taxa and location data (to provide sensitivity to the models varying random intercepts), but set the climate change covariate data to zero (in this case, change in extreme heat and change in drought). We then subtracted these counterfactual predictions from the observed trends to define ‘Difference in annual rate of change (%)’, whereby a positive value indicates carnivore populations would be in better shape (fewer declines) under the counterfactual scenario, and vice-versa. We summarise counterfactual scenarios by reporting the median Difference in annual rate of change and 95% quantiles across the observed 1123 populations.Socioeconomic development and non-linearity in carnivore trendsGiven the large effect of human development change on carnivore population trends within our counterfactual scenarios, we further explore the potential impacts of human development change (i.e. changes in the socioeconomic standards of society) on the dynamics of potential carnivore abundance change. Specifically, we test how changing the rate of human development growth of a hypothetical low human development country could impact carnivore abundances. We test this by simulating time series of human development change between the years 1960 and 2020 along three common development pathways for low human development countries, each given: (1) a mean rate of change in human development (%) defined as Slow (1.25%), Moderate (1.5%) and Fast (1.75%); (2) a shared deceleration rate set to −0.02% per year—a key feature of the human development data is that as human development grows, its growth rate decreases; and (3) a shared initial human development value which we set as 0.2 (a hypothetical low human development country) at year 1960 (Fig. 4a). All our selected parameter values are representative of the human development data (Supplementary Fig. 2), with the Moderate pathway being largely typical for a country with an initial human development value of 0.2, while Slow and Fast represent plausible extremes.We then used our fitted model (Fig. 2) to evaluate how the three pathways of Change in Human development would affect annual abundance of a hypothetical carnivore. This involved predicting the annual rate of change in abundance using the Change in human development pathways and the marginal effect of the Change in human development parameter from the fitted model—setting all other covariates in the model to zero, which in our z-transformed variables represents the mean. We then used the predicted annual rates of change in abundance to project carnivore abundance up to the year 2020, from an arbitrary baseline abundance of 100 in the year 1960 (Fig. 4c). These projections capture the 95% credible intervals around the human development change model coefficient, and assume constant and average values for all other effects (e.g. primary habitat loss or climate change).Reporting summaryFurther information on research design is available in the Nature Portfolio Reporting Summary linked to this article. More

At December’s United Nations Convention on Biological Diversity summit (COP15), an insidious threat emerged to national parks — even as scientists argued for expanding protected areas. The World Travel & Tourism Council wants commercial tourism to be allowed to build developments in national parks globally, without obligation to help finance park conservation (see go.nature.com/3x2fsi9). This would undermine existing private tourism developments that do support conservation.
Competing Interests
The authors declare no competing interests. More

StatementsWe report our study in accordance with ARRIVE guidelines.Structure of artificial turf of ICUA schematic illustration of a ground plan of the artificial turf sports field of the ICU is shown in Fig. 1. This artificial turf was installed in 2013 by Japanese company B. The field is surrounded by ditches, and there are three drains that connect to sewer pipes. The artificial turf field of TGU was installed in 2011 by Japanese company C.Characterization of rubber tips of artificial turf field of ICU and TGURT were collected from the artificial fields of ICU and TGU. RT for the artificial turf field of ICU were made of residual part of rubber for making tires, window frames and windshields of automobiles. RT of ICU consists of a mixture of EPDM (ethylene-propylene-diene) and SBR (styrene-butadiene rubber) (personal communication from a Japanese company B). The RT of TGU was made of rubber of the residual part of rubber for making tires, window frames, etc. (personal communication from a Japanese company C). Information on raw material of the RT was not manifested.RT collected from the fields (ICU and TGU) was sieved to estimate the particle sizes. The RT of the ICU varied from 0.053 to 3.35 mm, and that of TGU varied from 0.212 to 3.35 mm. The specific gravity of the RT was obtained as follows: A certain amount of RT was weighed and poured into a 10 ml graduated cylinder containing some water. The total volume of the RT was obtained by measuring the rise in the meniscus of the water. The specific gravities of the tested RT were 1.28 (ICU) and 1.28 (TGU). Elemental analyses of RT (ICU and TGU) were conducted using micro-PIXE line analysis47, and calcium, sulfur, zinc, and iron were detected, but lead was under the detection limit from the RT of both ICU and TGU.Sampling of sediments in the ditches of the fieldTo examine the migration of RT from the field to the ditches, approximately 200 g of sediments in the ditches was sampled at four different sites, D1–D4 (Fig. 1), in the ICU. The ditch surrounding the field is made by connecting U-shaped concrete blocks and concrete lids. The inner width, length, and depth of the block are 24, 60, and 24 cm, respectively. The size of the lid is 33 × 60 × 4.5 cm with 1.5 × 9.0 cm snicks at short sides, which make an opening of the ditch of 3.0 × 9.0 cm size between two lids.Each sample was weighed (wet weight) and washed with water using a fine sieve to remove the soil. After the removal of the soil, the sediment was dried, and RT was collected manually. The collected RT was weighed, and the percentages of RT in the sediments were calculated (weight/weight).Goldfish and crucian carpA common variety of goldfish C. auratus of different sizes were obtained from a fish merchant in Saitama Prefecture and from a pet shop in Tokyo and then kept in the ICU. Approximately 200 fish of four different sizes (large, body weight (BW) ~ 100 g; medium, BW, ~ 30 g; small-medium, BW, ~ 15 g; small, BW ~ 4.0 g) were kept in three 800-L stock tanks maintained at 20 °C under a 16-h light/8-h dark (16 L/8 D) photoperiod (lights on at 06:00). Small body size fish were kept in a floating cage in one of the stock tanks. The fish were fed commercial floating goldfish feed (Itosui) once a day ad libitum. The fish stock tanks had circulation filtration systems equipped with sand filters. The filter was cleaned every week to maintain the water quality. The health condition of the fish was judged by their appetite. All the experimental fish (mixed sex) in the present study were kept in stock tanks for over two weeks before they were used for experiments. A total of 127 goldfish were used for the present study. The sample size of each experiment was determined by the results of preliminary experiments. Our preliminary survey confirmed that the fish feed we used did not contain RT-like substances. Therefore, the sample sizes of the control groups (goldfish) were smaller than those of the experimental groups to sacrifice fewer fish. All goldfish and crucian carp experiments were conducted in the ICU.Approximately 30 wild juvenile crucian carp C. auratus subsp. 2 weighing 1.4–4.6 g were obtained from a fish merchant in Saitama Prefecture and kept in an 800-L stock tank in the same conditions as that for goldfish. A total of 16 crucian carp were used for the present study.For the experiments, fish were transferred from the stock tanks to experimental 60-L glass aquaria, which were maintained at 20 °C under a 16-h light/8-h dark (16 L/8 D) photoperiod (lights on at 06:00). The experimental aquaria had a running water system, and dechlorinated tap water was added at 20 ml/min. Plastic box filters were also set to each experimental aquarium to maintain water quality. When stock fish were transferred to experimental aquaria, fish were randomly allocated to the aquaria. All the methods for using goldfish and crucian carp were performed in accordance with the guidelines of the Animal Experimentation Committee of International Christian University. The conduct of the present study was approved by the Animal Experimentation Committee of International Christian University.Co-ingestion of feed and RT by goldfish of three different body sizesWe examined whether RT are ingested by goldfish with feed and whether the body size of fish affects the ingestion of RT using three different body sizes of fish, large, medium, and small. First, we conducted an experiment using large body size fish (N = 24; BW, 91.9 ± 21.6 g, mean and SD). Three goldfish of large body size were transferred from the stock tank to the experimental 60-L glass aquarium and kept for three days for acclimation of the environment and sinking fish feed. Fish were fed 3.0 g of large-size feed (Japan Pet Design Co. Ltd.) once a day. On the fourth day, fish were fed a mixture of RT collected from the field (ICU, 300 mg) and large feed (3.0 g). Control fish were fed only fish feed. At 90 min after feeding, the fish were transferred to a pail containing 0.05% 2-phenoxyethanol solution and deeply anesthetized. After body weight measurement, fish were dissected. We observed the intestine to determine whether RT was ingested. When RT was observed in the intestine, we collected the tips and counted the number of tips in each fish. The experimental tests were repeated eight times, and the data were combined.Second, we conducted an experiment using medium body size fish (N = 24; BW, 30.4 ± 12.4 g). Three goldfish of medium body size were transferred from the stock tank to the experimental 60-L glass aquarium and kept for three days for acclimation. Fish were fed 1.0 g of medium-size feed (Kyorin) once a day. On the fourth day, fish were fed a mixture of RT (ICU, 300 mg) and medium feed (1.0 g). Control fish were fed only fish feed. At 90 min after feeding, fish were anesthetized and dissected, and then the intestine was observed as described above. The experimental tests were repeated eight times, and the data were combined.Third, we conducted an experiment using small body size fish (N = 40; BW, 4.4 ± 1.5 g). Four goldfish of small body size were transferred from the stock tank to the experimental 60-L glass aquarium and kept for three days for acclimation. Fish were fed 0.5 g of small-size feed (Kyorin) once a day. On the fourth day, fish were fed a mixture of RT (ICU, 300 mg) and small feed (0.5 g). Because of the small size of fish, RT of small size particles (212–500 µm) were collected with sieves and used for the tests. Control fish were fed only fish feed. At 90 min after feeding, fish were anesthetized and dissected, and then the intestine was observed as described above. The experimental tests were repeated ten times, and the data were combined.In the first three experiments, all three control groups showed no ingestion of RT. From the results of the three experiments, it was clear that our experimental system was not contaminated with RT. Therefore, we omitted making control groups for further experiments to decrease the number of fish sacrificed from the standpoint of fish welfare.Fourth, we examined whether RT collected from TGU was ingested by goldfish. We conducted an experiment using large body size fish (N = 12; BW, 140.3 ± 27.0 g). Three goldfish of large body size were transferred from the stock tank to the experimental 60-L glass aquarium and kept for three days for acclimation. Fish were fed 3.0 g of large-size feed once a day. On the fourth day, fish were fed a mixture of RT (TGU, 300 mg) and large feed (3.0 g). At 90 min after feeding, fish were anesthetized and dissected, and then the intestine was observed as described above. The experimental tests were repeated four times, and the data were combined.We conducted an additional experiment with a similar design to those of the four experiments to take photographs of the fish and RT using fish of small-medium body size (N = 9; BW, 12.8 ± 2.7 g). Three fish were transferred from the stock tank to the experimental 60-L glass aquarium and kept for two days for acclimation. Fish were given 0.5 g of medium-size feed once a day. On the third day, fish were given a mixture of RT of ICU (30 pieces; size 0.5–1.0 mm) and medium feed (0.5 g). At 60 min after feeding, fish were anesthetized and dissected, and photographs of RT in the intestine were taken. The experimental tests were repeated three times, and the data were combined.Active ingestion of RT by goldfishWe examined whether goldfish actively ingest RT when RT are given without fish feed using large body size fish (N = 9; BW, 122.4 ± 20.8 g). Three fish were transferred from the stock tank to the experimental 60-L glass aquarium and kept for three days for acclimation. Fish were fed 3.0 g of large-size feed once a day. On the fourth day, fish were given 300 mg of RT (ICU) on the bottom of the aquarium. At 90 min after the placement of RT, fish were anesthetized and dissected, and then the intestine was observed as described above. The experimental tests were repeated three times, and the data were combined.Retention and elimination of ingested RT in the intestine of goldfishWe examined how long RT was retained in the intestine using large body size goldfish (N = 9; BW, 101.6 ± 11.4 g). Three goldfish were transferred from the stock tank to the experimental 60-L glass aquarium and kept for three days for acclimation. Fish were fed 3.0 g of large-size feed once a day. On Day 4, fish were given 1.0 g of RT (ICU). At 90 min after the placement of RT, each fish was individually transferred to three experimental 60-L glass aquaria. Then, each fish was fed 1.0 g of the feed. At 24 and 48 h (Day 5 and Day 6) after the transfer, we collected feces from fish and some water from the bottom of the aquaria. We observed whether RT was eliminated from the fish into the aquaria. When RT was observed in the feces and the bottom of the aquarium, we collected the RT and counted the number of RT. On Day 5, after the RT observation, each fish was fed 1.0 g of the feed. On Day 6, after RT observation in feces and water, the fish were anesthetized and dissected. We observed whether the intestine retained RT. The experimental tests were repeated three times, and the data were combined.Ingestion of RT by wild crucian carpWe examined whether wild Japanese crucian carp ingest RT. The experiment was conducted using juvenile crucian carp (N = 16, BW, 2.8 ± 0.9 g). Sixteen fish were transferred from the stock tank to three experimental 60-L glass aquaria (5 or 6 fish per aquarium) and kept for six days for acclimation. Fish were fed with 0.2 g of small-size feed once a day. On the seventh day, fish were fed a mixture of RT (ICU, 30 mg) and the small feed (0.2 g) or RT alone (30 mg). Because of the small size of fish, RT of small size particles (212–500 µm) were collected with sieves and used for the test. Control fish were fed only fish feed (0.2 g). At 6 h after feeding, fish were anesthetized and dissected, and then the intestine was observed as described above. More

There is a growing impetus to leverage our fundamental understanding of microbial community assembly towards applied problems. With microbes contributing to diverse physiological, biogeochemical, and agricultural processes, the potential to control and optimise microbial communities holds promise for interventions ranging from industrial and environmental remediation to human medicine and biofuel production [1, 2]. Realising this goal is contingent on high fidelity between theory, experiments, and the natural dynamics of target systems.Theoretical and experimental research in microbial community optimisation has largely proceeded along two parallel paths. Theoretical approaches leverage mathematical models and metabolic networks to predict which species combinations are stable and how they can optimise a given function (e.g., maximum biomass, waste degradation or host health) [3,4,5,6,7]. Experimental studies often take a combinatorial approach, iteratively assembling different species combinations in vitro and evaluating their stability and functional attributes [8,9,10,11]. Both theory and experiments are valuable but they are also susceptible to their own modus operandi that may limit their correspondence and their translation to real-world systems. On the one hand, theoretical approaches typically adopt the analytical tractability of steady state dynamics, where microbial consumers and the resources on which they depend are assumed to establish a stable equilibrium. On the other hand, experimental approaches almost exclusively embrace the high-throughput efficiency of serial-batch culture, where consumers and resources are made to fluctuate over several orders of magnitude with each serial passage. This raises an important question: should we expect unity in the composition of optimised communities emerging under continuous resource supply (e.g., chemostat) versus the discrete pulsed resource supply of, for example, serial-batch culture?To explore how microbial community composition varies under contrasting resource supply dynamics, we performed simulations of a classical resource-competition model:$$frac{{dN_i}}{{dt}} = N_ileft( {mathop {sum}limits_{j = 1}^n {mu _{ij}left( {R_j} right) – m} } right)$$
(1)
$$frac{{dR_j}}{{dt}} = {Psi}_jleft( {R_j} right) – mathop {sum }limits_{i = 1}^n Q_{ij}mu _{ij}left( {R_j} right)N_i,$$
(2)
where Ni is the population density of consumer i, Rj is the concentration of resource j, μij(Rj) is the per capita functional response of consumer i, m is the per capita mortality rate due to dilution, Ψj(Rj) is the resource supply function, and Qij is the resource quota of consumer i on resource j (amount of resource per unit consumer). The consumer functional response is given by the Monod function, (mu _{ij}(R_j) = mu _{max_{ij}}frac{{R_j}}{{K_{s_{ij}} + R_j}}) , where (mu _{max_{ij}}) is the maximum growth rate and (K_{s_{ij}}) is the half saturation constant for consumer i on resource j.To set up the simulations, we randomly sampled the parameters of the Monod growth functions, (μmax and Ks) for five species competing for five substitutable resources (essential resources are treated separately in the supplementary information, with similar findings). In one set of parametrisations (n = 100 unique competitor combinations) we used both random μmax and Ks, and in another set (n = 100) we imposed a trade-off in maximum growth rate and substrate affinity (( {frac{{mu _{max}}}{{K_s}}} )) (Fig. 1a). The rationale for imposing a trade-off is that metabolic theory predicts that organisms that invest energy into a high maximum growth rate will have lower substrate affinities and vice versa [12, 13]. To ensure reasonable growth rates relative to the time-scale of resource pulsing, we sampled μmax such that minimum doubling times spanned from 21 to 52 min (when all resources are non-limiting). For each of the random competitor combinations, we simulated resources under continuous or pulsed resource supply with resource replenishment every 1/2, 1, 2, 4, 12, or 24 h. Under pulsed resource supply, Ψj(Rj) and m are removed from Eq. (1) and (2) and replaced by discontinuous resource pulsing and cell transfer at fixed intervals. The total resource flux (and mortality) was held constant under all frequencies of resource supply i.e., less frequent replenishment corresponds to larger resource pulses (see Supplementary Information for full model/simulation specifications).Fig. 1: Quantifying compositional overlap between communities assembled under continuous vs. pulsed resource supply.a Per capita growth responses (Monod functions) from a single iteration of the model assuming a trade-off between maximum growth rate and resource affinity (colours correspond to individual consumers). b Time series of consumers in a under different resource supply regimes. Numbers above individual panels reflect pulsing intervals in hours. The amplitude of population fluctuations increases with longer intervals between pulses, with distinct phases of growth, saturation, and instantaneous mortality visible at a finer temporal resolution (Fig. S10). c Example measure of compositional overlap (Jaccard similarity index) between communities assembled under continuous resource supply (far left panel in b) vs. pulsing every two hours (centre panel in b).Full size imageAfter allowing the competitors to reach a steady state (time-averaged over 24 h under pulsed treatments), we quantified the correspondence between the continuous supply treatment and the pulsed treatments using the Jaccard similarity index, (Jleft( {A,B} right) = frac{{left| {A cap B} right|}}{{left| {A cup B} right|}}) (0 ≤ J(A,B) ≤ 1), where the numerator gives the number of species (max = 5) that persist under continuous (A) and pulsed (B) resource supply, and the denominator gives the number of species (max = 5) that persist under continuous or pulsed resource supply (Fig. 1b, c).Under both sets of simulations (with and without enforcing a trade-off between maximal growth rate and resource affinity), we observe that the similarity in final community composition between continuous and pulsed resource supply decays with increasingly large intervals between resource replenishment (Fig. 2a). When no trade-off is imposed between maximum growth rate and resource affinity (orange line in Fig. 2a) the mean compositional similarity is only 0.68 when resources are pulsed every 2 h and down to 0.41 when resources are pulsed every 24 h (typical of serial-batch culture). The rate of decay in the Jaccard index is more severe when a trade-off is imposed between maximum growth rate and substrate affinity, to the extent that once pulsing intervals reach four hours there is almost zero overlap in community composition (blue line in Fig. 2a).Fig. 2: Impact of resource supply regime on community composition and abundance weighted mean trait values.a Compositional overlap (Jaccard similarity) between communities under continuous versus pulsed resource supply. Orange lines, points and circles denote model parametrisations with random sampling of both μmax and Ks; blue lines, points and circles denote model parametrisations with a trade-off imposed between μmax and resource affinity (( {frac{{mu _{max}}}{{K_s}}} )). Simulation parameters provided in the Supplementary Information. b Mean trait values for affinity and μmax averaged for each consumer across the five resources and weighted by their final abundance at the end of a simulation (cont. = continuous). In both a and b, small points (jittered for clarity) give the result of an individual simulation; large circles indicate the corresponding mean.Full size imageEcological theory provides an intuitive explanation for these observations. When resources are more continuously supplied, the better competitor is the one that can sustain a positive growth rate at the lowest concentrations of a limiting resource (i.e., has a higher resource affinity or lower R* in the language of resource competition theory [14]). In contrast, under increasingly pulsed resource supply, the better competitor is the one that can grow rapidly at higher resource concentrations. Having a high resource affinity (low R*) is of little benefit if resource concentrations fluctuate over large amplitudes because it only confers an ephemeral competitive advantage in the brief period before the resource is completely depleted (ahead of the next resource pulse). Instead, a high maximum growth rate is optimal because it allows the consumer to grow rapidly and quickly deplete a shared limiting resource. This high maximum growth strategy is, however, sub-optimal under continuous resource supply because a low R* strategist can draw the resource down and hold it at a concentration at which the maximum growth strategist is unable to maintain a positive growth rate.Looking at the mean trait values for resource affinity and μmax weighted by each consumer’s final abundance, it is indeed apparent that consumers with a higher affinity (averaged across the five resources) are favoured under continuous resource supply, while consumers with high maximum growth rates are favoured under pulsing intervals of increasing length (Fig. 2b). Enforcing this trade-off, therefore, leads to the rapid decline in compositional similarity we observe under resource pulsing. Notably, it also leads to a richness peak at intermediate pulsing intervals, where these alternative strategies have a higher probability of coexisting [15] (Fig. S1). At the same time, we still observe a decline in compositional similarity when μmax and Ks are randomly sampled independently of each other simply because the trade-off between maximum growth and resource affinity will emerge occasionally by chance. Two experimental tests of microbial community composition under continuous versus pulsed resource supply are consistent with these observations [16, 17].To evaluate the sensitivity of these observations to different assumptions, we ran additional simulations under various alternative model parameterisations and formulations. In brief, comparable trends to those described above are observed when: i) maximum growth rates are faster or slower than those presented in the main text (Figs. S2, S3); ii) all resources are assumed to be essential to growth (following Liebig’s law of the minimum) (Fig. S4); iii) a weaker trade-off is imposed between maximum growth and affinity (Figs. S5, S6); or iv) mortality is continuous rather than intermittent (Figs. S7, S8). We also investigated the relationship between observed compositional overlap and the dynamical stability under continuous resource supply, anticipating that more stable communities would tend to be more resistant to compositional shifts under resource pulsing. The reality appears more nuanced, namely that weaker dynamical stability at the limit of constant resource supply is associated with higher variance in compositional overlap under continuous vs. pulsed conditions (Fig. S9). In other words, systems with weaker stability are less predictable. A wide range of other microbial traits and trade-offs may interact unpredictably with the relationship between resource supply and community composition. The potential modulating role of system instabilities generated by cross-feeding interactions, non-convex trade-off functions, and the evolution of specialist versus generalist strategies present several especially valuable lines of enquiry [18,19,20].Although these observations are germane to any consumer-resource system, our emphasis here is on the emerging field of microbial community optimisation, where the practical implications are especially timely and important; namely, the resource supply regime must be tailored to the community being optimised. For example, wastewater treatment might be more appropriately modelled under continuous resource supply [21], whereas fermented food and beverage production may be more closely allied to the pulsed resource dynamics observed in batch culture [22]. Resource supply might also be manipulated to favourably modify the competitive hierarchy in an existing community (e.g., by regulating the rate of nutrient supply to the gut through meal timing). Indeed, there is emerging evidence that feeding frequency can drive significant changes in gut microbiota composition [23, 24]. Thus, resource supply dynamics should be considered both a constraint in the design of novel microbial communities and as a tuning mechanism for the optimisation of preexisting communities like those found in the human gut. More

Study site, field experiment, and soil samplingThis study relied on an experimental field station that simulates atmospheric CO2 enrichment and warming (Fig. 1a). The experimental site is located at Kangbo village (31°30′N, 120°33′E), Guli Township, Changshu municipality in Jiangsu Province, China. The climate type is a subtropical monsoon climate with a mean annual precipitation between 1100–1200 mm and an annual average temperature of approximately 16 °C. The high rainfall and temperature mainly occur from May through September25. The soils are Gleyic Stagnic Anthrosols derived from clayey lacustrine deposits. The main properties of the topsoil (0–15 cm) before the field experiment were as follows: pH (H2O) 7.0, bulk density 1.2 g cm−3, and contents of organic C and total N of 16.0 g kg−1 and 1.9 g kg−1, respectively.The FACE system has been well described in previous publications26,53. The climate change treatments were set according to the representative concentration pathway (RCP) scenario that modeled CO2 atmospheric concentration and temperature elevation in approximately 30–40 years. Elevated atmospheric CO2 concentration and warming of crop canopy air were maintained 24 h a day during the crop-growing period. Each treatment was replicated in three rings with the same infrastructure, and the rings were arranged in a split row design (Fig. 1b). All rings were buffered by adjacent open fields to avoid any treatment cross-over. Rice and wheat cultivations of all plots were managed with local conventional practices. Soil samples for qSIP incubation were collected in June 2020.Evaluation of nutrient pulses in soil after water additionTo determine if there is a nutrient pulse after soil rewetting, we estimated the dynamics of several biochemical characteristics (i.e., CO2 fluxes, hydrolase activity, DOC, DN, TC, and TN) in soils before and during the incubation. Soil samples for nutrient flux measurements were collected from the free-air CO2 enrichment and warming experimental station in July 2022.To measure the CO2 production rate, soil (2.00 g) was set in a 29 mm diameter glass vial (Volume 23 mL). Water (400 μL) was added evenly to soil in each vial. Gas samples were then collected from each incubation vial at multiple time points (i.e., 3 h, 6 h, 9 h, 12 h, 24 h, 72 h, and 144 h after water addition), and CO2 concentrations were determined in all gas samples with a Trace Gas Chromatograph (Agilent 7890, Santa Clara, CA, USA). Other biochemical characteristics were measured in parallel incubations, with destructive sampling. Briefly, soil (40.00 g) was set in a plastic jar, and 8 mL water was added evenly to the soil. All the incubation vials and jars were placed in the dark at 25 °C and soil moisture content was adjusted twice a day. Destructive sampling was conducted after 1, 3, and 6 days of incubation. Including prewet soil (the soil before water addition), a total of 48 soil samples were obtained. Soil samples were weighed before and after being oven dried at 105 °C, and soil moisture content was measured using the gravimetric method. Fluorescein Diacetate (FDA) hydrolase activity of soil was measured by using soil FDA hydrolase activity assay kit (Solarbio®, Beijing, China) according to the manufacturer’s protocol. Soil total carbon and total nitrogen were measured with a C/N elemental analyzer (multi EA® 5000, analytikjena, Germany). Dissolved organic carbon and dissolved nitrogen were analyzed using a TOC/TN analyzer (multi N/C® 3100, analytikjena, Germany) after extraction with distilled water.Simulating nutrient pulse after rewetting dry soil by adding 18O-waterTo determine whether and how microbial growth varied in response to pulse events after long-term acclimation to warming and CO2 enrichment, we estimated the population-specific growth rates of active microbes by conducting a 18O-water incubation experiment combined with DNA quantitative stable isotope probing (DNA-qSIP) (Fig. 1c). The incubation conditions were similar to those reported in a previous study6. In brief, approximately 60 g fresh soil of each treatment were sieved (2 mm) and air-dried (24 h at room temperature) immediately after transport to the laboratory. Then, triplicate samples of dry soils (2.00 g) were incubated in the dark at room temperature in sterile plastic aerobic culture tubes (17 × 100 mm) with 400 μL of 98 atom% H218O or natural abundance water (H216O) for 6 days, with harvests at four time points (T = 0, 1, 3, 6 d after water addition). DNA was extracted from the soils of 0 d incubation treatment immediately after water addition (~30 s interval), representing the prewet treatment. At each harvest, soils were destructively sampled and immediately stored at −80 °C. A total of 84 soil samples (4 climate change treatments × 3 replicates at 0 h with H216O addition +  4 treatments × 3 subsequent time points × 3 replicates × 2 types of H2O addition) were collected.DNA extraction and isopycnic centrifugationTotal DNA from all the collected soil samples was extracted using the FastDNA™ SPIN Kit for Soil (MP Biomedicals, Cleveland, OH, USA) according to the manufacturer’s instructions. The concentration of extracted DNA was determined fluorometrically using Qubit® DNA HS (High Sensitivity) Assay Kits (Yeasen Biotechnology, Shanghai, China) on a Qubit® 4 fluorometer (Thermo Scientific™, Waltham, MA, USA). The DNA samples of day 1, day 3, and day 6 were used for isopycnic centrifugation, and the detailed pipeline was described previously with minor modifications6. Briefly, 3 μg DNA were added into 1.85 g mL−1 CsCl gradient buffer (0.1 M Tris-HCl, 0.1 M KCl, 1 mM EDTA, pH = 8.0) with a final buoyant density of 1.718 g mL−1. Approximately 5.1 mL of the solution was transferred to an ultracentrifuge tube (Beckman Coulter QuickSeal, 13 mm × 51 mm) and heat-sealed. All tubes were spun in an Optima XPN-100 ultracentrifuge (Beckman Coulter) using a VTi 65.2 rotor at 177,000 × g at 18 °C for 72 h with minimum acceleration and braking.Immediately after centrifugation, the contents of each ultracentrifuge tube were separated into 20 fractions (~250 μL each fraction) by displacing the gradient medium with sterile water at the top of the tube using a syringe pump (Longer Pump, LSP01‐2A, China). The buoyant density of each fraction was measured using a digital hand-held refractometer (Reichert, Inc., Buffalo, NY, USA) from 10 μL volumes. Fractionated DNA was precipitated from CsCl by adding 500 μL 30% polyethylene glycol (PEG) 6000 and 1.6 M NaCl solution, incubated at 37 °C for 1 h, and then washed twice with 70% ethanol. The DNA of each fraction was then dissolved in 30 μL of Tris‐EDTA buffer. Detailed information (company names and catalog numbers) of the reagents and consumables for qSIP experiment is provided in Supplementary Data 2.Quantitative PCR and sequencingTotal 16S rRNA gene copies for DNA samples of all the fractions and the day-0 soils were quantified using the primers for V4-V5 regions: 515F (5′‐GTG CCA GCM GCC GCG G‐3′) and 907R (5′‐CCG TCA ATT CMT TTR AGT TT‐3′)54. Plasmid standards were prepared by inserting a copy of purified PCR product from soil DNA into Escherichia coli. The E. coli was then cultured, followed by plasmid extraction and purification. The concentration of plasmid was measured using Qubit DNA HS Assay Kits. Standard curves were generated using 10‐fold serial dilutions of the plasmid. Each reaction was performed in a 25-μL volume containing 12.5 μL SYBR Premix Ex Taq (TaKaRa Biotechnology, Otsu, Shiga, Japan), 0.5 μL of forward and reverse primers (10 μM), 0.5 μL of ROX Reference Dye II (50×), 1 μL of template DNA, and 10 μL of sterile water. A two-step thermocycling procedure was performed, which consisted of 30 s at 95 °C, followed by 40 cycles of 5 s at 95 °C, 34 s at 60 °C (at which time the fluorescence signal was collected). Following qPCR cycling, melting curves were obtained to ensure that the results were representative of the target gene. Average PCR efficiency was 96% and the average slope was −3.38, with all standard curves having R2 ≥ 0.99.The DNA of day-0 samples (unfractionated) and the fractionated DNA of fractions with buoyant density between 1.695 and 1.735 g/mL were selected for 16S rRNA amplicon sequencing by using the same primers of qPCR (i.e., 515F/907R). Eleven out of 20 fractions from each ultracentrifuge tube (density between 1.695 and 1.735 g/mL) were selected because they contained more than 99% gene copy numbers of the 20 fractions. A total of 804 DNA samples (12 unfractionated DNA samples + 72 × 11 fractionated DNA samples) were sequenced using the NovaSeq6000 platform (Genesky Biotechnologies, Shanghai, China).The sequences were quality-filtered using the USEARCH v.11.055. In brief, sequences  0.5 were removed. Chimeras were identified and removed. Subsequently, high-quality sequences were clustered into operational taxonomic units (OTUs) using the UPARSE algorithm at a 97% identity threshold, and the most abundant sequence from each OTU was selected as a representative sequence. The taxonomic affiliation of the representative sequence was determined using the RDP classifier (version 16)56. In total, 51,127,459 reads of the bacterial 16S rRNA gene and 11,898 OTUs were obtained. The 16S rRNA amplicon sequences were uploaded to the National Genomics Data Center (NGDC) Genome Sequence Archive (GSA) with accession number CRA006507.Quantitative stable isotope probing calculationsWe used the amount of 18O incorporated into DNA to estimate the growth rates of active taxa24,57. The density shifts of OTUs between 16O and 18O treatments were calculated following the qSIP procedures24. Briefly, the number of 16S rRNA gene copies per OTU in each density fraction was calculated by multiplying the OTU’s relative abundance (acquisition by sequencing) by the total number of 16S rRNA gene copies (acquisition by qPCR). Then, the GC content and molecular weight of a particular taxon were calculated. Further, the change in 18O isotopic composition of 16S rRNA genes for each taxon was estimated. We assumed an exponential growth model over the course of the incubations, and absolute population growth rates were estimated over each of the three time intervals of the incubation: 0–1, 0–3, and 0–6 d corresponding to the samplings at days 1, 3, and 6. The absolute growth rate is a function of the rate of appearance of 18O-labeled 16S rRNA genes. Therefore, the growth rate (g) of taxon i was calculated as:$${g}_{i}={{{{{rm{ln}}}}}}left(frac{{N}_{{{{{{rm{TOTAL}}}}}}{it}}}{{N}_{{{{{{rm{LIGHT}}}}}}{it}}}right)times frac{1}{t}$$
(1)
Where NTOTALit is the number of total gene copies for taxon i and NLIGHTit represents the unlabeled 16S rRNA gene abundances of taxon i at the end of the incubation period (time t). NLIGHTit is calculated by a function with four variables: NTOTALit, molecular weights of DNA (taxon i) in the 16O treatment (MLIGHTi) and in the 18O treatment (MLABi), and the maximum molecular weight of DNA that could result from assimilation of H218O (MHEAVYi)24. We further calculated the average growth rates (represented by the production of new 16S rRNA gene copies of each taxon per g dry soil per day) over each of the three time intervals of the incubation: 0–1, 0–3, and 0–6 d, using the following equation39:$$frac{d{N}_{i}}{{dt}}={N}_{{{{{{rm{TOTAL}}}}}}{it}}left(1-{e}^{-{g}_{i}t}right)times frac{1}{t}$$
(2)
Where t is the incubation time (d). All data calculations were performed using the qSIP package (https://github.com/bramstone/qsip) in R (v. 3.6.2).Grouping of taxa into growth strategiesWe compared the average growth rates of taxa at three time intervals (n = 3 in each time interval) and classified the species into rapid, intermediate, and slow growth strategies based on the timing of the maximum growth rate (Fig. 1d): (1) Rapid responders: Species had the highest growth rates by 1 day of the incubation; (2) Intermediate responders: Species had the highest growth rates at the 3-day incubation; (3) Slow responders: Species had the highest growth rates at the 6-day incubation. The taxa with growth rates significantly greater than zero can be divided into one of three strategies in each treatment.Analyses of phylogenetic conservationPhylogenetic tree analyses were performed in Galaxy/DengLab (http://mem.rcees.ac.cn:8080/) with PyNAST Alignment and FastTree functions58,59. The trees were visualized and edited using iTOL60. To estimate the phylogenetic patterns of growth strategies, phylogenetic dispersion (D) was calculated by the function ‘phylo.d’ of package “caper” in R (v. 3.6.2). The D statistic equal to 1 means the observed trait has a phylogenetically random distribution across the tips of the phylogeny, and 0 means the observed trait is as clumped as if it had evolved by Brownian motion29. Increasing phylogenetic clumping in the binary trait is indicated by values of D decreasing from 1. The p values were obtained by performing 1000 permutations to test D for significant departure from 1 (random distribution). Furthermore, the phylogenetic signal metrics Blomberg’s K and Pagel’s λ, were used to test for significantly nonrandom phylogenetic distributions using the package “phylosignal” in R. The p values of both indices were used to test the significance of phylogenetic signals.The nearest taxon index (NTI) was calculated to determine the degree of phylogenetic clustering as described previously5. The mean nearest taxon distance (MNTD) were calculated in the “picante” package of R61. The values of NTI are equivalent to the negative output of the standardized effect size (SES) of the observed MNTD distances, which test whether the distribution of growth strategies across different phylogenetic groups is random or nonrandom. NTI values > 0 and their p values < 0.05 represent phylogenetic clustering, while NTI values < 0 and p values > 0.95 indicate phylogenetic over-dispersion. The p values of NTI between 0.05 and 0.95 represent random phylogenetic distributions. The data were converted into binary matrices (1 s and 0 s) before all phylogenetic analyses. For the OTUs that were present in more than one treatment and had differing responses (in the analyses when all climate treatments are combined, i.e., in Table 1 and Supplementary Fig. 6), the OTU was classified to the respective growth responder when that taxa exhibited that specific growth strategy in any treatment.Quantification of explained variance in the distribution patterns of strategiesThe phylogenetic distance matrix obtained from the phylogenetic tree and three binary matrices (including four climate treatments) for three growth strategies were decomposed by the package ‘FactoMineR’ in R. The distribution matrices (binary) of three growth responders in the four climate treatments were used to represent the impact of CO2 and temperature on bacterial growth. To predict the microbial growth strategy (i.e., “y”, a 1 or 0 indicating the membership in a response category), variance partitioning was performed by using the first four phylogenetic principal components (PCs) (accounting for over 80% of phylogenetic variance) and two environment PCs (accounting for over 65% of habitat variance). The fraction of the variance explained by phylogenetic constraints and environmental acclimation was calculated by the package ‘car’ in R.Statistical analysesUncertainty of growth rates (95% confidence interval) was estimated using a bootstrapping procedure with 1000 iterations6. The cumulative growth rates at phylum-level were estimated as the sum of taxon-specific growth rates of those OTUs affiliated to the same phylum. The per capita growth rates of each OTU were calculated by dividing absolute growth rates by the total 16S rRNA gene abundance of taxon i. Ecological processes of active populations (e.g., density-dependence) after rewetting dry soil were estimated by correlating species-specific fitness (refers to per capita growth rates of each OTU) with initial population size using linear regression analyses (R v.3.6.2, ‘lm’ function). The beta diversity of the bacterial community at 0 d incubation was visualized by unconstrained principal coordinate analysis (PCoA) and tested by permutational multivariate analysis of variance (PERMANOVA) based on Bray-Curtis distance, using the vegan package in R (v. 3.6.2)62. Comparisons between climate scenarios were tested by two-way ANOVA and LSD post hoc tests (SPSS 19 for Windows, IBM Corp., Armonk, NY, USA).Reporting summaryFurther information on research design is available in the Nature Portfolio Reporting Summary linked to this article. More

The University of California, Los Angeles, suspended ecologist Priyanga Amarasekare without salary or benefits for one year, and will cut her salary by 20% for two more years.Credit: Al Seib/Los Angeles Times via Getty

In April of last year, the Ecological Society of America awarded Priyanga Amarasekare one of the highest honours in the field of ecology: the Robert H. MacArthur Award. A little over two months later, the University of California, Los Angeles (UCLA), placed Amarasekare on a one-year suspension without pay or benefits, and forbid her from accessing her laboratory, maintaining her insect colonies, managing her grants or contacting students. Now scientists from around the world, who call Amarasekare a “highly distinguished ecologist”, “a committed teacher and outstanding mentor” and a “tireless advocate for under-represented groups”, are calling for her reinstatement.
Scientists question Max Planck Society’s treatment of women leaders
The precise allegations that led to her suspension are unknown. UCLA has declined to release them, and barred Amarasekare from discussing the matter publicly. But long-standing tensions between Amarasekare and the university are no secret. A native of Sri Lanka and one of two women of colour who have tenure in the ecology and evolution department, she has previously accused the university of discrimination for repeatedly denying her promotions that were granted to colleagues. Former students and faculty members who are familiar with the situation think that Amarasekare’s suspension was retaliation for speaking out.Some 315 scientists raised concerns about her suspension in a petition that was delivered to the university on 23 January, arguing that Amarasekare “has long been denied significant advancement within her department, out of keeping with her contributions to the field”. Moreover, the sanctions levied against Amarasekare — including the one-year suspension and 20% salary reduction for an additional two years — represent “the kind of punishment normally applied only to the most egregious wrongdoings”, including scientific misconduct and sexual harassment violations, the petitioners write.In the absence of compelling evidence to the contrary, the scientists ask that UCLA rescind the disciplinary actions and fully compensate Amarasekare.Officials with UCLA say that the university “supports freedom of expression and does not condone retaliation of any sort”. They declined to discuss the accusations against or in support of Amarasekare, saying the university is “bound to respect the privacy of the numerous individuals involved in this matter”. Amarasekare also declined to comment.A confusing decisionColleagues told Nature that Amarasekare is the rare ecologist whose research spans the theoretical, computational and experimental realms. One project in her laboratory that touches on all of these areas focuses on the impact of climate change on insect communities. “She’s really several years ahead of everybody else,” says Andy Dobson, an ecologist at Princeton University in New Jersey who led the petition. Dobson has written letters to support Amarasekare’s various applications for promotion at UCLA and says he has been baffled by the university’s decisions. “She complained, and most of what’s happened seems to be a reaction against that,” he says.
Legal win for US scientist bolsters others caught in China crackdown
Nature spoke to several former students and faculty members who defended Amarasekare in administrative hearings in September 2021. Although none knew the specific details of the charges against her, they all thought she had been targeted for speaking out against what she saw as discrimination within the department. In particular, they said Amarasekare vented about her own experience at UCLA on a departmental e-mail listserve created to discuss issues of racism and discrimination in the aftermath of the killing of George Floyd, whose death in May 2020 sparked national protests.“That’s why she got into trouble. She ended up criticizing pretty much the entire department — with good reason,” says Marcel Vaz, an ecologist at Wilkes University in Wilkes-Barre, Pennsylvania, who was a graduate student in the department at the time. He and other students came forward to support her. “We demanded some explanation,” Vaz says, “but we never got any feedback.”Peter Kareiva, a former UCLA faculty member who spoke on Amarasekare’s behalf during the administrative proceedings, calls her a brilliant scientist as well as a terrific teacher and student mentor. Kareiva witnessed Amarasekare raise uncomfortable issues and challenge internal policies in faculty meetings. He says she might have made mistakes in terms of “facilitating harmony” among fellow faculty members, but that her goal was always to improve the department.
How a scandal in spider biology upended researchers’ lives
“I am still incredulous by the punishment levied,” says Kareiva, who now serves as president of the Aquarium of the Pacific in Long Beach, California.It is unclear what happens next, but scientists and former students and faculty members contacted by Nature are concerned about the impact on Amarasekare’s current students, the disruption of federally funded research and the potentially irretrievable loss of time-sensitive experiments that could provide insights regarding the ecological impacts of climate change.As the recipient of the MacArthur award, Amarasekare is expected to discuss this research when she delivers her keynote address at the Ecological Society of America’s annual meeting in Portland, Oregon, in August. More