Ecology

1.
Haken, J. Transnational Crime in the Developing World (Global Financial Integrity, 2011).
2.
Patel, N. G. et al. Quantitative methods of identifying the key nodes in the illegal wildlife trade network. Proc. Natl Acad. Sci. USA 112, 7948–7953 (2015).
CAS  PubMed  Article  PubMed Central  Google Scholar 

3.
Bager Olsen, M. T. et al. Thirty-six years of legal and illegal wildlife trade entering the USA. Oryx https://doi.org/10.1017/S0030605319000541 (2019).

4.
Tittensor, D. P. et al. Evaluating the relationships between the legal and illegal international wildlife trades. Conserv. Lett. https://doi.org/10.1111/conl.12724 (2020).

5.
Harfoot, M. et al. Unveiling the patterns and trends in 40 years of global trade in CITES-listed wildlife. Biol. Conserv. 223, 47–57 (2018).
Article  Google Scholar 

6.
Scheffers, B. R., Oliveira, B. F., Lamb, I. & Edwards, D. P. Global wildlife trade across the tree of life. Science 76, 71–76 (2019).
Article  CAS  Google Scholar 

7.
Maxwell, S. L., Fuller, R. A., Brooks, T. M. & Watson, J. E. M. Biodiversity: the ravages of guns, nets and bulldozers. Nature 536, 143–145 (2016).
CAS  PubMed  Article  PubMed Central  Google Scholar 

8.
Nielsen, M. R., Meilby, H., Smith-Hall, C., Pouliot, M. & Treue, T. The importance of wild meat in the global south. Ecol. Econ. 146, 696–705 (2018).
Article  Google Scholar 

9.
’t Sas-Rolfes, M., Challender, D. W. S., Hinsley, A., Veríssimo, D. & Milner-Gulland, E. J. Illegal wildlife trade: scale, processes, and governance. Annu. Rev. Environ. Resour. 44, 201–228 (2019).
Article  Google Scholar 

10.
Cooney, R. et al. From poachers to protectors: engaging local communities in solutions to illegal wildlife trade. Conserv. Lett. 10, 367–374 (2017).
Article  Google Scholar 

11.
Bodmer, R. E. & Lozano, E. P. Rural development and sustainable wildlife use in Peru. Conserv. Biol. 15, 1163–1170 (2001).
Article  Google Scholar 

12.
McClenachan, L., Cooper, A. B. & Dulvy, N. K. Rethinking trade-driven extinction risk in marine and terrestrial megafauna. Curr. Biol. 26, 1640–1646 (2016).
CAS  PubMed  Article  PubMed Central  Google Scholar 

13.
Wittemyer, G. et al. Illegal killing for ivory drives global decline in African elephants. Proc. Natl Acad. Sci. USA 111, 13117–13121 (2014).
CAS  PubMed  Article  PubMed Central  Google Scholar 

14.
Brook, S., Van Coeverden De Groot, P. J., Mahood, S. & Long, B. Extinction of the Javan Rhinoceros (Rhinoceros sondaicus) from Vietnam (WWF, 2011).

15.
Heinrich, S. et al. Where did all the pangolins go? International CITES trade in pangolin species. Glob. Ecol. Conserv. 8, 241–253 (2016).
Article  Google Scholar 

16.
Cowlishaw, G., Mendelson, S. & Rowcliffe, J. M. Evidence for post-depletion sustainability in a mature bushmeat market. J. Appl. Ecol. 42, 460–468 (2005).
Article  Google Scholar 

17.
Hutton, J. M. & Webb, G. in The Trade in Wildlife: Regulation for Conservation (ed. Oldfield, S.) Ch. 11 (Earthscan, 2003).

18.
Harris, J. B. C. et al. Using market data and expert opinion to identify overexploited species in the wild bird trade. Biol. Conserv. 187, 51–60 (2015).
Article  Google Scholar 

19.
Milner-Gulland, E. J. & Clayton, L. The trade in babirusas and wild pigs in North Sulawesi, Indonesia. Ecol. Econ. 42, 165–183 (2002).
Article  Google Scholar 

20.
Thuiller, W. et al. Vulnerability of African mammals to anthropogenic climate change under conservative land transformation assumptions. Glob. Change Biol. 12, 424–440 (2006).
Article  Google Scholar 

21.
Benítez-López, A. et al. The impact of hunting on tropical mammal and bird populations. Science 356, 180–183 (2017).
PubMed  Article  CAS  PubMed Central  Google Scholar 

22.
Gibson, L. et al. Primary forests are irreplaceable for sustaining tropical biodiversity. Nature 478, 378–381 (2011).
CAS  PubMed  Article  PubMed Central  Google Scholar 

23.
CITES Trade Statistics Derived from the CITES Trade Database (CITES, 2020).

24.
Phelps, J. & Webb, E. L. ‘Invisible’ wildlife trades: Southeast Asia’s undocumented illegal trade in wild ornamental plants. Biol. Conserv. 186, 296–305 (2015).
Article  Google Scholar 

25.
Davies, G., Schulte-Herbrüggen, B., Kümpel, N. F. & Mendelson, S. Hunting and trapping in Gola Forests, south-eastern Sierra Leone: bushmeat from farm, fallow and forest. Bushmeat Livelihoods Wildl. Manage. Poverty Reduct. https://doi.org/10.1002/9780470692592.ch1 (2008).

26.
Linder, J. M. & Oates, J. F. Differential impact of bushmeat hunting on monkey species and implications for primate conservation in Korup National Park, Cameroon. Biol. Conserv. 144, 738–745 (2011).
Article  Google Scholar 

27.
Gilroy, J. J. & Edwards, D. P. Source–sink dynamics: a neglected problem for landscape-scale biodiversity conservation in the tropics. Curr. Landsc. Ecol. Rep. 2, 51–60 (2017).
Article  Google Scholar 

28.
Watson, J. E. M. et al. Catastrophic declines in wilderness areas undermine global environment targets. Curr. Biol. 26, 2929–2934 (2016).
CAS  PubMed  Article  PubMed Central  Google Scholar 

29.
Marshall, H. et al. Spatio-temporal dynamics of consumer demand driving the Asian songbird crisis. Biol. Conserv. 241, 108237 (2020).
Article  Google Scholar 

30.
Harris, J. B. C. et al. Measuring the impact of the pet trade on Indonesian birds. Conserv. Biol. 31, 394–405 (2017).
PubMed  Article  PubMed Central  Google Scholar 

31.
Carrasco, L. R., Chan, J., Mcgrath, F. L. & Nghiem, L. T. P. Biodiversity conservation in a telecoupled world. Ecol. Soc. 22, 24 (2017).
Article  Google Scholar 

32.
Blundell, A. G. & Mascia, M. B. Discrepancies in reported levels of international wildlife trade. Conserv. Biol. 19, 2020–2025 (2005).
Article  Google Scholar 

33.
Nelson, A. et al. A suite of global accessibility indicators. Sci. Data 6, 266 (2019).
PubMed  PubMed Central  Article  Google Scholar 

34.
Rao, M., Zaw, T., Htun, S. & Myint, T. Hunting for a living: wildlife trade, rural livelihoods and declining wildlife in the Hkakaborazi National Park, North Myanmar. Environ. Manage. 48, 158–167 (2011).
PubMed  Article  PubMed Central  Google Scholar 

35.
Symes, W. S., Edwards, D. P., Miettinen, J., Rheindt, F. E. & Carrasco, L. R. Combined impacts of deforestation and wildlife trade on tropical biodiversity are severely underestimated. Nat. Commun. 9, 4052 (2018).
PubMed  PubMed Central  Article  CAS  Google Scholar 

36.
Hinsley, A. et al. Building sustainability into the Belt and Road Initiative’s traditional Chinese medicine trade. Nat. Sustain. 3, 96–100 (2020).
Article  Google Scholar 

37.
Lechner, A. M., Chan, F. K. S. & Campos-Arceiz, A. Biodiversity conservation should be a core value of China’s Belt and Road Initiative. Nat. Ecol. Evol. 2, 408–409 (2018).
PubMed  Article  PubMed Central  Google Scholar 

38.
Farhadinia, M. S. et al. Belt and Road Initiative may create new supplies for illegal wildlife trade in large carnivores. Nat. Ecol. Evol. 3, 1267–1268 (2019).
PubMed  Article  Google Scholar 

39.
Courchamp, F. et al. Rarity value and species extinction: the anthropogenic Allee effect. PLoS Biol. 4, 2405–2410 (2006).
CAS  Article  Google Scholar 

40.
Jetz, W. & Freckleton, R. P. Towards a general framework for predicting threat status of data-deficient species from phylogenetic, spatial and environmental information. Phil. Trans. R. Soc. B 370, 20140016 (2015).
PubMed  Article  PubMed Central  Google Scholar 

41.
Dulac, J. Global Land Transport Infrastructure Requirements: Estimating Road and Railway Infrastructure Capacity and Costs to 2050 (IEA, 2013).

42.
Vilela, T. et al. A better Amazon road network for people and the environment. Proc. Natl Acad. Sci. USA 117, 7095–7102 (2020).
CAS  PubMed  Article  PubMed Central  Google Scholar 

43.
Challender, D. W. S., Harrop, S. R. & MacMillan, D. C. Towards informed and multi-faceted wildlife trade interventions. Glob. Ecol. Conserv. 3, 129–148 (2015).
Article  Google Scholar 

44.
Papworth, S., Milner-Gulland, E. J. & Slocombe, K. Hunted woolly monkeys (Lagothrix poeppigii) show threat-sensitive responses to human presence. PLoS ONE 8, e62000 (2013).
CAS  PubMed  PubMed Central  Article  Google Scholar 

45.
Toledo-Aceves, T., Garcia-Franco, J. G. & Lopez-Barrera, F. Bromeliad rain: an opportunity for cloud forest management. Ecol. Manage. 329, 129–136 (2014).
Article  Google Scholar 

46.
Challender, D. W. S. et al. Mischaracterization of wildlife trade threat. Science (30 October 2019).

47.
Leung, B. et al. Clustered versus catastrophic global vertebrate declines. Nature https://doi.org/10.1038/s41586-020-2920-6 (2020).

48.
Tierney, M. et al. Use it or lose it: measuring trends in wild species subject to substantial use. Oryx 48, 420–429 (2014).
Article  Google Scholar 

49.
Jachmann, H. Monitoring law-enforcement performance in nine protected areas in Ghana. Biol. Conserv. 141, 89–99 (2008).
Article  Google Scholar 

50.
Cardador, L., Tella, J. L., Anadón, J. D., Abellán, P. & Carrete, M. The European trade ban on wild birds reduced invasion risks. Conserv. Lett. https://doi.org/10.1111/conl.12631 (2019).

51.
Grames, E. M., Stillman, A. N., Tingley, M. W. & Elphick, C. S. An automated approach to identifying search terms for systematic reviews using keyword co-occurrence networks. Methods Ecol. Evol. 10, 1645–1654 (2019).
Article  Google Scholar 

52.
R Core Team R: A Language and Environment for Statistical Computing (R Foundation for Statistical Computing, 2019).

53.
Espinosa-Andrade, S. R. Road development, bushmeat extraction and jaguar conservation in Yasuni Biosphere Reserve Ecuador PhD Thesis, Univ. Florida (2012); https://ufdc.ufl.edu/UFE0044108/00001/citation

54.
Schoppe, S., Matillano, J., Cervancia, M. & Acosta, D. Conservation needs of the critically endangered Philippine forest turtle, Siebenrockiella leytensis, in Palawan, Philippines. Chelonian Conserv. Biol. 9, 145–153 (2010).
Article  Google Scholar 

55.
Wilman, H. et al. EltonTraits 1.0: species-level foraging attributes of the world’s birds and mammals. Ecology 95, 2027 (2014).
Article  Google Scholar 

56.
Slavenko, A., Tallowin, O. J. S., Itescu, Y., Raia, P. & Meiri, S. Late Quaternary reptile extinctions: size matters, insularity dominates. Glob. Ecol. Biogeogr. 25, 1308–1320 (2016).
Article  Google Scholar 

57.
Aquino, R. & Calle, A. Evaluation of the conservation status of the game mammals: a comparative model in communities of the Pacaya Samira National Reserve (Loreto, Peru). Rev. Peru. Biol. 10, 163–174 (2003).
Google Scholar 

58.
Aquino, R., Lopez, L., Garcia, G., Charpentier, E. & Arevalo, I. Conservation status and threats to atelids in the northeastern Peruvian Amazon. Primate Conserv. 30, 21–29 (2016).
Google Scholar 

59.
Carrillo, E., Wong, G. & Cuarón, A. D. Monitoring mammal populations in Costa Rican protected areas under different hunting restrictions. Conserv. Biol. 14, 1580–1591 (2000).
Article  Google Scholar 

60.
Cronin, D. T. The Impact of Bushmeat Hunting on the Primates of Bioko Island, Equatorial Guinea (Drexel Univ., 2013).

61.
Dasgupta, S. & Hilaluddin Differential effects of hunting on populations of hornbills and imperial pigeons in the rainforests of the eastern Indian Himalaya. Indian Forester 138, 902–909 (2012).
Google Scholar 

62.
De Thoisy, B., Renoux, F. & Julliot, C. Hunting in northern French Guiana and its impact on primate communities. Oryx 39, 149–157 (2005).
Article  Google Scholar 

63.
Fay, J. M. An elephant (Loxodonta africana) survey using dung counts in the forests of the Central African Republic. J. Trop. Ecol. 7, 25–36 (1991).
Article  Google Scholar 

64.
Gamble, T. & Simons, A. M. Comparison of harvested and nonharvested painted turtle populations. Wildl. Soc. Bull. 32, 1269–1277 (2004).
Article  Google Scholar 

65.
Gonzalez, J. A. Harvesting, local trade, and conservation of parrots in the northeastern Peruvian Amazon. Biol. Conserv. 114, 437–446 (2003).
Article  Google Scholar 

66.
Gray, T. N. E. & Phan, C. Habitat preferences and activity patterns of the larger mammal community in Phnom Prich Wildlife Sanctuary, Cambodia. Raffles Bull. Zool. 59, 311–318 (2011).
Google Scholar 

67.
Hall, J. S. et al. A survey of elephants (Loxodonta africana) in the Kahuzi-Biega National Park lowland sector and adjacent forest in eastern Zaire. Afr. J. Ecol. 35, 213–223 (1997).
Article  Google Scholar 

68.
Klemens, M. W. & Moll, D. An assessment of the effects of commercial exploitation on the pancake tortoise, Malacochersus tornieri, in Tanzania. Chelonian Conserv. Biol. 1, 197–206 (1995).
Google Scholar 

69.
Kümpel, N. F., Milner-Gulland, E. J., Rowcliffe, J. M. & Cowlishaw, G. Impact of gun-hunting on diurnal primates in continental Equatorial Guinea. Int. J. Primatol. 29, 1065–1082 (2008).
Article  Google Scholar 

70.
Magige, F. J., Holmern, T., Stokke, S., Mlingwa, C. & Røskaft, E. Does illegal hunting affect density and behaviour of African grassland birds? A case study on ostrich (Struthio camelus). Biodivers. Conserv. 18, 1361–1373 (2009).
Article  Google Scholar 

71.
Maldonado, A. M. & Peck, M. R. Research and in situ conservation of owl monkeys enhances environmental law enforcement at the Colombian–Peruvian border. Am. J. Primatol. 76, 658–669 (2014).
PubMed  Article  PubMed Central  Google Scholar 

72.
Maldonado, A. M., Nijman, V. & Bearder, S. K. Trade in night monkeys Aotus spp. in the Brazil–Colombia–Peru tri-border area: international wildlife trade regulations are ineffectively enforced. Endanger. Species Res. 9, 143–149 (2009).
Article  Google Scholar 

73.
Muchaal, P. K. & Ngandjui, G. Impact of village hunting on wildlife populations in the Western Dja Reserve, Cameroon. Conserv. Biol. 13, 385–396 (1999).
Article  Google Scholar 

74.
Nuñez-iturril, G. & Howe, H. F. Bushmeat and the fate of trees with seeds dispersed by large primates in a lowland rain forest in western Amazonia. Biotropica 39, 348–354 (2007).
Article  Google Scholar 

75.
O’Brien, S. et al. Decline of the Madagascar radiated tortoise Geochelone radiata due to overexploitation. Oryx 37, 338–343 (2003).
Google Scholar 

76.
Patrick, D. A., Shirk, P., Vonesh, J. R., Harper, E. B. & Howell, K. M. Abundance and roosting ecology of chameleons in the Eastern Arc Mountains of Tanzania and potential effects of harvesting. Herpetol. Conserv. Biol. 6, 422–431 (2011).
Google Scholar 

77.
Poulsen, J. R., Clark, C. J. & Bolker, B. M. Decoupling the effect of logging and hunting on an Afrotropical animal community. Ecol. Appl. 21, 1819–1836 (2011).
CAS  PubMed  Article  Google Scholar 

78.
Remis, M. J. & Kpanou, J. B. Primate and ungulate abundance in response to multi-use zoning and human extractive activities in a central African reserve. Afr. J. Ecol. 49, 70–80 (2010).
Article  Google Scholar 

79.
Rist, J., Milner-Gulland, E. J., Cowlishaw, G. & Rowcliffe, J. M. The importance of hunting and habitat in determining the abundance of tropical forest species in Equatorial Guinea. Biotropica 41, 700–710 (2009).
Article  Google Scholar 

80.
Rovero, F., Mtui, A. S., Kitegile, A. S. & Nielsen, M. R. Hunting or habitat degradation? Decline of primate populations in Udzungwa Mountains, Tanzania: an analysis of threats. Biol. Conserv. 146, 89–96 (2012).
Article  Google Scholar 

81.
Segura, A. & Acevedo, P. The importance of protected and unprotected areas for the Mediterranean spur-thighed tortoise demography in northwest Morocco. Amphib. Reptilia https://doi.org/10.6084/m9.figshare.7751783.v1 (2019).

82.
Sung, Y.-H., Karraker, N. E. & Hau, B. C. H. Demographic evidence of illegal harvesting of an endangered Asian turtle. Conserv. Biol. 27, 1421–1428 (2013).
PubMed  Article  PubMed Central  Google Scholar 

83.
Topp-Jorgensen, E., Nielsen, M. R., Marshall, A. R. & Pedersen, U. Relative densities of mammals in response to different levels of bushmeat hunting in the Udzungwa Mountains, Tanzania. Trop. Conserv. Sci. 2, 70–87 (2009).
Article  Google Scholar 

84.
Yasuoka, H. The sustainability of duiker (Cephalophus spp.) hunting for the Baka hunter–gatherers in southeastern Cameroon. Afr. Study Monogr. 33, 95–120 (2006).
Google Scholar 

85.
Protected Planet: The World Database on Protected Areas (WDPA) (UNEP-WCMC and IUCN, 2020).

86.
QGIS Development Team QGIS Geographic Information System v.3.12.0 (Open Source Geospatial Foundation Project, 2020).

87.
User Manual for the World Database on Protected Areas and World Database on Other Effective Area-Based Conservation Measures: 1.6 (UNEP-WCMC, 2019).

88.
Hedges, L. V., Gurevitch, J. & Curtis, P. S. The meta-analysis of response ratios in experimental ecology. Ecology 80, 1150–1156 (1999).
Article  Google Scholar 

89.
Smithson, M. & Verkuilen, J. A better lemon squeezer? Maximum-likelihood regression with beta-distributed dependent variables. Psychol. Methods 11, 54–71 (2006).
PubMed  Article  PubMed Central  Google Scholar 

90.
Lajeunesse, M. J. in Handbook of Meta-analysis in Ecology and Evolution (eds Koricheva, J. et al.) Ch. 13 (Princeton Univ. Press, 2013); https://doi.org/10.23943/princeton/9780691137285.003.0013

91.
Rubin, D. B. & Schenker, N. Multiple imputation in health-care databases: an overview and some applications. Stat. Med. 10, 585–598 (1991).
CAS  PubMed  Article  PubMed Central  Google Scholar 

92.
Sweeting, M. J., Sutton, A. J. & Lambert, P. C. What to add to nothing? Use and avoidance of continuity corrections in meta-analysis of sparse data. Stat. Med. 23, 1351–1375 (2004).
PubMed  Article  PubMed Central  Google Scholar 

93.
Viechtbauer, W. Conducting meta-analyses in R with the metafor. J. Stat. Softw. 36, 1–48 (2010).
Article  Google Scholar 

94.
Hothorn, T. et al. multcomp: Simultaneous inference in general parametric models. R package version 1.4-15 (2016).

95.
Cochran, W. G. The combination of estimates from different experiments. Biometrics 10, 101–129 (1954).
Article  Google Scholar 

96.
Hoaglin, D. C. Misunderstandings about Q and ‘Cochran’s Q test’ in meta-analysis. Stat. Med. 35, 485–495 (2016).
PubMed  Article  PubMed Central  Google Scholar 

97.
McFadden, D. Quantitative Methods for Analyzing Travel Behaviour of Individuals: Some Recent Developments Discussion Paper No. 474 (Cowles Foundation, 1977).

98.
Geary, R. C. The frequency distribution of the quotient of two normal variates. J. R. Stat. Soc. 93, 442–446 (1930).
Article  Google Scholar 

99.
Lajeunesse, M. J. Bias and correction for the log response ratio in ecological meta-analysis. Ecol. Soc. Am. 96, 2056–2063 (2015).
Google Scholar 

100.
Rosenthal, R. The file drawer problem and tolerance for null results. Psychol. Bull. 86, 638–641 (1979).
Article  Google Scholar 

101.
Rosenberg, M. S. The file-drawer problem revisited: a general weighted method for calculating fail-safe number in meta-analysis. Evolution 59, 464–468 (2005).
PubMed  Article  PubMed Central  Google Scholar  More

1.
Llewellyn LE. Saxitoxin, a toxic marine natural product that targets a multitude of receptors. Nat Prod Rep. 2006;23:200–18.
CAS  PubMed  Article  Google Scholar 
2.
Selander E, Thor P, Toth G, Pavia H. Copepods induce paralytic shellfish toxin production in marine dinoflagellates. Proc R Soc B. 2006;273:1673–80.
CAS  PubMed  Article  Google Scholar 

3.
Turner JT, Tester PA. Toxic marine phytoplankton, zooplankton grazers, and pelagic food webs. Limnol Oceanogr. 1997;42:1203–13.
Article  Google Scholar 

4.
Smetacek V. A watery arms race. Nature. 2001;441:745.
Article  Google Scholar 

5.
Xu J, Kiørboe T. Toxic dinoflagellates produce true grazer deterrents. Ecology. 2018;99:2240–9.
PubMed  Article  Google Scholar 

6.
Cusick KD, Widder EA. Bioluminescence and toxicity as driving factors in harmful algal blooms: ecological functions and genetic variability. Harmful Algae. 2020;98:101850.
CAS  PubMed  Article  Google Scholar 

7.
Pančić M, Kiørboe T. Phytoplankton defence mechanisms: traits and trade-offs: defensive traits and trade-offs. Biol Rev. 2018;93:1269–303.
PubMed  Article  Google Scholar 

8.
John EH, Flynn KJ. Growth dynamics and toxicity of Alexandrium fundyense (Dinophyceae): the effect of changing N:P supply ratios on internal toxin and nutrient levels. Eur J Phycol. 2000;35:11–23.
Google Scholar 

9.
Selander E, Cervin G, Pavia H. Effects of nitrate and phosphate on grazer-induced toxin production in Alexandrium minutum. Limnol Oceanogr. 2008;53:523–30.
CAS  Article  Google Scholar 

10.
Blossom HE, Markussen B, Daugbjerg N, Krock B, Norlin A, Hansen PJ. The cost of toxicity in microalgae: direct evidence from the dinoflagellate Alexandrium. Front Microbiol. 2019;10:1065.
PubMed  PubMed Central  Article  Google Scholar 

11.
Brown ER, Kubanek J. Harmful alga trades off growth and toxicity in response to cues from dead phytoplankton. Limnol Oceanogr. 2020;65:1723–33.

12.
Windust AJ, Wright JLC, McLachlan JL. The effects of the diarrhetic shellfish poisoning toxins, okadaic acid and dinophysistoxin-1, on the growth of microalgae. Mar Biol. 1996;126:19–25.
CAS  Article  Google Scholar 

13.
Legrand C, Rengefors K, Fistarol GO, Granéli E. Allelopathy in phytoplankton—biochemical, ecological and evolutionary aspects. Phycologia. 2003;42:406–19.
Article  Google Scholar 

14.
John E, Flynn K. Modelling changes in paralytic shellfish toxin content of dinoflagellates in response to nitrogen and phosphorus supply. Mar Ecol Prog Ser. 2002;225:147–60.
CAS  Article  Google Scholar 

15.
Lundholm N, Krock B, John U, Skov J, Cheng J, Pančić M, et al. Induction of domoic acid production in diatoms—types of grazers and diatoms are important. Harmful Algae. 2018;79:64–73.
CAS  PubMed  Article  Google Scholar 

16.
Bergkvist J, Selander E, Pavia H. Induction of toxin production in dinoflagellates: the grazer makes a difference. Oecologia. 2008;156:147–54.
PubMed  Article  Google Scholar 

17.
Griffin JE, Park G, Dam HG. Relative importance of nitrogen sources, algal alarm cues and grazer exposure to toxin production of the marine dinoflagellate Alexandrium catenella. Harmful Algae. 2019;84:181–7.
CAS  PubMed  Article  Google Scholar 

18.
Selander E, Berglund EC, Engström P, Berggren F, Eklund J, Harðardóttir S, et al. Copepods drive large-scale trait-mediated effects in marine plankton. Sci Adv. 2019;5:eaat5096.
CAS  PubMed  PubMed Central  Article  Google Scholar 

19.
Rhoades DF. Evolution of plant chemical defense against herbivores. Herbivores: their interaction with secondary plant metabolites. New York: Academic Press;1979. p 1–55.

20.
Karban R. The ecology and evolution of induced resistance against herbivores: induced resistance against herbivores. Funct Ecol. 2011;25:339–47.
Article  Google Scholar 

21.
Strauss SY, Rudgers JA, Lau JA, Irwin RE. Direct and ecological costs of resistance to herbivory. TREE. 2002;17:278–85.
Google Scholar 

22.
Agrawal AA. Current trends in the evolutionary ecology of plant defence. Funct Ecol. 2011;25:420–32.
Article  Google Scholar 

23.
Pančić M, Torres RR, Almeda R, Kiørboe T. Silicified cell walls as a defensive trait in diatoms. Proc R Soc B. 2019;286:20190184.
PubMed  Article  CAS  Google Scholar 

24.
Grønning J, Kiørboe T. Diatom defence: grazer induction and cost of shell‐thickening. Funct Ecol. 2020;34:1790–1801.

25.
Kiørboe T, Andersen KH. Nutrient affinity, half-saturation constants and the cost of toxin production in dinoflagellates. Ecol Lett. 2019;22:558–60.
PubMed  Article  Google Scholar 

26.
Wang X, Wang Y, Ou L, He X, Chen D. Allocation costs associated with induced defense in Phaeocystis globosa (Prymnesiophyceae): the effects of nutrient availability. Sci Rep. 2015;5:10850.
CAS  PubMed  PubMed Central  Article  Google Scholar 

27.
Zhu X, Wang J, Chen Q, Chen G, Huang Y, Yang Z. Costs and trade-offs of grazer-induced defenses in Scenedesmus under deficient resource. Sci Rep. 2016;6:22594.
CAS  PubMed  PubMed Central  Article  Google Scholar 

28.
Redfield AC. The biological control of chemical factors in the environment. Am Sci. 1958;46:205–21.
CAS  Google Scholar 

29.
Boyer GL, Sullivan JJ, Andersen RJ, Harrison PJ, Taylor FJR. Effects of nutrient limitation on toxin production and composition in the marine dinoflagellate Protogonyaulax tamarensis. Mar Biol. 1987;96:123–8.
CAS  Article  Google Scholar 

30.
Leong SCY, Murata A, Nagashima Y, Taguchi S. Variability in toxicity of the dinoflagellate Alexandrium tamarense in response to different nitrogen sources and concentrations. Toxicon. 2004;43:407–15.
CAS  PubMed  Article  Google Scholar 

31.
Chakraborty S, Pančić M, Andersen KH, Kiørboe T. The cost of toxin production in phytoplankton: the case of PST producing dinoflagellates. ISME J. 2019;13:64–75.
CAS  PubMed  Article  Google Scholar 

32.
Andersson L. Trends in nutrient and oxygen concentrations in the Skagerrak-Kattegat. J Sea Res. 1996;35:63–71.
CAS  Article  Google Scholar 

33.
Tiselius P, Belgrano A, Andersson L, Lindahl O. Primary productivity in a coastal ecosystem: a trophic perspective on a long-term time series. J Plankton Res. 2016;38:1092–102.
CAS  Article  Google Scholar 

34.
Kiørboe T, Nielsen TG. Regulation of zooplankton biomass and production in a temperate, coastal ecosystem. 1. Copepods. Limnol Oceanogr. 1994;39:493–507.
Article  Google Scholar 

35.
Selander E, Kubanek J, Hamberg M, Andersson MX, Cervin G, Pavia H. Predator lipids induce paralytic shellfish toxins in bloom-forming algae. Proc Natl Acad Sci USA. 2015;112:6395–400.
CAS  PubMed  Article  Google Scholar 

36.
Hansen PJ. The red tide dinoflagellate Alexandrium tamarense: effects on behaviour and growth of a tintinnid ciliate. Mar Ecol Prog Ser. 1989;53:105–16.
Article  Google Scholar 

37.
Berdalet E, Peters F, Koumandou VL, Roldán C, Guadayol Ò, Estrada M. Species-specific physiological response of dinoflagellates to quantified small-scale turbulence 1. J Phycol. 2007;43:965–77.
Article  Google Scholar 

38.
Fischer R, Andersen T, Hillebrand H, Ptacnik R. The exponentially fed batch culture as a reliable alternative to conventional chemostats. Limnol Oceanogr Meth. 2014;12:432–40.
Article  Google Scholar 

39.
Flynn K, Jones KJ, Flynn KJ. Comparisons among species of Alexandrium (Dinophyceae) grown in nitrogen- or phosphorus-limiting batch culture. Mar Biol. 1996;126:9–18.
CAS  Article  Google Scholar 

40.
Brandenburg KM, Wohlrab S, John U, Kremp A, Jerney J, Krock B, et al. Intraspecific trait variation and trade-offs within and across populations of a toxic dinoflagellate. Ecol Lett. 2018;21:1561–71.
PubMed  Article  Google Scholar 

41.
Hillebrand H, Dürselen C-D, Kirschtel D, Pollingher U, Zohary T. Biovolume calculation for pelagic and benthic microalgae. J Phycol. 1999;35:403–24.
Article  Google Scholar 

42.
Schnetger B, Lehners C. Determination of nitrate plus nitrite in small volume marine water samples using vanadium(III)chloride as a reduction agent. Mar Chem. 2014;160:91–8.
CAS  Article  Google Scholar 

43.
Asp TN, Larsen S, Aune T. Analysis of PSP toxins in Norwegian mussels by a post-column derivatization HPLC method. Toxicon. 2004;43:319–27.
CAS  PubMed  Article  Google Scholar 

44.
Turner A, Tölgyesi L. Determination of Paralytic Shellfish Toxins and Tetrodotoxin in Shellfish using HILIC/MS/MS. Application Note No. 5994-0967EN. [Internet]. 2019. Available from: https://www.agilent.com/en/solutions/food-testing-agriculture/seafood-testing.

45.
Franco JM, Fernández P, Reguera B. Toxin profiles of natural populations and cultures of Alexandrium minutum Halim from Galician (Spain) coastal waters. J Appl Phycol. 1994;6:275–9.
CAS  Article  Google Scholar 

46.
Xu J, Hansen PJ, Nielsen LT, Krock B, Tillmann U, Kiørboe T. Distinctly different behavioral responses of a copepod, Temora longicornis, to different strains of toxic dinoflagellates, Alexandrium spp. Harmful Algae. 2017;62:1–9.
PubMed  Article  Google Scholar 

47.
Wood S, Scheipl F. gamm4: Generalized additive mixed models using mgcv and lme4. R package version 0.2-6. [Internet]. 2020. Available from: https://cran.r-project.org/.

48.
Kuznetsova A, Brockhoff PB, Christensen RHB. lmerTest package: tests in linear mixed effects models. J Stat Soft 2017;82:1–26.

49.
Wohlrab S, Selander E, John U. Predator cues reduce intraspecific trait variability in a marine dinoflagellate. BMC Ecol. 2017;17:8.
PubMed  PubMed Central  Article  CAS  Google Scholar 

50.
Driscoll WW, Hackett JD, Ferrière R. Eco-evolutionary feedbacks between private and public goods: evidence from toxic algal blooms. Ecol Lett. 2016;19:81–97.
PubMed  Article  Google Scholar 

51.
Flynn K, Franco J, Fernandez P, Reguera B, Zapata M, Wood G, et al. Changes in toxin content, biomass and pigments of the dinoflagellate Alexandrium minutum during nitrogen refeeding and growth into nitrogen or phosphorus stress. Mar Ecol Prog Ser. 1994;111:99–109.
CAS  Article  Google Scholar 

52.
Tillmann U, John U. Toxic effects of Alexandrium spp. on heterotrophic dinoflagellates: an allelochemical defence mechanism independent of PSP-toxin content. Mar Ecol Prog Ser. 2002;230:47–58.
CAS  Article  Google Scholar 

53.
Tillmann U, Hansen PJ. Allelopathic effects of Alexandrium tamarense on other algae: evidence from mixed growth experiments. Aquat Micro Ecol. 2009;57:101–12.
Article  Google Scholar 

54.
Teegarden GJ, Campbell RG, Anson DT, Ouellett A, Westman BA, Durbin EG. Copepod feeding response to varying Alexandrium spp. cellular toxicity and cell concentration among natural plankton samples. Harmful Algae. 2008;7:33–44.
Article  Google Scholar 

55.
Peter KH, Sommer U. Interactive effect of warming, nitrogen and phosphorus limitation on phytoplankton cell size. Ecol Evol. 2015;5:1011–24.
PubMed  PubMed Central  Article  Google Scholar 

56.
Garcia NS, Bonachela JA, Martiny AC. Interactions between growth-dependent changes in cell size, nutrient supply and cellular elemental stoichiometry of marine. Synechococcus ISME J. 2016;10:2715–24.
CAS  PubMed  Article  Google Scholar 

57.
Kiørboe T. Turbulence, phytoplankton cell size, and the structure of pelagic food webs. Adv Mar Biol. 1993;29:1–72.
Article  Google Scholar 

58.
Lindemann C, Fiksen Ø, Andersen KH, Aksnes DL. Scaling laws in phytoplankton nutrient uptake affinity. Front Microbiol. 2016;3:1–6.
Google Scholar 

59.
Edwards KF, Thomas MK, Klausmeier CA, Litchman E. Allometric scaling and taxonomic variation in nutrient utilization traits and maximum growth rate of phytoplankton. Limnol Oceanogr. 2012;57:554–66.
Article  Google Scholar 

60.
Lürling M, Van Donk E. Grazer-induced colony formation in Scenedesmus: are there costs to being colonial? Oikos. 2000;88:111–8.
Article  Google Scholar 

61.
Selander E, Jakobsen HH, Lombard F, Kiorboe T. Grazer cues induce stealth behavior in marine dinoflagellates. Proc Natl Acad Sci USA. 2011;108:4030–4.
CAS  PubMed  Article  PubMed Central  Google Scholar 

62.
Kiørboe T. A mechanistic approach to plankton ecology. Princeton, NJ: Princeton University Press; 2008. pp. 128–129.

63.
Tollrian R, Harvell CD. The ecology and evolution of inducible defenses. Princeton, NJ: Princeton University Press; 1999. pp. 1–383.

64.
Leao T, Castelão G, Korobeynikov A, Monroe EA, Podell S, Glukhov E, et al. Comparative genomics uncovers the prolific and distinctive metabolic potential of the cyanobacterial genus Moorea. Proc Natl Acad Sci USA. 2017;114:3198–203.
CAS  PubMed  Article  PubMed Central  Google Scholar 

65.
Züst T, Agrawal AA. Trade-Offs Between Plant Growth and Defense Against Insect Herbivory: An Emerging Mechanistic Synthesis. Annu Rev Plant Biol. 2017;68:513–34.
PubMed  Article  CAS  PubMed Central  Google Scholar  More

Continuous and discrete quantity information are important in guiding animal behaviour in virtually all aspects of life. The capacity to evaluate continuous (uncountable) quantities, such as length, area, weight, or duration, is widespread and can be found in organisms with relatively simple nervous systems, such as annelids, crustaceans, or gastropds1,2,3. This quantitative information takes part in decision-making processes in different contexts. For example, animals may gauge body sizes of rivals or prospective mates, assess distances from home, or estimate the extent of a food patch4,5,6.
Several vertebrates, from teleost fishes to primates, can also process discrete (countable) information. For example, many species are capable of accurately estimating the number of elements in a set and comparing the numerosity of different sets7,8. Studies conducted in nature or in the laboratory have shown that numerical abilities serve important adaptive functions. For example, in guppies, New Zealand robins, and macaques, quantity discrimination is used to select the patch containing the larger number of food items9,10,11. Conversely, some predators, namely lions and striped field mice, use this ability to select the smallest prey groups because they are more vulnerable to predation12,13. Various group-living mammals, including chimpanzees, lions, and hyenas, gauge the relative number of opponents before deciding whether to attack or withdraw14,15,16. Gregarious fish use the same ability to select the social group that provides the best protection from predators17,18,19. Some species, including eastern mosquitofish, brown-headed cowbirds, and American coots, use their quantitative abilities to increase reproductive success20,21,22.
Cognitive psychologists have shown that in these cases it is not necessary to assume the existence of a true numerical estimation system because an animal can use continuous cues, such as the amount of movement, the cumulative surface occupied by items, or the convex hull of the set as a proxy for number23,24. Inferring the existence of a numerical system requires a series of careful laboratory control experiments in which the animal is subjected to numerical tasks, while the access to non-numerical information is simultaneously prevented8,25. This process is not always straightforward and studies often fail to reach a firm conclusion even after numerous experiments are performed. In fact, convincing evidence of the presence of a numerical system exists only for a small fraction of the species investigated (e.g., guppy26, chicken27, and rhesus monkeys10).
It is not known whether numerical abilities have similar selective advantages in other phyla and whether numerical systems are widespread outside the vertebrate group. To date, this issue has been investigated only in a handful of species, and there is convincing evidence of a true numerical system for only one of them, the honeybee28,29,30. Honeybees, Apis mellifera, can be trained to discriminate different numbers of dots to obtain a food reward31,32. They are able to accomplish this task even when main continuous cues are controlled, thus it is suggested that they possess a numerical system analogous to that of vertebrates. Honeybees can also use ordinal information and learn the correct position in a sequence of artificial flowers when distance cues are made irrelevant33. Similar evidences have been recently provided for another social bee, the bumblebee, Bombus terrestris34,35. The function of cardinal and ordinal numerical abilities in social bees is unclear, but it has been suggested that they mainly serve to recognise flowers from the number of petals and to learn the location of food around their hives, respectively.
Circumstantial evidence suggests the ability to estimate the quantity of conspecifics in three other arthropod species. The juvenile spiders of Portia africana have been reported to take into account the number of competitors present when choosing between two patches of food36. Males of the coleopteran Tenebrio molitor are able to discriminate different numbers of females based on the odours they emit37. Ants (Formica xerophila) perceiving themselves as part of a large group are more aggressive towards another species than ants perceiving themselves as isolated individuals38. Controls are difficult to perform in these types of experiments, and it is unknown whether these three species are actually estimating the number of individuals or they are using other types of information as a proxy of number.
Recently, a mollusc, the cuttlefish, was observed to prefer the larger quantity of shrimps up to 4 versus 5 items39. Although authors manipulate some continuous cues (i.e., density and total activity of preys), it is unclear whether cuttlefish are really counting prey or are using other cues, such as the cumulative area occupied by shrimps or the convex hull of the groups.
Theba pisana is a small terrestrial snail inhabiting the dunes of the Mediterranean coasts. Similar to most snails, it is active mainly at night. This species has a considerable thermal tolerance, with an upper lethal limit that lies, depending on exposure time, between 46 °C and 50 °C40. However, during sunny days, the sandy ground can reach temperatures that largely exceed this lethal limit (up to 75 °C). To avoid these adverse conditions at sunrise, dune snails climb the stem of tall vegetation, where the temperature rarely exceeds 30 °C, and remain inactive until night. If placed on the ground during the day, these snails rapidly regain an elevated position by orienting towards nearby stems and climbing on them (Fig. 1a; Supplementary Video S1). At our site of capture, snails were collected mainly from vertical, unbranched stems of live or dead inedible plants and herbs (e.g., Puccinellia palustris, P. distans, and Juncus maritimus).
Figure 1

(a) Example of a dune snail T. pisana climbing on the stem of tall vegetation. (b) The circular arena used for investigating quantity discrimination ability in laboratory.

Full size image

Zanforlin showed that it is possible to simulate this behaviour in the laboratory41. After placing dune snails on a brightly lit arena, they rapidly orient towards a black cardboard shape on a white background and climb on it. With this setup, it was possible to study shape preference by placing two shapes at 60° angle from the centre and releasing the snail from the centre of the arena. He found that, confronted with similar geometric figures (e.g., two rectangles), snails oriented consistently towards the stimulus with the largest area. When area was kept constant, no particular preference for shape was observed, although there was a tendency to prefer the figure with a longer perimeter or with wider axes.
In all the experiments of the former study, snails were required to choose between two single shapes. In nature, however, stems are frequently arranged in clusters. All things being equal, there are several potential advantages in heading towards a large cluster of stems. In a cluster, there is greater probability of finding the stem with the most suitable features, such as a correct diameter or an optimal orientation to shelter from wind and sun40. In addition, not all the stems are accessible due to the presence of intricate or thorny vegetation at the base. Heading towards a group of stems increases the chances that at least one stem can be reached and climbed. Furthermore, most predators (mainly passerine birds, wall lizards, and rats) are small and catch only one or few preys at a time, and hence, sheltering in clusters could determine a dilution effect on predation risk42,43.
Based on the above considerations, we made the prediction that natural selection in T. pisana should favour the ability to discriminate between a single stem and a cluster and discriminate among clusters, based on the quantity of stems. The aim of the first experiment was to test this hypothesis. In the laboratory, we simulated stems used by dune snails as refuges by using black vertical bars on a white background (Fig. 1b; Supplementary Video S2). As we found that dune snails discriminate rather accurately between quantities of stems, in a series of subsequent experiments, we investigated the mechanism involved. Specifically, we tried to figure out if snails were using a true numerical system or if they used continuous quantitative information that co-varied with numerosity, such as the cumulative area occupied by items, their density or the convex hull they spanned.
Experiment 1: discrimination of the quantity of refuges
A previous study on dune snails investigated the choice between single objects that differed in shape and size41. However, based on their ecology, we predict that snails searching for protection from the heat also should focus on number and should move towards the largest available group of stems. In Experiment 1a, we studied whether dune snails prefer a group of refuges to a single one (Fig. 2a), and in Experiment 1b, we measured their accuracy to discriminate among groups of refuges differing in numerosity (Fig. 2b). To obtain reference data about snails’ general discriminatory abilities, in Experiment 1c, we measured the accuracy of dune snails to discriminate two equally shaped objects that differ in surface area (Fig. 3c).
Figure 2

(a–c) Stimuli used in Experiment 1a, 1b and 1c, respectively.

Full size image

Figure 3

(a) Percentage of snails choosing the stimulus with larger quantity of bars in Experiment 1a and 1b. Snails showed a significant preference for larger quantity up to 4 versus 5 bars. There was a significant difference amongst the numerical ratios (P  More

Data
Data were collected in 2015 through a survey deployed as part of the Feed the Future Buena Milpa project (http://www.cimmyt.org/project-profile/buena-milpa/). The project’s goal was to reduce food insecurity and malnutrition by fostering sustainable, resilient, and innovative maize-based farming systems in the WHG.
We conducted the survey, with support from local researchers and agronomy students of the University of San Carlos (USAC), in summer of 2015 in 989 maize-growing farm households in 64 communities of 16 municipalities of the WHG. The number of surveys at the departmental level was: Totonicapán (226), Quiche (350), Quetzaltenango (187) and Huehuetenango (226) (See Supplementary Table 1 and Supplementary Fig. 1 for details). Criteria to select communities were: location within the targeted municipalities in the Buena Milpa project, the active work of project partners within the communities, and their distribution within four selected watersheds that included farming systems at different altitudes, ranging from 1400 to 3200 masl. For household selection, enumerators walked radial transects to survey household members, choosing only households that explicitly agreed to participate and had agricultural land on at least part of which maize was grown. The survey was a closed questionnaire with 139 questions in 18 sections including 5 project themes: milpa-maize germplasm improvement, natural resource conservation in farming system, farming system diversification, agricultural innovation systems and social inclusion.
Milpa diversity and extent
We used survey findings regarding the previous year’s crop production (2014) for all 989 surveyed households to understand the importance and diversity of the milpa system. The 989 households reported crop production on a total of 1,541 plots, 1,324 of which included maize. The other 217 plots grew potato (85 plots), coffee (50), vegetables (31), bean (19), fruit trees (12), faba bean (7), forestry trees (6), pea (2), oats, wheat, etc. Given the study’s focus on smallholder farmers, we discarded other 119 plots that had unrealistically large values for plot size or maize production levels for the region (i.e. more than ten times the average plot size or calculated maize yield,  > 2 ha and  > 11 ton ha−1 year−1).
For the resulting 1205 plots we constructed a tree depicting maize-system diversity in the WHG, with each node indicating the main cropping associations. The tree was nested, so the first node displays all maize plots, with successive splits for monocrop vs intercropped maize and with crop names presented according to how often they are grown (overall: maize  > common beans  > potatoes  > squash  > faba beans  > fruit trees  > vegetables). So, in plots where maize is grown with potatoes and common beans, the association is termed maize-bean-potatoes.
Milpa and food security
We assessed maize yield differences for monocropping and intercropping to detect a yield penalty or advantage for maize (see below), including survey information only for households from which we had available yield data for all crops. In the survey, total production for each crop was recorded regardless of the number of plots on which it was grown and therefore, for 398 households (40.2% of the sample) that had more than one plot under maize or any of the milpa crops, it was impossible to calculate the yield and had to be discarded from the analysis due to a lack of available plot-level information. The results were further screened for complete crop information and unrealistic values on crop production levels (i.e. ten times higher than the average yield for the different crops—See Supplemental Table 2), resulting in a usable sample of 368 plots.
For statistical analysis, only those crop combinations with a sample size equal to or greater than 9 plots were selected, resulting in 357 plots with, in descending order, the following numbers of 300 plots sown to each cropping combination: maize monocrop (163), maize-bean (109), maize-bean-squash (30), maize-bean/faba (12), maize-potato (13), maize-squash (11), maize-bean-potato (10), maize-faba (9). Other crop combinations with very low sample sizes were maize-squash-faba (4), maize-potato-faba (3), maize-bean-squash-faba (2), maize-bean-potato-faba (1) and maize-bean-potato-squash (1), making it difficult to include them in further statistical analyses.
Maize yields for plots under monocropped maize (163) were first compared to maize yields from intercropped plots (194). To choose the most appropriate statistical test, we checked if the outcome variable, maize yield, met the assumptions required for a parametric test. Although we had independent samples, large sample sizes, and homogeneous variances (F = 1.1809, p-value = 0.267), maize yield proved to be non-normally distributed after Shapiro-Wilks normality test was significant (W = 0.911, p-value  1]{ }}}{{text{N}}}{ } times {overline{text{s}}},{[18]}$$

where (N) is the number of nutrients considered (14), (s_{i}) is the fraction of potentially nourished male adults by a given nutrient, and (overline{s}) is the average of potentially nourished persons for all nutrients. To calculate s we multiplied the yield of each crop reported in the survey (kg ha−1) by the nutrient composition of each crop, using the content per 100 g of the 14 different nutrients for each crop, as per the INCAP Food Composition Table for Central America32 and using the INCAP RDA for an adult male33 (See Supplementary Tables 3 and 4). All values were calculated per hectare and year. PNA levels across cropping associations were also compared. First, we checked if PNA data met the assumptions required for a parametric test. PNA variances proved to be non-homogeneous (F = 14.5, p-value =   More

1.
Lynam, C. P. et al. Have jellyfish in the Irish Sea benefited from climate change and overfishing?. Glob. Change Biol. 17, 767–782 (2011).
ADS  Article  Google Scholar 
2.
Condon, R. H. et al. Recurrent jellyfish blooms are a consequence of global oscillations. Proc. Natl. Acad. Sci. 110, 1000–1005 (2013).
ADS  CAS  PubMed  Article  Google Scholar 

3.
Lynam, C. P., Hay, S. J. & Brierley, A. S. Jellyfish abundance and climatic variation: Contrasting responses in oceanographically distinct regions of the North Sea, and possible implications for fisheries. J. Mar. Biol. Assoc. 85, 435–450 (2005).
Article  Google Scholar 

4.
Lynam, C. P. et al. Jellyfish overtake fish in a heavily fished ecosystem. Curr. Biol. 16, R492-493 (2006).
CAS  PubMed  Article  Google Scholar 

5.
Hays, G. C., Doyle, T. K. & Houghton, J. D. R. A paradigm shift in the trophic importance of jellyfish?. Trends Ecol. Evol. 33, 874–884 (2018).
PubMed  Article  Google Scholar 

6.
Gibbons, M. J. & Richardson, A. J. Patterns of jellyfish abundance in the North Atlantic. In Jellyfish Blooms: Causes, Consequences, and Recent Advances: Proceedings of the Second International Jellyfish Blooms Symposium, held at the Gold Coast, Queensland, Australia, 24–27 June, 2007 (eds. Pitt, K. A. & Purcell, J. E.) 51–65 (Springer Netherlands, 2009). https://doi.org/10.1007/978-1-4020-9749-2_4.

7.
Attrill, M. J., Wright, J. & Edwards, M. Climate-related increases in jellyfish frequency suggest a more gelatinous future for the North Sea. Limnol. Oceanogr. 52, 480–485 (2007).
ADS  Article  Google Scholar 

8.
Brodeur, R. D., Sugisaki, H. & Hunt, G. L. Jr. Increases in jellyfish biomass in the Bering Sea: Implications for the ecosystem. Mar. Ecol. Prog. Ser. 233, 89–103 (2002).
ADS  Article  Google Scholar 

9.
Pauly, D., Christensen, V., Dalsgaard, J., Froese, R. & Torres, F. Fishing down marine food webs. Science 279, 860–863 (1998).
ADS  CAS  PubMed  Article  Google Scholar 

10.
Purcell, J. E. & Arai, M. N. Interactions of pelagic cnidarians and ctenophores with fish: A review. Hydrobiologia 451, 27–44 (2001).
Article  Google Scholar 

11.
Robinson, K. L. et al. Jellyfish, forage fish, and the world’s major fisheries. Oceanography 27, 104–115 (2014).
Article  Google Scholar 

12.
Uye, S. Blooms of the giant jellyfish Nemopilema nomurai: A threat to the fisheries sustainability of the East Asian Marginal Seas. Plankton Benthos Res. 3, 125–131 (2008).
Article  Google Scholar 

13.
Wright, R. M., Le Quéré, C., Buitenhuis, E., Pitois, S. & Gibbons, M. Unique role of jellyfish in the plankton ecosystem revealed using a global ocean biogeochemical model. Biogeosci. Discuss. https://doi.org/10.5194/bg-2020-136 (2020).
Article  Google Scholar 

14.
Jackson, J. B. C. Ecological extinction and evolution in the brave new ocean. Proc. Natl. Acad. Sci. 105, 11458–11465 (2008).
ADS  CAS  PubMed  Article  Google Scholar 

15.
Kintner, A. & Brierley, A. S. Cryptic hydrozoan blooms pose risks to gill health in farmed North Atlantic salmon (Salmo salar). J. Mar. Biol. Assoc. 99, 539–550 (2019).
Article  Google Scholar 

16.
Flynn, B. A. et al. Temporal and spatial patterns in the abundance of jellyfish in the northern Benguela upwelling ecosystem and their link to thwarted pelagic fishery recovery. Afr. J. Mar. Sci. 34, 131–146 (2012).
Article  Google Scholar 

17.
Luo, J. Y. et al. Gelatinous zooplankton-mediated carbon flows in the global oceans: A data-driven modeling study. Glob. Biogeochem. Cycles 34, e2020GB006704 (2020).
ADS  CAS  Article  Google Scholar 

18.
Behrenfeld, M. J. et al. Climate-driven trends in contemporary ocean productivity. Nature 444, 752–755 (2006).
ADS  CAS  PubMed  Article  Google Scholar 

19.
Hays, G. C., Richardson, A. J. & Robinson, C. Climate change and marine plankton. Trends Ecol. Evol. 20, 337–344 (2005).
PubMed  Article  Google Scholar 

20.
Richardson, A. J. & Schoeman, D. S. Climate impact on plankton ecosystems in the northeast Atlantic. Science 305, 1609–1612 (2004).
ADS  CAS  PubMed  Article  Google Scholar 

21.
Suikkanen, S. et al. Climate change and eutrophication induced shifts in northern summer plankton communities. PLoS ONE 8, e66475 (2013).
ADS  CAS  PubMed  PubMed Central  Article  Google Scholar 

22.
Wiafe, G., Yaqub, H. B., Mensah, M. A. & Frid, C. L. J. Impact of climate change on long-term zooplankton biomass in the upwelling region of the Gulf of Guinea. ICES J. Mar. Sci. 65, 318–324 (2008).
Article  Google Scholar 

23.
Reid, P. C., Colebrook, J. M., Matthews, J. B. L. & Aiken, J. The Continuous Plankton Recorder: Concepts and history, from Plankton Indicator to undulating recorders. Prog. Oceanogr. 58, 117–173 (2003).
ADS  Article  Google Scholar 

24.
Edwards, M. et al. Plankton, jellyfish and climate in the North-East Atlantic. MCCIP Sci. Rev. 2020, 322–353. https://doi.org/10.14465/2020.arc15.plk (2020).
Article  Google Scholar 

25.
ICES. Report of the Working Group on the Celtic Seas Ecoregion (WGCSE), 11–19 May 2011, Copenhagen, Denmark. (2011).

26.
Bartolino, V. et al. Herring assessment working group for the area south of 62° N (HAWG). (2019) https://doi.org/10.17895/ices.pub.5460.

27.
ICES Advice Book 5. https://www.ices.dk/sites/pub/Publication%20Reports/Advice/2007/may/her-nirs.pdf (2007).

28.
Beaugrand, G. The North Sea regime shift: Evidence, causes, mechanisms and consequences. Prog. Oceanogr. 60, 245–262 (2004).
ADS  Article  Google Scholar 

29.
Gregory, B., Christophe, L. & Martin, E. Rapid biogeographical plankton shifts in the North Atlantic Ocean. Glob. Change Biol. 15, 1790–1803 (2009).
ADS  Article  Google Scholar 

30.
deYoung, B. et al. Regime shifts in marine ecosystems: Detection, prediction and management. Trends Ecol. Evol. 23, 402–409 (2008).
PubMed  Article  Google Scholar 

31.
Bastian, T. et al. Large-scale sampling reveals the spatio-temporal distributions of the jellyfish Aurelia aurita and Cyanea capillata in the Irish Sea. Mar. Biol. 158, 2639–2652 (2011).
Article  Google Scholar 

32.
Houghton, J. D. R., Doyle, T. K., Davenport, J. & Hays, G. C. Developing a simple, rapid method for identifying and monitoring jellyfish aggregations from the air. Mar. Ecol. Prog. Ser. 314, 159–170 (2006).
ADS  Article  Google Scholar 

33.
Bastian, T., Lilley, M. K. S., Beggs, S. E., Hays, G. C. & Doyle, T. K. Ecosystem relevance of variable jellyfish biomass in the Irish Sea between years, regions and water types. Estuar. Coast. Shelf Sci. 149, 302–312 (2014).
ADS  Article  Google Scholar 

34.
Heckerman, D., Geiger, D. & Chickering, D. M. Learning Bayesian networks: The combination of knowledge and statistical data. Mach. Learn. 20, 197–243 (1995).
MATH  Google Scholar 

35.
Milns, I., Beale, C. M. & Smith, V. A. Revealing ecological networks using Bayesian network inference algorithms. Ecology 91, 1892–1899 (2010).
PubMed  Article  Google Scholar 

36.
Mitchell, E. G. & Neutel, A.-M. Feedback spectra of soil food webs across a complexity gradient, and the importance of three-species loops to stability. Theor. Ecol. 5, 153–159 (2012).
Article  Google Scholar 

37.
Olff, H. et al. Parallel ecological networks in ecosystems. Philos. Trans. R. Soc. B Biol. Sci. 364, 1755–1779 (2009).
Article  Google Scholar 

38.
Mitchell, E. G., Whittle, R. & Griffths, H. J. Benthic ecosystem cascade effects in Antarctica using Bayesian network inference. Commun. Biol. 3, 582 (2020).
PubMed  PubMed Central  Article  Google Scholar 

39.
Yu, J., Smith, V. A., Wang, P. P., Hartemink, A. J. & Jarvis, E. D. Advances to Bayesian network inference for generating causal networks from observational biological data. Bioinformatics 20, 3594–3603 (2004).
CAS  PubMed  Article  Google Scholar 

40.
Yu, J., Smith, V. A., Wang, P. P., Hartemink, E. J. & Jarvis, E. D. Using Bayesian network inference algorithms to recover molecular genetic regulatory networks. In Prof. of Int. (2002).

41.
Smith, V. A., Yu, J., Smulders, T. V., Hartemink, A. J. & Jarvis, E. D. Computational inference of neural information flow networks. PLOS Comput. Biol. 2, e161 (2006).
ADS  PubMed  PubMed Central  Article  CAS  Google Scholar 

42.
Mitchell, E. G. & Butterfield, N. J. Spatial analyses of Ediacaran communities at Mistaken Point. Paleobiology 44, 40–57 (2018).
Article  Google Scholar 

43.
Mitchell, E. G., Durden, J. M. & Ruhl, H. A. First network analysis of interspecific associations of abyssal benthic megafauna reveals potential vulnerability of abyssal hill community. Prog. Oceanogr. 187, 102401 (2020).
Article  Google Scholar 

44.
Mitchell, E. G. & Harris, S. Mortality, population and community dynamics of the glass sponge dominated community “The Forest of the Weird” from the RIDGE seamount, Johnston Atoll, Pacific Ocean. Front. Mar. Sci. 7, 872 (2020).
Article  Google Scholar 

45.
Reid, P. C., Borges, M. D. F. & Svendsen, E. A regime shift in the North Sea circa 1988 linked to changes in the North Sea horse mackerel fishery. Fish. Res. 50, 163–171 (2001).
Article  Google Scholar 

46.
Brierley, A. S. et al. Acoustic observations of jellyfish in the Namibian Benguela. Mar. Ecol. Prog. Ser. 210, 55–66 (2001).
ADS  Article  Google Scholar 

47.
Brierley, A. S. et al. Towards the acoustic estimation of jellyfish abundance. Mar. Ecol. Prog. Ser. 295, 105–111 (2005).
ADS  Article  Google Scholar 

48.
MacLennan, D. N. & Simmonds, E. J. Fisheries Acoustics (Springer, Berlin, 2013).
Google Scholar 

49.
Planque, B. & Fromentin, J. Calanus and environment in the eastern North Atlantic. I. Spatial and temporal patterns of C. finmarchicus and C. helgolandicus. Mar. Ecol. Prog. Ser. 134, 101–109 (1996).
ADS  Article  Google Scholar 

50.
Batten, S. D. et al. CPR sampling: The technical background, materials and methods, consistency and comparability. Prog. Oceanogr. 58, 193–215 (2003).
ADS  Article  Google Scholar 

51.
Richardson, A. J. et al. Using continuous plankton recorder data. Prog. Oceanogr. 68, 27–74 (2006).
ADS  Article  Google Scholar 

52.
John, E. H. et al. Continuous plankton records stand the test of time: Evaluation of flow rates, clogging and the continuity of the CPR time-series. J. Plankton Res. 24, 941–946 (2002).
Article  Google Scholar 

53.
Beaugrand, G., Ibañez, F., Lindley, J. A. & Reid, P. C. Diversity of calanoid copepods in the North Atlantic and adjacent seas: Species associations and biogeography. Mar. Ecol. Prog. Ser. 232, 179–195 (2002).
ADS  Article  Google Scholar 

54.
Yu, J. Developing Bayesian Network Inference Algorithms to Predict Causal Functional Pathways in Biological Systems (Duke University, Durham, 2005).
Google Scholar 

55.
Chickering, D. M. Learning Bayesian Networks is NP-Complete. In Learning from Data: Artificial Intelligence and Statistics V (eds. Fisher, D. & Lenz, H.-J.) 121–130 (Springer, Berlin, 1996). https://doi.org/10.1007/978-1-4612-2404-4_12.

56.
Rayner, N. A. et al. Global analyses of sea surface temperature, sea ice, and night marine air temperature since the late nineteenth century. J. Geophys. Res. Atmos. 108, (2003).

57.
Jones, P. D., Jonsson, T. & Wheeler, D. Extension to the North Atlantic oscillation using early instrumental pressure observations from Gibraltar and south-west Iceland. Int. J. Climatol. 17, 1433–1450 (1997).
Article  Google Scholar 

58.
Mitchell, E. G. Functional programming through deep time: Modeling the first complex ecosystems on earth. ACM SIGPLAN Not. 46, 28–31 (2011).
MATH  Article  Google Scholar 

59.
Jones, S. P. Haskell 98 Language and Libraries: The Revised Report (Cambridge University Press, Cambridge, 2003).
Google Scholar 

60.
Friedman, N., Geiger, D. & Goldszmidt, M. Bayesian network classifiers. Mach. Learn. 29, 131–163 (1997).
MATH  Article  Google Scholar 

61.
Dickey-Collas, M., Nash, R. D. M. & Brown, J. The location of spawning of Irish sea herring (Clupea harengus). J. Mar. Biol. Assoc. 81, 713–714 (2001).
Article  Google Scholar 

62.
Nash, R. D. M. & Geffen, A. J. Seasonal and interannual variation in abundance of Calanus finmarchicus (Gunnerus) and Calanus helgolandicus (Claus) in inshore waters (west coast of the Isle of Man) in the central Irish Sea. J. Plankton Res. 26, 265–273 (2004).
Article  Google Scholar 

63.
Hurrell, J. W. & Deser, C. North Atlantic climate variability: The role of the North Atlantic Oscillation. J. Mar. Syst. 78, 28–41 (2009).
Article  Google Scholar 

64.
Brotz, L., Cheung, W. W. L., Kleisner, K., Pakhomov, E. & Pauly, D. Increasing jellyfish populations: trends in Large Marine Ecosystems. In Jellyfish Blooms IV: Interactions with Humans and Fisheries (eds. Purcell, J. et al.) 3–20 (Springer Netherlands, 2012). https://doi.org/10.1007/978-94-007-5316-7_2.

65.
Purcell, J. E. Jellyfish and ctenophore blooms coincide with human proliferations and environmental perturbations. Annu. Rev. Mar. Sci. 4, 209–235 (2012).
ADS  Article  Google Scholar 

66.
Pitt, K. A., Lucas, C. H., Condon, R. H., Duarte, C. M. & Stewart-Koster, B. Claims that anthropogenic stressors facilitate jellyfish blooms have been amplified beyond the available evidence: A systematic review. Front. Mar. Sci. 5, 451 (2018).
Article  Google Scholar 

67.
Sanz-Martín, M. et al. Flawed citation practices facilitate the unsubstantiated perception of a global trend toward increased jellyfish blooms. Glob. Ecol. Biogeogr. 25, 1039–1049 (2016).
Article  Google Scholar 

68.
Dunne, J. A., Williams, R. J. & Martinez, N. D. Network structure and biodiversity loss in food webs: Robustness increases with connectance. Ecol. Lett. 5, 558–567 (2002).
Article  Google Scholar 

69.
Gardner, M. R. & Ashby, W. R. Connectance of large dynamic (cybernetic) systems: Critical values for stability. Nature 228, 784–784 (1970).
ADS  CAS  PubMed  Article  Google Scholar 

70.
Lawton, J. H. & Brown, V. K. Redundancy in ecosystems. In Biodiversity and Ecosystem Function (eds. Schulze, E.-D. & Mooney, H. A.) 255–270 (Springer, Berlin, 1994). https://doi.org/10.1007/978-3-642-58001-7_12.

71.
Thébault, E. & Loreau, M. Trophic interactions and the relationship between species diversity and ecosystem stability. Am. Nat. 166, E95–E114 (2005).
PubMed  Article  Google Scholar 

72.
Wang, S. & Loreau, M. Biodiversity and ecosystem stability across scales in metacommunities. Ecol. Lett. 19, 510–518 (2016).
PubMed  PubMed Central  Article  Google Scholar 

73.
Van Voris, P., O’Neill, R. V., Emanuel, W. R. & Shugart, H. H. Functional complexity and ecosystem stability. Ecology 61, 1352–1360 (1980).
Article  Google Scholar 

74.
Graham, W. M. & Kroutil, R. M. Size-based prey selectivity and dietary shifts in the jellyfish, Aurelia aurita. J. Plankton Res. 23, 67–74 (2001).
Article  Google Scholar 

75.
Marques, R., Bonnet, D., Carré, C., Roques, C. & Darnaude, A. M. Trophic ecology of a blooming jellyfish (Aurelia coerulea) in a Mediterranean coastal lagoon. Limnol. Oceanogr. n/a.

76.
Anninsky, B. E., Finenko, G. A., Datsyk, N. A. & Kıdeyş, A. E. Trophic ecology and assessment of the predatory impact of the Moon jellyfish Aurelia aurita (Linnaeus, 1758) on zooplankton in the Black Sea (2020) https://doi.org/10.21411/cbm.a.96dd01aa.

77.
Widmer, C. L., Fox, C. J. & Brierley, A. S. Effects of temperature and salinity on four species of northeastern Atlantic scyphistomae (Cnidaria: Scyphozoa). Mar. Ecol. Prog. Ser. 559, 73–88 (2016).
ADS  Article  Google Scholar 

78.
Watson, D. I. & Barnes, D. K. A. Temporal and spatial components of variability in benthic recruitment, a 5-year temperate example. Mar. Biol. 145, 201–214 (2004).
Article  Google Scholar 

79.
Arrhenius, F. & Hansson, S. Food consumption of larval, young and adult herring and sprat in the Baltic Sea. Mar. Ecol. Prog. Ser. 96, 125–137 (1993).
ADS  Article  Google Scholar 

80.
Williams, R., Conway, D. V. P. & Hunt, H. G. The role of copepods in the planktonic ecosystems of mixed and stratified waters of the European shelf seas. Hydrobiologia 292, 521–530 (1994).
Article  Google Scholar 

81.
Gowen, R. J., Mills, D. K., Trimmer, M. & Nedwell, D. B. Production and its fate in two coastal regions of the Irish Sea: The influence of anthropogenic nutrients. Mar. Ecol. Prog. Ser. 208, 51–64 (2000).
ADS  Article  Google Scholar 

82.
Scorrano, S., Aglieri, G., Boero, F., Dawson, M. N. & Piraino, S. Unmasking Aurelia species in the Mediterranean Sea: An integrative morphometric and molecular approach. Zool. J. Linn. Soc. 180, 243–267 (2017).
Google Scholar 

83.
Haussermann, V., Dawson, M. N. & Forsterra, G. First record of the moon jellyfish, Aurelia for Chile. Spixana 32, 3–7 (2009).
Google Scholar  More

Study area
The field work was carried out at the Sturt Meadows calcrete aquifer (28°41′ S 120°58′ E) located on Sturt Meadows pastoral station, Western Australia, ~ 42 km from the settlement of Leonora (833 km northeast of Perth, see Fig. 1a). The study area is a calcrete aquifer lying in the Raeside paleodrainages in the Yilgarn region of Western Australia (Fig. 1a). The vegetation of the area is Acacia woodlands, primarily Acacia aneura (F.Muell. ex Benth.), and is subjected to combined grazing pressure from domestic stock, feral animals and macropods. The aquifer is accessible through a bore grid comprising 115 bore holes of between 5 and 11 m depth (Fig. 1b).
Figure 1

Map of the Sturt Meadows calcrete. (a) Location within the Yilgarn craton region and detailed paleodrainages/cacretes in the area and (b) the grid map showing the location of the boreholes sampled for stygofauna together with probe samples, water samples (in light blue) and the combination of both. Map was produced in ArcGIS Desktop 10.679 and edited in Adobe Illustrator 25.080.

Full size image

Three sampling campaigns were carried out, two of them (LR1: 26/07/2017 and LR2: 07/11/2017) corresponding to low rainfall periods24 and one during the wet season (high rainfall, HR; two consecutive days of sampling collection: 17-18/03/2018) (Supplementary Fig. 1). The well-studied stygofaunal community of the area is composed of 11 main stygofaunal taxa belonging to five Classes: Oligochaeta (family Tubificidae (Vejdovský 1884)), subcohort Hydrachnidia, Maxillopoda (two species of harpacticoids: Novanitocrella cf. aboriginesi (Karanovic, 2004), Schizopera cf. austindownsi (Karanovic, 2004) and four species of cyclopoids: Halicyclops kieferi (Karanovic, 2004), Halicyclops cf. ambiguous (Kiefer, 1967), Schizopera slenderfurca (Karanovic & Cooper, 2012) and Fierscyclops fiersi (De Laurentiis et al., 2001)), Malacostraca: Amphipoda (species Scutachiltonia axfordi (King, 2012), Yilgarniella sturtensis (King, 2012) and Stygochiltonia bradfordae (King, 2012)) and Insecta: Coleoptera: Dytiscidae (species Paroster macrosturtensis (Watts & Humphreys 2006), Paroster mesosturtensis (Watts & Humphreys 2006) and Paroster microsturtensis (Watts & Humphreys 2006) and respective larvae).
Field work procedures and sample preparation
Given the sensitivity of the hydrological dynamics in shallow calretes23,25, extensive water extraction along the bores was avoided, and preliminary tests on the bores with the highest water depth were carried out to quantify potential risk of dewatering the calcrete. During the field campaigns LR2 and HR, 20 water samples in total (two samples for stable isotope analysis on DOC (Dissolved Organic Carbon) and DIC (Dissolved Inorganic Carbon), three samples for radiocarbon analysis on DOC, one sample for radiocarbon analysis on DIC, and two samples for stable isotope and radiocarbon analyses on POC (Particulate Organic Carbon)) were collected from bores D13 and W4 (Fig. 1b), which are representative of the two main geological conformations of the area—calcrete (W4) and clay (D13) (Supplementary Fig. 2)—and host stable hydrological and biotic conditions7. Water samples were collected using a submersible centrifugal pump (GEOSub 12 V Purging Pump) after wells were purged of three well-volumes and stabilisation of in-field parameters was observed, according to the methodology in Bryan et al.26.
Samples for 14CDIC analysis were filtered through 0.45 μm filters and collected in 1 L high density poly-ethylene (HDPE) bottles. δ13CDIC samples were filtered through 0.2 μm filters, collected in 12 mL glass vials (Exetainers) and refrigerated after sampling. δ 13CDOC samples were filtered through 0.2 μm filters, collected in 60 mL HDPE bottles and frozen after sampling.14CDOC samples were filtered through 0.2 μm filters, collected in 3 L HDPE bottles and frozen after sampling.
In order to investigate 14C and δ 13C content of POC, two extra liters were collected from the same bores (D13, W4) and kept frozen (− 20 °C) until further analyses. 14CPOC δ13CPOC samples were then filtered through pre-combusted GF/F filters (12 h at 450 °C), washed with 1.2 N HCl to remove any inorganic carbon, and subsequently dried at 60 °C for 24 h. All samples were closed with sealing tape after collection to limit atmospheric exchange and kept in darkness.
Temperature, pH, ORP, salinity, DO and depth were measured in situ using a portable Hydrolab Quanta Multi‐Probe Meter across 30 bores during LR1, LR2 and HR23 (presented in Supplementary Table 8). Adult and larval stygofaunal specimens were collected from the same bores by hauling a weighted plankton net (mesh 100 μm27) five times through the water column (Fig. 1b). All biological samples were kept frozen (− 20 °C) in darkness until laboratory processing. Individual organisms were counted and identified (and consequently separated) to the lowest taxonomic level via optical microscopy and reference to specific taxonomic keys. Plant material, sediment samples and fauna were each separated during the sorting in the laboratory and each taxon pooled according to sampling campaign (LR1, LR2 or HR) and subsequently washed with Milli-Q water to remove surface impurities from their bodies. Sediment samples were soaked in acid (0.1 N HCl) to remove inorganic carbon, and together with the other samples were then oven dried at 60 °C overnight and ground until obtaining a homogeneous fine powder and stored at − 20 °C until further analyses.
Previous investigations on the ecological niche trends at Sturt Meadows indicated that all stygofauna characterize similar niche occupations under low rainfall regimes (LR1 and LR2)23. Stygofaunal specimens from the two low rainfall sampling events were combined to form sample LR to address the competing requirements between isotopic detection limits, analytical replicates and cost, while maintaining the main taxonomic and biological classifications7.
Bulk isotope and 14C analyses
Water δ13CDIC and δ13CPOC isotopic ratios were analysed by Isotope Ratio Mass Spectrometer—WABC at The University of Western Australia using a GasBench II coupled with a Delta XL Mass Spectrometer (Thermo-Fisher Scientific)—and the results, with a precision of ± 0.10‰, were reported as ‰ deviation from the NBS19 and NSB18 international carbonate standard28.
δ13CDOC isotopic ratios of waters were analysed via Liquid Chromatography Isotope Ratio Mass Spectrometer (LC-IRMS) at the La Trobe Institute for Molecular Sciences (LIMS, La Trobe University, Melbourne, Australia) comprising an Accela 600 pump connected to a Delta V Plus Isotope Ratio Mass Spectrometer via a Thermo Scientific LC Isolink (Thermo Scientific).
C and N bulk SIA on homogenised samples of sediment, roots, stygofauna and copepods (cyclopoids and harpacticoids) were performed at the Australian Nuclear Science and Technology Organisation (ANSTO, Sydney, Australia). Samples were loaded into tin capsules and analysed with a continuous flow isotope ratio mass spectrometer (CF-IRMS), model Delta V Plus (Thermo Scientific Corporation, U.S.A.), interfaced with an elemental analyser (Thermo Fisher Flash 2000 HT EA, Thermo Electron Corporation, USA) following the procedure published by Mazumder et al.29.
For radiocarbon analyses, samples (sediment, roots, copepods, ants, stygofauna) were treated with 1 M HCl for 2 h to remove all possible carbonate contamination. These pre-treated samples together with 14CPOC, 14CDOC and 14CDIC samples were subjected to CO2 extraction and graphitization following the methodology published by Hua et al.30. 14C content of samples was determined by means of the Accelerator Mass Spectrometry (AMS) at ANSTO.
Carbon CSIA
Carbon CSIA followed the procedure described in Saccò et al.7. Samples of roots and stygofaunal specimens were hydrolysed under vacuum with 0.5 to 1 mL of amino acid-free 6 M HCl (Sigma-Aldrich) at 110 °C for 24 h. The protein hydrolysates were dried overnight in a rotary vacuum concentrator and stored in a freezer. Prior to analysis, the samples were dissolved in Milli-Q water and 10 μL of 1-mmol solution of 2-aminoisobutyric acid (Sigma-Aldrich) as internal standard. The sample stock had a concentration of approximately 8 to 10 mg/mL, which was further diluted as needed. Single amino acid carbon isotope analysis was carried out at the La Trobe Institute for Molecular Sciences (LIMS, La Trobe University, Melbourne, Australia) using an Accela 600 pump connected to a Delta V Plus Isotope Ratio Mass Spectrometer via a Thermo Scientific LC Isolink (Thermo Scientific).
The amino acids were separated using a mixed mode (reverse phase/ion exchange) Primesep A column (2.1 × 250 mm, 100 °C, 5 μm, SIELC Technologies) following the chromatographic method described in Mora et al.31 after Smith et al.32. Mobile phases are those described in Mora et al.33. Percentage of Phases B and C in the conditioning run, as well as flow rate of the analytical run and timing of onset of 100% Phase C were adjusted as needed. Samples were injected onto the column in the 15 μL—partial loop or no waste—injection mode, and measured in duplicate or triplicate.
To elucidate carbon flows through the stygofaunal community we focused on the essential amino acids Valine (Val), Phenylalanine (Phe) and Arginine (Arg), as these compounds must be integrated through diet and cannot be synthetised internally by the fauna14,34. In addition, to distinguish between terrestrial and aquatic carbon sources, the ratio between Val and Phe signals (δ13CVal-Phe), a widely employed index in archaeology and freshwater biology35, was calculated for roots, water mites, aquatic worms, amphipods and beetles (larvae and adults).
Microbial taxonomic and functional gene analyses
Consumer amphipods (Scutachiltonia axfordi (AM1), Yilgarniella sturtensis (AM2), S. bradfordae (AM3)), cyclopoids and harpacticoids, together with predator stygobiotic beetles (Paroster macrosturtensis (B), P. mesosturtensis (M) and P. microsturtensis (S)) (see Saccò el at 7. for further details on the trophic characterisation of the stygofaunal community at Sturt Meadows), were used for gut microbiome bacterial 16S metabarcoding and microbial functional analysis. A total of 16 AM1, 16 AM2, 16 AM3, 20 cyclopoids and 20 harpaticoids and 20 of each one of the three Paroster species (B, M and S), were sorted into duplicates of stygobiotic pools of 3–5 individuals from both LR and HR events for DNA extraction. Prior to DNA extraction stygobitic animals (3–5 individuals per pool; n = 40) were placed in a petri dish containing ultrapure water and UV sterilized in a UV oven for 10 min to eliminate any bacterial species that may be contained on the exoskeleton as this study targeted the gut microbiome. Immediately post-UV treatment, the animals were placed into tissue lysis tubes with 180 μL tissue lysis buffer (ATL) and 20 μL Proteinase K and homogenised using Minilys tissue homogeniser (ThermoFisher Scientific, Australia) at high speed for 30 s. Lysis tubes, inclusive of two laboratory controls, were incubated at 56 °C using an agitating heat block (Eppendorf ThermoStat C, VWR, Australia) for 5 h.
Following the incubation, the analytical procedure was adapted from Saccò et al.22 and DNA extraction was carried out using DNeasy Blood and Tissue Kit (Qiagen; Venlo, Netherlands) and eluted off the silica column in 30–50 μL AE buffer. The quality and quantity of DNA extracted from each stygobitic pool was measured using quantitative PCR (qPCR), targeting the bacterial 16S gene. PCR reactions were used to assess the quality and quantity of the DNA target of interest via qPCR (Applied Biosystems [ABI], USA) in 25 μL reaction volumes consisting of 2 mM MgCl2 (Fisher Biotec, Australia), 1 × PCR Gold Buffer (Fisher Biotec, Australia), 0.4 μM dNTPs (Astral Scientific, Australia), 0.1 mg bovine serum albumin (Fisher Biotec, Australia), 0.4 μM of each primer (Bact16S_515F and Bact16S_806R36,37), and 0.2 μL of AmpliTaq Gold (AmpliTaq Gold, ABI, USA), and 2 μL of template DNA (Neat, 1/10, 1/100 dilutions). The cycling conditions were: initial denaturation at 95 °C for 5 min, followed by 40 cycles of 95 °C for 30 s, 52 °C for 30 s, 72 °C for 30 s, and a final extension at 72 °C for 10 min.
DNA extracts that successfully yielded DNA of sufficient quality, free of inhibition, as determined by the initial qPCR screen (detailed above), were assigned a unique 6–8 bp multiplex identifier tag (MID-tag) with the bacterial 16S primer set. Independent MID-tag qPCR for each stygobiotic pool were carried out in 25 μL reactions containing 1 × PCR Gold Buffer, 2.5 mM MgCl2, 0.4 mg/mL BSA, 0.25 mM of each dNTP, 0.4 μM of each primer, 0.2 μL AmpliTaq Gold and 2–4 μL of DNA as determined by the initial qPCR screen. The cycling conditions for qPCR using the MID-tag primer sets were as described above. MID-tag PCR amplicons were generated in duplicate and the library was pooled in equimolar ratio post-PCR for DNA sequencing. The final library was size selected (160–600 bp) using Pippin Prep (Sage Sciences, USA) to remove any MID-tag primer-dimer products that may have formed during amplification. The final library concentration was determined using a QuBitTM 4 Fluorometer (Thermofischer, Australia) and sequenced using a 300 cycle V2 kit on an Illumina MiSeq platform (Illumina, USA).
MID-tag bacterial 16S sequence reads obtained from the MiSeq were sorted (filtered) back to the stygobitic pool based on the MID-tags assigned to each DNA extract using Geneious v10.2.538. MID-tag and primer sequences were trimmed from the sequence reads allowing for no mismatch in length or base composition.
Filtered reads were then input into a containerised workflow comprising USEARCH39 and BLASTN40, which was run on a high-throughput HPC cluster at Pawsey supercomputing centre. The fastx-uniques, unoise3 (with minimum abundance of 8) and otutab commands of USEARCH were applied to generate unique sequences, ZOTUs (zero-radius Operational Taxonomic Units) and abundance table, respectively. The ZOTUs were compared against the nucleotide database using the following parameters in BLASTN: perc_identity ≥ 94, evalue ≤ 1e−3, best_hit_score_edge 0.05, best_hit_overhang 0.25, qcov_hsp_perc 100, max_target_seqs = 5. An in-house Python script was used to assign the ZOTUs to their lowest common ancestor (LCA)41. The threshold for dropping a taxonomic assignment to LCA was set to perc_identity ≥ 96 and the difference between the % of identity of the two hits when their query coverage is equal was set to 1. Results on the microbial taxonomic diversity detected in ground water samples from a previous study on carbon inputs in the aquifer22 were incorporated in this work to allow qualitative comparison with the stygofaunal gut diversity.
To investigate functional activity involved in carbon cycling, the 16S metabarcoding data were fed to the Phylogenetic Investigation of Communities by Reconstruction of Unobserved States 2 (PICRUSt2) software package to generate predicted metagenome profiles42. These profiles were clustered into Kyoto Encyclopedia of Genes and Genomes (KEGG) Orthologs (KOs)43 and MetaCyc pathway abundances44 focusing on the relative abundances of four carbon metabolisms: carbon fixation in prokaryotes, carbohydrates, lipids and amino acid metabolisms. Relative abundance of pathways linked with methane, nitrogen and sulfur metabolisms were also investigated.
Statistical analyses
The Phyloseq package in R45,46 was used to plot the ZOTU abundance for the styfofaunal specimens at the order level under LR and HR periods. The Statistical Analysis of Metagenomic Profiles (STAMP) bioinformatics software package was used to carry out Principal Components Analysis (PCA) to assess the differences between functional genomic fingerprints based on the KEGG orthologs function prediction between copepods (C and H) and amphipods (AM1, AM2 and AM3), and determine statistically significant results from the PICRUSt2 output47. Clustering patterns according to rainfall periods (LR and HR) and major consumers taxonomic groups (cyclopoids, harpacticoids and amphipods) were assessed through Permutational multivariate analysis of variance (PERMANOVA, R-package46 ‘vegan’) and pairwise post hoc pairwise multilevel comparisons48.
For comparison of potential shifts in abundances of microbial metabolic pathways within groundwater samples, copepods and amphipods across rainfall periods, analysis of variance (ANOVA) was performed on the abundance data (two replicates per each group) on the predicted pathways depicting carbon fixation, carbohydrate, lipid, amino acid, methane, nitrogen and sulfur metabolisms. ANOVAs coupled with Tukey’s HSD pairwise comparisons (R-package46 ‘stats’) were employed to inspect significant differences between bulk SIA (δ13C and δ15N) and essential amino acid (δ13CPhe, δ13CArg, δ13CVal and δ13CVal-Phe) data from the stygofaunal taxa within the different rainfall conditions (LR and HR). PERMANOVAs (R-package46 ‘vegan’) were also performed to investigate the potential clustering trends within the stygofaunal taxa across rainfall periods from the combination of radiocarbon (Δ14C) and carbon SIA (δ13C) isotopic fingerprints.
SIMM (Stable Isotope Mixing Models, R-package46 ‘simmr’) were used to estimate dietary proportions of copepods and amphipods within a Bayesian framework. Due to the lack of species-specific trophic discrimination factors for stygofauna, we employed the widely accepted values of 3.4 ± 2‰ for nitrogen and 0.5 ± 1‰ for carbon49. Markov chain Monte Carlo (MCMC) algorithms were used for simulating posterior distributions in SIMM, and MCMC convergence was evaluated using the Gelman-Rubin diagnostic by using 1.1 as a threshold value for analysis validation. More

Interseasonal VGC dominates peak-to-late-season growth
We first examined the VGC effect at the seasonal scale, using satellite-derived Normalized Difference Vegetation Index (NDVI, see “Methods”) for the 1982–2016 period. We defined the dormancy season (DS) and three periods of the growing season, i.e., EGS, peak growing season (PGS), and late growing season (LGS), based on phenological metrics (see “Methods”). The partial autocorrelation calculated for NDVI time series of two consecutive seasons, after factoring out concurrent and preceding climatic impacts, provides an estimate of the interseasonal VGC effects (see “Methods”). At the hemispheric scale, our analyses show a significant (p  More

State variables
The first step in describing the macroscopic properties of the swarm is to define a set of state variables that fully characterizes the state of the system. The equation of state then links these state variables in a functional relation. In classical thermodynamics, a complete set of state variables is given by the conjugate pairs of pressure P and volume V, temperature T and entropy S, and, if the number of particles is not fixed, chemical potential μ and number of particles N. We use an analogous set of state variables here to characterize swarms. The most straightforward state variable to define is the number of individuals N, which is given simply by the number of midges in the swarm at a given time (note that midges that are not swarming simply sit on the walls or floor of the laboratory enclosure). The volume V of the swarm can be straightforwardly defined and computed as the volume of the convex hull enclosing all the midges. Note that while N and V are not independently controllable quantities, the ratio N/V is empirically approximately constant in large swarms24, meaning that the “thermodynamic” limit (that is, (N to infty) and (V to infty) with (frac{N}{V} to rho)) is approached in our swarms25. In typical swarming events, N changes on a time scale that is very slow compared to the swarm dynamics; thus, a chemical potential is not needed to describe the instantaneous state of the swarm. Note, though, that since the number of midges varies between measurements that may be separated by many days, N remains a relevant state variable for capturing swarm-to-swarm variability.
The remaining three state variables are somewhat more subtle, but can be defined by building on previous work. It has been explicitly shown26 that a virial relation based on the kinetic energy and an effective potential energy holds for laboratory swarms of Chironomus riparius. For particles moving in a potential, this virial relation can be used to define a pressure26. As we have shown previously, swarming midges behave as if they are trapped in a harmonic potential well that binds them to the swarm, with a spring constant k(N) that depends on the swarm size24,26 (Fig. 1b). The difference between the kinetic energy and this harmonic potential energy thus allows us to compute a pressure4,26,27, which is conceptually similar to the swim pressure defined in other active systems28. The virial theorem thus provides a link between kinetic energy, potential energy, and a field that plays the role of a pressure, when coupled with the observation that individual midges to a good approximation behave as if they are moving in a harmonic potential24,26. We can write this virial pressure P (per unit mass, assuming a constant mass per midge) as

$$P = frac{1}{3NV}mathop sum limits_{i = 1}^{N} left( {v_{i}^{2} – frac{1}{2}langle krangle r_{i}^{2} } right),$$

where N is the number of midges in the swarm, V is the swarm volume, vi is the velocity of midge i, ri is its distance from the swarm centre of mass, and (langle krangle = langle – {mathbf{a}}_{i} cdot hat{mathbf{r}}_{i} / r_{i} rangle) is the effective spring constant of the emergent potential well that binds midges to the swarm. In this expression, ai is the acceleration of midge i, (hat{mathbf{r}}_{i}) is the unit vector pointing from a midge towards the instantaneous centre of mass of the swarm (defined as (1/Nsumnolimits_{i = 1}^{N} {{mathbf{r}}_i })) and averages are taken over the individuals in the swarm. This spring constant depends on the swarm size N (Fig. 1b). We note that we have previously simply used the directly computed potential energy (- langle {mathbf{a}}_{i} cdot {mathbf{r}}_{i} rangle) to define the pressure4,27; here, we instead average the potential terms and fit them to a power law in N (Fig. 1b) to mitigate the contribution of spurious instantaneous noise in the individual positions that would be enhanced by differentiating them twice to compute accelerations. We use this power law to determine the spring constant k instantaneously at each time step.
The results from the two methods for computing the pressure are similar and consistent, but the method we use here is less prone to noise. Physically, this pressure P can be interpreted as the additional spatially variable energy density required to keep the midges bound to the swarm given that their potential energy varies in space but their mean velocity (and therefore kinetic energy) does not. Thus, compared to a simple passive particle moving in a harmonic well, midges have more kinetic energy than expected at the swarm edges; this pressure compensates for the excess kinetic energy. This pressure should be viewed as a manifestation of the active nature of the midges (similar to a swim pressure28), since the kinetic energy is an active property of each individual midge and the potential energy is an emergent property of the swarm.
We can define a Shannon-like entropy S via its definition in terms of the joint probability distributions of position and velocity. This entropy is defined as

$$S = – mathop smallint limits_{ – infty }^{infty } p(x,;v)log_{2} p(x,;v)dxdv,$$

where p(x,v) is the joint probability density function (PDF) of midge position and velocity. S here is measured in bits, as it is naturally an information entropy. Empirically, we find that the position and velocity PDFs are nearly statistically independent for all components and close to Gaussian, aside from the vertical component of the position (Fig. 1c–f). However, the deviation from Gaussianity in this component (which occurs because of the symmetry breaking due to the ground) does not significantly affect the estimate of the entropy; thus, we approximate it as Gaussian as well. Making these approximations, we can thus analytically write the (extensive) entropy as

$$S = frac{3N}{{ln 2}}ln left( {2pi esigma_{x} sigma_{v} } right),$$

where (sigma_{x}) and (sigma_{v}) are the standard deviations of the midge positions and velocities, respectively. In practice, we calculated (sigma_{v}) by averaging the instantaneous root-mean-square values of all three velocity components rather than a time-averaged value; the difference between these components was always less than 10%. This expression makes it more clear why the Gaussian approximation for the vertical component of the position is reasonable here: only the mean and variance of the PDFs are required to compute the entropy, and these low moments are very similar for the true data and the Gaussian estimate.
Although there is no obvious definition of temperature for a swarm, we can define one starting from the entropy, since temperature (when scaled by a Boltzmann constant) can be defined as the increase in the total physical energy of the system due to the addition of a single bit of entropy. Given our definitions, adding a single bit of entropy (that is, setting (S to S + 1)) for constant (sigma_{x}) and N (that is, a swarm of fixed number and spatial size) is equivalent to setting (sigma_{v} to 2^{1/(3N)} sigma_{v} .) Adding this entropy changes the total energy of the system by an amount

$$frac{3}{2}sigma_{v}^{2} Nleft( {2^{frac{2}{3N}} – 1} right) equiv k_{B}^{*} T,$$

which we thus define as the temperature (k_{B}^{*} T). Even though this temperature is nominally a function of the swarm size N, it correctly yields an intensive temperature as expected in the limit of large N, as the explicit N-dependence vanishes in that limit since (lim_{n to infty } k_{B}^{*} T = sigma_{v}^{2} ln 2). In practice, this limit is achieved very rapidly: we find that this temperature is nearly independent of N for N larger than about 20, consistent with our earlier results on the effective “thermodynamic limit” for swarms25. The effective Boltzmann constant (k_{B}^{*}) is included here to convert between temperature and energy, though we note that we cannot set its value, as there is no intrinsically preferred temperature scale.
Equipartition
With these definitions in hand, we can evaluate the suitability of these quantities for describing the macroscopic state of midge swarms. First, we note that proper state variables ought to be independent of the swarm history; that is, they ought to describe only the current state of the system rather than the protocol by which that state was prepared. Although this property is difficult to prove incontrovertibly, none of the definitions of our state variables have history dependence. We further find that when these state variables are modulated (see below), their correlation times are very short, lending support to their interpretation as true state variables. We can also compare the relationships between these state variables and the swarm behaviour to what would be expected classically. In equilibrium thermodynamics, for example, temperature is connected to the number of degrees of freedom (d.o.f.) in a system via equipartition, such that each d.o.f. contributes an energy of (frac{1}{2}k_{B}^{*} T). We can write the total energy E of a swarm as the sum of the kinetic energy (E_{k} (t) = frac{1}{2}v^{2}) and potential energy (E_{p} (t) = frac{1}{2}k(N)r(t)^{2}) for all the individuals, where r is the distance of a midge to the swarm centre of mass, v is the velocity of a midge, and k(N) is the effective spring constant. Surprisingly, even though individual midges are certainly not in equilibrium due to their active nature, we find that the total energy is linear in both T and N (Fig. 2a), and that there is no apparent anisotropy, suggesting that equipartition holds for our swarms. This result is highly nontrivial, especially given that our definition of T does not contain the spring constant k(N), which is only determined empirically from our data. Moreover, the slope of the (E/k_{B}^{*} T) curve is well approximated as (9/2)N, implying that each midge has 9 effective d.o.f. (or 6 after discounting the factor of ({text{ln}}2) in our definition of (k_{B}^{*} T)) These d.o.f. can be identified as 3 translational and 3 potential modes, given that the potential well in which the midges reside is three-dimensional. These results demonstrate the surprising applicability of equilibrium thermodynamics for describing the macroscopic state of swarms29.
Figure 2

Equipartition and the equation of state. (a) The total energy of the system (E) normalized by (k_{B}^{*} T) as a function of swarm size (blue) along with the kinetic energy (E_{k}) (yellow) and potential energy (E_{p}) (blue). The total normalized energy of the system is well approximated by (9/2)N (black dashed line), indicating that each individual midge contributes ((9/2)k_{B}^{*} T) to E and thus has 9 degrees of freedom (6 after discounting the factor of ({text{ln}}2) in our definition of (k_{B}^{*} T)). The deviations from that behaviour for the largest swarms can be attributed to a growing uncertainty in the energy due to the smaller number of experiments with such large swarms. (b) A portion of our ensemble of data of the measured pressure (blue). The yellow line is the reconstruction of the pressure from our equation of state. The inset shows a zoomed-in portion of the data to highlight the quality of the reconstruction. (c) PDF of the pressure for our entire data ensemble23. The statistics of the directly measured pressure (blue) and reconstructed pressure from the equation of state have nearly identical statistics for the full dynamic range of the signal.

Full size image

Equation of state
The fundamental relation in any thermodynamic system is the equation of state that expresses how the state variables co-vary. Equations of state are thus the foundation for the design and control of thermodynamic systems, because they describe how the system will respond when a subset of the state variables are modulated. Any equation of state can be written in the form (P = f(V, ;T,;N)) for some function f. Although the form of f is a priori unknown, it can typically be written as a power series in V, T, and N, in the spirit of a virial expansion. We fit the equation of state to our data assuming the functional form

$$P = f(V, ;k_{B}^{*} T,;N) = c_{4} V^{{c_{1} }} (k_{B}^{*} T)^{{c_{2} }} N^{{c_{3} }},$$

and using nonlinear least-squares regression. We chose to fit to the pressure for convenient analogy with a thermodynamic framework, but any other variable would have been an equivalent possibility. We note that when fitting, we normalized all the state variables by their root-mean-square values so that they were all of the same order of magnitude. These normalization pre-factors do not change the exponents, but are instead simply absorbed into (c_{4}). Thus, to leading order, we assume (P = f(V,;k_{B}^{*} T,;N) propto V^{{c_{1} }} (k_{B}^{*} T)^{{c_{2} }} N^{{c_{3} }}) and fit this relation to the swarm pressure (Fig. 2b,c), obtaining c1 = − 1.7, c2 = 2, and c3 = 1, with uncertainties on the order of 1%. Although the expression for the pressure does depend on three parameters in a nonlinear fashion, the resulting estimates for these parameters are remarkably stable and consistent across all measurements. Hence, we arrive at the equation of state (PV^{1.7} propto N(k_{B}^{*} T)^{2} .)
This equation of state reveals aspects of the nature of swarms, particularly when compared with the linear equation of state for an ideal gas (where (PV = Nk_{B} T)). In both cases, for example, to maintain a fixed pressure and volume, smaller systems need to be hotter; but this requirement is less severe for swarms since the temperature is squared, meaning that midges have to speed up less than ideal gas molecules do. Likewise, to maintain a fixed temperature, volume expansion must be counteracted by a reduction in pressure; but midges must lower the pressure more than a corresponding ideal gas, which is reflective of the decrease of the swarm spring constant with size.
Thermodynamic cycling
Beyond such reasoning, however, the true power of an equation of state in thermodynamics lies in specifying how the state variables will change when some are varied but the system remains in the same state, such as in an engine. To demonstrate that our equation of state similarly describes swarms, it is thus necessary to drive them away from their natural state. Although it is impossible to manipulate the state variables directly in this system of living organisms as one would do with a mechanical system, we have shown previously that time-varying acoustic30 and illumination27 stimulation lead to macroscopic changes in swarm behaviour. Here we therefore build on these findings and use interlaced illumination changes and acoustic signals to drive swarms along four distinct paths in pressure–volume space, analogous to a thermodynamic engine cycle. The stimulation protocol is sketched in Fig. 3a. The “on” state of the acoustic signal is telegraph noise (see Experimental details), while the “off” state is completely quiet. The illumination signal simply switches between two different steady light levels. Switching between the four states of “light-high and sound-on,” “light-high and sound-off,” “light-low and sound-off,” and “light-low and sound-on” with a 40-s period (Fig. 3a) produces the pressure–volume cycle shown in Fig. 3b. We suspect that the loops in the cycle stem from the swarm’s typical “startle” response after abrupt changes in environmental conditions, followed by a rapid relaxation to a steady state27,30.
Figure 3

Thermodynamic cycling of a midge swarm with (langle Nrangle = 27). Schematic of the perturbation cycle showing the illumination (solid) and sound (dashed) signal timings. The symbols indicate the switching points identified in (b). (b,c) Phase-averaged swarm behaviour during the perturbation cycle plotted in the pressure–volume phase plane for (b) the pressure signal as measured and (c) as reconstructed using our equation of state. (leftlangle {,} rightrangle_{phi }) denotes a phase average of a quantity over a full cycle. The four different states of the perturbation signal are indicated. The data has been averaged using a moving 3.5-s window for clarity. The swarm behaviour moves in a closed loop in this phase plane during this cycling, as would be expected for an engine in equilibrium thermodynamics, and the equation of state holds throughout even though it was developed only for unperturbed swarms. (d) Phase-averaged pressure (langle Prangle_{phi }) of the swarm during a continuous cycle through the four light and sound states. The blue line shows the directly measured pressure and the yellow line shows the reconstruction using the equation of state.

Full size image

In addition to the pressure and volume, we can also measure the other state variables as we perturb the swarms. Given that we do not observe any evidence of a phase transition, we would expect that our equation of state, if valid, should hold throughout this cycle. To check this hypothesis, we used the measured V, T, and N values during unperturbed experiments along with the equation of state to predict the scaling exponents, and in turn the pressure P. We then use these baseline, unperturbed exponents and V, T, and N during the interlaced perturbations to predict a pressure P. This pressure prediction matches the measured signal exceptionally well (Fig. 3c,d) even though the equation of state was formulated only using data from unperturbed swarms, highlighting the quality of this thermodynamic analogy. Although we might expect that a strong enough perturbation might lead to qualitatively different behaviour (if the swarm went through the analog of a phase transition31), our results give strong support to the hypothesis that our equation of state should hold for any perturbation that does not drive such a transition. More