Philippa Kaur

Harvey, P. H. & Pagel, M. D. The comparative method in evolutionary biology. Vol. 239 (Oxford University Press, 1991).Oyston, J. W., Hughes, M., Wagner, P. J., Gerber, S. & Wills, M. A. What limits the morphological disparity of clades? Interface Focus 5, 0042 (2015).Article 

Google Scholar 
Jetz, W., Thomas, G. H., Joy, J. B., Hartmann, K. & Mooers, A. O. The global diversity of birds in space and time. Nature 491, 444–448 (2012).CAS 
PubMed 
Article 

Google Scholar 
Webb, C. O. Exploring the phylogenetic structure of ecological communities: an example for rain forest trees. Am. Naturalist 156, 145–155 (2000).Article 

Google Scholar 
Purvis, A., Gittleman, J. L. & Brooks, T. Phylogeny and conservation. (Cambridge University Press, 2005).Page, R. D. M. Parallel phylogenies: reconstructing the history of host-parasite assemblages. Cladistics 10, 155–173 (1994).Article 

Google Scholar 
Weaver, S. C. & Vasilakis, N. Molecular evolution of dengue viruses: contributions of phylogenetics to understanding the history and epidemiology of the preeminent arboviral disease. Infect., Genet. Evolution 9, 523–540 (2009).CAS 
Article 

Google Scholar 
Tassy, P. Trees before and after Darwin. J. Zool. Syst. Evolut. Res. 49, 89–101 (2011).Article 

Google Scholar 
Heather, J. M. & Chain, B. The sequence of sequencers: The history of sequencing DNA. Genomics 107, 1–8 (2016).CAS 
PubMed 
Article 

Google Scholar 
Pyron, R. A. Post-molecular systematics and the future of phylogenetics. Trends Ecol. Evolution 30, 384–389 (2015).Article 

Google Scholar 
Sansom, R. S. & Wills, M. A. Differences between hard and soft phylogenetic data. Proc. R. Soc. B: Biol. Sci. 284, 20172150 (2017).Article 

Google Scholar 
Scotland, R. W., Olmstead, R. G. & Bennett, J. R. Phylogeny reconstruction: the role of morphology. Syst. Biol. 52, 539–548 (2003).PubMed 
Article 

Google Scholar 
Regier, J. C. et al. Arthropod relationships revealed by phylogenomic analysis of nuclear protein-coding sequences. Nature 463, 1079–1083 (2010).CAS 
PubMed 
Article 

Google Scholar 
Callender-Crowe, L. M. & Sansom, R. S. Osteological characters of birds and reptiles are more congruent with molecular phylogenies than soft characters are. Zool. J. Linn. Soc. 194, 1–13 (2022).Article 

Google Scholar 
Wahlberg, N. et al. Synergistic effects of combining morphological and molecular data in resolving the phylogeny of butterflies and skippers. Proc. R. Soc. B: Biol. Sci. 272, 1577–1586 (2005).CAS 
Article 

Google Scholar 
He, L. et al. A molecular phylogeny of selligueoid ferns (Polypodiaceae): Implications for a natural delimitation despite homoplasy and rapid radiation. Taxon 67, 237–249 (2018).Article 

Google Scholar 
Fernández, R., Edgecombe, G. D. & Giribet, G. Phylogenomics illuminates the backbone of the Myriapoda Tree of Life and reconciles morphological and molecular phylogenies. Sci. Rep. 8, 1–7 (2018).
Google Scholar 
Eme, L., Spang, A., Lombard, J., Stairs, C. W. & Ettema, T. J. G. Archaea and the origin of eukaryotes. Nat. Rev. Microbiol. 15, 711–723 (2017).CAS 
PubMed 
Article 

Google Scholar 
Asher, R. J., Bennett, N. & Lehmann, T. The new framework for understanding placental mammal evolution. BioEssays 31, 853–864 (2009).CAS 
PubMed 
Article 

Google Scholar 
Shoshani, J. & McKenna, M. C. Higher taxonomic relationships among extant mammals based on morphology, with selected comparisons of results from molecular data. Mol. Phylogenetics Evolution 9, 572–584 (1998).CAS 
Article 

Google Scholar 
Beck, R. M. D. & Baillie, C. Improvements in the fossil record may largely resolve current conflicts between morphological and molecular estimates of mammal phylogeny. Proc. R. Soc. B: Biol. Sci. 285, 20181632 (2018).Article 

Google Scholar 
Zou, Z. T. & Zhang, J. Z. Morphological and molecular convergences in mammalian phylogenetics. Nat. Commun. 7, 1–9 (2016).
Google Scholar 
Hillis, D. M. Molecular versus morphological approaches to systematics. Annu. Rev. Ecol. Syst. 18, 23–42 (1987).Article 

Google Scholar 
Thompson, N. Alfred Russell Wallace Contributions to the theory of Natural Selection, 1870, and Charles Darwin and Alfred Wallace, ‘On the Tendency of Species to form Varieties’ (Papers presented to the Linnean Society 30th June 1858). (Routledge, 2004).Croizat, L. Panbiogeography; or an introductory synthesis of zoogeography, phytogeography, and geology, with notes on evolution, systematics, ecology, anthropology, etc., Vol. 1, 2a & 2b (Published by the author, Caracas., 1958).Means, J. C. & Marek, P. E. Is geography an accurate predictor of evolutionary history in the millipede family Xystodesmidae? PeerJ 5, e3854 (2017).PubMed 
PubMed Central 
Article 

Google Scholar 
Wills, M. A., Barrett, P. M. & Heathcote, J. F. The modified gap excess ratio (GER*) and the stratigraphic congruence of dinosaur phylogenies. Syst. Biol. 57, 891–904 (2008).PubMed 
Article 

Google Scholar 
Fisher, D. C. Stratocladistics: integrating temporal data and character data in phylogenetic inference. Annu. Rev. Ecol., Evolution Syst. 39, 365–385 (2008).Article 

Google Scholar 
Lazarus, D. B. & Prothero, D. R. The role of stratigraphic and morphologic data in phylogeny. J. Paleontol. 58, 163–172 (1984).
Google Scholar 
Camerini, J. R. Evolution, biogeography, and maps: an early history of Wallace’s Line. Isis 84, 700–727 (1993).CAS 
PubMed 
Article 

Google Scholar 
Upchurch, P., Hunn, C. A. & Norman, D. B. An analysis of dinosaurian biogeography: evidence for the existence of vicariance and dispersal patterns caused by geological events. Proc. R. Soc. B: Biol. Sci. 269, 613–621 (2002).Article 

Google Scholar 
Ferreira, G. S., Bronzati, M., Langer, M. C. & Sterli, J. Phylogeny, biogeography and diversification patterns of side-necked turtles (Testudines: Pleurodira). R. Soc. Open Sci. 5, 171773 (2018).PubMed 
PubMed Central 
Article 

Google Scholar 
Ronquist, F. & Sanmartín, I. Phylogenetic methods in biogeography. Annu. Rev. Ecol., Evolution, Syst. 42, 441–464 (2011).Article 

Google Scholar 
IUCN. The IUCN Red List of Threatened Species. Version 2019-2., https://www.iucnredlist.org (2019).GBIF.org. GBIF Home Page, https://www.gbif.org/ (2019).Uetz, P., Freed, P., Aguilar, R. & Hošek, J. The reptile database., http://www.reptiledatabase.org (2019).Archie, J. W. Homoplasy excess ratios: new indices for measuring levels of homoplasy in phylogenetic systematics and a critique of the consistency index. Syst. Zool. 38, 253–269 (1989).Article 

Google Scholar 
Wilkinson, M. On phylogenetic relationships within Dendrotriton (Amphibia: Caudata: Plethodontidae) is there sufficient evidence? Herpetological J. 7, 55–65 (1997).
Google Scholar 
O’Connor, A. & Wills, M. A. Measuring stratigraphic congruence across trees, higher taxa, and time. Syst. Biol. 65, 792–811 (2016).PubMed 
PubMed Central 
Article 

Google Scholar 
Colless, D. H. Review of phylogenetics: the theory and practice of phylogenetic systematics. Syst. Zool. 31, 100–104 (1982).Article 

Google Scholar 
Lartillot, N. & Philippe, H. Improvement of molecular phylogenetic inference and the phylogeny of Bilateria. Philos. Trans. R. Soc. B: Biol. Sci. 363, 1463–1472 (2008).Article 

Google Scholar 
Sansom, R. S., Choate, P. G., Keating, J. N. & Randle, E. Parsimony, not Bayesian analysis, recovers more stratigraphically congruent phylogenetic trees. Biol. Lett. 14, 20180263 (2018).PubMed 
PubMed Central 
Article 

Google Scholar 
Rosa, B. B., Melo, G. A. & Barbeitos, M. S. Homoplasy-based partitioning outperforms alternatives in Bayesian analysis of discrete morphological data. Syst. Biol. 68, 657–671 (2019).PubMed 
Article 

Google Scholar 
Lucena, D. A. & Almeida, E. A. Morphology and Bayesian tip-dating recover deep Cretaceous-age divergences among major chrysidid lineages (Hymenoptera: Chrysididae). Zool. J. Linn. Soc. 194, 36–79 (2022).Article 

Google Scholar 
O’Reilly, J. E. et al. Bayesian methods outperform parsimony but at the expense of precision in the estimation of phylogeny from discrete morphological data. Biol. Lett. 12, 20160081 (2016).PubMed 
PubMed Central 
Article 

Google Scholar 
Smith, M. R. Bayesian and parsimony approaches reconstruct informative trees from simulated morphological datasets. Biol. Lett. 15, 20180632 (2019).PubMed 
PubMed Central 
Article 

Google Scholar 
Wiens, J. The role of morphological data in phylogeny reconstruction. Syst. Biol. 53, 653–661 (2004).PubMed 
Article 

Google Scholar 
O’Leary, M. A. & Kaufman, S. G. MorphoBank 3.0: Web application for morphological phylogenetics and taxonomy., http://www.morphobank.org (2012).de Queiroz, A. & Gatesy, J. The supermatrix approach to systematics. Trends Ecol. Evolution 22, 34–41 (2007).Article 

Google Scholar 
Wilkinson, M. A comparison of two methods of character construction. Cladistics 11, 297–308 (1995).Article 

Google Scholar 
Brazeau, M. D. Problematic character coding methods in morphology and their effects. Biol. J. Linn. Soc. 104, 489–498 (2011).Article 

Google Scholar 
Drummond, A. J., Ho, S. Y. W., Phillips, M. J. & Rambaut, A. Relaxed phylogenetics and dating with confidence. PLoS Biol. 4, e88 (2006).PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
O’Reilly, J. E., Puttick, M. N., Pisani, D. & Donoghue, P. C. Probabilistic methods surpass parsimony when assessing clade support in phylogenetic analyses of discrete morphological data. Palaeontology 61, 105–118 (2018).PubMed 
Article 

Google Scholar 
Keating, J. N., Sansom, R. S., Sutton, M. D., Knight, C. G. & Garwood, R. J. Morphological phylogenetics evaluated using novel evolutionary simulations. Syst. Biol. 69, 897–912 (2020).PubMed 
PubMed Central 
Article 

Google Scholar 
Makarenkov, V. et al. Weighted bootstrapping: a correction method for assessing the robustness of phylogenetic trees. BMC Evolut. Biol. 10, 1–16 (2010).Article 
CAS 

Google Scholar 
Stayton, C. T. The definition, recognition, and interpretation of convergent evolution, and two new measures for quantifying and assessing the significance of convergence. Evolution 69, 2140–2153 (2015).PubMed 
Article 

Google Scholar 
Sattler, R. Homology – a continuing challenge. Syst. Bot. 9, 382–394 (1984).Article 

Google Scholar 
Jenner, R. A. & Schram, F. R. The grand game of metazoan phylogeny: rules and strategies. Biol. Rev. 74, 121–142 (1999).Article 

Google Scholar 
Pisani, D. & Wilkinson, M. Matrix representation with parsimony, taxonomic congruence, and total evidence. Syst. Biol. 51, 151–155 (2002).PubMed 
Article 

Google Scholar 
Arcila, D. et al. Testing the utility of alternative metrics of branch support to address the ancient evolutionary radiation of tunas, stromateoids, and allies (Teleostei: Pelagiaria). Syst. Biol. 70, 1123–1144 (2021).PubMed 
Article 

Google Scholar 
Felsenstein, J. Phylogenies and the comparative method. Am. Naturalist 125, 1–15 (1985).Article 

Google Scholar 
Bremer, K. Branch support and tree stability. Cladistics 10, 295–304 (1994).Article 

Google Scholar 
Johnson, W. E. et al. The late Miocene radiation of modern Felidae: a genetic assessment. Science 311, 73–77 (2006).CAS 
PubMed 
Article 

Google Scholar 
Van der Made, J. Biogeography and climatic change as a context to human dispersal out of Africa and within Eurasia. Quat. Sci. Rev. 30, 1353–1367 (2011).Article 

Google Scholar 
May, F., Rosenbaum, B., Schurr, F. M. & Chase, J. M. The geometry of habitat fragmentation: Effects of species distribution patterns on extinction risk due to habitat conversion. Ecol. Evolution 9, 2775–2790 (2019).Article 

Google Scholar 
Swofford, D. L. et al. Bias in phylogenetic estimation and its relevance to the choice between parsimony and likelihood methods. Syst. Biol. 50, 525–539 (2001).CAS 
PubMed 
Article 

Google Scholar 
Jaeger, J. J. & Martin, M. African marsupials – vicariance or dispersion? Nature 312, 379–379 (1984).Article 

Google Scholar 
Smith, B. T. et al. The drivers of tropical speciation. Nature 515, 406–409 (2014).CAS 
PubMed 
Article 

Google Scholar 
Simkanin, C. et al. Exploring potential establishment of marine rafting species after transoceanic long-distance dispersal. Glob. Ecol. Biogeogr. 28, 588–600 (2019).Article 

Google Scholar 
Raxworthy, C. J., Forstner, M. R. J. & Nussbaum, R. A. Chameleon radiation by oceanic dispersal. Nature 415, 784–787 (2002).CAS 
PubMed 
Article 

Google Scholar 
Stehli, F. G. & Webb, S. D. The great American biotic interchange., Vol. 4 (Springer Science & Business Media, 2013).Ronquist, F. Dispersal-vicariance analysis: A new approach to the quantification of historical biogeography. Syst. Biol. 46, 195–203 (1997).Article 

Google Scholar 
Ricklefs, R. E. & Bermingham, E. The concept of the taxon cycle in biogeography. Glob. Ecol. Biogeogr. 11, 353–361 (2002).Article 

Google Scholar 
Ma, H. An analysis of the equilibrium of migration models for biogeography-based optimization. Inf. Sci. 180, 3444–3464 (2010).Article 

Google Scholar 
Yiming, L., Niemelä, J. & Dianmo, L. Nested distribution of amphibians in the Zhoushan archipelago, China: can selective extinction cause nested subsets of species? Oecologia 113, 557–564 (1998).CAS 
PubMed 
Article 

Google Scholar 
Crisci, J. V., Katinas, L. & Posadas, P. Historical Biogeography: An Introduction. (Harvard University Press, 2003).Chen, R. et al. Adaptive innovation of green plants by horizontal gene transfer. Biotechnol. Adv. 46, 107671 (2021).CAS 
PubMed 
Article 

Google Scholar 
Schönknecht, G., Weber, A. P. & Lercher, M. J. Horizontal gene acquisitions by eukaryotes as drivers of adaptive evolution. BioEssays 36, 9–20 (2014).PubMed 
Article 
CAS 

Google Scholar 
Smith, A. B. Echinoderm phylogeny: morphology and molecules approach accord. Trends Ecol. Evolution 7, 224–229 (1992).CAS 
Article 

Google Scholar 
Bateman, R. M., Hilton, J. & Rudall, P. J. Morphological and molecular phylogenetic context of the angiosperms: contrasting the ‘top-down’ and ‘bottom-up’ approaches used to infer the likely characteristics of the first flowers. J. Exp. Bot. 57, 3471–3503 (2006).CAS 
PubMed 
Article 

Google Scholar 
Morris, J. L. et al. The timescale of early land plant evolution. Proc. Natl Acad. Sci. 115, E2274–E2283 (2018).CAS 
PubMed 
PubMed Central 

Google Scholar 
Richter, S. The Tetraconata concept: hexapod-crustacean relationships and the phylogeny of Crustacea. Org. Diversity Evolution 2, 217–237 (2002).Article 

Google Scholar 
Dunn, C. W. et al. Broad phylogenomic sampling improves resolution of the animal tree of life. Nature 452, 745–749 (2008).CAS 
PubMed 
Article 

Google Scholar 
Caravas, J. & Friedrich, M. Of mites and millipedes: recent progress in resolving the base of the arthropod tree. BioEssays 32, 488–495 (2010).CAS 
PubMed 
Article 

Google Scholar 
Howard, R. J. et al. The Ediacaran origin of Ecdysozoa: integrating fossil and phylogenomic data. J. Geol. Soc. https://doi.org/10.1144/jgs2021-107 (2022).Newman, M. E. J. A model of mass extinction. J. Theor. Biol. 189, 235–252 (1997).CAS 
PubMed 
Article 

Google Scholar 
Cobbett, A., Wilkinson, M. & Wills, M. A. Fossils impact as hard as living taxa in parsimony analyses of morphology. Syst. Biol. 56, 753–766 (2007).PubMed 
Article 

Google Scholar 
Ruta, M., Krieger, J., Angielczyk, K. & Wills, M. A. The evolution of the tetrapod humerus: morphometrics, disparity, and evolutionary rates. Earth Environ. Sci. Trans. R. Soc. Edinb. 109, 351–369 (2018).
Google Scholar 
Puttick, M. N., Thomas, G. H. & Benton, M. J. High rates of evolution preceded the origins of birds. Evolution 68, 1497–1510 (2014).PubMed 
PubMed Central 
Article 

Google Scholar 
Sansom, R. S. & Wills, M. A. Fossilization causes organisms to appear erroneously primitive by distorting evolutionary trees. Sci. Rep. 3, 1–5 (2013).Article 

Google Scholar 
Brinkworth, A., Sansom, R. & Wills, M. A. Phylogenetic incongruence and homoplasy in the appendages and bodies of arthropods: why broad character sampling is best. Zool. J. Linn. Soc. 187, 100–116 (2019).Article 

Google Scholar 
Brown, J. W. & Smith, S. A. The past sure is tense: on interpreting phylogenetic divergence time estimates. Syst. Biol. 67, 340–353 (2018).PubMed 
Article 

Google Scholar 
Barba-Montoya, J., Dos Reis, M. & Yang, Z. H. Comparison of different strategies for using fossil calibrations to generate the time prior in Bayesian molecular clock dating. Mol. Phylogenetics Evolution 114, 386–400 (2017).CAS 
Article 

Google Scholar 
Sanderson, M. J. & Donoghue, M. J. Patterns of variation in levels of homoplasy. Evolution 43, 1781–1795 (1989).PubMed 
Article 

Google Scholar 
Alroy, J. Fossilworks: Gateway to the Paleobiology Database, http://fossilworks.org (2019).Benton, M. J. The Fossil Record 2. (Chapman & Hall, 1993).Cohen, K. M., Harper, D. A. T. & Gibbard, P. L. ICS International Chronostratigraphic Chart 2021/02, http://www.stratigraphy.org/ (2021).Gradstein, F. & Ogg, J. Geologic time scale 2004–why, how, and where next! Lethaia 37, 175–181 (2004).Article 

Google Scholar 
Rohde, R. A. The GeoWhen Database, (2005).O’Leary, M. A. et al. The placental mammal ancestor and the post–K-Pg radiation of placentals. Science 339, 662–667 (2013).PubMed 
Article 
CAS 

Google Scholar 
Kluge, A. G. A concern for evidence and a phylogenetic hypothesis of relationships among Epicrates (Boidae, Serpentes). Syst. Biol. 38, 7–25 (1989).Article 

Google Scholar 
Tolson, P. J. Phylogenetics of the boid snake genus Epicrates and Caribbean vicariance theory. Occasional Pap. Mus. Zool., Univ. Mich. 715, 1–68 (1987).
Google Scholar 
Clopper, C. J. & Pearson, E. S. The use of confidence or fiducial limits illustrated in the case of the binomial. Biometrika 26, 404–413 (1934).Article 

Google Scholar  More

Drainage area and branching ratio: a matter of scaleGeomorphological and ecological viewpoints on river networks generally differ owing to discordant definitions of the fundamental unit (the node) used to analyze them. From a geomorphological perspective, the determination of a river network entails the definition of an observational scale. Real river networks can be extracted from digital elevation models (DEMs) via algorithms for flow direction determination such as D8 (i.e., each pixel drains towards the lowest of its 8 nearest neighbors53). After the outlet location has been specified (and hence the upstream area A spanned by the river network), the first observational scale required is thus the pixel length l of the DEM, which defines the extent of a network node. A second scale is then needed to distinguish the portion of the drainage network effectively belonging to the channel network. The simplest but still widely used method53 defines channels as those pixels whose drainage area exceeds a threshold value AT. Hydrologically based criteria to determine the appropriate value for AT exist54; however, for the sake of simplicity, we here consider AT as a free parameter.BBTs and RBNs are random constructs, and as such they do not satisfy the optimality criterion of minimizing total energy expenditure, which is the fundamental physical process shaping fluvial landscapes. Furthermore, neither of these networks is a spanning tree, which is a key attribute of real fluvial landforms10: in fact, in both BBTs and RBNs, the extent of the drained domain is not defined. As a result, the drainage area at an arbitrary network node cannot in principle be attributed, unless by using the number of upstream nodes as a proxy. This has practical implications from an ecological viewpoint because drainage area is the master variable controlling several attributes of a river, such as width, depth, discharge, or slope3,55, which in turn impact habitat characteristics and the ecology of organisms therein56.In BBTs and RBNs, branching probability p has been defined35,38,45,46,47 as the probability that a network node is branching, i.e. connected to two upstream nodes. As such, the branching probability of a realized river network (be it a real river or a synthetic construct) could be evaluated as the ratio between the number of links NL constituting a network and the total number of network nodes N; if a unit distance between two adjacent nodes is assumed, the denominator equals the total network length. We note that the former definition of branching probability only holds in the context of the generation of a synthetic random network; it is in fact improper to refer to a “probability” when analyzing the properties of a realized river network. We clarify this aspect by introducing the concept of branching ratio pr for the latter definition (pr = NL/N). Moreover, in the case of BBTs, p and pr do not coincide (see Methods). Importantly, p and pr have no parallel in the literature on fluvial forms, nor do they refer to any of the well-studied measures of rivers’ fractal character.The choice of different observational scales for the same drainage network results in different values of NL and N, and hence of pr. Remarkably, the very same drainage network can result in river networks that virtually assume any value of pr (ranging from 0 to 1) and N (up to the upper bound A) depending on the choice of AT and A (the latter corresponding to a given l value when measured in the number of pixels; Fig. 1d–i); networks with low AT/A ratios result in high N (Fig. 2a), while networks with low AT result in high pr (Fig. 2b). Furthermore, pr does not identify the inherent (i.e., scale-independent) branching character of a given river network in relation to other river networks. In fact, by extracting different river networks at various scales (i.e., various AT values) and assessing the rivers’ rank in terms of pr, one observes that rivers that look more “branching” (i.e., have higher pr) than others for a given AT value can become less “branching” for a different AT value (Fig. 3). We therefore conclude that branching probability is a non-descriptive property of a river network, which by no means describes its inherent branching character, and depends on the observational scale.Fig. 2: Variation of N and pr as a function of observational scales for OCNs and real river networks.a Expected value of number of network nodes N as a function of threshold area AT and total drained area A (from Eq. (1)); the white dots indicate the values of AT and A used to generate the OCNs used in this analysis. b Expected value of branching ratio pr as a function of AT and A (from Eq. (1)); symbols as in a.Full size imageFig. 3: Values of branching ratio as a function of AT for the 50 real river networks analyzed in this study.a Natural values of pr in logarithmic scale. b z-normalized branching ratios (i.e., for each AT value, values of pr are normalized so that they have null mean and unit standard deviation), which better shows how rivers rank differently in terms of pr for different observation scales (i.e., AT). Lines connect dots relative to the same river. For visual purposes, rivers that rank first, second, second-to-last or last in at least one of the AT groups are displayed in colors; the other rivers are displayed in grey.Full size imageScaling is also crucial when looking at river networks from an ecological perspective. In this case, the relevant scale determining the dimension l of a node is the extent of habitat within which individuals (due to e.g. physical constrains) can be assigned to a single population57,58; the riverine connectivity and ensuing dispersal among these populations give rise to a metapopulation at the river network level. The specific spatial scale largely depends on the targeted species (e.g. being larger for fish than for aquatic insects), and it is conceivably much larger than (or, at least, it has no reason to be equal to) the pixel size of the DEM on which the river network is extracted. Since the evaluation of pr depends on the number of nodes N, which, in turn, is defined based on the scale length l, the resulting pr of a river network under this perspective would depend on the characteristics of the target taxa, which is inconsistent with the alleged role of pr as a scale-invariant property of river networks.Note also that using the ecological definition of l (i.e., spatial range of a local population) to discretize a real river network into N nodes, and from there calculate the branching ratio pr = NL/N, is problematic. Indeed, this would imply an elongation of all links shorter than l (which constitute a non-negligible fraction of the total links, under the assumption of exponential distribution of link lengths51), hence preventing a correct estimation of the connectivity patterns (i.e., distances between nodes) and the resulting ecological metrics of the river network (see section Ecological implications).From an ecological perspective, it could be reasonable to consider AT as a parameter expressing how a particular taxon perceives the suitable landscape, rather than a value to be determined from geomorphological arguments: for instance, large fishes inhabit wide and deep river reaches, and do not access small headwaters56. In this case, imposing a large AT would result in a coarser, less branching network constituted by few main channels (Fig. 1f, i), which could mimic the potentially available habitat for such species. Conversely, aquatic insects inhabit also small headwaters17,59, therefore their perceived landscape would resemble the finely resolved networks of Fig. 1d, g, characterized by low AT and higher (apparent) pr.Topology and scaling of river networks and random analoguesTo verify the topological (i.e., Horton’s laws on bifurcation and length ratios) and scaling (i.e., probability distribution of drainage areas) relationships of the different network types, we extracted from DEMs 50 real river networks encompassing a wide range of drainage areas (Fig. 4), and we generated 50 OCNs, 50 RBNs and 50 BBTs of comparable size (see Methods).Fig. 4: Location of real river basins used in the analysis.River basins are shown in dark grey; countries in light grey. Rivers’ numbering is sorted in ascending order according to drainage area values.Full size imageTypical values3,7,60 for the bifurcation ratio RB lie between 3 and 5, while length ratios (RL) range between 1.5 and 3.5. As expected, we observed that the real rivers and OCNs used in our analysis have RB and RL values within the aforementioned ranges (Fig. 5a, b). The same is true for RBNs, while the RB and RL values found for BBTs are lower than the typical ranges. This finding holds regardless of the scale (subsumed by AT) at which real river networks and OCNs are extracted (Supplementary Figs. 1 and 2). Remarkably, BBTs fail to satisfy Horton’s laws despite the statistical inevitability of such laws for any network argued by ref. 61. To this regard, we note that the networks analyzed by ref. 61 did not include constructs where all paths from the source nodes to the outlet have the same length, which is the defining feature of BBTs (Fig. 1a).Fig. 5: Comparison of topological and scaling properties of the different networks.a Scaling of number of network links Nω as a function of stream order ω for the various network types (rivers and OCNs obtained with AT = 20 pixels; RBNs and BBTs derived accordingly – see Methods). b Mean link length Lω (in units of l) as a function of ω. Networks used are as in panel a. c Scaling of drainage areas: probability P[A ≥ a] to randomly sample a node with drainage area A ≥ a as a function of a. The displayed trend lines are fitted on the ensemble values for the 50 network replicates, by excluding nodes with drainage area larger than 2000 pixels (cutoff value marked with a black solid line). The scaling coefficients β reported correspond to the slopes of the fitted trend lines. Extended details on all panels are provided in the Supplementary Methods.Full size imageWhile the power-law scaling of areas in OCNs (Fig. 5c) has an exponent β ≈ 0.45 that closely resembles the one found for the real rivers (β ≈ 0.46) and within the typically observed range8,10 β = 0.43 ± 0.02, drainage areas of RBNs scale as a power law with an exponent β ≈ 0.51, which departs from the observed range. Conversely, BBTs do not show any power-law scaling of areas. Scaling exponents of drainage areas fitted separately for each real river network yielded values in the range 0.36÷0.57 (Supplementary Table 1). In particular, we observed that these values tend to the expected range β = 0.43 ± 0.02 for increasing values of A, expressed in number of pixels (Supplementary Fig. 3), hence implying that highly resolved catchments are required in order to properly estimate β. Interestingly, the observed values of Horton ratios and scaling exponent β for RBNs are compatible with the values RB = 4, RL = 2, β = 0.5 predicted for Shreve’s random topology model3,60,62, which is actually equivalent to a RBN with infinite links.Ecological implicationsWe compared the different network types via two metrics that express the ecological value of a landscape for a metapopulation: the coefficient of variation of a metapopulation CVM and the metapopulation capacity λM. The coefficient of variation of a metapopulation63 is a measure of metapopulation stability (a metapopulation being more stable the lower CVM is), while the metapopulation capacity42,64 expresses the potential for a metapopulation to persist in the long run (persistence being more likely the higher λM is). Both measures are among the most universal metrics describing dynamics of spatially fragmented populations24,40. In order to assess the impact of the two landscape features mostly affecting metapopulation dynamics, i.e. spatial connectivity and spatial distribution of habitat patches, we calculated these metrics for the four network types under two different scenarios: uniform (CVM,U, λM,U) and non-uniform (CVM,H, λM,H) spatial distribution of habitat patch sizes. In the first scenario, CVM,U and λM,U assess stability and persistence (respectively) of a metapopulation solely based on pairwise distances between network nodes; in the second scenario, CVM,H and λM,H depend on the interplay between pairwise distances and spatially heterogeneous habitat availability (namely, downstream nodes being larger than upstream ones).We found that the values of CVM (be it derived with uniform (CVM,U) or nonuniform (CVM,H) distributions of patch sizes) obtained for OCNs match strikingly well those of real rivers (Fig. 6). These CVM values are consistently lower than those found for RBNs, while values of CVM for BBTs are even higher. Notably, this result holds for different values of AT (and hence different pr values) at which real rivers and OCNs are extracted (Fig. 6a–c; g–i), and for values of mean dispersal distance α (see Methods) spanning multiple orders of magnitude (Supplementary Figs. 4–7).Fig. 6: Comparison of values of metapopulation metrics across river network types and observational scales (AT).a–c CVM,U. d–f λM,U. g–i CVM,H. j–l λM,H. Boxplot elements are as follows: center line, median; notches, (pm 1.58cdot {{{{{{{rm{IQR}}}}}}}}/sqrt{50}), where IQR is the interquartile range; box limits, upper and lower quartiles; whiskers, extending up to the most extreme data points that are within ±1.5 ⋅ IQR; circles, outliers. Metapopulation metric values were obtained by setting α = 100 l. Note that in Eq. (1), given A = 40, 000, AT = 20 results in E[N] ≈ 4574, E[pr] ≈ 0.228; AT = 100 yields E[N] ≈ 2231, E[pr] ≈ 0.098; AT = 500 results in E[N] ≈ 1088, E[pr] ≈ 0.042.Full size imageFor a constant α value, the CVM of real rivers, OCNs and RBNs decreases as the resolution at which the network is extracted increases (i.e., AT decreases; see Fig. 6 and Supplementary Figs. 4–7). This is expected63, since a decrease in AT corresponds to an increase in N (Fig. 2a), leading to a decrease in CVM. Indeed, a larger ecosystem, constituted of more patches, has the potential to include a larger (and more diverse) number of subpopulations, which increases stability at a metapopulation level through statistical averaging–a phenomenon widely known as the portfolio effect65. We also found that BBT networks do not generally follow the above-described pattern of decreasing CVM with increasing N; rather, the CVM of BBTs increases with N when the mean dispersal distance α is set to intermediate to high values (Fig. 6 and Supplementary Figs. 5–7), and only when α is very low (e.g. α = 10 l as in Supplementary Fig. 4) and a uniform patch-size distribution is assumed does CVM,U follow the expected decreasing trend with increasing N.However, we need to warn against the conclusion that river networks with higher values of pr (and hence lower AT, see Fig. 2b) are inherently associated with higher metapopulation stability. Indeed, our result was obtained by changing the scale at which we observed the same river networks, and not by increasing the river networks’ size. If the number of network nodes (and, consequently, the branching ratio pr) is determined by the scale at which the landscape is observed, one cannot directly assume that any of such nodes is a node (or patch) in the ecological sense, i.e. the geographical span of a local population: the extent of such patches should be determined based on the mobility characteristics of the focus species, and should be independent of the scale at which the river network is observed. In contrast, we note that, if different river networks spanning different catchment areas (say, in km2) are compared, all of them extracted from the same DEM (same l and same AT in km2), then the larger river network will appear more branching (i.e., have larger pr). Indeed, by selecting catchments with larger A (in km2) for fixed l and AT (in km2), one moves towards the top-left corner of Fig. 2a, b (i.e., perpendicular to the level curves AT/A). The apparent higher “branchiness” of the river network with larger A will result in lower values of CVM; however, the higher metapopulation stability of the larger network will not be due to its (alleged) inherent more branching character, but only dictated by its larger habitat availability.We observed that metapopulation capacity λM values of OCNs (be it evaluated under uniform (λM,U) or non-uniform (λM,H) patch-size distribution assumption) are the closest to those of real rivers, while RBNs (and even more so BBTs) generally overestimate λM with respect to real rivers and OCNs (Fig. 6d–f; j–l). This result holds irrespective of the choice of AT and for intermediate to high values of α (Supplementary Figs. 5–7). When the mean dispersal distance is instead set to very low values (α = 10 l – Supplementary Fig. 4) and the river network is extracted at a high resolution (i.e., low AT), the metapopulation capacity of OCNs under assumption of uniform patch-size distribution (λM,U) is underestimated with respect to that of real rivers. A likely explanation for this apparent mismatch is that, for low values of AT, the number of nodes N tends to be somewhat higher for the extracted river networks used in this analysis than for OCNs (Supplementary Fig. 8), and the effect of the different dimensionality of real rivers and OCNs in the metapopulation capacity estimation tends to be more evident as the mean dispersal distance decreases. Interestingly, such mismatch is absent when a non-uniform patch size distribution is assumed, as λM,H values for OCNs match those for real rivers regardless of the mean dispersal distance value and the river network resolution (Fig. 6; Supplementary Figs. 4–7).The OCN construct encapsulates both random and deterministic processes, the former related to the stochastic nature of the OCN generation algorithm, and the latter pertaining to the minimization of total energy expenditure that characterizes OCN configurations. As such, OCNs reproduce the aggregation patterns of real river networks. From an ecological viewpoint, this implies that both pairwise distances between nodes and the distribution of patch sizes (expressed as a function of drainage areas, or of a proxy thereof such as the number of nodes upstream) are much closer to those of real networks than is the case for fully random synthetic networks as BBTs and RBNs. In particular, BBTs and (to a lesser extent) RBNs tend to underestimate pairwise distances with respect to real rivers and OCNs, as documented by a comparison of mean pairwise distances across network types (Supplementary Fig. 9a–c). Our analysis shows that the connectivity structure of these random networks (subsumed by the matrix of pairwise distances) is too compact with respect to that of real rivers, which leads to an overestimation of the role of dispersal in increasing the ability of a metapopulation to persist in the long run, but also an increased likelihood of synchrony among the different local populations, which results in higher instability.Comparison of patch size distributions among the network types expressed in terms of CVM,0 (i.e., the portion of CVM,H that uniquely depends on the distribution of patch sizes and not on pairwise distances) shows that, while for coarsely resolved networks (AT = 500) no clear differences in CVM,0 emerged, for highly resolved networks (AT = 20) BBTs heavily underestimate the CVM,0 of real rivers and OCNs, while RBNs slightly overestimate it (Supplementary Fig. 9d–f). As a result of the interplay of differences in distance matrices and patch size distributions, BBTs and (to a lesser extent) RBNs generally tend to overestimate the coefficient of variation of a metapopulation and the metapopulation capacity of real rivers and OCNs in both scenarios of uniform and non-uniform patch size distribution. The only exception to this trend occurs for the metapopulation capacity λM,H of very large BBTs (corresponding to AT = 20) in the case of very high dispersal distances (α = 1000 l – Supplementary Fig. 7): here, the patch-size effect (i.e., underestimation of CVM,0) predominates over the distance effect (i.e., overestimation of mean dij), resulting in an underestimation of λM,H with respect to real rivers and OCNs.Our results were derived under a number of simplifying assumptions. In particular, we acknowledge that, while the distance matrix of a landscape and the distribution of patch sizes have in general important implications for metapopulation dynamics, other factors not considered here, such as Euclidean between-patch distance48, fat-tailed dispersal kernel66 and density-dependent dispersal67 could also play a relevant role in this respect. However, it needs to be noted that, especially with regards to the assessment of the Moran effect in metapopulation synchrony (i.e., increased synchrony in local fluvial populations that are geographically close but not flow-connected48), the use of OCNs allows integration of Euclidean distances in a metapopulation model, while this is not possible for RBNs and BBTs, where Euclidean distances are not defined. Moreover, if a larger degree of realism is required for a specific ecological modelling study, such as heterogeneity in abiotic factors (e.g. water temperature or flow rates), the use of OCNs as model landscapes allows a direct integration of these variables, as they can conveniently be expressed as functions of drainage area3,55. In contrast, this is not possible for RBNs or BBTs, because only OCNs verify the scaling of areas (Fig. 5c), while RBNs and BBTs lack a proper definition of drainage areas.Our comparison of synthetic and real river networks showed that riverine metapopulations are more stable and less invasible than what would be predicted by random network analogues. Conversely, the use of OCNs as model landscapes allows capturing not only the scaling features of real rivers, but also drawing ecological conclusions that are in line with those that could be observed in real river networks. We thus support the use of OCNs as analogues of real river networks in theoretical and applied ecological modelling studies. While we found that BBTs are highly inaccurate in reproducing ecological metrics of real river networks and should be therefore discarded altogether in future modelling applications, RBNs show a certain degree of similarity with OCNs and real river networks in this respect; moreover, RBNs (as is the case for any random tree61) satisfy Horton’s laws on bifurcation and length ratios. A relevant advantage of RBNs over OCNs is that their generation algorithm is at least one order of magnitude faster49. Therefore, we acknowledge that RBNs could be considered as a suitable surrogate for real river networks as null models in cases where a large number of network replicates is required. To this end, we encourage researchers exploiting synthetic river networks (whether they be OCNs or RBNs) to always clarify the observational scales (that is, total area drained, size of a node, area drained by a headwater) subsumed by the synthetic network and which give rise to a certain complexity measure (i.e., branching ratio). Only in such a way could the predictions from these studies be compared with real river networks.In conclusion, our results advocate a tighter integration between physical (geomorphology, hydrology) and biological (ecology) disciplines in the study of freshwater ecosystems, and particularly in the perspective of a mechanistic understanding of drivers of persistence and loss of biodiversity. More

Houborg, R., Fisher, J. B. & Skidmore, A. K. Advances in remote sensing of vegetation function and traits. Int. J. Appl. Earth Obs. Geoinf. 43, 1–6 (2015).
Google Scholar 
Bannari, A., Morin, D., Bonn, F. & Huete, A. A review of vegetation indices. Remote Sens. Rev. 13, 95–120 (1995).Article 

Google Scholar 
Gao, X., Huete, A. R., Ni, W. & Miura, T. Optical–biophysical relationships of vegetation spectra without background contamination. Remote Sens. Environ. 74, 609–620 (2000).Article 

Google Scholar 
Huete, A. R. A soil-adjusted vegetation index (SAVI). Remote Sens. Environ. 25, 295–309 (1988).Article 

Google Scholar 
Badgley, G., Field, C. B. & Berry, J. A. Canopy near-infrared reflectance and terrestrial photosynthesis. Sci. Adv. 3, e1602244 (2017).Article 

Google Scholar 
Gamon, J. A. et al. A remotely sensed pigment index reveals photosynthetic phenology in evergreen conifers. Proc. Natl Acad. Sci. USA 113, 13087–13092 (2016).Article 

Google Scholar 
Joiner, J. et al. Estimation of terrestrial global gross primary production (GPP) with satellite data-driven models and eddy covariance flux data. Remote Sens. 10, 1346 (2018).Article 

Google Scholar 
Piao, S. et al. Characteristics, drivers and feedbacks of global greening. Nat. Rev. Earth Environ. 1, 14–27 (2020).Article 

Google Scholar 
Tian, F. et al. Evaluating temporal consistency of long-term global NDVI datasets for trend analysis. Remote Sens. Environ. 163, 326–340 (2015).Article 

Google Scholar 
Fan, X. & Liu, Y. A global study of NDVI difference among moderate-resolution satellite sensors. ISPRS J. Photogramm. Remote Sens. 121, 177–191 (2016).Article 

Google Scholar 
AghaKouchak, A. et al. Remote sensing of drought: progress, challenges and opportunities. Rev. Geophys. 53, 452–480 (2015).Article 

Google Scholar 
Anyamba, A. & Tucker, in Remote Sensing of Drought: Innovative Monitoring Approaches Ch. 2 (eds Wardlow, B. D., Anderson, M. C. & Verdin, J. P.) (Taylor & Francis, 2012).Veraverbeke, S. et al. Hyperspectral remote sensing of fire: state-of-the-art and future perspectives. Remote Sens. Environ. 216, 105–121 (2018).Article 

Google Scholar 
Tucker, C. J. Red and photographic infrared linear combinations for monitoring vegetation. Remote Sens. Environ. 8, 127–150 (1979).Article 

Google Scholar 
Rouse, J. W., Haas, R. H., Schell, J. A. & Deering, D. W. Monitoring vegetation systems in the Great Plains with ERTS. NASA Spec. Publ. 351, 309 (1974).
Google Scholar 
Rouse, J. W., Haas, R. H., Schell, J. A., Deering, D. W. & Harlan, J. C. Monitoring the vernal advancement and retrogradation (green wave effect) of natural vegetation. NASA/GSFCT Type III Final Report, 371 (NASA, 1974).Gutman, G., Skakun, S. & Gitelson, A. Revisiting the use of red and near-infrared reflectances in vegetation studies and numerical climate models. Sci. Remote Sens. 4, 100025 (2021).Article 

Google Scholar 
Jackson, R. D. & Huete, A. R. Interpreting vegetation indices. Prev. Vet. Med. 11, 185–200 (1991).Article 

Google Scholar 
Richardson, A. J. & Wiegand, C. Distinguishing vegetation from soil background information. Photogramm. Eng. Remote Sens. 43, 1541–1552 (1977).
Google Scholar 
Baret, F., Guyot, G. & Major, D. in 12th Canadian Symposium on Remote Sensing Geoscience and Remote Sensing Symposium 1355–1358 (IEEE, 1989).Qi, J., Chehbouni, A., Huete, A. R., Kerr, Y. H. & Sorooshian, S. A modified soil adjusted vegetation index. Remote Sens. Environ. 48, 119–126 (1994).Article 

Google Scholar 
Chen, J. M. Evaluation of vegetation indices and a modified simple ratio for boreal applications. Can. J. Remote Sens. 22, 229–242 (1996).Article 

Google Scholar 
Brown, L., Chen, J. M., Leblanc, S. G. & Cihlar, J. A shortwave infrared modification to the simple ratio for LAI retrieval in boreal forests: an image and model analysis. Remote Sens. Environ. 71, 16–25 (2000).Article 

Google Scholar 
Pinty, B. & Verstraete, M. GEMI: a non-linear index to monitor global vegetation from satellites. Vegetatio 101, 15–20 (1992).Article 

Google Scholar 
Huete, A. et al. Overview of the radiometric and biophysical performance of the MODIS vegetation indices. Remote Sens. Environ. 83, 195–213 (2002).Article 

Google Scholar 
Kaufman, Y. J. & Tanre, D. Atmospherically resistant vegetation index (ARVI) for EOS-MODIS. IEEE Trans. Geosci. Remote Sens. 30, 261–270 (1992).Article 

Google Scholar 
Jiang, Z., Huete, A. R., Didan, K. & Miura, T. Development of a two-band enhanced vegetation index without a blue band. Remote Sens. Environ. 112, 3833–3845 (2008).Article 

Google Scholar 
Jin, H. & Eklundh, L. A physically based vegetation index for improved monitoring of plant phenology. Remote Sens. Environ. 152, 512–525 (2014).Article 

Google Scholar 
Yang, P., van der Tol, C., Campbell, P. K. & Middleton, E. M. Fluorescence Correction Vegetation Index (FCVI): A physically based reflectance index to separate physiological and non-physiological information in far-red sun-induced chlorophyll fluorescence. Remote Sens. Environ. 240, 111676 (2020).Article 

Google Scholar 
Badgley, G., Anderegg, L. D., Berry, J. A. & Field, C. B. Terrestrial gross primary production: Using NIRV to scale from site to globe. Glob. Change Biol. 25, 3731–3740 (2019).Article 

Google Scholar 
Camps-Valls, G. et al. A unified vegetation index for quantifying the terrestrial biosphere. Sci. Adv. 7, eabc7447 (2021).Article 

Google Scholar 
Roberts, D. A., Roth, K. L. & Perroy, R. L. in Hyperspectral Remote Sensing of Vegetation Ch. 14 (eds Thenkabail, P. S., Lyon, J. G. & Huete, A.) (CRC, 2016).Gitelson, A. A., Vina, A., Ciganda, V., Rundquist, D. C. & Arkebauer, T. J. Remote estimation of canopy chlorophyll content in crops. Geophys. Res. Lett. 32, L08403 (2005).Article 

Google Scholar 
Gitelson, A. & Merzlyak, M. N. Spectral reflectance changes associated with autumn senescence of Aesculus hippocastanum L. and Acer platanoides L. leaves. Spectral features and relation to chlorophyll estimation. J. Plant Physiol. 143, 286–292 (1994).Article 

Google Scholar 
Dash, J. & Curran, P. The MERIS terrestrial chlorophyll index. Int. J. Remote Sens. 25, 5403–5413 (2004).Article 

Google Scholar 
Penuelas, J., Baret, F. & Filella, I. Semi-empirical indices to assess carotenoids/chlorophyll a ratio from leaf spectral reflectance. Photosynthetica 31, 221–230 (1995).
Google Scholar 
Peñuelas, J., Gamon, J., Fredeen, A., Merino, J. & Field, C. Reflectance indices associated with physiological changes in nitrogen- and water-limited sunflower leaves. Remote Sens. Environ. 48, 135–146 (1994).Article 

Google Scholar 
Merzlyak, M. N., Gitelson, A. A., Chivkunova, O. B. & Rakitin, V. Y. Non-destructive optical detection of pigment changes during leaf senescence and fruit ripening. Physiol. Plant. 106, 135–141 (1999).Article 

Google Scholar 
Gitelson, A. A., Merzlyak, M. N. & Chivkunova, O. B. Optical properties and nondestructive estimation of anthocyanin content in plant leaves. Photochem. Photobiol. 74, 38–45 (2001).Article 

Google Scholar 
van den Berg, A. K. & Perkins, T. D. Nondestructive estimation of anthocyanin content in autumn sugar maple leaves. HortScience 40, 685–686 (2005).Article 

Google Scholar 
Gamon, J. & Surfus, J. Assessing leaf pigment content and activity with a reflectometer. New Phytol. 143, 105–117 (1999).Article 

Google Scholar 
Gao, B.-C. NDWI — a normalized difference water index for remote sensing of vegetation liquid water from space. Remote Sens. Environ. 58, 257–266 (1996).Article 

Google Scholar 
Xiao, X., Boles, S., Liu, J., Zhuang, D. & Liu, M. Characterization of forest types in Northeastern China, using multi-temporal SPOT-4 VEGETATION sensor data. Remote Sens. Environ. 82, 335–348 (2002).Article 

Google Scholar 
Xiao, X. et al. Satellite-based modeling of gross primary production in an evergreen needleleaf forest. Remote Sens. Environ. 89, 519–534 (2004).Article 

Google Scholar 
Yilmaz, M. T., Hunt, E. R. Jr & Jackson, T. J. Remote sensing of vegetation water content from equivalent water thickness using satellite imagery. Remote Sens. Environ. 112, 2514–2522 (2008).Article 

Google Scholar 
Cheng, Y.-B., Ustin, S. L., Riaño, D. & Vanderbilt, V. C. Water content estimation from hyperspectral images and MODIS indexes in Southeastern Arizona. Remote Sens. Environ. 112, 363–374 (2008).Article 

Google Scholar 
Serrano, L., Penuelas, J. & Ustin, S. L. Remote sensing of nitrogen and lignin in Mediterranean vegetation from AVIRIS data: decomposing biochemical from structural signals. Remote Sens. Environ. 81, 355–364 (2002).Article 

Google Scholar 
Filella, I. et al. PRI assessment of long-term changes in carotenoids/chlorophyll ratio and short-term changes in de-epoxidation state of the xanthophyll cycle. Int. J. Remote Sens. 30, 4443–4455 (2009).Article 

Google Scholar 
Gamon, J., Penuelas, J. & Field, C. A narrow-waveband spectral index that tracks diurnal changes in photosynthetic efficiency. Remote Sens. Environ. 41, 35–44 (1992).Article 

Google Scholar 
Cheng, R. et al. Decomposing reflectance spectra to track gross primary production in a subalpine evergreen forest. Biogeosciences 17, 4523–4544 (2020).Article 

Google Scholar 
Seyednasrollah, B. et al. Seasonal variation in the canopy color of temperate evergreen conifer forests. New Phytol. 229, 2586–2600 (2021).Article 

Google Scholar 
Merton, R. in Proceedings of the Seventh Annual JPL Airborne Earth Science Workshop 12–16 (NASA, 2004).Naidu, R. A., Perry, E. M., Pierce, F. J. & Mekuria, T. The potential of spectral reflectance technique for the detection of Grapevine leafroll-associated virus-3 in two red-berried wine grape cultivars. Comput. Electron. Agric. 66, 38–45 (2009).Article 

Google Scholar 
Chen, Y. et al. Generation and evaluation of LAI and FPAR products from Himawari-8 Advanced Himawari imager (AHI) data. Remote Sens. 11, 1517 (2019).Article 

Google Scholar 
Zhu, Z. et al. Global data sets of vegetation leaf area index (LAI)3g and fraction of photosynthetically active radiation (FPAR)3g derived from global inventory modeling and mapping studies (GIMMS) normalized difference vegetation index (NDVI3g) for the period 1981 to 2011. Remote Sens. 5, 927–948 (2013).Article 

Google Scholar 
Liu, Y., Liu, R. & Chen, J. M. Retrospective retrieval of long-term consistent global leaf area index (1981–2011) from combined AVHRR and MODIS data. J. Geophys. Res. 117, G04003 (2012).
Google Scholar 
Croft, H. et al. The global distribution of leaf chlorophyll content. Remote Sens. Environ. 236, 111479 (2020).Article 

Google Scholar 
Bayat, B. et al. Toward operational validation systems for global satellite-based terrestrial essential climate variables. Int. J. Appl. Earth Obs. Geoinf. 95, 102240 (2021).
Google Scholar 
Cui, Y., Song, L. & Fan, W. Generation of spatio-temporally continuous evapotranspiration and its components by coupling a two-source energy balance model and a deep neural network over the Heihe River Basin. J. Hydrol. 597, 126176 (2021).Article 

Google Scholar 
Ali, I., Greifeneder, F., Stamenkovic, J., Neumann, M. & Notarnicola, C. Review of machine learning approaches for biomass and soil moisture retrievals from remote sensing data. Remote Sens. 7, 16398–16421 (2015).Article 

Google Scholar 
Gitelson, A. A. et al. Remote estimation of leaf area index and green leaf biomass in maize canopies. Geophys. Res. Lett. 30, 1248 (2003).Article 

Google Scholar 
Huang, M. et al. Air temperature optima of vegetation productivity across global biomes. Nat. Ecol. Evol. 3, 772–779 (2019).Article 

Google Scholar 
Wang, S. et al. Recent global decline of CO2 fertilization effects on vegetation photosynthesis. Science 370, 1295–1300 (2020).Article 

Google Scholar 
Morton, D. C. et al. Amazon forests maintain consistent canopy structure and greenness during the dry season. Nature 506, 221–224 (2014).Article 

Google Scholar 
Jiang, Z. et al. Analysis of NDVI and scaled difference vegetation index retrievals of vegetation fraction. Remote Sens. Environ. 101, 366–378 (2006).Article 

Google Scholar 
Haboudane, D., Miller, J. R., Pattey, E., Zarco-Tejada, P. J. & Strachan, I. B. Hyperspectral vegetation indices and novel algorithms for predicting green LAI of crop canopies: modeling and validation in the context of precision agriculture. Remote Sens. Environ. 90, 337–352 (2004).Article 

Google Scholar 
Wu, C., Wang, L., Niu, Z., Gao, S. & Wu, M. Nondestructive estimation of canopy chlorophyll content using Hyperion and Landsat/TM images. Int. J. Remote Sens. 31, 2159–2167 (2010).Article 

Google Scholar 
Wang, R. & Gamon, J. A. Remote sensing of terrestrial plant biodiversity. Remote Sens. Environ. 231, 111218 (2019).Article 

Google Scholar 
Ustin, S. L. & Gamon, J. A. Remote sensing of plant functional types. New Phytol. 186, 795–816 (2010).Article 

Google Scholar 
Hilker, T. et al. Vegetation dynamics and rainfall sensitivity of the Amazon. Proc. Natl Acad. Sci. USA 111, 16041–16046 (2014).Article 

Google Scholar 
Zhang, Y., Commane, R., Zhou, S., Williams, A. P. & Gentine, P. Light limitation regulates the response of autumn terrestrial carbon uptake to warming. Nat. Clim. Change 10, 739–743 (2020).Article 

Google Scholar 
Weber, M. et al. Exploring the use of DSCOVR/EPIC satellite observations to monitor vegetation phenology. Remote Sens. 12, 2384 (2020).Article 

Google Scholar 
Ganguly, S., Friedl, M. A., Tan, B., Zhang, X. & Verma, M. Land surface phenology from MODIS: characterization of the Collection 5 global land cover dynamics product. Remote Sens. Environ. 114, 1805–1816 (2010).Article 

Google Scholar 
Gray, J., Sulla-Menashe, D. & Friedl, M. A. User Guide to Collection 6 MODIS Land Cover Dynamics Product (MCD12Q2) (NASA, 2019).Wang, S., Zhang, Y., Ju, W., Qiu, B. & Zhang, Z. Tracking the seasonal and inter-annual variations of global gross primary production during last four decades using satellite near-infrared reflectance data. Sci. Total Environ. 755, 142569 (2021).Article 

Google Scholar 
Tian, F. et al. Calibrating vegetation phenology from Sentinel-2 using eddy covariance, PhenoCam, and PEP725 networks across Europe. Remote Sens. Environ. 260, 112456 (2021).Article 

Google Scholar 
Yin, G., Verger, A., Filella, I., Descals, A. & Peñuelas, J. Divergent estimates of forest photosynthetic phenology using structural and physiological vegetation indices. Geophys. Res. Lett. 47, e2020GL089167 (2020).Article 

Google Scholar 
Qin, Y. et al. Carbon loss from forest degradation exceeds that from deforestation in the Brazilian Amazon. Nat. Clim. Change 11, 442–448 (2021).Article 

Google Scholar 
Samanta, A. et al. Amazon forests did not green-up during the 2005 drought. Geophys. Res. Lett. 37, L05401 (2010).Article 

Google Scholar 
Shi, Y., Huang, W., Luo, J., Huang, L. & Zhou, X. Detection and discrimination of pests and diseases in winter wheat based on spectral indices and kernel discriminant analysis. Comput. Electron. Agric. 141, 171–180 (2017).Article 

Google Scholar 
Zhang, Z., Liu, M., Liu, X. & Zhou, G. A new vegetation index based on multitemporal Sentinel-2 images for discriminating heavy metal stress levels in rice. Sensors 18, 2172 (2018).Article 

Google Scholar 
Yengoh, G. T., Dent, D., Olsson, L., Tengberg, A. E. & Tucker III, C. J. Use of the Normalized Difference Vegetation Index (NDVI) to Assess Land Degradation at Multiple Scales: Current Status, Future Trends, and Practical Considerations (Springer, 2015).Potter, C. S. et al. Terrestrial ecosystem production: a process model based on global satellite and surface data. Glob. Biogeochem. Cycles 7, 811–841 (1993).Article 

Google Scholar 
Running, S. W. et al. A continuous satellite-derived measure of global terrestrial primary production. Bioscience 54, 547–560 (2004).Article 

Google Scholar 
Yuan, W. et al. Deriving a light use efficiency model from eddy covariance flux data for predicting daily gross primary production across biomes. Agric. For. Meteorol. 143, 189–207 (2007).Article 

Google Scholar 
Chen, M. et al. Quantification of terrestrial ecosystem carbon dynamics in the conterminous United States combining a process-based biogeochemical model and MODIS and AmeriFlux data. Biogeosciences 8, 2665–2688 (2011).Article 

Google Scholar 
Xiao, J. et al. A continuous measure of gross primary production for the conterminous United States derived from MODIS and AmeriFlux data. Remote Sens. Environ. 114, 576–591 (2010).Article 

Google Scholar 
Jiang, C., Guan, K., Wu, G., Peng, B. & Wang, S. A daily, 250 m, and real-time gross primary productivity product (2000–present) covering the contiguous United States. Earth Syst. Sci. Data Discuss. 2020, 1–28 (2020).
Google Scholar 
Schubert, P. et al. Modeling GPP in the Nordic forest landscape with MODIS time series data — comparison with the MODIS GPP product. Remote Sens. Environ. 126, 136–147 (2012).Article 

Google Scholar 
Zeng, Y. et al. A practical approach for estimating the escape ratio of near-infrared solar-induced chlorophyll fluorescence. Remote Sens. Environ. 232, 111209 (2019).Article 

Google Scholar 
Baldocchi, D. D. et al. Outgoing near infrared radiation from vegetation scales with canopy photosynthesis across a spectrum of function, structure, physiological capacity and weather. J. Geophys. Res. 125, e2019JG005534 (2020).
Google Scholar 
Dechant, B. et al. Canopy structure explains the relationship between photosynthesis and sun-induced chlorophyll fluorescence in crops. Remote Sens. Environ. 241, 111733 (2020).Article 

Google Scholar 
Rahman, A. F., Gamon, J. A., Fuentes, D. A., Roberts, D. A. & Prentiss, D. Modeling spatially distributed ecosystem flux of boreal forest using hyperspectral indices from AVIRIS imagery. J. Geophys. Res. Atmos. 106, 33579–33591 (2001).Article 

Google Scholar 
Zhu, Z. et al. Comment on “Recent global decline of CO2 fertilization effects on vegetation photosynthesis”. Science 373, eabg5673 (2021).Article 

Google Scholar 
Doughty, R. et al. Small anomalies in dry-season greenness and chlorophyll fluorescence for Amazon moist tropical forests during El Niño and La Niña. Remote Sens. Environ. 253, 112196 (2021).Article 

Google Scholar 
Huang, N. et al. Spatial and temporal variations in global soil respiration and their relationships with climate and land cover. Sci. Adv. 6, eabb8508 (2020).Article 

Google Scholar 
Huang, N., He, J.-S. & Niu, Z. Estimating the spatial pattern of soil respiration in Tibetan alpine grasslands using Landsat TM images and MODIS data. Ecol. Indic. 26, 117–125 (2013).Article 

Google Scholar 
Neale, C. M., Gonzalez-Dugo, M. P., Serrano-Perez, A., Campos, I. & Mateos, L. Cotton canopy reflectance under variable solar zenith angles: implications of use in evapotranspiration models. Hydrol. Process. 35, e14162 (2021).Article 

Google Scholar 
Chen, J. M. & Liu, J. Evolution of evapotranspiration models using thermal and shortwave remote sensing data. Remote Sens. Environ. 237, 111594 (2020).Article 

Google Scholar 
Glenn, E. P., Huete, A. R., Nagler, P. L. & Nelson, S. G. Relationship between remotely-sensed vegetation indices, canopy attributes and plant physiological processes: what vegetation indices can and cannot tell us about the landscape. Sensors 8, 2136–2160 (2008).Article 

Google Scholar 
Cui, Y., Jia, L. & Fan, W. Estimation of actual evapotranspiration and its components in an irrigated area by integrating the Shuttleworth-Wallace and surface temperature-vegetation index schemes using the particle swarm optimization algorithm. Agric. For. Meteorol. 307, 108488 (2021).Article 

Google Scholar 
Glenn, E. P., Neale, C. M., Hunsaker, D. J. & Nagler, P. L. Vegetation index-based crop coefficients to estimate evapotranspiration by remote sensing in agricultural and natural ecosystems. Hydrol. Process. 25, 4050–4062 (2011).Article 

Google Scholar 
French, A. N. et al. Satellite-based NDVI crop coefficients and evapotranspiration with eddy covariance validation for multiple durum wheat fields in the US Southwest. Agric. Water Manag. 239, 106266 (2020).Article 

Google Scholar 
Lotsch, A., Friedl, M. A., Anderson, B. T. & Tucker, C. J. Coupled vegetation-precipitation variability observed from satellite and climate records. Geophys. Res. Lett. 30, 1774 (2003).Article 

Google Scholar 
Nezlin, N. P., Kostianoy, A. G. & Li, B.-L. Inter-annual variability and interaction of remote-sensed vegetation index and atmospheric precipitation in the Aral Sea region. J. Arid Environ. 62, 677–700 (2005).Article 

Google Scholar 
Notaro, M., Liu, Z. & Williams, J. W. Observed vegetation–climate feedbacks in the United States. J. Clim. 19, 763–786 (2006).Article 

Google Scholar 
Fensholt, R. & Proud, S. R. Evaluation of earth observation based global long term vegetation trends — Comparing GIMMS and MODIS global NDVI time series. Remote Sens. Environ. 119, 131–147 (2012).Article 

Google Scholar 
Trishchenko, A. P., Cihlar, J. & Li, Z. Effects of spectral response function on surface reflectance and NDVI measured with moderate resolution satellite sensors. Remote Sens. Environ. 81, 1–18 (2002).Article 

Google Scholar 
Ustin, S. L. & Middleton, E. M. Current and near-term advances in Earth observation for ecological applications. Ecol. Process. 10, 1 (2021).Article 

Google Scholar 
Wang, D. et al. Impact of sensor degradation on the MODIS NDVI time series. Remote Sens. Environ. 119, 55–61 (2012).Article 

Google Scholar 
Zhang, Y., Song, C., Band, L. E., Sun, G. & Li, J. Reanalysis of global terrestrial vegetation trends from MODIS products: browning or greening? Remote Sens. Environ. 191, 145–155 (2017).Article 

Google Scholar 
Bhatt, R. et al. A consistent AVHRR visible calibration record based on multiple methods applicable for the NOAA degrading orbits. Part I: Methodology. J. Atmos. Ocean. Technol. 33, 2499–2515 (2016).Article 

Google Scholar 
Frankenberg, C., Yin, Y., Byrne, B., He, L. & Gentine, P. Comment on “Recent global decline of CO2 fertilization effects on vegetation photosynthesis”. Science 373, eabg2947 (2021).Article 

Google Scholar 
Los, S. O. Estimation of the ratio of sensor degradation between NOAA AVHRR channels 1 and 2 from monthly NDVI composites. IEEE Trans. Geosci. Remote Sens. 36, 206–213 (1998).Article 

Google Scholar 
Jiang, C. et al. Inconsistencies of interannual variability and trends in long-term satellite leaf area index products. Glob. Change Biol. 23, 4133–4146 (2017).Article 

Google Scholar 
de Beurs, K. M. & Henebry, G. M. Trend analysis of the Pathfinder AVHRR Land (PAL) NDVI data for the deserts of Central Asia. IEEE Geosci. Remote Sens. Lett. 1, 282–286 (2004).Article 

Google Scholar 
Wang, Z. et al. Large discrepancies of global greening: indication of multi-source remote sensing data. Global Ecol. Conserv. 34, e02016 (2022).Article 

Google Scholar 
Miura, T., Huete, A. R. & Yoshioka, H. Evaluation of sensor calibration uncertainties on vegetation indices for MODIS. IEEE Trans Geosci. Remote Sens. 38, 1399–1409 (2000).Article 

Google Scholar 
Lyapustin, A. et al. Scientific impact of MODIS C5 calibration degradation and C6+ improvements. Atmos. Meas. Tech. 7, 4353–4365 (2014).Article 

Google Scholar 
Buchhorn, M., Raynolds, M. K. & Walker, D. A. Influence of BRDF on NDVI and biomass estimations of Alaska Arctic tundra. Environ. Res. Lett. 11, 125002 (2016).Article 

Google Scholar 
Fensholt, R., Sandholt, I., Proud, S. R., Stisen, S. & Rasmussen, M. O. Assessment of MODIS sun-sensor geometry variations effect on observed NDVI using MSG SEVIRI geostationary data. Int. J. Remote Sens. 31, 6163–6187 (2010).Article 

Google Scholar 
Saleska, S. R. et al. Dry-season greening of Amazon forests. Nature 531, E4–E5 (2016).Article 

Google Scholar 
Lyapustin, A. I. et al. Multi-angle implementation of atmospheric correction for MODIS (MAIAC): 3. Atmospheric correction. Remote Sens. Environ. 127, 385–393 (2012).Article 

Google Scholar 
Norris, J. R. & Walker, J. J. Solar and sensor geometry, not vegetation response, drive satellite NDVI phenology in widespread ecosystems of the western United States. Remote Sens. Environ. 249, 112013 (2020).Article 

Google Scholar 
Roy, D. P. et al. A general method to normalize Landsat reflectance data to nadir BRDF adjusted reflectance. Remote Sens. Environ. 176, 255–271 (2016).Article 

Google Scholar 
Schaaf, C. B. et al. First operational BRDF, albedo nadir reflectance products from MODIS. Remote Sens. Environ. 83, 135–148 (2002).Article 

Google Scholar 
Didan, K., Munoz, A. B., Solano, R. & Huete, A. MODIS Vegetation Index User’s Guide (MOD13 Series) (Univ. Arizona, 2015).Wang, Z., Schaaf, C. B., Sun, Q., Shuai, Y. & Román, M. O. Capturing rapid land surface dynamics with Collection V006 MODIS BRDF/NBAR/Albedo (MCD43) products. Remote Sens. Environ. 207, 50–64 (2018).Article 

Google Scholar 
Saleska, S. R., Didan, K., Huete, A. R. & Da Rocha, H. R. Amazon forests green-up during 2005 drought. Science 318, 612 (2007).Article 

Google Scholar 
Vargas, M., Miura, T., Shabanov, N. & Kato, A. An initial assessment of Suomi NPP VIIRS vegetation index EDR. J. Geophys. Res. Atmos. 118, 12,301–12,316 (2013).Article 

Google Scholar 
Kobayashi, H. & Dye, D. G. Atmospheric conditions for monitoring the long-term vegetation dynamics in the Amazon using normalized difference vegetation index. Remote Sens. Environ. 97, 519–525 (2005).Article 

Google Scholar 
Jiang, C. & Fang, H. GSV: a general model for hyperspectral soil reflectance simulation. Int. J. Appl. Earth Obs. Geoinf. 83, 101932 (2019).
Google Scholar 
Verrelst, J., Schaepman, M. E., Malenovský, Z. & Clevers, J. G. Effects of woody elements on simulated canopy reflectance: Implications for forest chlorophyll content retrieval. Remote Sens. Environ. 114, 647–656 (2010).Article 

Google Scholar 
Huete, A. & Tucker, C. Investigation of soil influences in AVHRR red and near-infrared vegetation index imagery. Int. J. Remote Sens. 12, 1223–1242 (1991).Article 

Google Scholar 
Farrar, T., Nicholson, S. & Lare, A. The influence of soil type on the relationships between NDVI, rainfall, and soil moisture in semiarid Botswana. II. NDVI response to soil oisture. Remote Sens. Environ. 50, 121–133 (1994).Article 

Google Scholar 
Huete, A. & Warrick, A. Assessment of vegetation and soil water regimes in partial canopies with optical remotely sensed data. Remote Sens. Environ. 32, 155–167 (1990).Article 

Google Scholar 
Wang, C. et al. A snow-free vegetation index for improved monitoring of vegetation spring green-up date in deciduous ecosystems. Remote Sens. Environ. 196, 1–12 (2017).Article 

Google Scholar 
Myers-Smith, I. H. et al. Complexity revealed in the greening of the Arctic. Nat. Clim. Change 10, 106–117 (2020).Article 

Google Scholar 
Shen, M. et al. No evidence of continuously advanced green-up dates in the Tibetan Plateau over the last decade. Proc. Natl Acad. Sci. 110, E2329 (2013).
Google Scholar 
Hao, D. et al. Modeling anisotropic reflectance over composite sloping terrain. IEEE Trans. Geosci. Remote Sens. 56, 3903–3923 (2018).Article 

Google Scholar 
Matsushita, B., Yang, W., Chen, J., Onda, Y. & Qiu, G. Sensitivity of the enhanced vegetation index (EVI) and normalized difference vegetation index (NDVI) to topographic effects: a case study in high-density cypress forest. Sensors 7, 2636–2651 (2007).Article 

Google Scholar 
Wen, J. et al. Characterizing land surface anisotropic reflectance over rugged terrain: a review of concepts and recent developments. Remote Sens. 10, 370 (2018).Article 

Google Scholar 
Friedl, M. A., Davis, F. W., Michaelsen, J. & Moritz, M. Scaling and uncertainty in the relationship between the NDVI and land surface biophysical variables: an analysis using a scene simulation model and data from FIFE. Remote Sens. Environ. 54, 233–246 (1995).Article 

Google Scholar 
Tan, B. et al. The impact of gridding artifacts on the local spatial properties of MODIS data: implications for validation, compositing, and band-to-band registration across resolutions. Remote Sens. Environ. 105, 98–114 (2006).Article 

Google Scholar 
Wolfe, R. E. et al. Achieving sub-pixel geolocation accuracy in support of MODIS land science. Remote Sens. Environ. 83, 31–49 (2002).Article 

Google Scholar 
Ferreira, M. P. et al. Retrieving structural and chemical properties of individual tree crowns in a highly diverse tropical forest with 3D radiative transfer modeling and imaging spectroscopy. Remote Sens. Environ. 211, 276–291 (2018).Article 

Google Scholar 
Huete, A. R. et al. Amazon rainforests green-up with sunlight in dry season. Geophys. Res. Lett. 33, L06405 (2006).Article 

Google Scholar 
Herrmann, S. M. & Tappan, G. G. Vegetation impoverishment despite greening: a case study from central Senegal. J. Arid Environ. 90, 55–66 (2013).Article 

Google Scholar 
Wang, X. et al. No consistent evidence for advancing or delaying trends in spring phenology on the Tibetan Plateau. J. Geophys. Res. Biogeosci. 122, 3288–3305 (2017).Article 

Google Scholar 
Donnelly, A., Yu, R. & Liu, L. Comparing in situ spring phenology and satellite-derived start of season at rural and urban sites in Ireland. Int. J. Remote Sens. 42, 7821–7841 (2021).Article 

Google Scholar 
Templ, B. et al. Pan European Phenological database (PEP725): a single point of access for European data. Int. J. Biometeorol. 62, 1109–1113 (2018).Article 

Google Scholar 
Fu, Y. H. et al. Declining global warming effects on the phenology of spring leaf unfolding. Nature 526, 104–107 (2015).Article 

Google Scholar 
Chen, X. & Yang, Y. Observed earlier start of the growing season from middle to high latitudes across the Northern Hemisphere snow-covered landmass for the period 2001–2014. Environ. Res. Lett. 15, 034042 (2020).Article 

Google Scholar 
Alatorre, L. C. et al. Temporal changes of NDVI for qualitative environmental assessment of mangroves: shrimp farming impact on the health decline of the arid mangroves in the Gulf of California (1990–2010). J. Arid Environ. 125, 98–109 (2016).Article 

Google Scholar 
Jacquemoud, S. & Baret, F. PROSPECT: a model of leaf optical properties spectra. Remote Sens. Environ. 34, 75–91 (1990).Article 

Google Scholar 
Wu, S. et al. Quantifying leaf optical properties with spectral invariants theory. Remote Sens. Environ. 253, 112131 (2021).Article 

Google Scholar 
Wang, Z. et al. Mapping foliar functional traits and their uncertainties across three years in a grassland experiment. Remote Sens. Environ. 221, 405–416 (2019).Article 

Google Scholar 
Van Leeuwen, W. & Huete, A. Effects of standing litter on the biophysical interpretation of plant canopies with spectral indices. Remote Sens. Environ. 55, 123–138 (1996).Article 

Google Scholar 
Dechant, B. et al. NIRvP: a robust structural proxy for sun-induced chlorophyll fluorescence and photosynthesis across scales. Remote Sens. Environ. 268, 112763 (2022).Article 

Google Scholar 
Zeng, Y. et al. Estimating near-infrared reflectance of vegetation from hyperspectral data. Remote Sens. Environ. 267, 112723 (2021).Article 

Google Scholar 
Claverie, M. et al. The Harmonized Landsat and Sentinel-2 surface reflectance data set. Remote Sens. Environ. 219, 145–161 (2018).Article 

Google Scholar 
Hantson, S. & Chuvieco, E. Evaluation of different topographic correction methods for Landsat imagery. Int. J. Appl. Earth Obs. Geoinf. 13, 691–700 (2011).
Google Scholar 
Zhang, H. K. et al. Characterization of Sentinel-2A and Landsat-8 top of atmosphere, surface, and nadir BRDF adjusted reflectance and NDVI differences. Remote Sens. Environ. 215, 482–494 (2018).Article 

Google Scholar 
Gao, F., Masek, J., Schwaller, M. & Hall, F. On the blending of the Landsat and MODIS surface reflectance: predicting daily Landsat surface reflectance. IEEE Trans. Geosci. Remote Sens. 44, 2207–2218 (2006).Article 

Google Scholar 
Zhu, X. et al. A flexible spatiotemporal method for fusing satellite images with different resolutions. Remote Sens. Environ. 172, 165–177 (2016).Article 

Google Scholar 
Luo, Y., Guan, K. & Peng, J. STAIR: A generic and fully-automated method to fuse multiple sources of optical satellite data to generate a high-resolution, daily and cloud-/gap-free surface reflectance product. Remote Sens. Environ. 214, 87–99 (2018).Article 

Google Scholar 
Houborg, R. & McCabe, M. F. Daily retrieval of NDVI and LAI at 3 m resolution via the fusion of CubeSat, Landsat, and MODIS data. Remote Sens. 10, 890 (2018).Article 

Google Scholar 
Kimm, H. et al. Deriving high-spatiotemporal-resolution leaf area index for agroecosystems in the US Corn Belt using Planet Labs CubeSat and STAIR fusion data. Remote Sens. Environ. 239, 111615 (2020).Article 

Google Scholar 
Kong, J. et al. Evaluation of four image fusion NDVI products against in-situ spectral-measurements over a heterogeneous rice paddy landscape. Agric. For. Meteorol. 297, 108255 (2021).Article 

Google Scholar 
Köhler, P. et al. Global retrievals of solar-induced chlorophyll fluorescence with TROPOMI: first results and intersensor comparison to OCO-2. Geophys. Res. Lett. 45, 10,456–10,463 (2018).Article 

Google Scholar 
Sun, Y. et al. OCO-2 advances photosynthesis observation from space via solar-induced chlorophyll fluorescence. Science 358, eaam5747 (2017).Article 

Google Scholar 
Joiner, J., Yoshida, Y., Vasilkov, A. & Middleton, E. First observations of global and seasonal terrestrial chlorophyll fluorescence from space. Biogeosciences 8, 637–651 (2011).Article 

Google Scholar 
Frankenberg, C. et al. New global observations of the terrestrial carbon cycle from GOSAT: patterns of plant fluorescence with gross primary productivity. Geophys. Res. Lett. 38, L17706 (2011).Article 

Google Scholar 
Qiu, B., Ge, J., Guo, W., Pitman, A. J. & Mu, M. Responses of Australian dryland vegetation to the 2019 heat wave at a subdaily scale. Geophys. Res. Lett. 47, e2019GL086569 (2020).
Google Scholar 
Magney, T. S. et al. Mechanistic evidence for tracking the seasonality of photosynthesis with solar-induced fluorescence. Proc. Natl Acad. Sci. USA 116, 11640–11645 (2019).Article 

Google Scholar 
Guanter, L. et al. Potential of the TROPOspheric Monitoring Instrument (TROPOMI) onboard the Sentinel-5 Precursor for the monitoring of terrestrial chlorophyll fluorescence. Atmos. Meas. Tech. 8, 1337–1352 (2015).Article 

Google Scholar 
Knyazikhin, Y. et al. Hyperspectral remote sensing of foliar nitrogen content. Proc. Natl Acad. Sci. USA 110, E185–E192 (2013).
Google Scholar 
Li, X. & Xiao, J. A global, 0.05-degree product of solar-induced chlorophyll fluorescence derived from OCO-2, MODIS, and reanalysis data. Remote Sens. 11, 517 (2019).Article 

Google Scholar 
Zeng, Y. et al. Combining near-infrared radiance of vegetation and fluorescence spectroscopy to detect effects of abiotic changes and stresses. Remote Sens. Environ. 270, 112856 (2022).Article 

Google Scholar 
Shi, J. et al. Microwave vegetation indices for short vegetation covers from satellite passive microwave sensor AMSR-E. Remote Sens. Environ. 112, 4285–4300 (2008).Article 

Google Scholar 
Talebiesfandarani, S. et al. Microwave vegetation index from multi-angular observations and its application in vegetation properties retrieval: theoretical modelling. Remote Sens. 11, 730 (2019).Article 

Google Scholar 
Wigneron, J.-P. et al. SMOS-IC data record of soil moisture and L-VOD: historical development, applications and perspectives. Remote Sens. Environ. 254, 112238 (2021).Article 

Google Scholar 
Zhang, Y., Zhou, S., Gentine, P. & Xiao, X. Can vegetation optical depth reflect changes in leaf water potential during soil moisture dry-down events? Remote Sens. Environ. 234, 111451 (2019).Article 

Google Scholar 
Frappart, F. et al. Global monitoring of the vegetation dynamics from the vegetation optical depth (VOD): a review. Remote Sens. 12, 2915 (2020).Article 

Google Scholar 
Xiao, J., Fisher, J. B., Hashimoto, H., Ichii, K. & Parazoo, N. C. Emerging satellite observations for diurnal cycling of ecosystem processes. Nat. Plants 7, 877–887 (2021).Article 

Google Scholar 
Hashimoto, H. et al. New generation geostationary satellite observations support seasonality in greenness of the Amazon evergreen forests. Nat. Commun. 12, 684 (2021).Article 

Google Scholar 
Somkuti, P. et al. Solar-induced chlorophyll fluorescence from the Geostationary Carbon Cycle Observatory (GeoCarb): An extensive simulation study. Remote Sens. Environ. 263, 112565 (2021).Article 

Google Scholar 
Gorelick, N. et al. Google Earth Engine: planetary-scale geospatial analysis for everyone. Remote Sens. Environ. 202, 18–27 (2017).Article 

Google Scholar 
Richardson, A. D., Braswell, B. H., Hollinger, D. Y., Jenkins, J. P. & Ollinger, S. V. Near-surface remote sensing of spatial and temporal variation in canopy phenology. Ecol. Appl. 19, 1417–1428 (2009).Article 

Google Scholar 
Daughtry, C. S. Discriminating crop residues from soil by shortwave infrared reflectance. Agron. J. 93, 125–131 (2001).Article 

Google Scholar  More

Data UsedWe used cadastral maps from 1858 to reconstruct the LULC structure of Aksu and Kestel for the mid-nineteenth century. General Staff of the Ottoman Army produced these maps in 1:10,000 scale. These maps were one of the earliest attempts of creating cadastral maps in the Ottoman Empire. The images of historical maps scanned at 1270 dpi resolutions are provided by the Turkish Presidency State Archives of the Republic of Turkey – Department of Ottoman Archives (Map collection, HRT.h, 561–567). Individual tiles of cadastral maps are of a 1:2,000 scale. However, these maps are kept separated from their accompanying cadastral registers or documentation regarding their production process in the archives. There are no studies on the production of these cadastral maps, but few studies used them until now35,36.The LULC structures of Aksu and Kestel for the mid-twentieth century were generated using aerial photographs from June 23, 1955, with a scale of 1:30,000. These aerial photographs were captured by the US Navy Photographic Squadron VJ-62 (established on April 10, 1952, re-designated to VAP-62 on July 1956, and disestablished on October 15, 1969). The squadron conducted mapping and special photographic projects worldwide37. Lastly, the VHR satellite images of WorldView-3 (WV-3) with 0.3 m of spatial resolution, collected on September 6, 2020, were used to show the up-to-date LULC patterns of Aksu and Kestel.MethodologyFigure 2 shows the flowchart of steps followed in this study to detect the LULC changes. The workflow includes three phases: preprocessing, LULC mapping, and statistical analysis of LULC changes.Figure 2Flowchart of the processing steps for the LULC change analysis for Kestel.Full size imageData preprocessingOrthorectification is the first important step in ensuring accurate spatial positioning among the multi-temporal and multi-source images, eliminating geometric distortions, and defining all data sets on a common projection system. To align the multi-modal geospatial datasets, we first performed the orthorectification of the satellite images and then we used the orthorectified satellite images as reference for the georeferencing of the cadastral maps and aerial photographs.Satellite imagery orthorectificationWe first pan-sharpened the WV-3 images by applying the PANSHARP2 algorithm38 to fuse the panchromatic (PAN) image of 0.3 m spatial resolution with four multispectral bands (R, G, B, and near-infrared (NIR)) of 1.2 m. We then geometrically corrected the pan-sharpened (PSP) WV-3 imageries using an ALOS Global Digital Surface Model with a horizontal resolution of approximately 30 m (ALOS World 3D – 30 m), rational polynomial coefficients (RPC) file, and additional five ground control points (GCPs) for the refinement. As a geometric model, we used the RPC model with zero-order polynomial adjustment39, and orthorectified images were registered to the Universal Traverse Mercator (UTM) Zone 35 N as the reference coordinate system.Georeferencing of scanned cadastral maps and aerial photographsWe used orthorectified WV-3 imageries as a reference for the geometric correction of the historical cadastral maps and the aerial photographs. We selected the spline (triangulation) transformation, a rubber sheeting method, useful for local accuracy and requiring a minimum of 10 control points, as the transformation method to determine the correct map coordinate location for each cell in the historical maps and aerial photographs. The spline transformation provides superior accuracies for the geometric correction of the historical geospatial data40,41.The lack of topographic properties of planimetric features in the historical cadastral maps and the inherent distortions of the aerial photographs due to terrain and camera tilts causes difficulties in precise georeferencing of these data sets. It increases the number of required ground control points (GCPs) for optimal image rectification. Adequate and homogenously distributed GCPs, identified from cadastral maps and aerial photographs, can ensure precise spatial alignment among different geospatial data. The best locations for GCPs were border intersections of agricultural fields and roads. As for artificial objects, places of worship and schools, which are highly probable that have remained unchanged, are other optimal locations for GCPs to perform the accurate geometric correction. Figure 3 displays samples of GCPs selected from cadastral maps and aerial photographs. We obtained 2.11 m or better overall RMSE (Root Mean Square Error) values for the geometric correction of the historical maps and aerial photographs.Figure 3Examples of GCPs selection (red crosses in blue circles) on (a), (c) Cadastral maps and their counterparts on (b), (d) Aerial photographs.Full size imageLULC classification schemeWe defined our classification scheme by analyzing the LULC classes distinguished in all three datasets (i.e., cadastral maps, aerial photographs, and WV-3 imageries). We used the classification scheme shown in Table 1. We also present codes and names of the land cover (LC) classes derived from Corine LC nomenclature42.Table 1 Classification scheme of the study.Full size tableThe legends provided on the historical cadastral maps of Aksu and Kestel delineate 15 LULC categories, which are: (1) buildings, (2) home gardens, (3) roads, (4) arable land, (5) gardens, (6) mulberry groves, (7) chestnut groves, (8) olive groves, (9) vegetable gardens, (10) forest, (11) courtyards, (12) vineyards, (13) arable fields, (14) cemeteries, (15) watercourses. Categorizing the land cover types of cadastral maps is limited with the classes available in the map legend. The legend of cadastral maps categorizes the forested area in one class named “forest”. Therefore, it was not possible to use third-level LC sub-categories in our classification schema for forest area which is separating forested areas into three subclasses (3.1.1, 3.1.2, and 3.1.3) according to the type of tree cover. Although some of the third-level LC sub-categories could be extracted from the cadastral map legend, it was not possible to extract all third level agricultural classes from single-band monochromatic aerial photographs. Although the spatial extent of fruit trees as a permanent crop could be determined from aerial photographs, it was not possible to classify these trees into different fruit types (e.g. 2.2.1 Vineyards, 2.2.2 Fruit trees and berry plantations, 2.2.3 Olive groves). Limitation on the number of forest classes is due to the historical cadastral map content; whereas limitation on the number of agricultural classes is mainly offset by the aerial photographs which have only one spectral band and we did not have any field survey or ancillary geographical data that could help the specific identification of fruit trees.Our primary focus is to find out agricultural land abandonment, deforestation/afforestation, urbanization, and rural depopulation within the historical periods. Therefore, most of the second level LULC classes are sufficient for our purpose. LULC changes within the third class level such as the conversion of third level agriculture classes among each other were not aimed to analyze in this research. Our datasets allow us to use Level 3 CORINE classes for the artificial surfaces. These classes are useful to determine residential area implications of rural depopulation or urbanization, one of the focused transformation types for our analysis.We re-organized and categorized the LULC types used in the cadastral maps, with minimum possible manipulation (only for 2.4 and 3.2 LC classes) according to the classification scheme, as shown in Table 2.Table 2 Correspondence between Corine Land Cover and historical cadastral maps nomenclature.Full size tableLULC mappingAfter aligning all geospatial data, we used the georeferenced cadastral maps, aerial photographs, and satellite images for the LULC mapping. We set the spatial extent of the selected regions based on boundaries digitized from the cadastral maps of 1858. Then we detected historical LULC changes within these extents for all geospatial datasets covering 1900 ha and 830 ha of the Aksu and Kestel regions, respectively. Figures 4 and 5 show the selected extents from the historical maps, aerial photographs, and satellite images of the Kestel and Aksu sites, respectively.Figure 4Geospatial dataset for the Kestel study region. (a) 1858 Cadastral map, (b) 1955 aerial photo, and (c) 2020 WV-3 satellite image (finer details shown in the inset images highlighted by Blue boxes).Full size imageFigure 5Geospatial dataset for the Aksu study region. (a) 1858 Cadastral map, (b) 1955 aerial photo, and (c) 2020 WV-3 satellite image (finer details shown in the inset images highlighted by red boxes).Full size imageDigitization of cadastral maps-1858 LULC mapsWe visually interpreted and manually digitized the geographic features on the historical maps and created vector data for each class. The road features in cadastral maps lack topological properties. They also include spatial errors possibly generated in the process of surveying and map production. Therefore, we cross-checked digitized road segments by visual inspection of the road data of the aerial photographs from 1955. We then further modified road polygons to represent accurate road widths. Afterward, we categorized vectorized features of the cadastral maps into the LULC classes defined in Table 1. Finally, we created the vectorized 1858 LULC map. Figure 6 presents the vectorized 1858 cadastral maps of Aksu and Kestel.Figure 6Vectorized cadastral maps of (a) Kestel and (b) Aksu with Red and green lines showing the vector boundaries.Full size imageObject-based image analysis of aerial photographs-1955 LULC mapsAt the second stage of LULC mapping, we performed the segmentation and classification of the aerial photographs using an object-based approach for generating the 1955 LULC map. The object-based image analysis (OBIA) approach in LULC mapping provides advantages over the traditional per-pixel techniques such as higher classification accuracy, depicting more accurate LULC change, and differentiating extra LULC classes33,43,44. We used the eCognition® software (Trimble Germany GmbH, Munich) to implement an object-based image analysis (OBIA). The OBIA approach contains two phases including the segmentation and classification phases that are performed to locate meaningful objects in an image and categorize the created objects, respectively.Multiple ancillary datasets have been used to support different phases of OBIA. The Open Street Map (OSM) vector data, an open-source geospatial dataset (http://www.openstreetmap.org/), has been utilized as ancillary vector data in OBIA to improve the classification of the remotely sensed images. Sertel et al. (2018) used OSM as a thematic layer for road extraction7. Since there are several limitations in extracting the roads from aerial imagery, the OSM road network data could be useful. A majority of unpaved roads in single-band aerial photographs can easily be misclassified as homogeneous areas of arable lands. Precise detection of the roads from monoband aerial photographs without multi-spectral information is difficult. Therefore, we overlaid the OSM road network data with the aerial photographs to extract the revised aerial road vectors through visual interpretation and manual digitization.We segmented the 1955 aerial photographs with the integration of 1858 LULC map produced from cadastral maps. We implemented the multi-resolution segmentation algorithm. In this segmentation method, a parameter called scale determines the size of resulting objects, and the shape and compactness parameters determine the boundaries of objects. The segmentation process of the aerial photographs was performed at multiple stages with various scale, shape, and compactness parameter values. At the initial stage, we segmented the regions according to the 1858 LULC map and we utilized large-scale parameters. The scale parameter was set to 100 and the shape parameter and the compactness were set as 0.7 and 0.3, respectively. At this stage, we focused on interpreting the objects that have not changed between 1858 and 1955. We classified the segments using the thematic layer attribute (LULC classes defined by the cadastral maps) with the highest coverage. Image objects in which the land surface has changed during 1858–1955 period were detected by visual interpretation and unclassified for further segmentation. We followed this approach to reduce the manual effort. We defined unchanged objects between 1858 and 1955 and assigned the same classes of 1858 LULC map to the objects in 1955 aerial photographs. We then segmented the remaining segments, the last time into smaller objects with the scale parameter set as 25, the shape parameter set as 0.2, and the compactness set as 0.8.We classified the remaining unclassified objects through the development of rulesets. An object can be described by several possible features as explanatory variables which are provided by eCognition. In the classification ruleset, different features and parameters can be defined to describe and extract object classes of interest and thresholds for each feature can be defined by the trial-and-error method. We tested sets of variables for the classification of the monoband aerial photographs. Object features such as the mean value of the monoband, texture after Haralick, distance to neighbor objects, shape features (e.g., rectangular fit and asymmetry), and extent features (e.g., area and length/width) were the most useful alternatives. The classification process of the parcels of the aerial photographs with LULC change started with the classification of roads constructed between 1858 and 1955 by utilizing the aerial road map. The watercourse class was the most difficult to classify since shrubs or trees mostly covered the watercourses. These areas were misclassified as forest or agricultural land. Therefore, experts in historical map reading with local geographical information performed the detection and classification of the water course class and interpreted by the cadastral map (1858) and the google map (2020). After roads and watercourses, we classified forest and agricultural lands using the optimal thresholds for the brightness feature. We calculated the thresholds using the single band of the aerial photograph combined with the area and rectangular fit features. The heterogeneous agricultural areas class principally occupied by agriculture with significant areas of natural grass and trees within the same object are separated from the arable lands using the standard deviation of the digital number (DN) values of the aerial photographs. The texture feature helped classify the permanent crops. The brightness, shape, asymmetry, and distance to road class features were the best-performing ones for classifying the remaining artificial surfaces. The manual interpretation was performed for the classification of sub-classes of artificial surface class, including the continuous/discontinuous urban fabric, industrial, commercial, and transport units, mine, dump and construction sites, and artificial, non-agricultural vegetated areas. Since these land use classes contain one or more land cover and land use categories (e.g., artificial non-agriculture land or industrial or commercial units), finding the optimal threshold and exact feature for distinguishing the subclasses of artificial surfaces is difficult. Especially in the case of using the single-band aerial photographs, manual interpretation was required.Object-based image analysis of satellite images-2020 LULC mapsWe segmented WV-3 satellite images using multi-resolution segmentation algorithm and ancillary geographic data. Similar to the aerial road map, the road network of the study region in 2020, named, WV-3 road map, was extracted by overlaying the OSM road data with the WV-3 satellite image. In the segmentation process of the WV-3 image, we used the vector boundaries of the classified aerial photograph (the 1955 LULC map) and the WV-3 road map as ancillary thematic layers. We opted for the same segmentation and classification approach used for the aerial photographs for the WV-3 image.Firstly, we segmented the satellite image into spectrally homogeneous objects using vector data of the 1955 LULC map by applying large-scale parameters. We implemented scale parameter values of 300, 200, 100, and 50 to find the optimal scale to classify objects that have not changed between 1955 and 2020. The best multi-resolution segmentation configuration was the scale of 100 and the shape and compactness parameters of 0.3 and 0.7, respectively. We classified the segments using the thematic layer attribute (LULC classes defined by the aerial maps) with the highest coverage. Segments with LULC change, e.g. the image objects in which the land surface has changed during 1955–2020 period were detected by visual interpretation and unclassified for further segmentation. As a result, we excluded the objects which were remained unchanged during 1955–2020 by assigning the prepared labels which were allocated in the previous step during the classification of 1955 aerial photographs. We then segmented the remaining objects into smaller objects to identify the changed areas in detail. At this step, the scale, shape, and compactness parameters were set as 25, 0.2, and 0.8, respectively.Except for the additional sets of variables utilized to classify the WV-3 images, we applied the rule-set developed for the classification of the aerial photograph for the classification of the remaining objects of 2020 satellite images. The additional sets of variables include the mean of G, B, R, and NIR and two spectral indices, the Normalized Difference Water Index (NDWI), and the Normalized Difference Vegetation Index (NDVI). NDVI was calculated as the normalized difference of reflectance values in the red and NIR bands; whereas , NDWI was determined as the normalized difference of reflectance values of the green and NIR bands. Through the logical conditions, objects having specified values of NDVI and NDWI can be assigned to vegetation and water classes, respectively. The use of NDVI facilitated the delineation of terrains covered by vegetation and the NDWI improved the extraction of water bodies due to its ability to separate water and non-water objects. We separated different sub-classes of agricultural areas and forests by using optimal thresholds for NDVI which were defined by a trial and error method. Also we utilized assigning the optimal threshold to NDWI to separate water bodies from other land covers. In addition, the mean blue band layer was useful in classifying the artificial surfaces. We assessed the accuracy of each classification using error matrices (overall, user’s and producer’s accuracies, and Kappa statistics)45,46.Estimating LULC changes and LULC conversionsAfter the production of LULC maps of Aksu and Kestel for 1858, 1955, and 2020, the vector data of the LULC maps were used to quantify the LULC conversions for two different periods which are 1858–1955 and 1955–2020. To compare the LULC maps of study areas between two different dates of each study period, we provided detailed “from-to” LULC change information by calculating the LULC change transition matrix computed using overlay functions in ArcGIS.We overlaid LULC maps of 1858 and 1955 and intersected the vector boundaries of the 1858 and 1955 LULC maps to determine the conversion types of LULC classes (from which class to which class). Similarly, to quantify the LULC changes between 1955 and 2020, we overlaid the 1955 and 2020 LULC maps. Then we created transition matrices and performed statistical analysis utilizing the matrices. Finally, we discussed the main LULC change types and the driving factors of the changes in the selected study areas. More

Moritz, M. A. et al. Learning to coexist with wildfire. Nature 515, 58–66 (2014).ADS 
CAS 
PubMed 

Google Scholar 
Bowman, D. M. et al. Vegetation fires in the Anthropocene. Nat. Rev. Earth Environ. 1, 500–515 (2020).ADS 

Google Scholar 
Fischer, A. P. et al. Wildfire risk as a socioecological pathology. Front. Ecol. Environ. 14, 276–284 (2016).
Google Scholar 
Steffensen, V. Fire Country: How Indigenous Fire Management Could Help Save Australia (Hardie Grant Books, 2020).
Google Scholar 
Australian Government, Royal Commission into National Natural Disaster Arrangements. Commonwealth Letters Patent—20 February, 2020. https://naturaldisaster.royalcommission.gov.au/publications/commonwealth-letters-patent-20-february-2020 (2020).Bowman, D. M. The impact of Aboriginal landscape burning on the Australian biota. New Phytol. 140, 385–410 (1998).CAS 
PubMed 

Google Scholar 
Roos, C. I., Williamson, G. J. & Bowman, D. M. Is anthropogenic pyrodiversity invisible in paleofire records?. Fire 2, 42 (2019).
Google Scholar 
Liebmann, M. J. et al. Native American depopulation, reforestation, and fire regimes in the Southwest United States, 1492–1900 CE. Proc. Natl. Acad. Sci. 113, E696–E704 (2016).CAS 
PubMed 
PubMed Central 

Google Scholar 
Fletcher, M.-S., Hamilton, R., Dressler, W. & Palmer, L. Indigenous knowledge and the shackles of wilderness. Proc. Natl. Acad. Sci. 118, e2022218118 (2021).CAS 
PubMed 
PubMed Central 

Google Scholar 
Thomson, D. F. Arnhem land: Explorations among an unknown people part I. The journey to Bennet Bay. Geogr. J. 112, 146–164. https://doi.org/10.2307/1789695 (1948).Article 

Google Scholar 
Yibarbuk, D. et al. Fire ecology and Aboriginal land management in central Arnhem Land, northern Australia: A tradition of ecosystem management. J. Biogeogr. 28, 325–343 (2001).
Google Scholar 
Jones, G. M. & Tingley, M. W. Pyrodiversity and biodiversity: A history, synthesis, and outlook. Divers. Distrib. 28, 386–403 (2021).
Google Scholar 
Steel, Z. L., Collins, B. M., Sapsis, D. B. & Stephens, S. L. Quantifying pyrodiversity and its drivers. Proc. R. Soc. B 288, 20203202 (2021).PubMed 
PubMed Central 

Google Scholar 
Bird, R. B., Tayor, N., Codding, B. F. & Bird, D. W. Niche construction and Dreaming logic: Aboriginal patch mosaic burning and varanid lizards (Varanus gouldii) in Australia. Proc. R. Soc. B: Biol. Sci. 280, 20132297 (2013).
Google Scholar 
Bowman, D. M., Walsh, A. & Prior, L. D. Landscape analysis of Aboriginal fire management in Central Arnhem Land, north Australia. J. Biogeogr. 31, 207–223 (2004).
Google Scholar 
Haynes, C. D. in Proceedings of the Ecological Society of Australia (Australia) (Darwin Institute of Technology).Murphy, B. P. & Bowman, D. M. The interdependence of fire, grass, kangaroos and Australian Aborigines: A case study from central Arnhem Land, northern Australia. J. Biogeogr. 34, 237–250 (2007).
Google Scholar 
Bowman, D., Garde, M. & Saulwick, A. in Histories of Old Ages: Essays in Honour of Rhys Jones (eds Anderson, A. et al.) 61–78 (Australian National University, 2001).Bowman, D. & Panton, W. Decline of Callitris intratropica RT Baker & HG Smith in the Northern Territory: Implications for pre-and post-European colonization fire regimes. J. Biogeogr. 20, 373–381 (1993).
Google Scholar 
Sharp, B. R. & Bowman, D. M. Patterns of long-term woody vegetation change in a sandstone-plateau savanna woodland, Northern Territory, Australia. J. Trop. Ecol. 20, 259–270 (2004).
Google Scholar 
Trauernicht, C., Murphy, B. P., Tangalin, N. & Bowman, D. M. Cultural legacies, fire ecology, and environmental change in the Stone Country of Arnhem Land and Kakadu National Park, Australia. Ecol. Evol. 3, 286–297 (2013).PubMed 
PubMed Central 

Google Scholar 
Edwards, A. C. & Russell-Smith, J. Ecological thresholds and the status of fire-sensitive vegetation in western Arnhem Land, northern Australia: Implications for management. Int. J. Wildland Fire 18, 127–146 (2009).
Google Scholar 
Yates, C. & Russell-Smith, J. Fire regimes and vegetation sensitivity analysis: An example from Bradshaw Station, monsoonal northern Australia. Int. J. Wildland Fire 12, 349–358 (2003).
Google Scholar 
McVicar, D. Reports Concerning Marketable Timbers and Forest Products of Several Regions of the North-West Part of the State (WA Forests Department, 1922).
Google Scholar 
Bowman, D. M., Price, O., Whitehead, P. J. & Walsh, A. The ‘wilderness effect’ and the decline of Callitris intratropica on the Arnhem Land Plateau, northern Australia. Aust. J. Bot. 49, 665–672 (2001).
Google Scholar 
Prior, L. D., McCaw, W. L., Grierson, P. F., Murphy, B. P. & Bowman, D. M. Population structures of the widespread Australian conifer Callitris columellaris are a bio-indicator of continental environmental change. For. Ecol. Manag. 262, 252–262 (2011).
Google Scholar 
Bowman, D. M., MacDermott, H. J., Nichols, S. C. & Murphy, B. P. A grass–fire cycle eliminates an obligate-seeding tree in a tropical savanna. Ecol. Evol. 4, 4185–4194 (2014).PubMed 
PubMed Central 

Google Scholar 
Lawes, M. J., Richards, A., Dathe, J. & Midgley, J. J. Bark thickness determines fire resistance of selected tree species from fire-prone tropical savanna in north Australia. Plant Ecol. 212, 2057–2069 (2011).
Google Scholar 
Bowman, D. M., Haverkamp, C., Rann, K. D. & Prior, L. D. Differential demographic filtering by surface fires: How fuel type and fuel load affect sapling mortality of an obligate seeder savanna tree. J. Ecol. 106, 1010–1022 (2018).
Google Scholar 
Trauernicht, C., Murphy, B. P., Prior, L. D., Lawes, M. J. & Bowman, D. M. Human-imposed, fine-grained patch burning explains the population stability of a fire-sensitive conifer in a frequently burnt northern Australia savanna. Ecosystems 19, 896–909 (2016).
Google Scholar 
Bininj Kunwok Regional Language Centre. https://bininjkunwok.org.au (2021).Cooke, P. M. Buffalo and tin, baki and Jesus. In Culture, Ecology and Economy of Fire Management in North Australian Savannas: Rekindling the Wurrk Tradition (eds Russell-Smith, J. et al.) 69–83 (Csiro Publishing, 2009).
Google Scholar 
Edwards, A. et al. Transforming fire management in northern Australia through successful implementation of savanna burning emissions reductions projects. J. Environ. Manag. 290, 112568 (2021).
Google Scholar 
Clarkson, C. et al. Human occupation of northern Australia by 65,000 years ago. Nature 547, 306–310 (2017).ADS 
CAS 
PubMed 

Google Scholar 
Tobler, R. et al. Aboriginal mitogenomes reveal 50,000 years of regionalism in Australia. Nature 544, 180–184 (2017).ADS 
CAS 
PubMed 

Google Scholar 
Press, T., Lea, D., Webb, A. & Alistair, G. Kakadu Natural and Cultural Heritage and Management (The Australian National University, 1995).
Google Scholar 
Murphy, B. P., Cochrane, M. A. & Russell-Smith, J. Prescribed burning protects endangered tropical heathlands of the Arnhem Plateau, northern Australia. J. Appl. Ecol. 52, 980–991 (2015).
Google Scholar 
Russell-Smith, J., Needham, S. & Brock, J. in Kakadu: Natural and Cultural Heritage and Management (eds A. J. Press et al.) 128–166 (Australian National University, 1995).Haynes, C., Ridpath, M. & Williams, M. A. Monsoonal Australia: Landscape, Ecology and Man in Northern Lowlands (CRC Press, 1991).
Google Scholar 
Bowman, D. M. et al. Biogeography of the Australian monsoon tropics. J. Biogeogr. 37, 201–216 (2010).
Google Scholar 
Williamson, G. J. et al. Measurement of inter-and intra-annual variability of landscape fire activity at a continental scale: The Australian case. Environ. Res. Lett. 11, 035003 (2016).ADS 

Google Scholar 
Russell-Smith, J. et al. Bushfires down under: Patterns and implications of contemporary Australian landscape burning. Int. J. Wildland Fire 16, 361–377. https://doi.org/10.1071/WF07018 (2007).Article 

Google Scholar 
Evans, J. & Russell-Smith, J. Delivering effective savanna fire management for defined biodiversity conservation outcomes: An Arnhem Land case study. Int. J. Wildland Fire 29, 386–400 (2019).
Google Scholar 
Corey, B. et al. Better biodiversity accounting is needed to prevent bioperversity and maximize co-benefits from savanna burning. Conserv. Lett. 13, e12685 (2020).
Google Scholar 
Crisp, M. D. et al. Turnover of southern cypresses in the post-Gondwanan world: Extinction, transoceanic dispersal, adaptation and rediversification. New Phytol. 221, 2308–2319 (2019).PubMed 

Google Scholar 
Prior, L. D. & Bowman, D. M. Classification of post-fire responses of woody plants to include pyrophobic communities. Fire 3, 15 (2020).
Google Scholar 
Brodribb, T. J. et al. Conservative water management in the widespread conifer genus Callitris. AoB Plants 5, plt052 (2013).PubMed Central 

Google Scholar 
Sakaguchi, S. et al. Climate, not Aboriginal landscape burning, controlled the historical demography and distribution of fire-sensitive conifer populations across Australia. Proc. R. Soc. B: Biol. Sci. 280, 20132182 (2013).
Google Scholar 
Allen, K. J. et al. Two climate-sensitive tree-ring chronologies from Arnhem Land, monsoonal Australia. Austral Ecol. 44, 581–596 (2019).
Google Scholar 
Baker, P. J., Palmer, J. G. & D’Arrigo, R. The dendrochronology of Callitris intratropica in northern Australia: Annual ring structure, chronology development and climate correlations. Aust. J. Bot. 56, 311–320 (2008).
Google Scholar 
Hammer, G. Site classification and tree diameter-height-age relationships for cypress pine in the Top End of the Northern Territory. Aust. For. 44, 35–41 (1981).ADS 

Google Scholar 
Prior, L., Bowman, D. & Brook, B. Growth and survival of two north Australian relictual tree species, Allosyncarpia ternata (Myrtaceae) and Callitris intratropica (Cupressaceae). Ecol. Res. 22, 228–236 (2007).
Google Scholar 
Stocker, G. Aspects of the Seeding Habits of Callitris intratropica (Forestry and Timber Bureau, 1966).
Google Scholar 
Bowman, D., Wilson, B. & Davis, G. Response of Callitris intratropica RT Baker & HG Smith to fire protection, Murgenella, northern Australia. Aust. J. Ecol. 13, 147–159 (1988).
Google Scholar 
Hawkins, P. Seed production and litter fall studies of Callitris columellaris. Aust. For. Res. 2, 3–16 (1966).
Google Scholar 
Lawes, M. J., Taplin, P., Bellairs, S. M. & Franklin, D. C. A trade-off in stand size effects in the reproductive biology of a declining tropical conifer Callitris intratropica. Plant Ecol. 214, 169–174 (2013).
Google Scholar 
Petty, A. M. & Bowman, D. M. A satellite analysis of contrasting fire patterns in aboriginal-and euro-Australian lands in tropical North Australia. Fire Ecol. 3, 32–47 (2007).
Google Scholar 
Bowman, D. & Prior, L. Impact of Aboriginal landscape burning on woody vegetation in Eucalyptus tetrodonta savanna in Arnhem Land, northern Australia. J. Biogeogr. 31, 807–817 (2004).
Google Scholar 
Trauernicht, C., Murphy, B. P., Portner, T. E. & Bowman, D. M. Tree cover–fire interactions promote the persistence of a fire-sensitive conifer in a highly flammable savanna. J. Ecol. 100, 958–968 (2012).
Google Scholar 
Russell-Smith, J. Recruitment dynamics of the long-lived obligate seeders Callitris intratropica (Cupressaceae) and Petraeomyrtus punicea (Myrtaceae). Aust. J. Bot. 54, 479–485 (2006).
Google Scholar 
Trauernicht, C., Brook, B. W., Murphy, B. P., Williamson, G. J. & Bowman, D. M. Local and global pyrogeographic evidence that indigenous fire management creates pyrodiversity. Ecol. Evol. 5, 1908–1918 (2015).PubMed 
PubMed Central 

Google Scholar 
D’Antonio, C. M. & Vitousek, P. M. Biological invasions by exotic grasses, the grass/fire cycle, and global change. Annu. Rev. Ecol. Syst. 23, 63–87 (1992).
Google Scholar 
Bowman, D. M., Franklin, D. C., Price, O. F. & Brook, B. W. Land management affects grass biomass in the Eucalyptus tetrodonta savannas of monsoonal Australia. Austral Ecol. 32, 446–452 (2007).
Google Scholar 
Cochrane, M. A. & Bowman, D. M. Manage fire regimes, not fires. Nat. Geosci. 14, 1–3 (2021).
Google Scholar 
Huffman, M. R. The many elements of traditional fire knowledge: Synthesis, classification, and aids to cross-cultural problem solving in fire-dependent systems around the world. Ecol. Soc. 18, 3 (2013).
Google Scholar 
Roos, C. I. et al. Native American fire management at an ancient wildland–urban interface in the Southwest United States. Proc. Natl. Acad. Sci. 118, e2018733118 (2021).CAS 
PubMed 
PubMed Central 

Google Scholar 
Long, J. W., Lake, F. K. & Goode, R. W. The importance of Indigenous cultural burning in forested regions of the Pacific West, USA. For. Ecol. Manag. 500, 119597 (2021).
Google Scholar 
Lake, F. K. et al. Returning fire to the land: Celebrating traditional knowledge and fire. J. For. 115, 343–353 (2017).ADS 

Google Scholar 
Petty, A. M., deKoninck, V. & Orlove, B. Cleaning, protecting, or abating? Making indigenous fire management “work” in northern Australia. J. Ethnobiol. 35, 140–162 (2015).
Google Scholar 
Bird, R. B. & Nimmo, D. Restore the lost ecological functions of people. Nat. Ecol. Evol. 2, 1050–1052 (2018).
Google Scholar 
Bowman, D. M. & Legge, S. Pyrodiversity—Why managing fire in food webs is relevant to restoration ecology. Restor. Ecol. 24, 848–853 (2016).
Google Scholar 
Trisos, C. H., Auerbach, J. & Katti, M. Decoloniality and anti-oppressive practices for a more ethical ecology. Nat. Ecol. Evol. 5, 1–8 (2021).
Google Scholar 
Department of Environment and Science, Q. G. Seasonal Surface Reflectance—Landsat, JRSRP Algorithm, Australia Coverage Dataset Version 1.0.0. https://portal.tern.org.au/seasonal-surface-reflectance-australia-coverage/22021 (2014).Key, C. & Benson, N. Landscape Assessment (LA) Sampling and Analysis Methods (USDA Forest Service, 2006).
Google Scholar 
Edwards, A. C., Russell-Smith, J. & Maier, S. W. A comparison and validation of satellite-derived fire severity mapping techniques in fire prone north Australian savannas: Extreme fires and tree stem mortality. Remote Sens. Environ. 206, 287–299 (2018).ADS 

Google Scholar 
Bowman, D. M., Zhang, Y., Walsh, A. & Williams, R. Experimental comparison of four remote sensing techniques to map tropical savanna fire-scars using Landsat-TM imagery. Int. J. Wildland Fire 12, 341–348 (2003).
Google Scholar 
Hesselbarth, M. H., Sciaini, M., With, K. A., Wiegand, K. & Nowosad, J. landscapemetrics: An open-source R tool to calculate landscape metrics. Ecography 42, 1648–1657 (2019).
Google Scholar 
Koenker, R. quantreg: Quantile Regression. R package version 5.05 (R Foundation for Statistical Computing). http://CRAN.R-project.org/package=quantreg (2013).Stokes, M. A. & Smiley, T. L. Introduction to Tree-Ring Dating (University of Chicago Press, 1968).
Google Scholar 
Geoscience Australia. Surface Geology of Australia 1:1 Million Scale Dataset 2012 Edition. https://data.gov.au/data/dataset/surface-geology-of-australia-1-1-million-scale-dataset-2012-edition. More

Our study reveals zinc isotopes to be a promising trophic indicator in sharks and other fishes in general, similar to previous studies featuring both terrestrial and marine mammals9,10,11,12,13. We analysed the Zn isotope values for extant sharks spanning captive/aquarium and wild individuals from various localities and found close correspondence with their respective trophic level. Further, Zn concentration and isotope composition suggest preservation of this biological signal in fossil specimens with little diagenetic alteration. A survey of fossil shark teeth spanning the Miocene–Pliocene reveal similar δ66Zn values and variation as found in related (e.g., congeneric) extant sharks with similar dentition and ecology. Our δ66Zn results indicate high trophic levels for Otodus and perhaps a trophic change in C. carcharias, the great white shark.Zinc isotopes in extant sharks and teleostsAs with mammals9,10,11,12,13, bioapatite δ66Zn values in wild extant elasmobranchs and teleosts show overall lower values with increasing trophic level (Fig. 1; Supplementary Fig. 1). Both δ66Zn and δ15Ncoll correlate with FishBase17 trophic levels, despite differences in species’ geographic origin and tissue types sampled (Spearman’s correlation r = −0.87, p = 5.65E–16, n = 48 and r = +0.42, p = 1.47E–5, n = 40, respectively). There is no statistically significant relationship between bioapatite δ66Zn values and δ13Ccoll, but there is one between wild fish δ66Zn and δ15Ncoll values from the same tooth or individual (R2 = 0.28, p = 6.89E–4, n = 38; Supplementary Fig. 2). Both proxies thus generally reflect trophic levels. Large apex predatory sharks (e.g., Carcharodon carcharias, Isurus oxyrinchus, and Lamna ditropis) have significantly more negative δ66Zn values than lower trophic level teleosts and the plankton-feeding basking shark (Cetorhinus maximus). In particular, the shortfin mako shark (I. oxyrinchus) and great white shark (Carcharodon carcharias), both apex predators18,19, have much lower δ66Zn values than in any previously recorded extant vertebrate species (enameloid up to –0.71 and –0.63‰, respectively). These low δ66Zn values are likely due to the larger number of trophic levels in the marine ecosystem than in terrestrial food webs (terrestrial mammal enamel lies typically between 0 and +1.6‰9,11) and perhaps differences between marine and terrestrial Zn isotope baselines.Fig. 1: Zinc isotope (δ66Zn) composition of extant elasmobranch and teleost fish teeth and gill raker.Specimens come from off the coast of KwaZulu-Natal (KZN) South Africa, New Jersey (NJ), California (CA), North Carolina (NC), Iceland (IS), Norway (NO), Florida (FL), Cyprus (CY), Massachusetts (MA), Alaska and Israel. Aquarium sharks are from the New York (NY) and Tokyo (TYO) Aquariums and the Eilat (Israel) Underwater Observatory Park. Pisciculture S. aurata individuals are numbered and plotted individually to visualise the homogeneity among control-fed individuals compared to wild elasmobranchs and teleosts. Silhouettes are not to scale. Measurement uncertainty is indicated at the 2 SD level. Samples are colour-coded following their genus, regardless of locality. Source data are provided as a Source Data file.Full size imageAbsolute enameloid δ66Zn values vary by up to 1.26‰ among the extant species analysed from various oceanographic areas (Fig. 1). Our results also demonstrate large variability in enameloid δ66Zn values among extant sharks within the same region; for example, there is a 0.88‰ difference between mean values of Carcharodon carcharias and the bull shark (Carcharhinus leucas; Fig. 1) from KwaZulu-Natal (South Africa, KZN). In contrast, enameloid δ66Zn from the same species (e.g., Carcharodon carcharias, Carcharhinus obscurus) demonstrate a low isotopic variability, independent of geographic location (Fig. 1).We observe uniformity in the enameloid δ66Zn values of five gilt-head bream (Sparus aurata) individuals fed on a controlled fish pellet diet in pisciculture cages located offshore of Central Israel, with values within the measurement uncertainty of each other (–0.01 ± 0.01‰). As with δ66Zn, the δ15Ncoll and δ13Ccoll values are distinct from those of wild teleost individuals caught nearby in Haifa Bay, reflecting the artificial pelleted diet of the pisciculture individuals (Fig. 1, Supplementary Figs. 3, 4). Strongly contrasting the homogenous control fed S. aurata δ66Zn values, we observe a higher δ66Zn variability among (and even within) wild and aquaria elasmobranch individuals (fed with wild-caught fish and cephalopods). For instance, two teeth of a single tiger shark (Galeocerdo cuvier) individual (–0.52 and –0.27‰, Fig. 1) have a variability higher than the total variability among the three KZN G. cuvier individuals. Galeocerdo cuvier is well known for its highly opportunistic prey selection20. Therefore, the δ66Zn value of bioapatite is likely highly responsive to an individual’s diet at the time of tissue formation, and as shark teeth form and replace continuously, enameloid δ66Zn values can vary among teeth of a single individual. Thus, although fish can absorb Zn via their gills, waterborne Zn absorption appears to have a negligible effect on elasmobranch tooth δ66Zn values, in line with Zn incorporated into soft and skeletal tissues in natural environments being predominantly derived from dietary gastrointestinal uptake7,8.Carcharhinus enameloid δ66Zn values are high relative to sharks with similar bulk δ15Ncoll values, which contrary to the here analysed Carcharhinus species more regularly consume pelagic prey offshore, oceanic and on the continental shelf (e.g., Galeocerdo cuvier)21. This discrepancy may relate to Carcharhinus species inhabiting neritic waters where they feed primarily on demersal/benthic, freshwater-brackish-coastal prey22,23,24,25. While the diet of KZN G. cuvier and Carcharodon carcharias can also include reef-associated or demersal prey, pelagic organisms are typically more important by mass, especially in adult individuals20,26. Zinc isotope variability among marine organisms and their tissues is largely unknown, currently limiting our ability to identify specific food items based on shark enameloid δ66Zn values beyond generally observed trophic level effects. Whether higher Carcharhinus enameloid δ66Zn values relate to specific prey species (and trophic level) or general differences in basal organic matter source between a primarily neritic food web compared to a more open marine pelagic food web remains unclear (Supplementary Discussion 1). However, we observe no difference in δ13Ccoll values that imply a more terrestrial carbon signal in the KZN Carcharhinus species relative to sympatric species, arguing against differences in the basal organic matter source amongst the KZN shark species (Supplementary Fig. 3).A previous study on Arctic marine mammal bones suggested a higher geographic independence of δ66Zn values from baseline variability compared to δ13Ccoll and δ15Ncoll values13. Likewise, fish taxa with similar diet composition, habitat use and/or trophic level, have a similar range of bioapatite δ66Zn values regardless of their geographic locality (Fig. 1), indicating that δ66Zn may allow worldwide dietary and trophic level comparability with limited marine baseline variation. Further studies will need to expand our knowledge on δ66Zn variability in extant marine vertebrates as well as the effects of baseline on marine vertebrate enameloid δ66Zn values, especially compared to dentine δ13Ccoll and δ15Ncoll values. Nevertheless, the high taxa-specific and perhaps baseline-independent δ66Zn values suggest δ66Zn is an independent indicator of trophic level and an asset for present and past food web reconstructions in the marine realm.Deep-time zinc isotope preservation in fossil enameloidFossil enameloid has δ66Zn values and Zn concentrations ([Zn]) in the range of extant elasmobranch species, arguing against significant diagenetic modification (Figs. 2, 3, Supplementary Fig. 5). Fossil shark teeth examined herein are from Germany, Malta, Japan, North Carolina (USA) and Florida (USA) covering the Early Miocene (Burdigalian, 20.4–16.0 Ma), Miocene-Pliocene transition (Messinian-Zanclean boundary, ca. 5.3 Ma), and the Early Pliocene (Zanclean, 5.3–3.6 Ma; Fig. 2; Supplementary Note 2). Importantly, extant and fossil elasmobranch enameloid δ66Zn values (–0.71 to +0.28‰ and –0.83 to +0.27‰, respectively) differ from: (1) previously reported values of terrestrial mammal enamel (0 to +1.6‰);9,11 and (2) sedimentary carbonate δ66Zn values of the fossil sites (+0.34 to +0.49‰, Supplementary Fig. 6, Supplementary Table 1). These differences support a preserved biological signal in fossil enameloid.Fig. 2: Zinc isotope (δ66Zn) composition of fossil elasmobranch enameloid.Teeth are from the Early Pliocene of Japan, North Carolina (NC) and Florida (FL), Pliocene to Miocene transition of Florida, the Early Miocene of North Carolina, Germany and Malta. For more details on the sample background, see Supplementary Data 1, Supplementary Note 2. Silhouettes are not to scale. Measurement uncertainty is indicated at the 2 SD level. Samples are colour-coded following their genus, regardless of locality. Source data are provided as a Source Data file.Full size imageFig. 3: Zinc isotope (δ66Zn) composition of fossil and extant elasmobranch enameloid of selected taxa combined from Figs. 1 and 2.Fossil teeth are from multiple locations and ages. Grey silhouettes indicate extant teeth. The boxes for n  > 5 represent the 25th–75th percentiles (with the median as a horizontal line) and the whiskers show the 10th–90th percentiles. Box plots (and n) do not include aquarium teeth (open squares). Otodus spp. includes all O. chubutensis (dark blue) and O. megalodon teeth (light blue) analysed and samples are otherwise colour-coded following their genus. Silhouettes are not to scale. For more details on the samples, see Figs. 1 and 2 and Supplementary Data 1. Source data are provided as a Source Data file. Measurement uncertainty is indicated at the 2 SD level.Full size imageWe observe the same within tooth Zn spatial concentration pattern in extant and Miocene tiger shark (Galeocerdo spp.) teeth, with Zn being more enriched in the outer enameloid than close to the enameloid-dentine junction. If significant diagenetic Zn exchange had occurred throughout the enameloid, this original Zn concentration pattern would not be preserved in the fossil tooth (Supplementary Fig. 7). Additionally, both extant and fossil shark enameloid show the same variation in [Zn] according to their taxonomy, with carcharhiniforms generally having higher [Zn] than lamniforms (Supplementary Fig. 5), again arguing against significant diagenetic enameloid Zn exchange. For the European Miocene sites, δ18OP analyses were also conducted on a subset of teeth, where their enameloid appears to be generally well-preserved as suggested by δ18OP values demonstrating species-specific relative in-vivo temperature ranges as expected compared to the habitat use of equivalent modern species27 (Supplementary Fig. 8).To discern the effects of diagenetic Zn alteration, we compare visually pristine appearing enameloid with areas sampled along fractures and dentine of the same tooth (Supplementary Figs. 6 and 9, Supplementary Table 2). Our results demonstrate that the diagenetic Zn exchange in fractured enameloid leads to higher δ66Zn values than in the pristine enameloid of the same tooth, whereas we observe no differences in enameloid δ66Zn profiles of modern teeth (Supplementary Figs. 9 and 10, Supplementary Table 3). Likewise, the diagenetically more susceptible fossil dentine shows higher δ66Zn values as reflected by a significantly higher and more variable dentine-enamel δ66Zn offset (+0.78 ± 0.33‰, n = 13) than observed for extant teeth (+0.22 ± 0.1‰, n = 23; Supplementary Discussion 2, Supplementary Figs. 6 and 11). For the fossil enameloid shown here, in-vivo δ66Zn values must be at least as low as their current values, indicating limited to no alteration. Consequently, δ66Zn analysis of fossil enameloid can enable deep-time dietary reconstructions.The homogeneity in δ66Zn values for the same species or genera independent of locality and geological age is a remarkable observation (Figs. 2, 3), not only limiting the likelihood of Zn diagenetic alteration but also arguing for minimal variability in habitat-specific food web baselines or a strong homogenisation of δ66Zn values at low trophic levels. There are still some limitations, such as the absence of reported δ66Zn values of marine non-mammalian vertebrates for comparison outside this study, the limited sample size for some species, and uncertainties regarding Zn isotope baseline variability. However, our extensive δ66Zn dataset includes not only multiple species from different localities and periods with distinct differences in dietary Zn uptake among extinct elasmobranch species, but also direct overlap in extant and fossil δ66Zn values of the same genus and/or lifestyle. This spatial and temporal coherence suggests that it may be possible to use the same interpretative framework on extant and fossil elasmobranch assemblages globally (Figs. 1, 2, 3), and our remaining discussion is based on this assumption.Zinc isotopes and ecology of Miocene-Pliocene sharksAbsolute and relative δ66Zn values among some taxonomic groups show no statistical variation with geologic age and locality (e.g., Carcharias spp., Galeocerdo spp.), indicating relatively stable trophic levels and ecological niches throughout time and space. For example, most extinct elasmobranchs with a slender tearing, grasping tooth morphology (e.g., Carcharias) have δ66Zn values that can be directly compared to modern equivalents (e.g., Carcharias taurus, Isurus oxyrinchus, Lamna ditropis). This type of dentition and corresponding tooth morphology are adapted to restrain small, active prey—like fish and cephalopods28,29,30. However, there are differences among the δ66Zn values for these types of elasmobranchs within the Early Miocene of Germany, with Mitsukurina lineata and Pseudocarcharias rigida having higher mean δ66Zn values compared to Araloselachus cuspidatus (Fig. 2). Indeed, post hoc Tukey pairwise comparisons draw out A. cuspidatus as distinct from most species for the Germany (Early Miocene) assemblage, including those with a similar grasping tooth morphology (Supplementary Table 4). Our δ66Zn values indicate that A. cuspidatus was likely a higher trophic level piscivore than M. lineata and P. rigida, supported by the larger tooth size of A. cuspidatus.Zinc isotope values within the Galeocerdo lineage show no statistical variability with age nor locality, suggesting tiger sharks occupied a similar trophic level and ecological role in the marine ecosystem since at least the Early Miocene (Fig. 3). Notably, our results imply that the increase in body size from G. aduncus to the modern G. cuvier did not change its overall trophic level, which is in line with the highly similar tooth morphology between the two species31.For Carcharhinus, the Early Miocene Malta assemblage is drawn out as statistically different from extant wild Carcharhinus spp. (Supplementary Table 5). Still, Carcharhinus spp. in both extant and fossil assemblages always have higher mean δ66Zn values than other sympatric predatory sharks and are drawn out as statistically different from sympatric shark species in each fossil assemblage (Figs. 2, 3, Supplementary Tables 6–8). Based on similarities in tooth morphology and δ66Zn values among extant and extinct Carcharhinus spp., we suggest that extinct taxa also primarily occupied a neritic-coastal habitat feeding upon demersal-benthic prey22,23,24,25. For the Carcharhinus teeth from Malta, this interpretation is supported by lower δ18OP values than sympatric species, indicating a higher water temperature or lower salinity: i.e., a shallow and/or brackish water habitat (Supplementary Fig. 8). Consequently, the uniformly higher δ66Zn values of extant and fossil Carcharhinus spp. indicate the consumption of food items distinct from other measured sympatric species already during the Early Miocene and Early Pliocene.Absolute δ66Zn values for Otodus spp., along with values relative to sympatric species, indicate megatooth sharks were apex predators feeding at a very high trophic level (Figs. 2, 3). In all Early Miocene assemblages, mean O. chubutensis δ66Zn values are among the lowest compared to sympatric species, including the lowest bioapatite δ66Zn value measured to date (–0.83‰). Mean O. chubutensis δ66Zn values are as low as extant Carcharodon carcharias (respectively, –0.57 ± 0.18‰, n = 19 and –0.57 ± 0.05‰, n = 4). Noteworthy, Games-Howell pairwise comparisons indicate the lower extant C. carcharias δ66Zn values as distinct from most fossil Carcharodon populations, possibly indicating a dietary shift in the Carcharodon lineage (Supplementary Table 9). Early Pliocene values from O. megalodon from Japan also demonstrate very low mean δ66Zn values (–0.62 ± 0.11‰, n = 5) that are statistically different from the Atlantic O. megalodon populations sampled from Florida and North Carolina, which have higher mean δ66Zn values (respectively, –0.34 ± 0.11‰, n = 11; –0.38 ± 0.11‰, n = 7; Fig. 2, Supplementary Table 10).Possible explanations for the observed spatial and temporal variability in Otodus and Carcharodon δ66Zn values in our study are differences in prey consumption (and trophic level) or baseline variation. Additionally, we cannot rule out other factors such as interpretive limitations due to sample sizes. For example, extant C. carcharias can exhibit some degree of dietary individuality32, yet we only have δ66Zn data from two individuals (4 teeth) from two localities. Still, the low δ66Zn values in both extant C. carcharias compared to other extant sharks is in line with the generally high trophic level estimates of this species18. Particularly for O. megalodon from Japan where we have only one species analysed, we cannot exclude the possibility of either differences in δ66Zn baseline or regionally different prey species. However, the absence of significant δ66Zn differences within many taxa amongst locations and geological ages implies negligible differences in δ66Zn food web baselines (Figs. 2, 3). Therefore, the observed spatial and temporal variability in δ66Zn values likely demonstrates true dietary differences amongst Otodus and Carcharodon populations both geographically and temporally, with important implications for each species’ feeding ecology and evolution both on a local and global scale.Otodus and Carcharodon in the Early Miocene are represented by O. chubutensis and C. hastalis, respectively, with statistically significant higher mean δ66Zn values from the latter (Figs. 3, 4, Supplementary Tables 11, 12). The mean δ66Zn value for all O. chubutensis is the lowest of all mean values recorded in our fossil shark dataset (Fig. 3), suggesting that O. chubutensis could occupy a higher trophic position than C. hastalis. Importantly, differences between δ66Zn values of O. chubutensis and C. hastalis do not appear related to a different ratio of juveniles to adults in either species, as our results do not record an ontogenetic diet shift (Supplementary Fig. 12). We observe no correlation between the total body length of Otodus spp., Carcharodon spp. and their respective δ66Zn values (Supplementary Fig. 12), likely, because each examined specimen had already surpassed the body size for which ontogenetic dietary shifts, if any, occur.Fig. 4: Results from post-hoc Games-Howell pairwise comparisons of δ66Zn values of enameloid from fossil and extant Otodus spp. and Carcharodon spp.All assemblages and ages are combined for a given species, except for extant C. carcharias. Extant C. carcharias teeth are indicated with grey silhouettes. a Includes all O. megalodon populations, whereas (b) excludes the Japanese (Pacific) population, focusing on Atlantic and Tethys/Paratethys populations only. The boxes for n  > 5 represent the 25th–75th percentiles (with the median as a horizontal line), and the whiskers show the 10th–90th percentiles. Significance level is indicated by “*” (p value < 0.05), “**” (p value < 0.005), “***” (p value < 0.0005) and “****” (p value < 0.00005). Measurement uncertainty is indicated at the 2 SD level. See also Supplementary Tables 11, 12. Otodus chubutensis (dark blue) and O. megalodon teeth (light blue) are coloured separately. All other samples are colour-coded following their genus. Source data are provided as a Source Data file. Silhouettes are not to scale.Full size imageWhen including only Otodus spp. from the Atlantic and Paratethys/Tethys regions, we observe a statistically significant difference between O. chubutensis and O. megalodon (Supplementary Table 12, Fig. 4b). During the Early Pliocene, the Otodus lineage represented by O. megalodon shows a considerable increase in the mean δ66Zn value for the Atlantic populations, hinting at a reduced trophic position for the megatooth shark lineage in the Atlantic. At the same time, the Early Pliocene C. carcharias remains at the same trophic level as C. hastalis (Figs. 2–4, Supplementary Tables 11, 12). Although the extant sample size is limited, our results are intriguing because the mean δ66Zn value for extant C. carcharias places it at a trophic level that would be higher than the Atlantic Early Pliocene O. megalodon (Figs. 3, 4, Supplementary Table 11, 12).Extant Carcharodon carcharias is a predatory shark whereby larger individuals regularly feed on high trophic level marine mammals33. Although Neogene Carcharodon and Otodus were likely opportunistic in their prey selection similar to many extant apex predatory sharks33, fossil evidence of bite marks suggests that both taxa fed largely on marine mammals such as cetaceans (mysticetes and odontocetes) and pinnipeds1,2,4,34,35,36,37,38. However, in the majority of cases, it remains unclear if these feeding events on mammals document active hunting or scavenging and how important each prey taxa were to their overall diet. Early Pliocene C. carcharias and O. megalodon δ66Zn data suggest that lower trophic level mammal prey such as mysticetes (and perhaps herbivorous sirenians) may have been an important food item for both species. Mysticetes are filter-feeders and likely to have higher tissue δ66Zn values than piscivorous odontocetes or pinnipeds, similar to the higher δ66Zn values in the plankton-feeding extant Cetorhinus maximus compared to piscivorous sharks (Fig. 1). Bite marks on Late Miocene–Early Pliocene mysticetes bones from both Carcharodon carcharias and O. megalodon1,4,34,38 corroborate at least occasional feeding events.Now extinct small- and medium-sized mysticetes (e.g., Cetotheriidae and various small-sized Balaenidae and Balaenopteridae) were abundant during the Early Pliocene39,40 and were thus available as prey for large sharks, i.e., Otodus megalodon4 and Carcharodon carcharias1. In contrast, Early Miocene cetacean fossils are dominated by toothed cetaceans, where the Early Miocene European and North American sites sampled in this study lack any mysticete remains41,42,43,44. The Early Miocene Otodus (and modern C. carcharias) lower δ66Zn values (higher trophic level) may partly be related to the lack of lower trophic level mammals (e.g., mysticetes) available as prey. Mysticetes became more abundant following a diversity plateau during the mid-Miocene39,45. Subsequently mysticetes remains become more prominent in the Late Miocene to Early Pliocene fossil assemblages from North Carolina and Florida studied herein43,44, where mysticetes were, together with other mammals (e.g., odontocetes), possibly preyed upon by O. megalodon and C. carcharias.For the Early Pliocene of North Carolina, where we have δ66Zn values for both Otodus megalodon and Carcharodon carcharias, our results suggest largely overlapping trophic levels for both species. Feeding at the same trophic level does not necessarily imply direct dietary competition, as both species could have specialised on different prey with similar trophic levels. However, at least some overlap in food items between both species is likely, as also indicated by fossil bite marks1,4,34,38. Extant predatory sharks typically feed on a wide range of food items33, and there is evidence for generalist feeding, as well as, in some cases, specialisation at lower trophic levels for extant C. carcharias32. Higher dietary individuality and the opportunistic nature of apex predators are possible explanations for the range of δ66Zn values observed in both species (–0.61 to –0.04‰ in Pliocene North Carolina).The extinction of Otodus megalodon could have been caused by multiple, compounding environmental and ecological factors46,47, including climate change and thermal limitations48, the collapse of prey populations4 and resource competition with Carcharodon carcharias15 and possibly other taxa not examined here (e.g., carnivorous odontocetes). The δ66Zn results presented here indicate the potential of trophic change, where we find evidence for a decrease in the mean trophic position from O. chubutensis to O. megalodon in the Atlantic and an increase in trophic position for C. carcharias from the Early Pliocene to its extant form. If these trophic dynamics are accurate, then there is a possibility for the competition of dietary resources between these two shark lineages15. Our results also support the hypothesis of Otodus size-driven co-evolution and co-extinction with mysticetes4, indicated, at least for the Atlantic assemblage, by a shift towards lower trophic level prey from the Early Miocene to the Early Pliocene within the Otodus lineage. In general, our study demonstrates δ66Zn to be a powerful, promising tool to investigate the trophic ecology, diet, evolution, and extinction of fossil marine vertebrates. More

Animal husbandryCarukia barnesi polyps were available in culture from the James Cook University Aquarium, spawned from medusa originally collected near Double Island, North Queensland, Australia (16° 43.5′ S, 145° 41.0′ E) in 2014 and 20158. Populations exponentially increase through asexual reproduction8. Detached buds and swimming polyps were collected from the main culture, and transferred into 6-well tissue culture plates in natural filtered seawater. Plates were maintained in darkness to inhibit algae growth at 27 °C in a constant temperature cabinet. Buds and swimming polyps were left to develop and attach to well bottoms, at which point they were then fed freshly hatched Artemia nauplii and water changed 2–3 times per week. Lids remained attached to tissue culture plates to negate water evaporation and maintain a stable salinity. Polyps were maintained in this way for a minimum of 4 months before experiments began, with all individuals matured with the ability to asexually reproduce further buds. To preserve water quality15 polyps were starved for two days prior to experiment start and were not fed for the duration of the trials. One day prior to the experiment start, all immature buds and polyps were removed from wells, leaving approximately 5–10 mature polyps attached to the substrate for analysis.Preparation of reagentsReagentsSix chemicals were trialed in the current study to induce metamorphosis in C. barnesi polyps. Four indole containing compounds were chosen that have previously been trialed with other cubozoan species: 5-methoxy-2-methyl-3-indoleacetic acid, 5-methoxyindole-2-carboxylic acid, 2-methylindole16 and 5-Methoxy-2-methylindole15,16. Along with the retinoic X receptor 9-cis-retinoic acid and lugols solution.Indole compound treatmentsChemical concentrations of indoles documented in the literature were used to conduct preliminary concentration tests. Fifty mM stock solutions were prepared with 100% ethanol, which was diluted with filtered seawater to the desired experimental concentrations: 50 μM16, 20 μM and 5 μM15. Due to high fatality rates at all of these concentrations when used in this study on C. barnesi, all concentrations were diluted. Fifty mM stock solutions of 5-methoxy-2-methyl-3-indoleacetic acid, 5-methoxyindole-2-carboxylic acid, 2-methylindole and 5-Methoxy-2-methylindole were prepared with 50% ethanol (50% Milli-Q® water) and stored at − 20 °C. Fifty mM stock solutions were diluted with filtered seawater to the experimental concentrations of 5 μM, 1 μM, 0.5 μM, 0.1 μM and 0.05 μM. The carrier solution of 50% ethanol (50% Milli-Q® water) was diluted to the equivalent of the experimental concentrations listed above for use as a control, and incorporated into data as concentration 0. Seventeen ml of solution was added to polyps to fill each well of a 6-well plate.Iodine treatment (lugols solution)Aqueous iodine in the form of Lugols solution (0.37% iodine and 0.74% potassium iodide (sigma product information)) was prepared with equivalent concentrations of moles iodine/iodide: 1.5 μM, 3 μM, 6 μM, 12 μM and 24 μM. Filtered seawater only was used a control for this treatment and incorporated into data as concentration 0. 17 ml of solution was added to polyps to fill each well of a 6-well plate.Retinoid treatmentTo reduce ethanol associated fatality of polyps 0.015% ethanol in Milli-Q® water was used to prepare a 1 mM stock solution of 9-cis-Retinoic acid. The 1 mM stock solution was diluted with filtered seawater to the experimental concentrations of 5 μM, 1 μM, 0.5 μM, 0.1 μM and 0.05 μM. The carrier solution of 0.015% ethanol (Milli-Q® water) was diluted to the equivalent of the experimental concentrations listed above for use as a control, and incorporated into data as concentration 0. 17 ml of solution was added to polyps to fill each well of a 6-well plate.Metamorphosis trialsPrimary trialsExperimental concentrations of reagents were added to C. barnesi polyps growing in the wells of sterile 6-well tissue culture plates. One plate was used per chemical, per concentration, in which five wells functioned as replicates containing the chemical being trialed, whilst the sixth well contained only the control medium. Five concentrations were run for each of six chemicals; 30 plates in total.The filtered seawater the polyps were growing in was exchanged for the experimental chemical on day 0, and was not changed for the duration of the trial. Lids remained attached to tissue culture plates to negate water evaporation and hence salinity changes.Polyps in each well were photographed each day through a dissection microscope over a period of 34 days. Results were then recorded from the photographs, categorised (Fig. 6) as the number of polyps which displayed:Tentacle migration: one of the key signs of metamorphosis in this species, polyp tentacles merge, migrating to form four distinct corners in a square shape8.Detached medusa: a medusa formed and detached from the polyp, recorded regardless of health.Mobile detached medusa: a healthy medusa formed and detached from the polyp, with the ability to swim.Polyp survival: this was then used to calculate the number of polyps which survived the treatment which did not metamorphose.Optimisation trialThe optimal chemical and concentration was then deduced by choosing the combination that produced the largest percentage of healthy detached medusa, in this case 5-methoxy-2-methylindole at 1 μM. A final trial was then run with this to determine if length of chemical exposure could optimize healthy medusa yield. Three replicates of a minimum of five polyps were used per treatment, in which in 1 μM of 5-methoxy-2-methylindole (in seawater) was added to polyps for 24, 48, 72, 96 and 120 h, before the solution was changed to fresh seawater. A sea water only control was also run. The total number of healthy detached medusa were recorded each day.Data analysisAll statistical analysis was conducted in IBM SPSS Statistics Ver28. Graphs were produced in Microsoft Excel 2016 and OriginPro Graphing and Analysis 2021.Primary trialsThe effect of chemical, concentration and time was analysed using a repeated measures three-way ANOVA for four sets of data gathered during the metamorphosis process: percentage of polyps to display tentacle migration, percentage of polyps to have medusa detach, percentage of polyps to have healthy swimming medusa detach, percentage survival of polyps that did not metamorphose. Percentage data was arcsine square root transformed prior to analysis. Mauchly’s Test of Sphericity indicated that the assumption of sphericity had been violated on all four sets of data and therefore, a Greenhouse–Geisser correction was used.Optimisation trialDifferences in the mean percentage of healthy medusa produced at different exposure times was analysed using ANOVA. Differences between means were elucidated using a Post hoc Tukey pairwise comparison test (Tukey HSD alpha 0.05). More

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain
the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in
Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles
and JavaScript. More