Philippa Kaur

Deforestation, in places such as the Amazon, contributes to biodiversity loss.Credit: Ivan Valencia/Bloomberg/Getty

Researchers are relieved that a pivotal summit to finalize a new global agreement to save nature will go ahead this year, after two-years of delays because of the pandemic. But they say the hard work of negotiating an ambitious deal lays ahead.The United Nations Convention on Biological Diversity (CBD) announced yesterday that the meeting will move from Kunming in China to Montreal in Canada. The meeting of representatives from almost 200 member states of the CBD — known as COP15 — will now run from 5 to 17 December. China will continue as president of the COP15 and Huang Runqiu, China’s minister of ecology and environment, will continue as chairman.Conservation and biodiversity scientists were growing increasingly concerned that China’s strict ‘zero COVID’ strategy, which uses measures such as lockdowns to quash all infections, would force the host nation to delay the meeting again. Researchers warned that another setback to the agreement, which aims to halt the alarming rate of species extinctions and protect vulnerable ecosystems, would be disastrous for countries’ abilities to meet ambitious targets to protect biodiversity over the next decade.“We are relieved and thankful that we have a firm date for these critically important biodiversity negotiations within this calendar year,” says Andrew Deutz, an expert in biodiversity law and finance at the Nature Conservancy, a conservation group in Virginia, US. “The global community is already behind in agreeing, let alone implementing, a plan to halt and reverse biodiversity loss by 2030,” he says.With the date now set, Anne Larigauderie, executive secretary of the Intergovernmental Platform on Biodiversity and Ecosystem Services, says the key to success in Montreal will be for the new global biodiversity agreement to focus on the direct and indirect drivers of nature loss, and the behaviors that underpin them. “Policy should be led by science, action adequately resourced and change should be transformative,” she adds.New locationThe decision to move the meeting came about after representatives of the global regions who make up the decision-making body of the COP reached a consensus to shift it to Montreal. China and Canada then thrashed out the details of how the move would work. The CBD has provisions that if a host country is unable to hold a COP, the meeting shifts to the home of the convention’s secretariat, Montreal.Announcing the decision, Elizabeth Mrema, executive secretary of the CBD, said in a statement, “I want to thank the government of China for their flexibility and continued commitment to advancing our path towards an ambitious post 2020 Global Biodiversity Framework.”In a statement, Runqiu said, “China would like to emphasize its continued strong commitment, as COP president, to ensure the success of the second part of COP 15, including the adoption of an effective post 2020 Global Biodiversity Framework, and to promote its delivery throughout its presidency.”China also agreed to pay for ministers from the least developed countries and small Island developing states to travel to Montreal to participate in the meeting.Work aheadPaul Matiku, an environmental scientist and head of Nature Kenya, a conservation organization in Nairobi, Kenya, says the move “is a welcome decision” after “the world lost patience after a series of postponements”.But he says that rich nations need to reach deeper into their pockets to help low- and middle-income countries — which are home to much of the world’s biodiversity — to implement the deal, including meeting targets such as protecting at least 30% of the world’s land and seas and reducing the rate of extinction. Disputes over funding already threaten to stall the agreement. At a meeting in Geneva in March, nations failed to make progress on the new deal because countries including Gabon and Kenya argued that the US$10 billion of funding per year proposed in the draft text of the agreement was insufficient. They called for $100 billion per year in aid.“The extent to which the CBD is implemented will depend on the availability of predictable, adequate financial flows from developed nations to developing country parties,” says Matiku.Talks on the agreement are resuming in Nairobi from 21-26 June, where Deutz hopes countries can find common ground on key issues such as financing before heading to Montreal. Having a firm date set for the COP15 will help push negotiations forward, he says.“Negotiators only start to compromise when they are up against a deadline. Now they have one,” he says. More

Study areaThe NCA encompasses an area of 8292 km2 and in 2020 had approximately 87,000 inhabitants23, who are primarily dependent on livestock for their livelihoods. It is a multiple-use area where people coexist with wildlife and livestock, and practise pastoralism with transhumance, characterised by seasonal movements of livestock for accessing resources such as grazing areas and water. The NCA comprises eleven administrative wards: Alailelai, Endulen, Eyasi, Laitole, Kakesio, Misigiyo, Ngorongoro, Naiyobi, Nainokanoka, Ngoile and Olbalbal (Fig. 1). The NCA was chosen for our study as it is known to be hyperendemic for anthrax4,17,20. In addition, informal consultations we held prior to the study, as well as tailored data collection at the community and household level, indicated that local communities have a good understanding of the disease in humans and animals, and of practices around carcass and livestock management that increase risks, particularly in certain locations and periods of the year24.Figure 1Locations of participatory mapping. Map showing the 11 administrative wards of the Ngorongoro Conservation Area in northern Tanzania and the locations where participatory mapping sessions took place (red dots). The maps were produced in QGIS 2.18.2 using data from the National Bureau of Statistics, Tanzania (http://www.nbs.go.tz/).Full size imageEthics approval and consent to participateThe study received approval from the National Institute for Medical Research, Tanzania, with reference number NIMRJHQ/R.8a/Vol. IX/2660; the Tanzania Commission for Science and Technology (numbers 2016-94-NA-2016-88 (O. R. Aminu), 2016-95-NA-2016-45 (T. L. Forde) and 2018-377-NA-2016-45 (T. Lembo)); Kilimanjaro Christian Medical University College Ethics Review Committee (certificate No. 2050); and the University of Glasgow College of Medical Veterinary & Life Sciences Ethics Committee (application number 200150152). Approval and permission to access communities and participants were also obtained from relevant local authorities. Written informed consent was obtained from all participants involved in the study. All data collected were analysed anonymously, ensuring the confidentiality of participants. All research activities were performed in accordance with relevant guidelines and regulations.Participatory mappingA participatory mapping approach based on methodology previously tested in East Africa25 was employed to define areas of anthrax risk for animals in the NCA based on community knowledge. Georeferenced maps of the NCA were produced using data from Google and DigitalGlobe (2016). The maps used datum Arc 1960/UTM zone 36S and grid intervals of 1000 km and were produced at 1:10,000 and 1:50,000 scales, in order to provide participants with a choice. Ten participatory mapping focus groups were held at ward administrative level (Fig. 1) in order to identify areas in the NCA that communities perceive as posing a high risk of anthrax. One mapping exercise was held in each ward. Ngoile and Olbalbal wards were covered at the same time and treated as one, as they had only recently (in 2015) been split from one ward (Olbalbal). Each session had between ten and thirteen participants, who consisted of village and ward administrators, animal health professionals (including community animal health workers and livestock field officers), community leaders, and selected community members. These participants represented members of the community concerned with animal health and owning livestock and, as such, were likely to hold in-depth knowledge relating to community experience of animal health and disease, including anthrax. Participants were recruited by consulting with animal health professionals as well as village and ward administrators, who gave permission to conduct the mapping sessions.The mapping sessions were conducted in Swahili and translated into English by an interpreter. Participants’ general knowledge of the area was first verified by testing whether they could correctly identify popular locations such as health centres, places of worship, markets and schools. Subsequently, participants discussed among themselves and came to a consensus about areas they considered to be at high risk of anthrax. Specifically, we asked them to identify locations they perceived as areas where they considered their animals to be at risk of being exposed to anthrax. These areas were drawn on the maps provided (Fig. 2). While they did not locate areas where the animals had succumbed to disease, we also asked for generic information on locations where anthrax outbreaks had occurred in the past to define areas that could be targeted for active surveillance of cases. In order to improve the fidelity of the data, participants defined risk areas in relation to their own locality (ward) and locations where their animals access resources. Therefore, the areas were not defined by administrative boundaries, as communities may access locations outside their wards, for instance for grazing or watering. The resulting maps were scanned, digitised and analysed as detailed in the following sections. Further detail on the participatory mapping process is provided in the Supplementary Methods (Additional File 1).Figure 2Participatory mapping of anthrax risk areas in the Ngorongoro Conservation Area. Images show (A) the set-up of a mapping session, (B) participants engaged during a session and (C) an example of a 1:50,000 scale map annotated by participants. The map was created with QGIS opensource mapping software. The basemap used was a scanned and geo-referenced full colour 1:50,000 scale topographic map produced by the Surveys & Mapping Division, Ministry of Lands, Housing & Human Settlements, Dar es Salaam, Tanzania. The grid is based on the Arc1960 UTM 36S projection and datum. The map was exported from QGIS in Acrobat Pdf format to enable it to be printed at suitable sizes for using in the fieldwork and to be manually annotated during the participatory mapping.Full size imageDigitisation of maps and generation of random pointsScanned maps were saved as PDF files and converted to high resolution TIFF files for digitisation in QGIS 2.18.2-Las Palmas free OpenSource software26. All maps were georeferenced with geographical coordinates during production and reference points were available to enable the precise mapping of all locations. The digitization was carried out using the QGIS digitizing tools and by creating polygon layers of the defined risk areas.Sourcing data on the environmental predictors of anthraxAvailable soil and environmental data (250 m grid) for Tanzania were obtained from various sources (Table 1). From the available data, we selected the following seven variables which have previously been shown to contribute to or explain the risk of anthrax based on the biology of B. anthracis (Table 1).Table 1 Environmental factors with potential to influence anthrax occurrence.Full size tableCation exchange capacity (CEC)Measured in cmol/kg, CEC is the total capacity of the soil to retain exchangeable cations such as Ca2+, Mg2+ etc. It is an inherent soil characteristic and is difficult to alter significantly. It influences the soil’s ability to hold on to essential nutrients and provides a buffer against soil acidification27. CEC has been reported to be positively correlated with anthrax risk. In addition, CEC is a proxy for calcium content, which may contribute to anthrax risk in a pH-dependent manner as explained below19,22.Predicted topsoil pH (pH)Soil pH below 6.0 (acidic soil) is thought to inhibit the viability of spores19 thus a positive effect of higher pH on the risk of anthrax is expected. It has been suggested that the exosporium of B. anthracis is negatively charged in soils with neutral to slightly alkaline pH. This negative charge attracts positively charged cations in soil, mainly calcium, enabling the spores to be firmly attached to soil particles and calcium to be maintained within the spore core, thereby promoting the viability of B. anthracis19,28.Distance to inland water bodies (DOWS)Both the distance from water and proximity to water may increase anthrax risk. Distance to inland water may indicate the degree to which an area is dry/arid. Anthrax outbreaks have been shown to occur in areas with very dry conditions19. Although anthrax occurrence has also been associated with high soil moisture, this relates more to the spore germination in the environment (a mechanism that is disputed) and the concentration of spores in moist humus that amount to an infectious dose18,29. Spores will survive much longer in soils with low moisture content19. Low moisture may also be associated with low vegetation which results in animals grazing close to the soil, increasing the risk of ingesting soil with spores. Hampson et al. reported that anthrax outbreaks occurred close to water sources in the Serengeti ecosystem of Tanzania in periods of heavy rainfall20, and Steenkamp et al. found that close proximity to water bodies was key to the transmission of B. anthracis spores in Kruger National Park, South Africa22. Water is an important resource for livestock and a large number of animals may congregate at water sources during dry seasons. The close proximity of a water source to a risk area may increase the chance of infection, particularly during periods of high precipitation which might unearth buried spores.Average enhanced vegetation index (EVI)Vegetation density may influence the likelihood of an animal ingesting soil or inhaling dust that may be contaminated with spores. Grazing animals are more likely to encounter bacteria in soil with low vegetation density20, although there is a possibility that spores can be washed onto higher vegetation by the action of water19. Vegetation index may also reflect the moisture content of soil. Arid/dry conditions favour the formation and resistance of spores in the environment, thus lower vegetation may be associated with the occurrence of anthrax.Average daytime land surface temperature (LSTD)Anthrax has been more commonly reported to occur in regions with warmer climates worldwide. Minett observed that under generally favourable conditions and at 32 °C to 37 °C, sporulation of B. anthracis occurs readily but vegetative cells are more likely to disintegrate at temperatures below 21 °C30. Another hypothesis for the association of high temperature with anthrax occurrence is altered host immune response to disease due to stress caused by elevated temperatures19. In addition, elevated temperatures are usually associated with arid areas where vegetation is low, limiting access to adequate nutrition, which in turn affects immunity. Similarly, in hotter climates where infectious diseases occur more often, host interactions with other pathogens may modulate immune response to anthrax31. In this case, a lower infectious and lethal dose of spores would be sufficient to cause infection and death, respectively19. Contact with and ingestion of soil, spores and abrasive pasture is also higher with low vegetation in hot and arid areas19,32. In boreal regions such as in northern Canada, where anthrax occurs in wood bison, and Siberia, the disease is more commonly reported in the summer19. We therefore hypothesised a positive effect of LSTD on the risk of anthrax.SlopeSpores of B. anthracis are hypothesized to persist more easily in flat landscapes that are characterised by shallow slopes19, as it is thought that wind and water may disperse spores more easily along areas with a higher slope gradient, thereby decreasing the density of spores to levels that may be insufficient to cause infection in a susceptible host. Therefore, we expected a negative relationship between slope and the risk of anthrax.Predicted topsoil organic carbon content (SOC)Organic matter (g/kg) may aid spore persistence by providing mechanical support. The negatively charged exosporium of spores is attracted to the positive charges on hummus-rich soil, thus anthrax is thought to persist in soil rich in organic matter18. Based on available evidence, we expected a positive effect of SOC on the risk of anthrax.Creating the datasetThe annotated and digitised maps yielded polygons of high-risk areas within the NCA (Fig. 3). After digitization, 5000 random points were generated33 to cover the 8292 km2 area of the NCA. This enabled us to obtain distinct points allowed by the 250 m grid resolution of the environmental variables. Points falling within the defined risk areas were selected to represent risk areas while those falling outside represented low-risk areas. Measures of the environmental characteristics associated with individual points were obtained with the ‘add Raster data to points’ feature in QGIS.Figure 3Ngorongoro Conservation Area map showing (A) defined risk areas (in red) and (B) distance to settlements. For analysis, 5000 random points were generated throughout the area; points falling within 4.26 km of human settlements (the average distance herds are moved from settlements in a day as determined through interviews of resident livestock owners) were retained for analysis (n = 2173, shown in blue in 3a). The maps were created in QGIS 2.18.2 using data from the National Bureau of Statistics, Tanzania (http://www.nbs.go.tz/).Full size imageIn order to focus on areas of greatest risk to humans and livestock and to exclude locations that are not accessible, only points within a certain range of distance from settlements were included (Fig. 3). On average, herders in the NCA move their livestock 4.26 km away from settlements for grazing and watering during the day (unpublished data obtained through a cross-sectional survey of 209 households). Thus, only points falling within this distance from settlements were selected, providing us with data on areas where infection is most likely to occur. Data on locations of settlements were obtained from satellite imagery and included permanent residences as well as temporary settlements (e.g. seasonal camps set up after long distance movement away from permanent settlements, typically in the dry season, in search of pasture and water). These data were collated from the Center for International Earth Science Information Network (CIESIN).After adjusting for accessibility of resource locations using the average distance moved by livestock, 2173 points were retained for analysis, of which 239 (11%) fell within high-risk areas.Data analysisAll statistical analyses were carried out in R (v 4.1.0) within the RStudio environment34. The aims of the statistical analysis were to infer the relationship between anthrax risk areas as determined through participatory mapping and the environmental factors identified in Table 1, and to use this relationship to make spatial predictions of anthrax risk across the study area. We achieved both aims by modelling the binary risk status (high or low) of the randomly generated points as a function of their environmental characteristics in a Bayesian spatial logit-binomial generalised linear mixed-effects model (GLMM), implemented in the package glmmfields35. Spatial autocorrelation (residual non-independence between nearby points) was accounted for by including spatial random effects in the GLMM. We chose relatively non-informative priors for the intercept and the covariates, using Student’s t-distributions centred at 0 and wide variances (intercept: df = 3, location = 0, scale = 10; betas: df = 3, location = 0, scale = 3). For the spatial Gaussian Process and the observation process scale parameters, we adopted the default glmmfields settings and used half-t priors (both gp_theta and gp_sigma: df = 3, location = 0, scale = 5), and 12 knots. To achieve convergence, the models were run for 5000 iterations35.First, univariable models were fitted to estimate unadjusted associations between each environmental factor (CEC, pH, DOWS, EVI, LSTD, slope, and SOC; Table 1; Supplementary Table S1) and high- and low-risk areas. Second, we constructed multivariable models by fitting multiple environmental variables (Supplementary Table S2). Three variables, SOC, slope and EVI showed a strongly right-skewed distribution and were therefore log-transformed prior to GLMM analysis to prevent excessive influence of outliers. All predictor variables were centred to zero mean and scaled to unit standard deviation for analysis, and odds ratios were rescaled back to the original units for ease of interpretation. Prior to fitting the multivariable GLMM, the presence of collinearity among the predictor variables—which were all continuous—was assessed using variance inflation factors (VIFs)36, calculated with the car package and illustrated using scatter plots (Supplementary Fig. S1)36. Three predictor variables showed a VIF greater than 3 (LSTD, ln EVI and pH with VIFs of 6.8, 4.2 and 3.5, respectively). Removal of LSTD and ln EVI reduced all VIFs to below 3, therefore these two variables were excluded from the multivariable regression analysis37.The model performance was assessed by calculating the area under the receiver operating characteristic curve. The predicted probability of being an anthrax high-risk area was determined and depicted on a map of the NCA using a regular grid of points generated throughout the NCA with one point sampled every 500 m.Consent for publicationPermission to publish was granted by the National Institute for Medical Research, Tanzania. More

Foote, M. The evolution of morphological diversity. Annu. Rev. Ecol. Syst. 28, 129–152 (1997).Article 

Google Scholar 
Félix, M. A. Phenotypic evolution with and beyond genome evolution. Curr. Top. Dev. Biol. 119, 291–347 (2016).PubMed 
Article 
CAS 

Google Scholar 
Carroll, S. B. Evo-devo and an expanding evolutionary synthesis: A genetic theory of morphological evolution. Cell 134, 25–36 (2008).CAS 
PubMed 
Article 

Google Scholar 
Harvey, P. & Pagel, M. The Comparative Method in Evolutionary Biology. (Oxford University Press, 1991).Huxley, J. S. & Teissier, G. Terminology of relative growth. Nature 137, 780–781 (1936).ADS 
Article 

Google Scholar 
Klingenberg, C. P. Size, shape, and form: Concepts of allometry in geometric morphometrics. Dev. Genes Evol. 226, 113–137 (2016).PubMed 
PubMed Central 
Article 

Google Scholar 
Russell, E. S. Form and Function: A Contribution to the History of Animal Morphology. (John Murray, 1916).Goswami, A. & Polly, P. D. Methods for studying morphological integration and modularity. Paleontol. Soc. Pap. 16, 213–243 (2010).Article 

Google Scholar 
Vidal-García, M., Byrne, P. G., Roberts, J. D. & Keogh, J. S. The role of phylogeny and ecology in shaping morphology in 21 genera and 127 species of Australo-Papuan myobatrachid frogs. J. Evol. Biol. 27, 181–192 (2014).PubMed 
Article 

Google Scholar 
Erwin, D. H. Disparity: Morphological pattern and developmental context. Palaeontology 50, 57–73 (2007).Article 

Google Scholar 
Fišer, C., Robinson, C. T. & Malard, F. Cryptic species as a window into the paradigm shift of the species concept. Mol. Ecol. 27, 613–635 (2018).PubMed 
Article 

Google Scholar 
Wilson, D. E. & Mittermeier, R. A. Handbook of the Mammals of the World: Volume 8: Insectivores. vol. 8 (Lynx Edicions, 2018).Jacquet, F. et al. Phylogeography and evolutionary history of the Crocidura olivieri complex (Mammalia, Soricomorpha): From a forest origin to broad ecological expansion across Africa. BMC Evol. Biol. 15, 71. https://doi.org/10.1186/s12862-015-0344-y (2015).Article 
PubMed 
PubMed Central 

Google Scholar 
Ceríaco, L. M. P. et al. Description of a new endemic species of shrew (Mammalia, Soricomorpha) from PrÍncipe Island (Gulf of Guinea). Mammalia 79, 325–341 (2015).Article 

Google Scholar 
Nicolas, V. et al. Multilocus phylogeny of the Crocidura poensis species complex (Mammalia, Eulipotyphla): Influences of the palaeoclimate on its diversification and evolution. J. Biogeogr. 46, 871–883 (2019).Article 

Google Scholar 
Konečný, A., Hutterer, R., Meheretu, Y. & Bryja, J. Two new species of Crocidura (Mammalia: Soricidae) from Ethiopia and updates on the Ethiopian shrew fauna. J. Vertebr. Biol. 69, 20064.1. https://doi.org/10.25225/jvb.20064 (2020).Article 

Google Scholar 
Couvreur, T. L. P. et al. Tectonics, climate and the diversification of the tropical African terrestrial flora and fauna. Biol. Rev. 96, 16–51 (2021).PubMed 
Article 

Google Scholar 
Mayr, E. & O’Hara, R. J. The biogeographic evidence supporting the Pleistocene forest refuge hypothesis. Evolution 40, 55–67 (1986).PubMed 
Article 

Google Scholar 
Wiens, J. J. & Graham, C. H. Niche conservatism: Integrating evolution, ecology, and conservation biology. Annu. Rev. Ecol. Evol. Syst. 36, 519–539 (2005).Article 

Google Scholar 
Smith, T. B., Wayne, R. K., Girman, D. J. & Bruford, M. W. A role for ecotones in generating rainforest biodiversity. Science 276, 1855–1857 (1997).CAS 
Article 

Google Scholar 
Needham, A. E. & Hardy, A. C. The form-transformation of the abdomen of the female pea-crab, Pinnotheres pisum Leach. Proc. R Soc. Lond. Ser. B Biol. Sci. 137, 115–136 (1950).ADS 
CAS 

Google Scholar 
Hanken, J. & Hall, B. K. The Skull, Volume 3: Functional and Evolutionary Mechanisms. (University of Chicago Press, 1993).Hautier, L., Lebrun, R. & Cox, P. G. Patterns of covariation in the masticatory apparatus of hystricognathous rodents: Implications for evolution and diversification. J. Morphol. 273, 1319–1337 (2012).PubMed 
Article 

Google Scholar 
Aristide, L. et al. Multiple factors behind early diversification of skull morphology in the continental radiation of New World monkeys. Evolution 72, 2697–2711 (2018).PubMed 
Article 

Google Scholar 
Hardin, G. The competitive exclusion principle. Science 131, 1292–1297 (1960).ADS 
CAS 
PubMed 
Article 

Google Scholar 
Denys, C. et al. Shrews (Mammalia, Eulipotyphla) from a biodiversity hotspot, Mount Nimba (West Africa), with a field identification key to species. Zoosystema 43, 729–757 (2021).Article 

Google Scholar 
Estevo, C. A., Nagy-Reis, M. B. & Nichols, J. D. When habitat matters: Habitat preferences can modulate co-occurrence patterns of similar sympatric species. PLoS One 12, e0179489. https://doi.org/10.1371/journal.pone.0179489 (2017).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Spaeth, P. A. Morphological convergence and coexistence in three sympatric North American species of Microtus (Rodentia: Arvicolinae). J. Biogeogr. 36, 350–361 (2009).Article 

Google Scholar 
Adams, D. C., Berns, C. M., Kozak, K. H. & Wiens, J. J. Are rates of species diversification correlated with rates of morphological evolution?. Proc. R. Soc. B Biol. Sci. 276, 2729–2738 (2009).Article 

Google Scholar 
Caumul, R. & Polly, P. D. Phylogenetic and environmental components of morphological variation: Skull, mandible, and molar shape in marmots (marmota, Rodentia). Evolution 59, 2460–2472 (2005).PubMed 
Article 

Google Scholar 
Da Silva, F. O. et al. The ecological origins of snakes as revealed by skull evolution. Nat. Commun. 9, 376. https://doi.org/10.1038/s41467-017-02788-3 (2018).ADS 
CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Hirano, T., Kameda, Y., Kimura, K. & Chiba, S. Substantial incongruence among the morphology, taxonomy, and molecular phylogeny of the land snails Aegista, Landouria, Trishoplita, and Pseudobuliminus (Pulmonata: Bradybaenidae) occurring in East Asia. Mol. Phylogenet. Evol. 70, 171–181 (2014).PubMed 
Article 

Google Scholar 
Ge, D., Yao, L., Xia, L., Zhang, Z. & Yang, Q. Geometric morphometric analysis of skull morphology reveals loss of phylogenetic signal at the generic level in extant lagomorphs (Mammalia: Lagomorpha). Contrib. Zool. 84, 267–284 (2015).Article 

Google Scholar 
Zou, Z. & Zhang, J. Morphological and molecular convergences in mammalian phylogenetics. Nat. Commun. 7, 12758. https://doi.org/10.1038/ncomms12758 (2016).ADS 
CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Ananjeva, N. B. Current state of the problems in the phylogeny of squamate reptiles (Squamata, Reptilia). Biol. Bull. Rev. 9, 119–128 (2019).Article 

Google Scholar 
Revell, L. J., Harmon, L. J. & Collar, D. C. Phylogenetic signal, evolutionary process, and rate. Syst. Biol. 57, 591–601 (2008).PubMed 
Article 

Google Scholar 
Klingenberg, C. P. & Marugán-Lobón, J. Evolutionary covariation in geometric morphometric data: Analyzing integration, modularity, and allometry in a phylogenetic context. Syst. Biol. 62, 591–610 (2013).PubMed 
Article 

Google Scholar 
Cardini, A. & Polly, P. D. Larger mammals have longer faces because of size-related constraints on skull form. Nat. Commun. 4, 2458. https://doi.org/10.1038/ncomms3458 (2013).ADS 
CAS 
Article 
PubMed 

Google Scholar 
Esquerré, D., Sherratt, E. & Keogh, J. S. Evolution of extreme ontogenetic allometric diversity and heterochrony in pythons, a clade of giant and dwarf snakes. Evolution 71, 2829–2844 (2017).PubMed 
Article 

Google Scholar 
Marroig, G. & Cheverud, J. M. Size as a line of least evolutionary resistance: Diet and adaptive morphological radiation in New World monkeys. Evolution 59, 1128–1142 (2005).PubMed 
Article 

Google Scholar 
Cornette, R., Tresset, A., Houssin, C., Pascal, M. & Herrel, A. Does bite force provide a competitive advantage in shrews? The case of the greater white-toothed shrew. Biol. J. Linn. Soc. 114, 795–807 (2015).Article 

Google Scholar 
Rodgers, G. M., Downing, B. & Morrell, L. J. Prey body size mediates the predation risk associated with being “odd”. Behav. Ecol. 26, 242–246 (2015).Article 

Google Scholar 
Damuth, J. Population density and body size in mammals. Nature 290, 699–700 (1981).ADS 
Article 

Google Scholar 
Verschuren, D. Decadal and century-scale climate variability in tropical Africa during the past 2000 years. In Past Climate Variability Through Europe and Africa (eds. Battarbee, R. W., Gasse, F. & Stickley, C. E.) 139–158 (Springer Netherlands, 2004). https://doi.org/10.1007/978-1-4020-2121-3_8.Smith, T. B., Schneider, C. J. & Holder, K. Refugial isolation versus ecological gradients. Genetica 112, 383–398 (2001).PubMed 
Article 

Google Scholar 
Brown, W. L. Jr. & Wilson, E. O. Character displacement. Syst. Biol. 5, 49–64 (1956).
Google Scholar 
Vogel, P. et al. Genetic identity of the critically endangered Wimmer’s shrew Crocidura wimmeri. Biol. J. Linn. Soc. 111, 224–229 (2014).Article 

Google Scholar 
Esselstyn, J. A. et al. Fourteen new, endemic species of shrew (genus Crocidura) from Sulawesi reveal a spectacular island radiation. Bull. Am. Mus. Nat. Hist. 454, 1–108 (2021).Article 

Google Scholar 
Evin, A., Bonhomme, V. & Claude, J. Optimizing digitalization effort in morphometrics. Biol. Methods Protoc. 5, bpaa023. https://doi.org/10.1093/biomethods/bpaa023 (2020).Article 
PubMed 
PubMed Central 

Google Scholar 
Blomberg, S. P., Garland, T. & Ives, A. R. Testing for phylogenetic signal in comparative data: Behavioral traits are more labile. Evolution 57, 717–745 (2003).PubMed 
Article 

Google Scholar 
Adams, D. C. A generalized K statistic for estimating phylogenetic signal from shape and other high-dimensional multivariate data. Syst. Biol. 63, 685–697 (2014).PubMed 
Article 

Google Scholar 
Kembel, S. W. et al. Picante: R tools for integrating phylogenies and ecology. Bioinformatics 26, 1463–1464 (2010).CAS 
PubMed 
Article 

Google Scholar 
Revell, L. J. phytools: Phylogenetic Tools for Comparative Biology (and Other Things). (2021).Fick, S. E. & Hijmans, R. J. WorldClim 2: New 1-km spatial resolution climate surfaces for global land areas. Int. J. Climatol. 37, 4302–4315 (2017).Article 

Google Scholar 
Pebesma, E. Simple features for R: Standardized support for spatial vector data. R J. 10, 439 (2018).Article 

Google Scholar 
Oksanen, J. et al. vegan: Community Ecology Package. (2020).Dray, S., Legendre, P. & Peres-Neto, P. R. Spatial modelling: A comprehensive framework for principal coordinate analysis of neighbour matrices (PCNM). Ecol. Model. 196, 483–493 (2006).Article 

Google Scholar 
Borcard, D., Gillet, F. & Legendre, P. Numerical Ecology with R. (Springer, 2018).Dray, S. et al. adespatial: Multivariate Multiscale Spatial Analysis. (2021).Collyer, M. & Adams, D. RRPP: Linear Model Evaluation with Randomized Residuals in a Permutation Procedure. (2021).Kassambara, A. rstatix: Pipe-Friendly Framework for Basic Statistical Tests. (2021).Borcard, D., Legendre, P. & Drapeau, P. Partialling out the spatial component of ecological variation. Ecology 73, 1045–1055 (1992).Article 

Google Scholar 
Rohlf, F. J. & Corti, M. Use of two-block partial least-squares to study covariation in shape. Syst. Biol. 49, 740–753 (2000).CAS 
PubMed 
Article 

Google Scholar 
Schlager, S., Jefferis, G. & Ian, D. Morpho: Calculations and Visualisations Related to Geometric Morphometrics. (2020). More

Breed, M. F. et al. The potential of genomics for restoring ecosystems and biodiversity. Nat. Rev. Genet. 20, 615–628 (2019).CAS 
PubMed 
Article 

Google Scholar 
Carpenter, S. R., Stanley, E. H. & Vander Zanden, M. J. State of the world’s freshwater ecosystems: Physical, chemical, and biological changes. Annu. Rev. Environ. Resour. 36, 75–99 (2011).Article 

Google Scholar 
Geist, J. Integrative freshwater ecology and biodiversity conservation. Ecol. Indic. 11, 1507–1516 (2011).Article 

Google Scholar 
Jeppesen, E., Søndergaard, M., Meerhoff, M., Lauridsen, T. L. & Jensen, J. P. Shallow lake restoration by nutrient loading reduction–some recent findings and challenges ahead. Hydrobiologia 584, 239–252 (2007).CAS 
Article 

Google Scholar 
Søndergaard, M. & Jeppesen, E. Anthropogenic impacts on lake and stream ecosystems, and approaches to restoration. J. Appl. Ecol. 44, 1089–1094 (2007).Article 

Google Scholar 
Marburg, A. E., Turner, M. G. & Kratz, T. K. Natural and anthropogenic variation in coarse wood among and within lakes. J. Ecol. 94, 558–568 (2006).Article 

Google Scholar 
Schindler, D. W. Recent advances in the understanding and management of eutrophication. Limnol. Oceanogr. 51, 356–363 (2006).ADS 
Article 

Google Scholar 
Lau, S. S. S. & Lane, S. N. Continuity and change in environmental systems: The case of shallow lake ecosystems. Prog. Phys. Geogr. Earth Environ. 25, 178–202 (2001).Article 

Google Scholar 
Brinkhurst, R. O. Distribution and abundance of Tubificid (Oligochaeta) species in Toronto harbour, Lake Ontario. J. Fish. Res. Board Can. 27, 1961–1969 (1970).Article 

Google Scholar 
Wood, L. W. & Chua, K. E. Glucose flux at the sediment-water interface of Toronto Harbour, Lake Ontario, with reference to pollution stress. Can. J. Microbiol. 19, 413–420 (1973).CAS 
PubMed 
Article 

Google Scholar 
Nriagu, J. O., Wong, H. K. T. & Snodgrass, W. J. Historical records of metal pollution in sediments of Toronto and Hamilton harbours. J. Gt. Lakes Res. 9(3), 365–373 (1983).CAS 
Article 

Google Scholar 
Toronto & Region Remedial Action Plan. Metro Toronto and Region Remedial Action Plan (1989).Dahmer, S. C., Matos, L. & Morley, A. Restoring Toronto’s waters: Progress toward delisting the Toronto and Region area of concern. Aquat. Ecosyst. Health Manag. 21, 229–233 (2018).Article 

Google Scholar 
Munawar, M., Norwood, W., McCarthy, L. & Mayfield, C. In situ bioassessment of dredging and disposal activities in a contaminated ecosystem: Toronto Harbour. Hydrobiologia https://doi.org/10.1007/978-94-009-1896-2_62 (1989).Article 

Google Scholar 
Dahmer, S. C., Matos, L. & Jarvie, S. Assessment of the degradation of aesthetics beneficial use impairment in the Toronto and region area of concern. Aquat. Ecosyst. Health Manag. 21, 276–284 (2018).Article 

Google Scholar 
Metro Toronto and Region Remedial Action Plan. Within Reach: 2015 Toronto an Region Remedial Action Plan Progress Report (2016).Burniston, D. & Waltho, J. Report on Sediment Quality in the Toronto Inner Harbour 2007 (2011).Elbrecht, V., Vamos, E. E., Meissner, K., Aroviita, J. & Leese, F. Assessing strengths and weaknesses of DNA metabarcoding-based macroinvertebrate identification for routine stream monitoring. Methods Ecol. Evol. 8, 1265–1275 (2017).Article 

Google Scholar 
Emilson, C. E. et al. DNA metabarcoding and morphological macroinvertebrate metrics reveal the same changes in boreal watersheds across an environmental gradient. Sci. Rep. 7, 12777 (2017).ADS 
PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Aylagas, E., Borja, Á., Muxika, I. & Rodríguez-Ezpeleta, N. Adapting metabarcoding-based benthic biomonitoring into routine marine ecological status assessment networks. Ecol. Indic. 95, 194–202 (2018).Article 

Google Scholar 
Bush, A. et al. Studying ecosystems with DNA metabarcoding: Lessons from biomonitoring of aquatic macroinvertebrates. Front. Ecol. Evol. 7, 434 (2019).Article 

Google Scholar 
Serrana, J. M., Miyake, Y., Gamboa, M. & Watanabe, K. Comparison of DNA metabarcoding and morphological identification for stream macroinvertebrate biodiversity assessment and monitoring. Ecol. Indic. 101, 963–972 (2019).Article 

Google Scholar 
Fernández, S., Rodríguez-Martínez, S., Martínez, J. L., Garcia-Vazquez, E. & Ardura, A. How can eDNA contribute in riverine macroinvertebrate assessment? A metabarcoding approach in the Nalón River (Asturias, Northern Spain). Environ. DNA 1, 385–401 (2019).Article 

Google Scholar 
Hajibabaei, M. et al. Watered-down biodiversity? A comparison of metabarcoding results from DNA extracted from matched water and bulk tissue biomonitoring samples. PLoS ONE 14, e0225409 (2019).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Baird, D. J. & Hajibabaei, M. Biomonitoring 2.0: A new paradigm in ecosystem assessment made possible by next-generation DNA sequencing. Mol. Ecol. 21, 2039–2044 (2012).PubMed 
Article 

Google Scholar 
Hajibabaei, M., Baird, D. J., Fahner, N. A., Beiko, R. & Golding, G. B. A new way to contemplate Darwin’s tangled bank: How DNA barcodes are reconnecting biodiversity science and biomonitoring. Philos. Trans. R. Soc. B. Biol. Sci. 371, 20150330 (2016).Article 
CAS 

Google Scholar 
Beermann, A. J., Zizka, V. M. A., Elbrecht, V., Baranov, V. & Leese, F. DNA metabarcoding reveals the complex and hidden responses of chironomids to multiple stressors. Environ. Sci. Eur. 30, 26 (2018).Article 

Google Scholar 
Bush, A. et al. DNA metabarcoding reveals metacommunity dynamics in a threatened boreal wetland wilderness. Proc. Natl. Acad. Sci. 117, 8539–8545 (2020).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Compson, Z. G. et al. Chapter Two—Linking DNA Metabarcoding and Text Mining to Create Network-Based Biomonitoring Tools: A Case Study on Boreal Wetland Macroinvertebrate Communities. In Advances in Ecological Research Vol. 59 (eds Bohan, D. A. et al.) 33–74 (Academic Press, 2018).
Google Scholar 
Fernandes, K. et al. DNA metabarcoding—A new approach to fauna monitoring in mine site restoration. Restor. Ecol. 26, 1098–1107 (2018).Article 

Google Scholar 
Fernandes, K. et al. Invertebrate DNA metabarcoding reveals changes in communities across mine site restoration chronosequences. Restor. Ecol. 27, 1177–1186 (2019).Article 

Google Scholar 
Poikane, S. et al. Benthic macroinvertebrates in lake ecological assessment: A review of methods, intercalibration and practical recommendations. Sci. Total Environ. 543, 123–134 (2016).ADS 
CAS 
PubMed 
Article 

Google Scholar 
Macher, J.-N. et al. Comparison of environmental DNA and bulk-sample metabarcoding using highly degenerate cytochrome c oxidase I primers. Mol. Ecol. Resour. 18, 1456–1468 (2018).CAS 
PubMed 
Article 

Google Scholar 
Marshall, N. T. & Stepien, C. A. Macroinvertebrate community diversity and habitat quality relationships along a large river from targeted eDNA metabarcode assays. Environ. DNA 2, 572–586 (2020).Article 

Google Scholar 
Metro Toronto and Region Remedial Action Plan. Updates on Actions 2013–2014. (2013).López-López, E. & Sedeño-Díaz, J. E. Biological indicators of water quality: The role of fish and macroinvertebrates as indicators of water quality. In Environmental Indicators (eds Armon, R. H. & Hänninen, O.) 643–661 (Springer Netherlands, 2015). https://doi.org/10.1007/978-94-017-9499-2_37.Chapter 

Google Scholar 
Berry, O. et al. A Comparison of Morphological and DNA Metabarcoding Analysis of Diets in Exploited Marine Fishes (2015).Sweeney, B. W., Battle, J. M., Jackson, J. K. & Dapkey, T. Can DNA barcodes of stream macroinvertebrates improve descriptions of community structure and water quality?. J. N. Am. Benthol. Soc. 30, 195–216 (2011).Article 

Google Scholar 
Banerji, A. et al. Spatial and temporal dynamics of a freshwater eukaryotic plankton community revealed via 18S rRNA gene metabarcoding. Hydrobiologia 818, 71–86 (2018).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Porter, T. M. et al. Widespread occurrence and phylogenetic placement of a soil clone group adds a prominent new branch to the fungal tree of life. Mol. Phylogenet. Evol. 46, 635–644 (2008).CAS 
PubMed 
Article 

Google Scholar 
Rosling, A. et al. Archaeorhizomycetes: Unearthing an ancient class of ubiquitous soil fungi. Science 333, 876–879 (2011).ADS 
CAS 
PubMed 
Article 

Google Scholar 
Mandaville, S. M. Benthic Macroinvertebrates in Freshwaters—Taxa Tolerance Values, Metrics, and Protocols, vol. 128. http://lakes.chebucto.org/H-1/tolerance.pdf (2002).Trzcinski, M. K. et al. The effects of food web structure on ecosystem function exceeds those of precipitation. J. Anim. Ecol. 85, 1147–1160 (2016).PubMed 
Article 

Google Scholar 
Liu, X. & Wang, H. Contrasting patterns and drivers in taxonomic versus functional diversity, and community assembly of aquatic plants in subtropical lakes. Biodivers. Conserv. 27(12), 3103–3118 (2018).Article 

Google Scholar 
Kovalenko, K. E., Brady, V. J., Ciborowski, J. J. H., Ilyushkin, S. & Johnson, L. B. Functional changes in littoral macroinvertebrate communities in response to watershed-level anthropogenic stress. PLoS ONE 9, e101499 (2014).ADS 
PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Luiza-Andrade, A., Montag, L. F. A. & Juen, L. Functional diversity in studies of aquatic macroinvertebrates community. Scientometrics 111, 1643–1656 (2017).Article 

Google Scholar 
MacMillan, G. A., Chételat, J., Heath, J. P., Mickpegak, R. & Amyot, M. Rare earth elements in freshwater, marine, and terrestrial ecosystems in the eastern Canadian Arctic. Environ. Sci. Process. Impacts 19, 1336–1345 (2017).CAS 
PubMed 
Article 

Google Scholar 
Pastorino, P. et al. Macrobenthic invertebrates as tracers of rare earth elements in freshwater watercourses. Sci. Total Environ. 698, 134282 (2020).ADS 
CAS 
PubMed 
Article 

Google Scholar 
Kulaš, A. et al. Ciliates (Alveolata, Ciliophora) as bioindicators of environmental pressure: A karstic river case. Ecol. Indic. 124, 107430 (2021).Article 

Google Scholar 
Persaud, D., Lomas, T., Boyd, D. & Mathai, S. Historical Development and Quality of the Toronto Waterfront Sediments (1985).Milani, D. & Grapentine, L. Assessment of Sediment Quality in the Bay of Quinte Area Of Concern (2000).Reynoldson, T. B., Bailey, R. C., Day, K. E. & Norris, R. H. Biological guidelines for freshwater sediment based on BEnthic Assessment of SedimenT (the BEAST) using a multivariate approach for predicting biological state. Aust. J. Ecol. 20(1), 198–219 (1995).Article 

Google Scholar 
Geller, J., Meyer, C., Parker, M. & Hawk, H. Redesign of PCR primers for mitochondrial cytochrome c oxidase subunit I for marine invertebrates and application in all-taxa biotic surveys. Mol. Ecol. Resour. 13, 851–861 (2013).CAS 
PubMed 
Article 

Google Scholar 
Leray, M. et al. A new versatile primer set targeting a short fragment of the mitochondrial COI region for metabarcoding metazoan diversity: Application for characterizing coral reef fish gut contents. Front. Zool. 10, 34 (2013).PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Zhan, A. et al. High sensitivity of 454 pyrosequencing for detection of rare species in aquatic communities. Methods Ecol. Evol. 4, 558–565 (2013).Article 

Google Scholar 
Gibson, J. et al. Simultaneous assessment of the macrobiome and microbiome in a bulk sample of tropical arthropods through DNA metasystematics. Proc. Natl. Acad. Sci. 111, 8007–8012 (2014).ADS 
CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Gibson, J. F. et al. Large-scale biomonitoring of remote and threatened ecosystems via high-throughput sequencing. PLoS ONE 10, e0138432 (2015).PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Porter, T. M. & Hajibabaei, M. METAWORKS: A flexible, scalable bioinformatic pipeline for multi-marker biodiversity assessments. bioRxiv https://doi.org/10.1101/2020.07.14.202960 (2020).Article 
PubMed 
PubMed Central 

Google Scholar 
Köster, J. & Rahmann, S. Snakemake—a scalable bioinformatics workflow engine. Bioinformatics 28, 2520–2522 (2012).PubMed 
Article 
CAS 

Google Scholar 
Anon. Conda. (2016).Porter, T. M. & Hajibabaei, M. Automated high throughput animal CO1 metabarcode classification. Sci. Rep. 8, 4226 (2018).ADS 
PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Porter, T. M. Eukaryote CO1 Reference set for the RDP Classifier (Zenodo, 2017) https://doi.org/10.5281/zenodo.4741447.Book 

Google Scholar 
Porter, T. M. SILVA 18S Reference Set for the RDP Classifier(Zenodo, 2018) https://doi.org/10.5281/zenodo.4741433.Book 

Google Scholar 
R Core Team. R: A Language and Environment for Statistical Computing (R Core Team, 2020).
Google Scholar 
Wickham, H. ggplot2: Elegant Graphics for Data Analysis (Springer-Verlag, 2009). https://doi.org/10.1007/978-0-387-98141-3.Book 
MATH 

Google Scholar 
Oksanen, J. et al. vegan: Community Ecology Package (2020).Komsta, L. & Novomestky, F. moments: Moments, cumulants, skewness, kurtosis and related tests (2015).U.S. Environmental Protection Agency. Freshwater Biological Traits Database (Final Report) EPA/600/R-11/038F. (2012)U.S. Environmental Protection Agency. Freshwater Biological Traits Database (2012).Schmidt-Kloiber, A. & Hering, D. An online tool that unifies, standardises and codifies more than 20,000 European freshwater organisms and their ecological preferences. Ecol. Indic. 53, 271–282 (2015).Article 

Google Scholar 
Moog, O. Fauna Aquatica Austriaca – Catalogue for autecological Classification of Austrian Aquatic Organisms (1995).Tachet, H., Bournaud, M., Richoux, P., Usseglio-Polatera, P. Invertébrés d’eau douce – systématique, biologie, écologie (2010).Nally, R. M. & Walsh, C. J. Hierarchical partitioning public-domain software. Biodivers. Conserv. https://doi.org/10.1023/B:BIOC.0000009515.11717.0b (2004).Article 

Google Scholar  More

Parra-Lopez, C., Groot, J. C. J., Carmona-Torres, C. & Rossing, W. A. H. Integrating public demands into model-based design for multifunctional agriculture: an application to intensive Dutch dairy landscapes. Ecol. Econ. 67(4), 538–551 (2008).Article 

Google Scholar 
Pinto-Correia, T., Guiomar, N., Guerra, C.A. et al. Assessing the ability of rural (2016)UPA. Desarrollo rural. Oportunidades Desaprovechadas La Tierra 254, 31–33 (2016).
Google Scholar 
Abreu, I., Nunes, J. M. & Mesias, F. J. Can rural development Be measured? Design and application of a synthetic index to Portuguese municipalities. Soc. Indic. Res. 145, 1107–1123 (2019).Article 

Google Scholar 
Doxiadis, C. A. Ekistics: an introduction to the science of human settlements (Oxford University Press, 1968).
Google Scholar 
Algeciras, J. A. R., Coch, H. & Perez, G. D. L. P. Human thermal comfort conditions and urban planning in hot-humid climates-The case of Cuba. Int. J. Biometeorol. 60, 1151–1164 (2016).Article 

Google Scholar 
Zhang, H., Zhang, S. & Liu, Z. Evolution and influencing factors of China’s rural population distribution patterns since 1990. PLoS ONE 15, e0233637 (2020).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Eurostat. Eurostat Regional Yearbook. 2019 Edition. Publications Office of the European Union, Luxembourg (2019)Overview of CAP Reform 2014–2020. Agricultural Policy Perspectives. Brief, no. 5, December 2013. European Commission.Marsden, T. & Sonnino, R. Rural development and the regional state: denying multifunctional agriculture in the UK. J. Rural Stud. 24(4), 422–431 (2008).Article 

Google Scholar 
Adamowicz, M. Normative Aspects of Rural Development Strategy and Policy in The European Union Normative Aspects of Rural Development Strategy and Policy in The European Union, (2018)Léon, Y. Rural development in Europe: a research frontier for agricultural economists. Eur. Rev. Agric. Econ. 32, 301–317 (2005).Article 

Google Scholar 
Biegańska, J., Środa-Murawska, S., Kruzmetra, Z. & Swiaczny, F. Peri-Urban development as a significant rural development trend. Quaest. Geogr. 37, 125–140 (2018).Article 

Google Scholar 
Lee I.-H. Change of rural development policy in South Korea After Korean War. J. Reg City Plan, 2021.Oh, Y.-Y. et al. The selection of proper resource and change of salinity in Helianthus tuberosus L. cultivated in Saemangeum reclaimed tidal land. Korean J. Environ. Agric. 37, 73–78 (2018).Article 

Google Scholar 
Yoon, J.-Y., Jeong, J.-H. & Choi, S.-K. Validation of reference genes for quantifying changes in physiological gene expression in apple tree under cold stress and virus infection. Radiat. Prot. Dosim. 26, 144–158 (2020).
Google Scholar 
Faradiba, F. & Zet, L. The impact of climate factors, disaster, and social community in rural development. J. Asian Finance Econ. Bus. 7, 707–717 (2020).Article 

Google Scholar 
Kaneko, M., Ohta, R., Vingilis, E. & Mathews, M. Thomas Robert freeman; systematic scoping review of factors and measures of rurality: toward the development of a rurality index for health care research in Japan. BMC Health Services Res. 21, 1–11 (2021).Article 

Google Scholar 
Yokoyama, S. Sustainable activities for rural development, New Frontiers in Regional Science: Asian Perspectives, (2019).Georgios, C., & Barraí, H. Social innovation in rural governance: a comparative case study across the marginalised rural EU. J. Rural Stud. (2021)Michalek, J. & Zarnekow, N. Application of the rural development index to analysis of rural regions in Poland and Slovakia. Soc. Indic. Res. 105, 1–37 (2012).Article 

Google Scholar 
Liu, Y., Wang, G. & Zhang, F. Spatio-temporal dynamic patterns of rural area development in eastern coastal China. Chin. Geogr. Sci. 23, 173–181 (2013).Article 

Google Scholar 
Kim, T.-H. & Yang, S.-R. Construction of the rural development index: the case of Vietnam. J. Rural Dev. 39, 113–142 (2016).
Google Scholar 
Houkai, W. Current top ten frontier issues [J]in the study of agriculture. Rural Areas Rural Econ. China 4, 2–6 (2019).
Google Scholar 
Hu, Y., Fu, R. & Jin, S. Ecological environment concern in the organic link between poverty alleviation and rural revitalization. Reform 10, 141–148 (2019).
Google Scholar 
Zhang H., Gao L., & Yan K. On the theoretical origin, main innovation and realization path of rural revitalization thought rural economy in China, (11), 2–16 (2018).Feiwei, S. & Zegan, X. Construction and empirical analysis of the index system of beautiful villages in Zhejiang Province. J. Huazhong Agric. Univ. (Social Sciences Edition) 02, 45-51+132 (2017).
Google Scholar 
Research Group of Shanghai Rural Revitalization Index. Construction and evaluation of the index system of rural revitalization in Shanghai. Sci. Dev. 9, 56–63 (2020).World Bank, Expanding the Measure of Wealth. Indicators of Environmentally Sustainable Development. World Bank, Washington, D.C. (1997)Hicks, D. A. The inequality-adjusted human development index: a constructive proposal. World Dev. 25, 1283–1298 (1997).ADS 
Article 

Google Scholar 
Zhao, R., Shao, C. & He, R. Spatiotemporal evolution of ecosystem health of China’s Provinces based on SDGs. Int. J Environ. Res. Public Health 18, 10569 (2021).PubMed 
PubMed Central 
Article 

Google Scholar 
Kageyama, A., Desenvolvimento rural : conceitos e aplicaçao ao caso brasileiro. UFRGS Editora, Porto Alegre (Brasil) (2008).Abreu, I. Construçao de um índice de desenvolvimento rural e sua aplicaçao ao Alto Alentejo. Instituto Polit´ecnico de Portalegre (2014)Castellani, V. & Sala, S. Sustainable performance index for tourism policy development. Tour. Manag. 31, 871–880 (2009).Article 

Google Scholar 
Hashemi, N. The role of ecotourism in sustainable rural development. J. Rural Dev. Stud. 13(3), 173–188 (2010).
Google Scholar 
Alkire, S., Conconi, A., & Seth, S. Multidimensional Poverty Index 2013: Brief methodological note and results. Oxford Poverty and Human Development Initiative (OPHI) (2014b)Alkire, S., Foster, J. & Santos, M. E. Where did identification go?. J. Econ. Inequal. 9, 501–505 (2011).Article 

Google Scholar 
Alkire, S. et al. Multidimensional poverty measurement and analysis (Oxford University Press, Oxford, 2015).MATH 
Book 

Google Scholar 
Alkire, S., & Seth, S. Identifying destitution through linked subsetsof multidimensionally poor: an ordinal approach, OPHI Working Paper 99. University of Oxford (2016)Li, X., Yang, H., Jia, J., Shen, Y. & Liu, J. Index system of sustainable rural development based on the concept of ecological livability. Environ. Impact Assess. Rev. 86, 1–12 (2021).
Google Scholar 
Guo, X. & Hu, Y. Construction of evaluation index system of rural revitalization level. Agric. Econ. Manag. 05, 5–15 (2020).
Google Scholar 
Kong, X. & Xia, J. The value logic relation and synergy path choice between rural revitalization strategy and rural integration development. J. Northwest Univ. (Philosophy and Social Sciences Edition) 49(02), 10–18 (2019).
Google Scholar 
Wen, T. Three hundred years: the context and development of Chinese rural construction. Open Age. (04) (2016)Ren, C. A Study on the basis, constraints and institutional supply of industrial prosperity. Acad. Acad. 07, 15–27 (2018).
Google Scholar 
Ye, X., Cheng, Y., Zhao, J. & Ning, X. Rural revitalization during the 14th Five-year plan: trend judgment, general thinking and safeguard mechanism. Rural Econ. 09, 1–9 (2020).
Google Scholar 
Zhu, Q. The sociological explanation of the prosperity of rural industry-industry in the context of rural revitalization. J. China Agric. Univ. (Social Sciences Edition) 35(03), 89–95 (2018).
Google Scholar 
Bithas, K. A bioeconomic approach to sustainability with ecological thresholds as an operational indicator. Environ. Sustain. Indic. 6, 100027 (2020).Article 

Google Scholar 
Zheng, X. The East Asian experience of Rural Revitalization and its Enlightenment to China—Take Japan and South Korea as an example. Lanzhou J. 11, 200–208 (2019).ADS 

Google Scholar 
Srivastava, P. K., Kulshreshtha, K., Monhanty, C. S., Pushpangadan, P. & Singh, A. Stakeholder-based SWOT analysis for successful municipal solid waste managementin Lucknow, India. Waste Manag 25(5), 531–537 (2005).CAS 
PubMed 
Article 

Google Scholar 
Alcon, F., Tapsuwan, S., Martínez-Paz, J. M., Brouwer, R. & Miguel, M. Forecasting deficit irrigation adoption using a mixed stakeholder assessment methodology. Technol. Forecast. Soc. Change 83, 183–193. https://doi.org/10.1016/j.techfore.2013.07.003 (2014).Article 

Google Scholar 
Chen, Y., Yang, G., Sweeney, S. & Feng, Y. Household biogas use in rural China: a study of opportunities and constraints. Renew. Sust. Energ. Rev. 14(1), 545–549 (2010).Article 

Google Scholar 
Yeh, C. H. A problem-based selection of multi-attribute decision-making methods. Int. Trans. Oper. Res. 9(2), 169–181 (2002).MATH 
Article 

Google Scholar 
Chakraborty, S., & Yeh, C. -H. A simulation comparison of normalization procedures for TOPSIS, In 2009 International Conference on Computers & Industrial Engineering, IEEE, 2009.Yoon, K. & Hwang, C. L. Multiple attribute decision making. Eur. J. Oper. Res. 4(4), 287–288 (1995).
Google Scholar 
Sharma, N., Khan, Z.A., Siddiquee, A.N., Wahid, M.A. Proc. Inst. Mech. Eng. Part C:J. Mech. Eng. Sci. 233 (2019) 1–10.Maniraj, S. & Thanigaivelan, R. Optimization of electrochemical micromachining process parameters for machining of AMCs with different % compositions of GGBS using Taguchi and TOPSIS methods. Trans. Indian Inst. Met. 72, 3057–3066 (2019).CAS 
Article 

Google Scholar 
Martinez-Fernandez, C. et al. Shrinking cities in Australia, Japan, Europe and the USA: From a global process to local policy responses. Prog. Plan. 105, 1–48 (2016).Article 

Google Scholar 
Mo, G. Green poverty reduction: the value orientation and realization path of ecological poverty alleviation in the battle against poverty—a series of studies on the mechanism of improving the performance of precision poverty alleviation. Modern Econ. Discuss. 11, 10–14 (2016).
Google Scholar 
Wu, L. Echanism and path selection of poverty alleviation in deep-poverty areas. Chin. Soft Sci. 7, 63–70 (2018).
Google Scholar 
Shengzu, Gu. et al. Countermeasure reflections on advancing the poverty relief in the 13th five-year plan. China Financial Res 2, 7–16 (2016).
Google Scholar 
Xie, Z. & Xunwu, J. Giving full play to the advantages of financial financing to help lift out of poverty. China Finance 3, 83–84 (2018).
Google Scholar 
Chen, W. A Way to realize the effective linkage between poverty relief and rural revitalization. Guizhou Soc. Sci. 1, 11–14 (2020).
Google Scholar 
Bijker, R. & Haartsen, T. More than counterurbanisation: migration to popular andless-popular rural areas in the Netherlands. Popul. Space Place 18, 643–657 (2012).Article 

Google Scholar  More

Current scientific definitions of pondsWe compiled existing scientific definitions of ponds by conducting a backwards and forwards search of papers referenced in or subsequently referencing three seminal pond papers8,17,18 (see “Methods”). We ultimately compiled 54 pond definitions from scientific literature (data available19). The variables most often included in definitions were surface area (91% of definitions), depth (48%), permanence (48%), origin (i.e., natural or human-made; 33%), and standing water (33%; Fig. 2a). When surface area or depth were included in definitions, they were often mentioned qualitatively (e.g., “small” and “shallow”). Of the 61% of definitions that included a maximum pond surface area, the range was 0.1 to 100 ha, the median was 2 ha, and all but two definitions were ≤ 10 ha (Fig. 2b). For depth, only 17% of studies provided a maximum depth cutoff, which ranged 2 to 8 m (Fig. 2c). Of the 26 definitions mentioning permanence, 22 stated that ponds could be temporary or permanent and only three indicated that ponds are exclusively permanent waterbodies. Of the 18 definitions mentioning origin, 17 mentioned that ponds could be natural or human-made with the remaining study indicating ponds can have diverse origins.Figure 2Summary of “pond” definitions from scientific literature including (a) presence of various morphological, biological, and physical characteristics in the definition as blue bars (n = 54 definitions total). Bold black lines indicate the number of definitions with surface area and depth values. Histograms of the upper limits from “pond” definitions for (b) surface area and (c) maximum depth.Full size imageOther important factors included in definitions related to morphometry. For example, 30% of definitions mentioned the potential for plants to colonize the entire basin, which relates to high light penetration (mentioned in 11% of definitions) and/or shallow depths. For example, Wetzel11 defines ponds as having enough light penetration that macrophyte photosynthesis can occur over the entire waterbody. As such, these conditions may be comparable to the littoral region of lakes (11% of definitions). Lastly, 7% of pond definitions mentioned mixing versus stratification, whereby ponds mix more than lakes20 yet less than shallow lakes due to a smaller fetch16.To assess if there was agreement in pond definitions among papers, we examined the number of times each definition was cited. Across the 54 definitions, there were 89 citations of 48 unique papers. Ultimately, most papers (75%) were only cited only once, indicating no consensus in pond definition. The most cited paper was Biggs et al.21, which accounted for 15% of citations. The next two most cited papers were Oertli et al.17 and Sondergaard et al.18, which were seminal papers included in our backwards-forwards search, and each comprised 8% of citations.International definitionsAt an international level, there is no consensus on how to discriminate among ponds, lakes, and wetlands. In North America, wetlands are generally considered to be shallow:  More

Harbach, C. J. et al. Delayed senescence in soybean: Terminology, research update, and survey results from growers. Plant Health Progress 17, 76–83 (2016).Article 

Google Scholar 
Hobbs, H. A. et al. Green stem disorder of soybean. Plant Dis. 90, 513–518 (2006).CAS 
PubMed 
Article 

Google Scholar 
Hill, C. B., Hartman, G. L., Esgar, R. & Hobbs, H. A. Field evaluation of green stem disorder in soybean cultivars. Crop Sci. 46, 879–885 (2006).Article 

Google Scholar 
Morita, K. et al. (2006) Effect of green stem on soiled bean index at harvest of soybean by combine harvester. Hokuriku Crop Sci. 41, 107–109 (2006) (in Japanese).
Google Scholar 
Ogiwara, H. Delayed leaf senescence. In: Agriculture, Forestry and Fisheries Research Council of Japan, ed. Soybean-technical development for improving national food self-sufficiency ratio. Annotated bibliography of Agriculture, Forestry, and Fisheries Research, vol. 27, 291–294 (2002). (in Japanese).Crafts-Brandner, S. J. & Egli, D. B. Sink removal and leaf senescence in soybean. Plant Physiol. 85, 662–666 (1987).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Crafts-Brandner, S. J., Below, F. E., Harper, J. E. & Hageman, R. H. Effects of pod removal on metabolism and senescence of nodulating and nonnodulating soybean isolines. Plant Physiol. 75, 311–317 (1984).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Egli, D. B. & Bruening, W. P. Depodding causes green-stem syndrome in soybean. Crop Manag. 5(1), 1–9. https://doi.org/10.1094/CM-2006-0104-01-RS (2006).Article 

Google Scholar 
Htwe, N. M. P. S. et al. Leaf senescence of soybean at reproductive stage is associated with induction of autophagy-related genes, GmATG8c, GmATG8i and GmATG4. Plant Prod. Sci. 14, 141–147 (2011).CAS 
Article 

Google Scholar 
Leopold, A. C., Niedergang-Kamien, E. & Janick, J. Experimental modification of plant senescence. Plant Physiol. 34, 570–573 (1959).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Mondal, M. H., Brun, W. A. & Brenner, M. L. Effects of sink removal on photosynthesis and senescence in leaves of soybean (Glycine max L.) plants. Plant Physiol. 61, 394–397 (1978).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Wittenbach, V. A. Effect of pod removal on leaf senescence in soybean. Plant Physiol. 70, 1544–1548 (1982).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Wittenbach, V. A. Effect of pod removal on leaf photosynthesis and soluble protein composition of field-grown soybeans. Plant Physiol. 73, 121–124 (1983).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Wittenbach, V. A. Purification and characterization of a soybean leaf storage glycoprotein. Plant Physiol. 73, 125–129 (1983).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Staswick, P. E. Developmental regulation and the influence of plant sinks on vegetative storage protein gene expression in soybean leaves. Plant Physiol. 89, 309–315 (1989).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Sato, J., Shiraiwa, T., Sakashita, M., Tsujimoto, Y. & Yoshida, R. The occurrence of delayed stem senescence in relation to trans-zeatin riboside level in the xylem exudate in soybeans grown under excess-wet and drought soil conditions. Plant Prod. Sci. 10, 460–467 (2007).Article 

Google Scholar 
Takehara, T. et al. Occurrence of delayed leaf senescence of soybean caused by Rhizoctonia aerial blight in Japan. Jpn. Agric. Res. Q. 50, 201–208 (2016).Article 

Google Scholar 
Boethel, D. J. et al. Delayed maturity associated with southern green stink bug (Heteroptera: Pentatomidae) injury at various soybean phenological stages. J. Econ. Entomol. 93, 707–712 (2000).CAS 
PubMed 
Article 

Google Scholar 
Islam, M. M. et al. Nitrogen manipulation affects leaf senescence during late seed filling in soybean. Acta Physiol. Plant. 39, 42 (2017).Article 
CAS 

Google Scholar 
Yamazaki, R., Katsube-Tanaka, T. & Shiraiwa, T. Effect of thinning and shade removal on green stem disorder in soybean. Plant Prod. Sci. 21, 83–92 (2018).CAS 
Article 

Google Scholar 
Yamazaki, R., Katsube-Tanaka, T., Kawasaki, Y., Katayama, K. & Shiraiwa, T. Effect of thinning on cultivar differences of green stem disorder in soybean. Plant Prod. Sci. 22, 311–318 (2019).CAS 
Article 

Google Scholar 
Board, J. E. & Tan, Q. Assimilatory capacity effects on soybean yield components and pod number. Crop Sci. 35, 846–851 (1995).Article 

Google Scholar 
Egli, D. B. Soybean reproductive sink size and short-term reductions in photosynthesis during flowering and pod set. Crop Sci. 50, 1971–1977 (2010).Article 

Google Scholar 
Wells, R., Schulze, L. L., Ashley, D. A., Boerma, H. R. & Brown, R. H. Cultivar differences in canopy apparent photosynthesis and their relationship to seed yield in soybean. Crop Sci. 22, 886–890 (1982).Article 

Google Scholar 
Islam, M. M. et al. Nitrogen redistribution and its relationship with the expression of GmATG8c during seed filling in soybean. J. Plant Physiol. 192, 71–74 (2016).CAS 
PubMed 
Article 

Google Scholar 
Zhao, X., Zheng, S. H. & Arima, S. Influence of nitrogen enrichment during reproductive growth stage on leaf nitrogen accumulation and seed yield in soybean. Plant Prod. Sci. 17, 209–217 (2014).CAS 
Article 

Google Scholar 
Brown, A. W. & Hudson, K. A. Transcriptional profiling of mechanically and genetically sink-limited soybeans. Plant Cell Environ. 40, 2307–2318 (2017).CAS 
PubMed 
Article 

Google Scholar 
Tranbarger, T. J., Franceschi, V. R., Hildebrand, D. F. & Grimes, H. D. The soybean 94-kilodalton vegetative storage protein is a lipoxygenase that is localized in paraveinal mesophyll cell vacuoles. Plant Cell 3, 973–987 (1991).CAS 
PubMed 
PubMed Central 

Google Scholar 
Melo, B. P. et al. Revisiting the soybean GmNAC superfamily. Front. Plant Sci. 9, 1864 (2018).PubMed 
PubMed Central 
Article 

Google Scholar 
Kim, H. J. et al. Gene regulatory cascade of senescence-associated NAC transcription factors activated by ETHYLENE-INSENSITIVE2-mediated leaf senescence signaling in Arabidopsis. J. Exp. Bot. 65, 4023–4036 (2014).PubMed 
PubMed Central 
Article 

Google Scholar 
Tucker, M. L., Burke, A., Murphy, C. A., Thai, V. K. & Ehrenfried, M. L. Gene expression profiles for cell wall-modifying proteins associated with soybean cyst nematode infection, petiole abscission, root tips, flowers, apical buds, and leaves. J. Exp. Bot. 58, 3395–3406 (2007).CAS 
PubMed 
Article 

Google Scholar 
Turner, G. W. et al. Experimental sink removal induces stress responses, including shifts in amino acid and phenylpropanoid metabolism, in soybean leaves. Planta 235, 939–954 (2012).CAS 
PubMed 
Article 

Google Scholar 
Roach, T. & Krieger-Liszkay, A. The role of the PsbS protein in the protection of photosystems I and II against high light in Arabidopsis thaliana. Biochim. Biophys. Acta Bioenerg. 1817, 2158–2165 (2012).CAS 
Article 

Google Scholar 
Horton, P., Ruban, A. V. & Walters, R. G. Regulation of light harvesting in green plants. Annu. Rev. Plant Physiol. Plant Mol. Biol. 47, 655–684 (1996).CAS 
PubMed 
Article 

Google Scholar 
Hutin, C. et al. Early light-induced proteins protect Arabidopsis from photooxidative stress. Proc. Natl. Acad. Sci. U.S.A. 100, 4921–4926 (2003).CAS 
PubMed 
PubMed Central 
Article 
ADS 

Google Scholar 
Wang, H. et al. Functional characterization of dihydroflavonol-4-reductase in anthocyanin biosynthesis of purple sweet potato underlies the direct evidence of anthocyanins function against abiotic stresses. PLoS ONE 8, e78484 (2013).CAS 
PubMed 
PubMed Central 
Article 
ADS 

Google Scholar 
Saravitz, D. M. & Siedow, J. N. The differential expression of wound-inducible lipoxygenase genes in soybean leaves. Plant Physiol. 110, 287–299 (1996).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Pimenta, M. R. et al. The stress-induced soybean NAC transcription factor GmNAC81 plays a positive role in developmentally programmed leaf senescence. Plant Cell Physiol. 57, 1098–1114 (2016).CAS 
PubMed 
Article 

Google Scholar 
Fujimoto, M. et al. Transcriptional switch for programmed cell death in pith parenchyma of sorghum stems. Proc. Natl. Acad. Sci. U.S.A. 115, 8783–8792 (2018).Article 
CAS 

Google Scholar 
Egli, D. B. Variation in leaf starch and sink limitations during seed filling in soybean. Crop Sci. 39, 1361–1368 (1999).CAS 
Article 

Google Scholar 
Board, J. E. & Harville, B. G. Late-planted soybean yield response to reproductive source/sink stress. Crop Sci. 38, 763–771 (1998).Article 

Google Scholar 
Fatichin, Zheng, S. H., Narasaki, K. & Arima, S. Genotypic adaptation of soybean to late sowing in southwestern Japan. Plant Prod. Sci. 16, 123–130 (2013).CAS 
Article 

Google Scholar 
Wakasugi, K. & Fujimori, S. Subsurface Water Level Control System “FOEAS” that promotes the full use of paddy fields. J. Jpn. Soc. Irrig. Drain. Rural Eng. 77, 705–708 (2009) (in Japanese).
Google Scholar 
Fehr, W. R. & Caviness, C. E. Stages of soybean development. Spec. Rep. 80. Iowa Agric. Home Econ. Exp. Stn. Iowa State Univ., Ames. (1977).Furuya, T. & Umezaki, T. Simplified distinction method of degree of delayed stem maturation of soybean plants. Jpn. J. Crop Sci. 62, 126–127 (1993) (in Japanese with English abstract).Article 

Google Scholar 
Kim, D. et al. TopHat2: Accurate alignment of transcriptomes in the presence of insertions, deletions and gene fusions. Genome Biol. 14, R36 (2013).PubMed 
PubMed Central 
Article 
CAS 

Google Scholar  More

In this section, we provide quantitative evaluation for both spatial coverage and temporal dynamics of ReaLSAT dataset.Spatial coverageSince the dataset was created using satellite imagery analysis, it can provide more comprehensive coverage than existing datasets. However, using an automated process also has its challenges. It can invariably lead to the detection of spurious waterbodies because of issues in data (e.g., due to errors in GSW maps used as inputs in ReaLSAT).To provide more insights into the types of lakes and potential issues in the spatial coverage of ReaLSAT, we randomly sampled 5,000 lakes out of 435,717 that are only present in ReaLSAT (i.e., not available in the HydroLAKES dataset). A human annotator used Google’s satellite imagery base layer to categorize these lakes. Figure 5a shows the geographical distribution of these lakes, and Fig. 5b shows the distribution of different lake types in the sample set. Out of the 5,000 lakes, the human annotator identified 2,019 traditional lakes and reservoirs where sufficient water was visible in the satellite imagery. Another 551 lakes in the sample set showed signs of a bowl-like depression but with no (or very little) water visible in the satellite imagery and were labeled as ephemeral. There were 861 other lakes that were tagged as farm ponds because they showed geometric patterns of farming in the imagery. This diversity of waterbody types discovered by ReaLSAT that were previously unreported by HydroLakes highlights one of the strengths of our approach. In limnology, the origin/type of lake is a very important regulator of ecosystem dynamics. For instance, reservoirs will have faster water flow/lower residence time than natural lakes, and therefore nutrient and carbon processing rates will differ; floodplain lakes may dry periodically, leading to the denudation of sediments; and farm ponds will likely have much higher rates of nutrient loading and methane production than non-agriculturally influenced lakes. Hence, capturing a more comprehensive range of waterbody categories can enable various scientific studies where knowing the origin/lake type could provide a critical understanding of the process.Fig. 5(a) Geographic location of 5000 randomly selected lakes used for manual evaluation of lake type. (b) Allocation of the 5000 manually referenced lakes to specific lake types. Regular implies a traditional lake or reservoir. Unverifiable implies that the lake type could not be identified based on the available Google Earth imagery.Full size imageAlong with the lentic water types discovered in the sampled set, we also found that ReaLSAT identified 603 river segments missed by our morphological score filter. As stated earlier, this is an inherent challenge with automated approaches that use a fixed score threshold for eliminating river segments. Another 239 lakes were tagged as wetlands because of significant vegetation inside and around the lake polygon. There were also 97 lakes that were adjacent to rivers, which were labeled as riverine or floodplain lakes that were formed as a result of river channels meandering over time. Furthermore, there were 59 lakes where the polygons represented only a small portion of a larger lake and were labeled as partial. Finally, for 571 polygons, there was not enough evidence to tag them in any of the above categories. Since Google imagery represents only a single snapshot in time, these 571 waterbodies could not be definitively labeled as spurious (hence, they were labeled as unverifiable), highlighting a limitation of this evaluation pipeline. In particular, a vast majority of these waterbodies appear to be ephemeral based on their surface area timeseries (completely dry for extended periods of time). Hence, if the satellite imagery layer is from one of these timesteps, the annotator would not be able to confirm the presence of the lake.To assess whether we would obtain a similar distribution of different waterbody categories in existing datasets, we performed a similar evaluation on another 5,000 lakes sampled from ReaLSAT where each polygon has some overlap (greater than 1 pixel) with a polygon from HydroLAKES. In this sampled set, the annotator identified 4,030 lakes as traditional lakes or reservoirs, 370 as ephemeral, 138 as farm ponds, 6 as river segments, 66 as wetlands, 95 as riverine or floodplain lakes, 20 as partial, and 275 as unverifiable.Compared to previous distribution, this set of 5,000 waterbodies contains relatively fewer river segments and wetlands polygons in HydroLAKES, because these categories were manually identified and removed during HydroLAKES database creation6. Similary, this set contains relatively few farm ponds because HydroLAKES was created by manual curation of existing static databases and hence does not contain new farm ponds that got created over the years.Temporal dynamicsTo assess the quality of surface extent maps, we performed a quantitative evaluation on a random selection of extent maps. These extent maps were compared against reference maps created by a human annotator using a semi-automated pixel classification procedure. This strategy of creating reference maps is used extensively in the remote sensing literature (e.g. see36,37,38,39). Next, we describe our evaluation process in detail.Sample selectionThere are 462,574 lakes out of 681,137 total lakes where the label updates (corrections and imputations) by the ORBIT approach have trust scores within our chosen thresholds (as described in the methods section). To evaluate these candidate lakes effectively, we focus on lake extent maps where the ORBIT approach resulted in a different map than the underlying GSW extent based map. Hence, we remove maps where no updates were made by the ORBIT approach (neither corrections nor imputations) from the candidate pool of extent maps used for evaluation. We also remove maps where the percentage of missing labels was more than 90% because these maps tend to suffer from significant cloud coverage. Hence, it would be challenging to generate reference maps. Since the GSW dataset has a significant amount of missing data for most places in the world before 2000, we evaluated maps only from 2000 onwards. These three filters left us with a total of 51,077,278 water extent maps considered for selection. Figure 6a shows the distribution of percentage pixels updated made by the ORBIT approach in these water extent maps. To evaluate the robustness of our approach in comparison to GSW maps, we randomly selected 10,000 water extent maps such that extents with significant updates are given higher weight to reduce the skew in distribution towards extents with relative less updates (Fig. 6b).Fig. 6Distribution of updates made by the ORBIT approach. (a) distribution using candidate water extents (b) distribution using randomly selected 10000 water extent maps for evaluation.Full size imageSample pruningFrom these randomly selected water extent maps, we removed maps for which a reference map could not be generated due to clouds or the inability of the annotator to distinguish between land and water. A final set of 2,095 water extent maps were considered for evaluation. Figure 7a shows the distribution of percentage updates in the final set of evaluation extents and Fig. 7b shows the geographical distribution of these extent maps.Fig. 7Summary of the dataset used for evaluating water extent maps. (a) Distribution of updates made by the ORBIT approach in the water extent maps selected for evaluation. (b) Geographical location of the lakes in the evaluation set.Full size imageReference map generationFor these water extent maps, we created ground truth reference maps using a semi-automatic labeling process37,38,39. Specifically, the annotator selects land and water samples to train an SVM (Support Vector Machine) classification model for each image. The annotator keeps adding samples until a stable map is generated. As a final step, the annotator masks out pixels affected by clouds, cloud shadows, and any other region where the annotator is not confident about the accuracy of the reference labels. This process enables a quick and robust generation of reference maps. Supplementary Fig. S7 shows one of the reference maps in the evaluation set. While this strategy of comparing maps is different from the traditional approach of comparing pixels (often selected using stratified sampling), it provides a much more exhaustive evaluation of surface extent maps. The reference maps used for evaluation in this study are also available as part of the dataset.ComparisonTo compare the extent maps generated by ReaLSAT with the reference maps, we used accuracy as the evaluation metric, a widely used metric to measure the quality of classification maps. Accuracy is simply defined as the ratio of pixels with correct labels over a total number of pixels. Specifically, we assign 1 to water pixels and 0 to land pixels. Since GSW based extent maps contain missing labels, they are assigned a value of 0.5 to reflect the uncertainty between land and water. Accuracy is then calculated as follows:$$Accuracy=1-frac{1}{Rast C}mathop{sum }limits_{i=1}^{R}mathop{sum }limits_{j=1}^{C}left|ReferenceMap[i,j]-PredictedMap[i,j]right|$$
(2)
where, R is the number of rows and C is the number of columns of the map.When the accuracy of RealSAT and GSW labels are compared, a vast majority of points lie above the diagonal 1:1 line, which implies that ReaLSAT labels were more accurate overall (Fig. 8a). In Fig. 8 the points are colored based on % of pixels where GSW labels were missing. To better show the improvement in RealSAT labeling, we plot the distribution of the difference in accuracy values between the two datasets as shown in Fig. 8b. A positive value indicates that the surface extent map from the ReaLSAT dataset had better accuracy than the map from the GSW dataset and vice versa. For ease of visualization, we plot this distribution after excluding cases where the accuracy from both datasets was equal. The positively skewed distribution demonstrates the efficacy of the ORBIT approach.Fig. 8Comparison of accuracy values using GSW labels vs ReaLSAT labels. (a) Scatter plot of accuracy values using GSW labels vs ReaLSAT labels. (b) Histogram of difference in accuracy between ReaLSAT labels vs GSW labels. Positive value represents cases where ReaLSAT labels were more accurate than GSW labels. (c) Histogram of difference in accuracy values for the scenario where pixels labelled as land by both products as well as ground truth were removed to reduce the skew of surrounding land pixel on the accuracy values.Full size imageNote that the shape of a lake will influence the number of land pixels surrounding it, which might bias the accuracy values. For example, the reference map shown in Supplementary Fig. S7 contains more than 70% of land pixels. To address this bias, we also calculated accuracy values after removing pixels that were labeled as land by both datasets as well as the ground truth. This variation allows a more strict evaluation of water extent maps. Figure 8c shows the distribution of the difference in accuracy values under this scenario (after excluding cases with equal accuracy). As shown, a vast majority of the distribution is still towards positive values. Furthermore, the distribution has a larger spread towards high positive values, suggesting significant improvement made by the ORBIT approach.From Fig. 8, we can see that for some cases ReaLSAT based extent maps are less accurate relative to GSW. As described earlier, violation of assumptions made by the ORBIT approach could lead to the observed poor performance. Out of 2,095 extent maps, GSW labels show better accuracy than ReaLSAT for 323 of them. On visual analysis of errors in these maps, we found that 165 maps are slightly different only at the lake’s boundary. We categorized the remaining extent maps based on the reason behind the observed poor performance. In particular, 45 maps have poor performance due to occlusion of water surface by algae, 18 maps contain farm ponds, 8 contain mining lakes, 27 maps have unreliable bathymetry, 30 maps have issues due to the weighting factor used by ORBIT approach, and 30 maps have class conditional missing data. All the reference maps and corresponding maps from GSW and ReaLSAT are provided with the dataset.Next, we describe some of these cases in detail.Impact of algae: It can be difficult to visually differentiate surface algae or floating aquatic plants from terrestrial vegetation40, as they have similar reflectance spectra. Therefore, surface algal blooms often get incorrectly labeled as land in the reference maps. However, in most cases, the appearance and disappearance of algae on a lake are independent of the bathymetry. Thus, algae pixels get detected as physically inconsistent by the ORBIT approach, and consequently, these pixels are updated based on the labels of other pixels without algae. In many cases, while the accuracy with respect to the reference map is poor (because algae get labeled as land), ReaLSAT based extent maps are closer to the true extent of the lake. For example, Supplementary Fig. S8 illustrates the impact of algae on the extent mapping of Center Lake, Texas. In this example, the bimodal distribution of fraction values (either low or high) reveals high confidence in lake persistence (Supplementary Fig. S8b). On Oct 22, 2008, false-color composite processing of LANDSAT-5 imagery reveals a strong vegetative signal on the west side of the lake (Supplementary Fig. S8c). Since we know that this is a lake, we can assume that the west side of the lake is experiencing a large surface algal bloom with a similar reflectance to the surrounding terrestrial landscape. Because of the strong vegetative reflectance signal, the semi-automated reference mapping labels the west side of the lake as land (Supplementary Fig. S8d), as does most GSW labels (Supplementary Fig. S8e). Conversely, the ReaLSAT extent map labels the west side of the lake as water (Supplementary Fig. S8f). However, we calculate accuracy based on the semi-automated reference map (Supplementary Fig. S8d). Due to this, the GSW extent map is considered more accurate than the ReaLSAT map, even though this is not true because the reference map is incorrectly labeled. Therefore, some negative accuracy values may be a misrepresentation of reality due to surface algal blooms.Impact of variable bathymetry: Even though we tried to remove lakes with unreliable bathymetry by using score-based filters defined in an earlier section, not all cases were removed. For example, agricultural ponds often have small sections that are connected and change shape based on agricultural needs. Supplementary Fig. S9 highlights an example of labeling issues on agricultural ponds in Mexico. In this area, satellite imagery and the GSW fraction map confirm the presence of agricultural ponds (Supplementary Fig. S9a,b). These individual ponds are filled and drained based on operational decisions and do not follow a consistent pattern of growing or shrinking. Thus, the ORBIT approach can introduce spurious updates in water extent maps for these farms. In the Landsat-5 imagery from 2009–10–08, some of the ponds are dry, while others are filled (Supplementary Fig. S9c). This distribution of water is evident from a visual inspection and is confirmed in the semi-automated reference map (Supplementary Fig. S9d). Due to the similar elevations between the individual pond sections, the ORBIT approach spuriously fills the remaining sections with water based on the incorrectly learned bathymetry (Supplementary Fig. S9f). While quantification of such uncertainties is outside the scope of this paper, we hope that the wider research community can use RealSAT to address such questions. In particular, changes in bathymetry of a lake can be identified using spatial-temporal patterns in the label corrections. Specifically, if the elevation of some pixels in a lake increases after a certain time (e.g., sediment deposits leading to increase the elevation of a pixel), they will appear as physically inconsistent to the ORBIT framework, and hence the labels for these locations will be changed from land to water much more frequently after this increase in elevation.Impact of bias in errors and missing data: As mentioned earlier in the methods section, based on our observation, the confidence of water labels is higher than land labels in the GSW dataset. To account for this bias, we used a weighting factor of 3 for the water class. While this weighting factor improves the ORBIT approach’s performance in most cases, this assumption leads to an overestimation of water for some lakes. For example, Supplementary Fig. S10 compares the water extent maps with and without the weighting factor for a small reservoir in eastern Brazil. As we can see, the GSW labels contain false positives, and due to the weighting factor of 3, ORBIT prefers to update the land labels to water which further increases the number of false positives, as shown in Supplementary Fig. S10e. However, if we use a weighting factor of 1 for this example, the ORBIT approach can effectively remove many of the false positives in the GSW map, as shown in Supplementary Fig. S10f.Similarly, apart from missing data due to clouds in the GSW dataset, there can also be missing values on pixels where the GSW classification model is not confident. Hence, for some water extent maps, class-dependent missing data (compared to missing data which is class independent) adversely impact the ORBIT approach. For example, Supplementary Fig. S11 shows a water extent map for Zhongleng Reservoir in China, where missing data along the eastern edges is not independent but has resulted from ambiguous pixels around the lake where the GSW’s approach was not confident. In such a scenario, the ORBIT approach heavily relies on information from nearby timesteps to infer labels for missing pixels, leading to errors in ReaLSAT maps if there is a significant variation in lake extent in nearby timesteps, as shown in Supplementary Fig. S11e. More