Ecology

DataSpatial gridWe created a grid whose units measure 250 m by 250 m based on the census tract layer for the city of Rio de Janeiro from the Instituto Brasileiro de Geografia e Estatística [Brazilian Institute of Geography and Statistics] (IBGE) website https://www.ibge.gov.br/geociencias/organizacao-do-territorio/malhas-territoriais. Uninhabited locations were excluded.Dengue cases on the gridDengue is a disease of compulsory notification in Brazil, and cases are notified at the Sistema de Informação de Agravos de Notificação [Information System on Diseases of Compulsory Declaration] (SINAN). Dengue cases notified in Rio de Janeiro between January 2010 and March 2015 were geocoded according to address of residency, and then counted for each grid unit by the Secretariat of Health of the city. We obtained the monthly dengue cases data aggregated at the grid level.Population on the gridThe population data is obtained from the Census 2010 (IBGE) (https://www.ibge.gov.br/estatisticas/downloads-estatisticas.html) and it is available at the census tract level. The census tract areas vary in size and can be bigger than the unit of the grid, primarily in the least densely populated zones of the city. To overcome this issue, we cropped from the census tract layer the areas classified as non-urbanized (such as water bodies, swamps, agricultural areas, green areas, beaches, rocky outcrops) in 2010 by the City Hall of Rio de Janeiro (layer available at http://www.data.rio/datasets/uso-do-solo-2010). The population of each census tract is distributed randomly (uniformly) in the areas obtained after deleting the non-urban areas. The population within the units is computed by adding the grid layer. To create the grid and edit the census tract layer we used QGIS (version 3.6.3)45, and to obtain the population in the grid we used the R software46 with the packages tidyverse47 and sf48. We verify the accuracy of our estimated population by comparison with the WordPop dataset49 (see detailed description and Supplementary Fig. 12 and Supplementary Note 2). We chose the WorldPop dataset because: (i) the estimates are also calculated based on census data and are available for 2010, (ii) the pixel size is 100 m, smaller than the size of our grid unit, and (iii) it is open access.Since the units are in fact small and most of them conserve their area of 250 m by 250 m (Supplementary Fig. 1A), we consider population density as the population of each unit. For consistency, we do not consider units with small effective areas and/or populations sizes less than, or equal to, 10 in our analysis. In total, 8954/20212 units were so excluded. This choice circumvents the problem of high sensitivity to random population distribution, and urban vs. non-urban classification, in very small and/or sparsely populated areas. It also facilitates model simulation and does not affect the peak ratio pattern (Supplementary Fig. 1B).Peak ratio and spatial aggregationSince units are small, we binned them into G groups and aggregated their times series of reported cases. The groups were generated according to two aspects: (1) the geographical location of the units as determined by the administrative divisions of the city (10 areas, 33 regions, and 160 neighborhoods); and (2) the population of the units based on quantiles in order to obtain equal size groups. We considered specifically four different partition levels, resulting in 12, 25, 50, and 100 groups with about 900, 450, 225, and 100 units, respectively (from a total number of 11,247 units for the whole city). Groups of unequal size can introduce different statistical effects (it is not the same, for example, to calculate a mean value using 1000 or 10 elements). To compare quantities across groups it is therefore prudent to define groups with the same number of elements. In particular, this consideration becomes important for a large number of groups. Since the population density distribution (number of individuals per unit) is not uniform, groups defined with “equidistant” boundaries would exhibit very different numbers of elements.Given a unit u, we define its time series ({{{{{{bf{v}}}}}}}_{{{{{{bf{u}}}}}}}={{c}_{u}({t}_{1}),{c}_{u}({t}_{2}),…,{c}_{u}({t}_{f})}), where ({c}_{u}({t}_{i})) is the number of reported cases of dengue at time ({t}_{i}) (i = 1, 2, …f) (and the bold symbol is used to indicate a vector). Thus, the aggregated time series is given by$${{{{{{bf{V}}}}}}}_{{{{{{bf{g}}}}}}}=mathop{sum}limits_{uin g}{{{{{{bf{v}}}}}}}_{{{{{{bf{u}}}}}}}={{C}_{g}({t}_{1})=mathop{sum}limits_{uin g}{c}_{u}({t}_{1}),{C}_{g}({t}_{2})=mathop{sum}limits_{uin g}{c}_{u}({t}_{2}),…,{C}_{g}({t}_{f})=mathop{sum}limits_{uin g}{c}_{u}({t}_{f})},$$with (g=1,2,…,G). Then, for each ({{{{{{bf{V}}}}}}}_{{{{{{bf{g}}}}}}}) we computed the ratio between the sizes of the second and first DENV4 peaks, that is$${{{{{rm{peakrati}}}}}}{{{{{{rm{o}}}}}}}_{{{{{{rm{g}}}}}}}=frac{{ma}{x}_{tin {season}2}{{C}_{g}({t}_{1}),{C}_{g}({t}_{2}),…,{C}_{g}({t}_{f})}}{{ma}{x}_{tin {season}1}{{C}_{g}({t}_{1}),{C}_{g}({t}_{2}),…,{C}_{g}({t}_{f})}}$$
(1)
(Supplementary Fig. 2).The deterministic SIR modelAlthough dengue is a vector-borne disease, for simplicity we omitted the explicit representation of the dynamics of the mosquito population, and treated vector transmission via the seasonality of the transmission rate26. Thus, for each unit u, the deterministic SIR model is based on the following traditional differential equations:$$frac{d{S}_{u}}{{dt}}=mu {N}_{u}-beta {S}_{u}frac{{I}_{u}}{{N}_{u}}-mu {S}_{u}$$$$frac{d{I}_{u}}{{dt}}=beta {S}_{u}frac{{I}_{u}}{{N}_{u}}-gamma {I}_{u}-mu {I}_{u}$$
(2)
$$frac{d{R}_{u}}{{dt}}={gamma I}_{u}-mu {R}_{u},$$where ({S}_{u},{I}_{u},{R}_{u}), are, respectively, the number of susceptible, infected, and recovered individuals, and ({N}_{u}) the number of inhabitants, of the spatial unit u. Parameter (mu) is the mortality rate (equal to the birth rate), and (gamma) is the recovery rate. The seasonal transmission rate is specified as (beta (t)={beta }_{0}(1+delta {{sin }},(omega t+phi ))). The units are considered independent of each other, and the initial conditions establish that the whole population of each unit is susceptible to the virus (({S}_{u}(t=0)={N}_{u}) and ({I}_{u}left(t=0right)={R}_{u}left(t=0right)=0forall u)). Transmission begins with one infected individual at a time ({t}_{0u}ge t=0) where ({t}_{0u}) is obtained from the data.Since the goal of this model is to examine the representative dynamics of different population densities, we binned the units according to their population into 12 groups, and computed the mean value of their number of inhabitants ({N}_{g}=langle {N}_{uin g}rangle) and of their arrival times of the infection ({t}_{0g}sim langle {t}_{0uin g}rangle) (where g = 1, …, 12). We then simulated the system considering the 12 sets ({{N}_{g},{t}_{0g}}) as given.The stochastic modelSince units will suffer local extinction of transmission, a major component of a stochastic implementation is the description of the local reintroduction of the virus, namely the arrival of a ‘spark’ or imported infection, in analogy to fire spread. Because space is described by a highly-resolved lattice, we considered that well-mixed transmission applies within each unit. Moreover, in lieu of  explicit spatial coupling between units, we postulated  the importation of infection through the specification of a spark rate.For this purpose, we constructed a binary representation of the time series of cases per month by defining the spatial units either as positive or negative according to whether they reported cases or not (Supplementary Fig. 3). Then, to derive a spark rate we explored the dynamics of the number of positive units as follows,$${U}^{{{mbox{+}}}}(t+{dt})={U}^{{{mbox{+}}}}(t)+{U}_{{{{{{{mathrm{new}}}}}}}}^{{{mbox{+}}}}(t,t+{dt})-{U}_{{{{{{{mathrm{extinct}}}}}}}}^{{{mbox{+}}}}(t,t+{dt})$$
(3)
The number of positive units at time ({t+dt}) is equal to the number of positive units at time t, plus the number of units that have been infected ({{U}_{{{{{{{mathrm{new}}}}}}}}}^{{{mbox{+}}}}(t,t+{dt})) between t and t + dt, minus the number of units that were infected at t but are no longer infected at t + dt (i.e., the number of ‘extinctions’ between t and t + dt, ({{U}_{{{{{{{mathrm{extinct}}}}}}}}}^{{{mbox{+}}}}(t,t+{dt}))).Since uninfected units (i.e., negative units) require the arrival of a spark to become positive, the following equation specifies the mean of ({{U}_{{{{{{{mathrm{new}}}}}}}}}^{{{mbox{+}}}}(t,t+{dt})) under the assumption that a small unit is unlikely to receive more than a single spark in a period of time dt$${{langle }}{U}_{{{{{{{mathrm{new}}}}}}}}^{{+}}(t,t+{dt}){{rangle }}simeq {N}_{{{{{{{mathrm{sparks}}}}}}}}(t,t+{dt})frac{{U}^{{-}}(t)}{U},$$
(4)
where ({N}_{{{{{{{mathrm{sparks}}}}}}}}(t,t+{dt})) is the number of sparks produced between t and t + dt, ({U}^{{{{-}}}}(t)) is the number of negative units at a time t, and (U) is the total number of units in the city ((U={U}^{{{mbox{+}}}}+{U}^{{{{-}}}})).By introducing Eq. (4) into Eq. (3) we obtain,$${U}^{{{mbox{+}}}}(t+{dt})simeq {U}^{{{mbox{+}}}}(t)+{N}_{{{{{{{mathrm{sparks}}}}}}}}(t,t+{dt})frac{{U}^{{{{-}}}}(t)}{U}-{{U}_{{{{{{{mathrm{extinct}}}}}}}}}^{{{mbox{+}}}}(t,t+{dt})$$
(5)
From Eq. (5) we can now compute the spark rate per unit ({{sigma }_{u}}^{{emp}}(t,t+{dt})) from the high-resolution incidence data as$${{sigma }_{u}}^{{emp}}(t,t+{dt})=frac{{N}_{{{{{{{mathrm{sparks}}}}}}}}(t,t+{dt})}{U}simeq frac{{U}^{{{mbox{+}}}}(t+{dt})-{U}^{{{mbox{+}}}}(t)+{U}_{{{{{{{mathrm{extinct}}}}}}}}(t,t+{dt})}{{U}^{{{{-}}}}(t)}$$
(6)
In order to address the effects of human density on the spark rate, we binned the spatial units according to their population into G groups. To avoid statistical effects due to group size, we considered population quantiles. Then, by applying Eq. (6) to each of these groups, we obtained an empirical spark rate per unit that depends on human density,$${sigma }_{uin g}^{{emp}}(t,t+{dt})={sigma }_{u}^{{emp}}(t,t+{dt}{{{{{rm{;}}}}}}{N}_{g}),$$
(7)
where ({N}_{g}={{langle }}{N}_{uin g}{{rangle }}) with g = 1, 2, …, G.SimulationsThe associated differential equations of the stochastic model are those shown on Eq. (2) but the transmission component has now an additional term ({sigma }_{u}) to describe the importation of infections.$$frac{d{S}_{u}}{{dt}}=mu {N}_{u}-left(beta {S}_{u}frac{{I}_{u}}{{N}_{u}}+{sigma }_{u}right)-mu {S}_{u}$$$$frac{d{I}_{u}}{{dt}}=left(beta {S}_{u}frac{{I}_{u}}{{N}_{u}}+{sigma }_{u}right)-gamma {I}_{u}-mu {I}_{u}$$
(8)
$$frac{d{R}_{u}}{{dt}}={gamma I}_{u}-mu {R}_{u}$$Since the inferred spark rate from the data (Eq. (7)) is obtained from observed infections, we computed the spark rate ({sigma }_{u}) as:$${sigma }_{uin g}={{{{{{mathrm{Poisson}}}}}}}({{sigma }_{uin g}}^{{emp}}/rho )$$
(9)
where (rho) is the reporting rate.The model shown on Eq. (8) was formulated as stochastic by incorporating demographic noise (with the different events represented as Poisson processes). It was implemented in R with the package pomp50. We also considered measurement error by assuming that the observed number of cases ({{C}_{u}}^{{obs}}) during a period of time T is,$${{C}_{u}}^{{obs}}left(Tright)={{{{{{mathrm{binomial}}}}}}}left(rho ,{C}_{u}left(Tright)right),$$
(10)
where ({C}_{u}(T)) is the number of cases computed in the unit u. We simulated the 11,247 units that compose the city of Rio de Janeiro, and aggregated the resulting time series as for the empirical data (see Peak ratio section).The parameters of the model are given in Supplementary Table 1. We relied on parameters estimated for dengue transmission in Rio de Janeiro by ref. 26. Those estimates were obtained for the aggregated city and for the emergence of DENV1. We use these parameters here as a sufficiently realistic set for illustrating and exploring the behavior of the stochastic model with population density. Moreover, with the exception of the spark rate, the model parameters were considered the same for all units. In particular, we applied a uniform reporting rate because access to the nearest public healthcare clinic does not show a dependency on population density (see Supplementary Note 1).Reporting summaryFurther information on research design is available in the Nature Research Reporting Summary linked to this article. More

PlantProbs.net. Nickel in plants and soil https://plantprobs.net/plant/nutrientImbalances/sodium.html (accessed Apr 28, 2021).Guodong Liu, E. H. Simonne, and Y. L. Nickel Nutrition in Plants | EDIS. EDis 2011.Liu, G. D. “A New Essential Mineral Element–Nickel.” Plants Nutr. Fertil. Sci. 2001.Kabata-Pendias, A.; Mukherjee, A. Trace Elements from Soil to Human; 2007.Kasprzak, K. S. Nickel advances in modern environmental toxicology. Environ. Toxicol. 11, 145–183 (1987).CAS 

Google Scholar 
Cempel, M. & Nikel, G. Nickel: A review of its sources and environmental toxicology. Polish J. Environ. Stud. 15, 375–382 (2006).CAS 

Google Scholar 
Bradl, H. B. Chapter Sources and origins of heavy metals. Interface Sci. Technol. 6, 1–27 (2005).CAS 
Article 

Google Scholar 
Von Burg, R. Nickel and some nickel compounds. J. Appl. Toxicol. 17, 425–431 (1997).Article 

Google Scholar 
Freedman, B. & Hutchinson, T. C. Pollutant inputs from the atmosphere and accumulations in soils and vegetation near a nickel–copper smelter at Sudbury, Ontario, Canada. Can. J. Bot. 58(1), 108–132. https://doi.org/10.1139/b80-014 (1980).CAS 
Article 

Google Scholar 
Manyiwa, T. et al. Heavy metals in soil, plants, and associated risk on grazing ruminants in the vicinity of Cu–Ni mine in Selebi-Phikwe, Botswana. Environ. Geochem. Health https://doi.org/10.1007/s10653-021-00918-x (2021).Article 
PubMed 

Google Scholar 
Kabata-Pendias. Kabata-Pendias A. 2011. Trace elements in soils and… – Google Scholar https://scholar.google.com/scholar?hl=en&as_sdt=0%2C5&q=Kabata-Pendias+A.+2011.+Trace+elements+in+soils+and+plants.+4th+ed.+New+York+%28NY%29%3A+CRC+Press&btnG= (accessed Nov 24, 2020).Almås, A., Singh, B., Agricultural, T. S.-N. J. of & 1995, undefined. The impact of nickel industry in Russia on concentrations of heavy metals in agricultural soils and grass in Soer-Varanger, Norway. agris.fao.org.Nielsen, G. D. et al. Absorption and retention of nickel from drinking water in relation to food intake and nickel sensitivity. Toxicol. Appl. Pharmacol. 154, 67–75 (1999).CAS 
Article 

Google Scholar 
Costa, M. & Klein, C. B. Nickel carcinogenesis, mutation, epigenetics, or selection. Environ. Health Perspect. 107, 2 (1999).Article 

Google Scholar 
Agyeman, P. C.; Ahado, S. K.; Borůvka, L.; Biney, J. K. M.; Sarkodie, V. Y. O.; Kebonye, N. M.; Kingsley, J. Trend Analysis of Global Usage of Digital Soil Mapping Models in the Prediction of Potentially Toxic Elements in Soil/Sediments: A Bibliometric Review. Environmental Geochemistry and Health. Springer Science and Business Media B.V. 2020. https://doi.org/10.1007/s10653-020-00742-9.Minasny, B. & McBratney, A. B. Digital soil mapping: A brief history and some lessons. Geoderma 264, 301–311. https://doi.org/10.1016/j.geoderma.2015.07.017 (2016).ADS 
Article 

Google Scholar 
McBratney, A. B., Mendonça Santos, M. L. & Minasny, B. On digital soil mapping. Geoderma 117(1–2), 3–52. https://doi.org/10.1016/S0016-7061(03)00223-4 (2003).ADS 
Article 

Google Scholar 
Deutsch.C.V. Geostatistical Reservoir Modeling,… – Google Scholar https://scholar.google.com/scholar?hl=en&as_sdt=0%2C5&q=C.V.+Deutsch%2C+2002%2C+Geostatistical+Reservoir+Modeling%2C+Oxford+University+Press%2C+376+pages.+&btnG= (accessed Apr 28, 2021).Olea, R. A. Geostatistics for engineers & earth scientists. Stoch. Environ. Res. Risk Assess. 14(3), 207–209. https://doi.org/10.1007/pl00009782 (2000).Article 

Google Scholar 
Gumiaux, C., Gapais, D. & Brun, J. P. Geostatistics applied to best-fit interpolation of orientation data. Tectonophysics 376(3–4), 241–259. https://doi.org/10.1016/j.tecto.2003.08.008 (2003).ADS 
Article 

Google Scholar 
Wadoux, A. M. J. C., Minasny, B. & McBratney, A. B. Machine learning for digital soil mapping: applications, challenges and suggested solutions. Earth-Sci Rev. https://doi.org/10.1016/j.earscirev.2020.103359 (2020).Article 

Google Scholar 
Tan, K. et al. Estimation of the spatial distribution of heavy metal in agricultural soils using airborne hyperspectral imaging and random forest. J. Hazard. Mater. 382, 120987. https://doi.org/10.1016/j.jhazmat.2019.120987 (2020).CAS 
Article 
PubMed 

Google Scholar 
Sakizadeh, M., Mirzaei, R. & Ghorbani, H. Support vector machine and artificial neural network to model soil pollution: a case study in Semnan Province, Iran. Neural Comput. Appl. 28(11), 3229–3238. https://doi.org/10.1007/s00521-016-2231-x (2017).Article 

Google Scholar 
Vega, F. A., Matías, J. M., Andrade, M. L., Reigosa, M. J. & Covelo, E. F. Classification and regression trees (CARTs) for modelling the sorption and retention of heavy metals by soil. J. Hazard. Mater. 167(1–3), 615–624. https://doi.org/10.1016/j.jhazmat.2009.01.016 (2009).CAS 
Article 
PubMed 

Google Scholar 
Sun, H. et al. Prediction of distribution of soil cd concentrations in Guangdong Province, China. Huanjing Kexue/Environmental Sci. 38(5), 2111–2124. https://doi.org/10.13227/j.hjkx.201611006 (2017).Article 

Google Scholar 
Woodcock, C. E. & Gopal, S. Fuzzy set theory and thematic maps: accuracy assessment and area estimation. Int. J. Geogr. Inf. Sci. 14(2), 153–172. https://doi.org/10.1080/136588100240895 (2000).Article 

Google Scholar 
Finke, P. A. Chapter 39 Quality assessment of digital soil maps: producers and users perspectives. Dev. Soil Sci. https://doi.org/10.1016/S0166-2481(06)31039-2 (2006).Article 

Google Scholar 
Pontius, R. G. & Cheuk, M. L. A generalized cross-tabulation matrix to compare soft-classified maps at multiple resolutions. Int. J. Geogr. Inf. Sci. 20(1), 1–30. https://doi.org/10.1080/13658810500391024 (2006).Article 

Google Scholar 
Grunwald, S. Multi-criteria characterization of recent digital soil mapping and modeling approaches. Geoderma 152(3–4), 195–207. https://doi.org/10.1016/j.geoderma.2009.06.003 (2009).ADS 
Article 

Google Scholar 
Nelson, M. A., Bishop, T. F. A., Triantafilis, J. & Odeh, I. O. A. An error budget for different sources of error in digital soil mapping. Eur. J. Soil Sci. 62, 417–430 (2011).Article 

Google Scholar 
McBratney, A. B., Minasny, B. & ViscarraRossel, R. Spectral soil analysis and inference systems: A powerful combination for solving the soil data crisis. Geoderma 136, 272–278 (2006).ADS 
CAS 
Article 

Google Scholar 
Stumpf, F. et al. Uncertainty-guided sampling to improve digital soil maps. CATENA 153, 30–38 (2017).Article 

Google Scholar 
Legates, D. R. & McCabe, G. J. Evaluating the use of ‘goodness-of-fit’ measures in hydrologic and hydroclimatic model validation. Water Resour. Res. 35, 233–241 (1999).ADS 
Article 

Google Scholar 
Sergeev, A. P. et al. High variation subarctic topsoil pollutant concentration prediction using neural network residual kriging. AIP Conf. Proc. 2017, 1836. https://doi.org/10.1063/1.4981963 (2017).CAS 
Article 

Google Scholar 
Subbotina, I. E. et al. Multilayer perceptron, generalized regression neural network, and hybrid model in predicting the spatial distribution of impurity in the topsoil of urbanized area. AIP Conf. Proc. https://doi.org/10.1063/1.5045410 (2018).Article 

Google Scholar 
Tarasov, D. A., Buevich, A. G., Sergeev, A. P. & Shichkin, A. V. High variation topsoil pollution forecasting in the Russian subarctic: using artificial neural networks combined with residual kriging. Appl. Geochemistry 88, 188–197. https://doi.org/10.1016/j.apgeochem.2017.07.007 (2018).CAS 
Article 

Google Scholar 
Tarasov, D.; Buevich, A.; Shichkin, A.; Subbotina, I.; Tyagunov, A.; Baglaeva, E. Chromium Distribution Forecasting Using Multilayer Perceptron Neural Network and Multilayer Perceptron Residual Kriging. In AIP Conference Proceedings; American Institute of Physics Inc., 2018; Vol. 1978, p 440019. https://doi.org/10.1063/1.5044048.John, K. et al. Hybridization of cokriging and gaussian process regression modelling techniques in mapping soil sulphur. CATENA 206, 2 (2021).Article 

Google Scholar 
Gribov, A. & Krivoruchko, K. Empirical Bayesian Kriging Implementation and Usage. Sci. Total Environ. https://doi.org/10.1016/j.scitotenv.2020.137290 (2020).Article 
PubMed 

Google Scholar 
Samsonova, V. P., Blagoveshchenskii, Y. N. & Meshalkina, Y. L. Use of empirical Bayesian kriging for revealing heterogeneities in the distribution of organic carbon on agricultural lands. Eurasian Soil Sci. 50(3), 305–311. https://doi.org/10.1134/S1064229317030103 (2017).ADS 
CAS 
Article 

Google Scholar 
Fabijańczyk, P., Zawadzki, J. & Magiera, T. Magnetometric assessment of soil contamination in problematic area using empirical bayesian and indicator kriging: a case study in upper Silesia, Poland. Geoderma 308, 69–77. https://doi.org/10.1016/j.geoderma.2017.08.029 (2017).ADS 
CAS 
Article 

Google Scholar 
John, K. et al. Mapping soil properties with soil-environmental covariates using geostatistics and multivariate statistics. Int. J. Environ. Sci. Technol. 2, 1–16. https://doi.org/10.1007/s13762-020-03089-x (2021).CAS 
Article 

Google Scholar 
Li, T. et al. Using self-organizing map for coastal water quality classification: Towards a better understanding of patterns and processes. Sci. Total Environ. 628–629, 1446–1459. https://doi.org/10.1016/j.scitotenv.2018.02.163 (2018).ADS 
CAS 
Article 
PubMed 

Google Scholar 
Wang, Z. et al. Elucidating the differentiation of soil heavy metals under different land uses with geographically weighted regression and self-organizing map. Environ. Pollut. https://doi.org/10.1016/j.envpol.2020.114065 (2020).Article 
PubMed 
PubMed Central 

Google Scholar 
Hossain Bhuiyan, M. A., Chandra Karmaker, S., Bodrud-Doza, M., Rakib, M. A. & Saha, B. B. Enrichment, sources and ecological risk mapping of heavy metals in agricultural soils of dhaka district employing SOM PMF and GIS Methods. Chemosphere https://doi.org/10.1016/j.chemosphere.2020.128339 (2021).Article 
PubMed 

Google Scholar 
Kebonye, N. M. et al. Self-organizing map artificial neural networks and sequential gaussian simulation technique for mapping potentially toxic element hotspots in polluted mining soils. J. Geochemical Explor. 222, 106680. https://doi.org/10.1016/j.gexplo.2020.106680 (2021).CAS 
Article 

Google Scholar 
Weather Spark. Average Weather in Frýdek-Místek, Czechia, Year Round – Weather Spark https://weatherspark.com/y/83671/Average-Weather-in-Frýdek-Místek-Czechia-Year-Round (accessed Sep 14, 2020).Kozák, J. Soil Atlas of the Czech Republic. 2010, 150.Vacek, O., Vašát, R. & Borůvka, L. Quantifying the pedodiversity-elevation relations. Geoderma 373, 114441. https://doi.org/10.1016/j.geoderma.2020.114441 (2020).ADS 
Article 

Google Scholar 
Krivoruchko, K. Empirical Bayesian Kriging; 2012; Vol. Fall 2012.Vapnik, V. The nature of statistical learning theory. Technometrics 38(4), 409. https://doi.org/10.2307/1271324 (1995).Article 
MATH 

Google Scholar 
Li, Z., Zhou, M., Xu, L. J., Lin, H. & Pu, H. Training sparse SVM on the core sets of fitting-planes. Neurocomputing 130, 20–27. https://doi.org/10.1016/j.neucom.2013.04.046 (2014).Article 

Google Scholar 
Cherkassky, V.; Mulier, F. Learning from Data: Concepts, Theory, and Methods: Second Edition; 2006. https://doi.org/10.1002/9780470140529.John, K. et al. Using machine learning algorithms to estimate soil organic carbon variability with environmental variables and soil nutrient indicators in an alluvial soil. Land 9(12), 1–20. https://doi.org/10.3390/land9120487 (2020).CAS 
Article 

Google Scholar 
Vohland, M., Besold, J., Hill, J. & Fründ, H. C. Comparing different multivariate calibration methods for the determination of soil organic carbon pools with visible to near infrared spectroscopy. Geoderma 166(1), 198–205. https://doi.org/10.1016/j.geoderma.2011.08.001 (2011).ADS 
CAS 
Article 

Google Scholar 
Fraser, S. J.; Dickson, B. L. A New Method for Data Integration and Integrated Data Interpretation: Self-Organising Maps; 2007.Melssen, W. J.; Smits, J. R. M.; Buydens, L. M. C.; Kateman, G. Using Artificial Neural Networks for Solving Chemical Problems Part II. Kohonen Self-Organising Feature Maps and Hopfield Networks. Chemometrics and Intelligent Laboratory Systems. Elsevier, Amsterdam, 1, 1994, pp 267–291. https://doi.org/10.1016/0169-7439(93)E0036-4.Kooistra, L. et al. The potential of field spectroscopy for the assessment of sediment properties in river floodplains. Anal. Chim. Acta 484(2), 189–200. https://doi.org/10.1016/S0003-2670(03)00331-3 (2003).CAS 
Article 

Google Scholar 
Li, L. et al. Methods for estimating leaf nitrogen concentration of winter oilseed rape (Brassica Napus L.) using in situ leaf spectroscopy. Ind. Crops Prod. 91, 194–204. https://doi.org/10.1016/j.indcrop.2016.07.008 (2016).CAS 
Article 

Google Scholar 
Różański, S. Ł, Kwasowski, W., Castejón, J. M. P. & Hardy, A. Heavy metal content and mobility in urban soils of public playgrounds and sport facility areas, Poland. Chemosphere 212, 456–466. https://doi.org/10.1016/j.chemosphere.2018.08.109 (2018).ADS 
CAS 
Article 
PubMed 

Google Scholar 
Bretzel, F. & Calderisi, M. Metal contamination in urban soils of coastal Tuscany (Italy). Environ. Monit. Assess. 118(1–3), 319–335. https://doi.org/10.1007/s10661-006-1495-5 (2006).CAS 
Article 
PubMed 

Google Scholar 
Jim, C. Y. Urban soil characteristics and limitations for landscape planting in hong kong. Landsc. Urban Plan. 40(4), 235–249. https://doi.org/10.1016/S0169-2046(97)00117-5 (1998).Article 

Google Scholar 
Birke, M.; Rauch, U.; Chmieleski, J. Environmental Geochemical Survey of the City of Stassfurt: An Old Mining and Industrial Urban Area in Sachsen-Anhalt, Germany. In Mapping the Chemical Environment of Urban Areas; John Wiley and Sons, 2011; pp 269–306. https://doi.org/10.1002/9780470670071.ch18.Khodadoust, A. P., Reddy, K. R. & Maturi, K. Removal of nickel and phenanthrene from kaolin soil using different extractants. Environ. Eng. Sci. 21(6), 691–704. https://doi.org/10.1089/ees.2004.21.691 (2004).CAS 
Article 

Google Scholar 
Jakovljevic, M.; Kostic, N.; Antic-Mladenovic, S. The Availability of Base Elements (Ca, Mg, Na, K) in Some Important Soil Types in Serbia; 2003. https://doi.org/10.2298/zmspn0304011j.Orzechowski, M.; Smolczynski, S. IN SOILS DEVELOPED FROM THE HOLOCENE DEPOSITS IN NORTH-EASTERN POLAND*; -, 2007; Vol. 15.Pongrac, P. et al. Mineral element composition of cabbage as affected by soil type and phosphorus and zinc fertilisation. Plant Soil 434(1–2), 151–165. https://doi.org/10.1007/s11104-018-3628-3 (2019).CAS 
Article 

Google Scholar 
Kingston, G.; Anink, M. C.; Clift, B. M.; Beattie, R. N. Potassium Management for Sugarcane on Base Saturated Soils in Northern New South Wales; 2009; Vol. 31.Santo, L. T., Nakahata, M. H., & Schell, V. P. Santo LT, Nakahata MH, Ito GP and Schell VP (2000)…. – Google Scholar https://scholar.google.com/scholar?hl=en&as_sdt=0%2C5&q=Santo+LT%2C+Nakahata+MH%2C+Ito+GP+and+Schell+VP+%282000%29.+Calcium+and+liming+trials+from+1994+to+1998+at+HC%26S.+Technical+supplement+to+Agronomy+Report+83%2C+Hawaiian+Agricultural+Research+Centre. (accessed May 16, 2021).Burgos, P., Madejón, E., Pérez-de-Mora, A. & Cabrera, F. Horizontal and vertical variability of soil properties in a trace element contaminated area. Int. J. Appl. Earth Obs. Geoinf. 10(1), 11–25. https://doi.org/10.1016/j.jag.2007.04.001 (2008).ADS 
Article 

Google Scholar 
Olinic, T. & Olinic, E. The effect of quicklime stabilization on soil properties. Agric. Agric. Sci. Procedia 10, 444–451. https://doi.org/10.1016/j.aaspro.2016.09.013 (2016).Article 

Google Scholar 
Madaras, M.; Lipavský, J. Interannual Dynamics of Available Potassium in a Long-Term Fertilization Experiment; 2009; Vol. 55. https://doi.org/10.17221/34/2009-pse.Madaras, M., Koubova, M. & Lipavský, J. Stabilization of available potassium across soil and climatic conditions of the Czech Republic. Arch. Agron. Soil Sci. 56(4), 433–449. https://doi.org/10.1080/03650341003605750 (2010).CAS 
Article 

Google Scholar 
Pulkrabová, J. et al. Is the long-term application of sewage sludge turning soil into a sink for organic pollutants?: Evidence from field studies in the Czech Republic. J. Soils Sedim. 19(5), 2445–2458. https://doi.org/10.1007/s11368-019-02265-y (2019).CAS 
Article 

Google Scholar 
Asare, M. O., Horák, J., Šmejda, L., Janovský, M. & Hejcman, M. A medieval hillfort as an island of extraordinary fertile archaeological dark earth soil in the Czech Republic. Eur. J. Soil Sci. 72(1), 98–113. https://doi.org/10.1111/ejss.12965 (2021).CAS 
Article 

Google Scholar 
Zádorová, T. et al. Identification of Neolithic to Modern Erosion-Sedimentation Phases Using Geochemical Approach in a Loess Covered Sub-Catchment of South Moravia Czech Republic. Geoderma 195–196, 56–69. https://doi.org/10.1016/j.geoderma.2012.11.012 (2013).ADS 
CAS 
Article 

Google Scholar 
Tlustoš, P. et al. Nutrient status of soil and winter wheat (Triticum Aestivum L.) in response to long-term farmyard manure application under different climatic and soil physicochemical conditions in the Czech Republic. Arch. Agron. Soil Sci. 64(1), 70–83. https://doi.org/10.1080/03650340.2017.1331297 (2018).Article 

Google Scholar 
Wang, Z. et al. Elucidating the differentiation of soil heavy metals under different land uses with geographically weighted regression and self-organizing map. Environ. Pollut. 260, 2 (2020).
Google Scholar 
Yan, P., Peng, H., Yan, L. & Lin, K. Spatial variability of soil physical properties based on GIS and geo-statistical methods in the red beds of the Nanxiong Basin, China. Polish J. Environ. Stud. 28, 2961–2972 (2019).Article 

Google Scholar 
Beguin, J., Fuglstad, G. A., Mansuy, N. & Paré, D. Predicting soil properties in the Canadian boreal forest with limited data: Comparison of spatial and non-spatial statistical approaches. Geoderma 306, 195–205 (2017).ADS 
CAS 
Article 

Google Scholar 
Adhikary, P. P., Dash, C. J., Bej, R. & Chandrasekharan, H. Indicator and probability kriging methods for delineating Cu, Fe, and Mn contamination in groundwater of Najafgarh Block, Delhi, India. Environ. Monit. Assess. 176, 663–676 (2011).CAS 
Article 

Google Scholar 
John, K. et al. Mapping soil properties with soil-environmental covariates using geostatistics and multivariate statistics. Int. J. Environ. Sci. Technol. 18, 3327–3342 (2021).CAS 
Article 

Google Scholar 
Eldeiry, A. A. & Garcia, L. A. Detecting soil salinity in alfalfa fields using spatial modeling and remote sensing. Soil Sci. Soc. Am. J. 72, 201–211 (2008).ADS 
CAS 
Article 

Google Scholar  More

Plant materials, growth conditions and salt treatmentsSoil samples were performed as in previous experiments with S. europaea25, seeds were collected at two maternal sites, the first of which represents natural salinity related to inland salt springs at the health resort of Ciechocinek (Cie) (52°53′N, 18°47′E) characterised by a high soil salinity of ca 100 dS m−1 (~ 1000 mM NaCl), and the second of which is associated with soda factory waste that affects the local environment in Inowrocław-Mątwy (Inw) (52°48′N, 18°15′E) and with a lower salinity of ca 55 dS m−1 (~ 550 mM NaCl). The complete soil description is reported in Piernik et al.51 and Szymanska et al.52,53. Populations are isolated by a distance of ca 40 km without any saline environment between them, however, they were somehow connected due to the presence of salt springs in the nineteenth century. The seeds came from one generation and were collected in early November 2018. The seeds were germinated and grown according to the same steps reported in Cárdenas-Pérez et al.25 with a slight modification in the number of salt treatments at 0, 200, 400, 600, 800 and 1000 mM NaCl. In total, 144 plants were cultivated, and, therefore, a complete randomised factorial design 26 was used, which included (12 plants × 6 treatments × 2 populations) with 14 response variables. After 2 months of development, anatomical analysis such as cell area (A), roundness (R) and maximum cell diameter (Cdiam) were estimated in 12 samples, whereas high and low methyl esterified pectins (HM-HGs and LM-HGs), proline (P), hydrogen peroxide (HP), total soluble protein (Prot), catalase activity (CAT), peroxidase activity (POD), chlorophyll a, b and total (Cha, Chb and TC), carotenoid (Carot) contents, as well as SeNHX1 and SeSOS1 gene expression, were all determined per triplicate (plants were randomly selected). The collection of plant material, comply with relevant institutional, national, and international guidelines and legislation, IUCN Policy Statement on Research Involving Species at Risk of Extinction and Convention on the Trade in Endangered Species of Wild Fauna and Flora. The voucher specimen of the plant material has been deposited in a publicly available herbarium of the Nicolaus Copernicus University in Toruń (Index Herbarium code TRN), deposition number not available (dr. hab. Agnieszka Piernik, prof. NCU undertook the formal identification of plant species, and permission to work with the seeds was provided by the Regional Director of Environmental Protection in Bydgoszcz, WOP.6400.12.2020.JC).Anatomical image analysisFrom the middle primary branch (fleshy segment shoot) of S. europaea plant treatments (0, 200, 400, 600, 800 and 1000 mM NaCl), slices of fresh tissue were obtained by cutting them with a sharp bi-shave blade. The thinner slices of approximately 0.5 mm were selected and used in the microstructure analysis. The size and shape of the stem-cortex cells from the fresh water-storing tissue were characterised by a light microscope (Olympus BX51, USA) connected to a digital camera (DP72 digital microscope camera) and digital acquisition software (DP2-BSW). The microscope images were captured at a magnification of 10 ×/0.30 in RGB scale and stored in TIFF format at 1280 × 1024 pixels. A total of 300 ± 50 cells from five individuals per treatment were analysed. Finally, the shape and size of the cells were obtained from the captured images. Cell image analysis (IA) was performed in ImageJ v. 1.47 (National Institutes of Health, Bethesda, MD, USA). The following anatomical parameters were obtained. Firstly, the cell area (A) was estimated as the number of pixels within the boundary. Secondly, the maximum cell’s diameter (Cdiam) was determined by the distance between the two points separated by the largest coordinates in different orientations, and the cell roundness (R) was obtained through the equation R = (4 A)/(π (Cdiam)2)—where a perfectly round cell has R = 1.0, while elongated cells will show an R → 0. Finally, the degree of succulence (S) in stem was calculated according to24 with slight change S = (Fresh Weight-Dry Weight)/stem Area, where the Area of the stem (As) was calculated as: As = π × r2, the diameter of the stems was obtained according to Cárdenas-Pérez et al.25.Immunolocalisation experimentsThe samples dissected from the middle segment of the shoot (3 individuals per treatment) were prepared for embedding in BMM resin (butyl methacrylate, methyl methacrylate, 0.5% benzoyl ethyl ether (Sigma) with 10 mM DDT (Thermo Fisher Scientific) according to Niedojadło et al.54. Next, specimens were cut on a Leica UCT ultramicrotome into serial semi-thin cross sections (1.5 µm) that were collected on Thermo Scientific Polysine adhesion microscope slides. Before immunocytochemical reaction, the resin was removed with two changes of acetone and washed in distilled water and PBS pH 7.2. After incubation with blocking solution containing 2% BSA (bovine serum albumin, Sigma) in PBS pH 7.2 for 30 min at room temperature, the sections were incubated with anti-pectin rat monoclonal primary antibody JIM7 (recognises partially methylesterified epitopes of homogalacturonan [HG] but does not bind to fully de-esterified HGs) or antibody LM19 (recognises partially methylesterified epitopes of HG and binds strongly to de-esterified HGs) (Plant Probes) diluted 1:50 in 0.2% BSA in PBS pH 7.2 overnight at 4 °C. After washing with PBS pH 7.2, the material was incubated with AlexaFluor 488-conjugated goat anti-rat secondary antibody (Thermo Fisher Scientific) diluted 1:1000 in 0.2% BSA in PBS pH 7.2 for 1 h at 37 °C. Finally, the sections were washed in PBS pH 7.2, dried at room temperature and covered with ProLongTMGold antifade reagent (Thermo Fisher Scientific). The control reactions were performed with the omission of incubation with primary antibodies. Semithin sections were analysed with an Olympus BX50 fluorescence microscope, with an UPlanFI 1009 (N.A. 1.3) oil immersion lens and narrow band filters (U-MNU, U-MNG). The results were recorded with an Olympus XC50 digital colour camera and CellB software (Olympus Soft Imaging Solutions GmbH, Germany).Fluorescence quantitative evaluationFor the quantitative measurement, each experiment was performed using consistent temperatures, incubation times and concentrations of antibodies. The aforementioned ImageJ (1.47v) software was used for image processing and analysis. The fluorescence intensity was measured for five semi-thin sections for each experimental population (Inowrocław and Ciechocinek) at the same magnification (100 ×) and the constant exposure time to ensure comparable results. The threshold fluorescence in the sample was established based on the autofluorescence of the control reaction. The level of signal intensity was expressed in arbitrary units (a.u.) as the mean intensity per μm2 according to Niedojadło et al.54.Biochemical analysisProline content (P) was measured according to Ábrahám et al.55. Five hundred milligrams of fresh stem material was minced on ice and homogenised with 3% aqueous sulfosalicylic acid solution (5 μl mg−1 fresh plant material), centrifuged at 18,000×g, 10 min at 4 °C, and the supernatant was collected. The reaction mixture: 100 μl of 3% sulphosalicylic acid, 200 μl of glacial acetic acid, 200 μl of acidic ninhydrin reagent and 100 μl of supernatant. Acidic ninhydrin reagent was prepared according to Bates et al.56. The standard curve for proline in the concentration range of 0 to 40 μg ml−1. The standard curve equation was y = 0.0467x − 0.0734, R2 = 0.963. P was expressed in mg of proline per gram of fresh weight. Hydrogen peroxide (HP) levels were determined according to the methods described by Velikova et al.57, and 500 mg of stem tissues were homogenised with 5 ml trichloroacetic acid 0.1% (w:v) in an ice bath. The homogenate was centrifuged (12,000×g, 4 °C, 15 min) and 0.5 ml of the supernatant was added to potassium phosphate buffer (0.5 ml) (10 mM, pH 7.0) and 2 ml of 1 M KI. The absorbance was read at 390 nm, and the HP content was given on a standard curve from 0 to 40 mM. The standard curve equation was y = 0.0188x + 0.046, R2 = 0.987. HP concentrations were expressed in nM per gram of fresh weight. Chlorophylls (Cha and Chb) and carotenoids were extracted from fresh plant stems (100 mg) using 80% acetone for 6 h in darkness, and then centrifuged at 10,000 rpm, 10 min. Supernatants were quantified spectrophotometrically. Absorbance was determined at 646, 663 and 470 nm and calculations were performed according to Lichtenthaler and Wellburn58, when 80% of acetone is used as dissolvent. Total chlorophyll content was calculated as the sum of chlorophyll a and b contents.Total CAT activity was determined spectrophotometrically by following the decline in A240 as H2O2 (ε = 39.9 M−1 cm−1) was catabolised, according to the method of Beers and Sizer59. Decrease in absorbance of the reaction at 240 nm was recorded after every 20 s. One unit CAT was defined as an absorbance change of 0.01 units min−1. Total POD activity was determined spectrophotometrically by monitoring the formation of tetraguaiacol (ε = 26.6 mM−1 cm−1) from guaiacol at A470 in the presence of H2O2 by the method of Chance and Maehly60. Increase in absorbance of the reaction solution at 470 nm was recorded after every 20 s. One unit of POD activity was defined as an absorbance change of 0.01 units min−1. Total soluble protein (Prot) content was measured according to Bradford61 using bovine serum albumin (BSA) as a protein standard. Fresh leaf samples (1 g) were homogenised with 4 ml Na-phosphate buffer (pH 7.2) and then centrifuged at 4 °C. Supernatant and dye were pipetted in spectrophotometer cuvettes and absorbances were measured using a UV–vis spectrophotometer (PG instruments T80) at 595 nm62. Prot was determined based on the standard curve y = 1.6565x + 0.0837, R2 = 0.982, for total soluble protein in the concentration range of 0 to 1.2 mg ml−1 BSA. Triplicates per treatment were used for each analysis.Total RNA isolationAfter 2 months of salt treatment, shoots of S. europaea plants (3 individuals per treatment) were washed several times with tap water and then three times with miliQ water. After drying, plant material was frozen in liquid nitrogen, and stored at − 80 °C. Total RNA isolation was performed using RNeasy Plant Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s protocol. The quality and quantity of RNA was checked on 1.5% agarose gels in TAE (Tris–HCl, acetic acid, EDTA, pH 8.3) buffer stained with ethidium bromide, and by spectrophotometric measurement (NanoDrop Lite, Thermo Fisher Scientific, Waltham, MA, USA).Cloning of SOS1 gene from S. europaea (SeSOS1)One microgram (1 µg) of total RNA isolated from shoots of S. europaea was primed with 0.5 µg of oligo (dT)20 primer for 5 min at 70 °C. Then 4 µl of ImProm-II 5 × reaction buffer, 2 mM MgCl2, 0.5 mM each dNTP, 20 U of recombinant RNasin ribonuclease inhibitor, and 1 µl of ImProm-II reverse transcriptase (Promega, Madison, WI, USA) were added to a final volume of 20 µl. The reaction was performed at 42 °C for 60 min. To design degenerate primers for SOS1, cDNA sequences from Arabidopsis thaliana (NM_126259.4), Lycopersicon esculentum (AJ717346.1), Mesembryanthemum crystallinum (EF207776.1), Oryza sativa (AY785147.1), Triticum aestivum (AY326952.3), Salicornia brachiata (EU879059.1) were obtained from NCBI GeneBank. The sequences were aligned using the Clustal Omega tool (https://www.ebi.ac.uk/Tools/msa/clustalo/) and three pairs of degenerate primes were designed (listed in Table 4). The PCR reaction mixture includes cDNA, 0.2 µM each primer, 0.2 mM each dNTP, 4 µl of 5 × HF buffer, and 0.5 U of Phusion High-Fidelity DNA polymerase (Thermo Fisher Scientific, Waltham, MA, USA) in a total volume of 20 µl. The thermal conditions were as follows: 98 °C for 30 s, 98 °C for 10 s, gradient between 48 °C and 56 °C for 20 s, 72 °C for 60 s, 32 cycle, final extension for 10 min at 72 °C. A pair of primers deg2_F and deg2_R yielded a PCR product with expected size. The PCR product was purified from agarose gel, cloned into pJET1.2 vector (Thermo Fisher Scientific, Waltham, MA, USA) according to manufacturer’s protocol and sequenced (Genomed, Warsaw, Poland). The obtained partial cDNA sequence was named SeSOS1 and deposited in NCBI GeneBank (acc. no. MZ707082).Table 4 Sequences of the primers used for cloning of SeSOS1 and quantitative real-time PCR.Full size tableReverse transcription reaction and quantitative real-time PCR (qPCR) SeNHX1 and SeSOS1 gene expression analysisPrior to reverse transcription reaction, RNA was treated with DNaseI (Thermo Fisher Scientific, Waltham, MA, USA). The cDNA was synthesised from 1.5 µg of total RNA using a mixture of 2.5 µM oligo(dT)20 primer and 0.2 µg of random hexamers with NG dART RT Kit (Eurx, Gdańsk, Poland) according to the manufacturer’s protocol. The reaction was performed at 25 °C for 10 min, followed by 50 min at 50 °C. The cDNA was stored at − 20 °C.The PCR reaction mixture includes 4 µl of 1/20 diluted cDNA, 0.5 µM gene-specific primers (Table 4) and 5 µl of LightCycler 480 SYBR Green I Master (Roche, Penzberg, Germany) in a total volume of 10 µl. Clathrin adaptor complexes (CAC) was used as a reference gene63. The reaction was performed in triplicate (technical replicates) in LightCycler 480 Instrument II (Roche, Penzberg, Germany). The thermal cycling conditions were as follows: 95 °C for 5 min, 95 °C for 10 s, 60 °C for 20 s, 72 °C for 20 s, 40 cycles. The SYBR Green I fluorescence signal was recorded at the end of the extension step in each cycle. The specificity of the assay was confirmed by the melt curve analysis i.e., increasing the temperature from 55 to 95 °C at a ramp rate 0.11 °C/s. The fold-change in gene expression was calculated using LightCycler 480 Software release 1.5.1.62 (Roche, Penzberg, Germany).Statistical and multivariate analysisIn order to determine the projection of the effect of salt treatment in plants we followed Cárdenas-Pérez et al.25 methodology. A principal component analysis (PCA) was developed using XLSTAT software version 2019.4.165. For this analysis, 14 variables were used, (A, Cdiam, R, Prot, CAT, POD, HM-HGs, LM-HGs, P, HP, Cha, Chb, TC, Carot), arranged in a matrix with the average values obtained from replicates of each treatment and population. A two-way ANOVA comparing treatments within populations and populations within treatments was conducted for all the results with the Holm–Sidak method. The data was fit with a modified three parameter exponential decay using SigmaPlot version 11.066. The relationships between variables were performed using a Pearson analysis, while a significance test (Kaisere Meyere Olkin) was performed in order to determine which variables had a significant correlation with each other (α = 0.05). Then, a 3D plot was developed using the three principal component factors according to the Kaiser criterion which stated that the factors below the unit are irrelevant. The three main factorial scores of the PCA from each sample were used to calculate the distance (D) between the two points (populations) under the same treatment P1 = (x1, y1, z1) and P2 = (x2, y2, z2) in 3D space of the PCA (Eq. 1).$$D ( {P_{1} ,, P_{2} } ) = sqrt {( {x_{2} – x_{1} } )^{2} + ( {y_{2} – y_{1} } )^{2} + ( {z_{2} – z_{1} } )^{2} }$$
(1)
where x, y, and z are the three main factorial scores in the PCA corresponding to the evaluated treatment in Inw and in Cie. Distances were used to evaluate and determine in which salt treatment the greatest differences between the populations were recorded. More

Wehner, R., Michel, B. & Antonsen, P. Visual navigation in insects: Coupling of egocentric and geocentric information. J. Exp. Biol. 199(1), 129–140 (1996).CAS 
PubMed 

Google Scholar 
Collett, M., Chittka, L. & Collett, T. S. Spatial memory in insect navigation. Curr. Biol. 23(17), R789–R800 (2013).CAS 
PubMed 

Google Scholar 
Cheng, K., Schultheiss, P., Schwarz, S., Wystrach, A. & Wehner, R. Beginnings of a synthetic approach to desert ant navigation. Behav. Proc. 102, 51–61 (2014).
Google Scholar 
Freas, C. A. & Schultheiss, P. How to navigate in different environments and situations: Lessons from ants. Front. Psych. 9, 841 (2018).
Google Scholar 
Wehner, R. Desert ant navigation: how miniature brains solve complex tasks. J. Comp. Physiol. A 189(8), 579–588 (2003).ADS 
CAS 

Google Scholar 
Wehner, R. The desert ant’s navigational toolkit: Procedural rather than positional knowledge. Navigation 55(2), 101–114 (2008).
Google Scholar 
Wehner, R. Desert Navigator (The Belknap Press of Harvard University Press, 2020).
Google Scholar 
Kohler, M. & Wehner, R. Idiosyncratic route-based memories in desert ants, Melophorus bagoti: How do they interact with path-integration vectors?. Neurobiol. Learn. Mem. 83(1), 1–12 (2005).PubMed 

Google Scholar 
Müller, M. & Wehner, R. Path integration provides a scaffold for landmark learning in desert ants. Curr. Biol. 20(15), 1368–1371 (2010).PubMed 

Google Scholar 
Mangan, M. & Webb, B. Spontaneous formation of multiple routes in individual desert ants (Cataglyphis velox). Behav. Ecol. 23(5), 944–954 (2012).
Google Scholar 
Schwarz, S., Wystrach, A. & Cheng, K. Ants’ navigation in an unfamiliar environment is influenced by their experience of a familiar route. Sci. Rep. 7(1), 1–10 (2017).
Google Scholar 
Graham, P. & Cheng, K. Ants use the panoramic skyline as a visual cue during navigation. Curr. Biol. 19, R935–R937 (2009).CAS 
PubMed 

Google Scholar 
Wystrach, A., Beugnon, G. & Cheng, K. Landmarks or panoramas: What do navigating ants attend to for guidance?. Front. Zool. 8(1), 21 (2011).PubMed 
PubMed Central 

Google Scholar 
Wehner, R., Meier, C. & Zollikofer, C. The ontogeny of foraging behaviour in desertants, Cataglyphis bicolor. Ecol. Entom. 29, 240–250 (2004).
Google Scholar 
Zeil, J. & Fleischmann, P. N. The learning walks of ants (Hymenoptera: Formicidae). Myrmecol. News. 29, 93–110 (2019).
Google Scholar 
Schultheiss, P. et al. Crucial role of ultraviolet light for desert ants in determining direction from the terrestrial panorama. Anim. Behav. 115, 19–28 (2016).
Google Scholar 
Freas, C. A., Wystrach, A., Narendra, A. & Cheng, K. The view from the trees: Nocturnal bull ants, Myrmecia midas, use the surrounding panorama while descending from trees. Front. Psych. 9, 1–10 (2018).
Google Scholar 
Freas, C. A. & Cheng, K. Landmark learning, cue conflict, and outbound view sequence in navigating desert ants. J. Exp. Psych. Anim. Learn. Cogn. 44(4), 409–421 (2018).
Google Scholar 
Freas, C. A. & Spetch, M. L. Terrestrial cue learning and retention during the outbound and inbound foraging trip in the desert ant, Cataglyphis bicolor. J. Comp. Physiol. A. 205(2), 177–189 (2019).
Google Scholar 
Narendra, A., Si, A., Sulikowski, D. & Cheng, K. Learning, retention and coding of nest-associated visual cues by the Australian desert ant, Myrmecia midas. Behav. Ecol. Sociobiol. 61(10), 1543–1553 (2007).
Google Scholar 
Zeil, J. Visual homing: an insect perspective. Curr. Opin. Neurobiol. 22(2), 285–293 (2012).CAS 
PubMed 

Google Scholar 
Zeil, J., Hofmann, M. I. & Chahl, J. S. Catchment areas of panoramic snapshots in outdoor scenes. J. Optic. Soc. Am. A. 20(3), 450 (2003).ADS 

Google Scholar 
Wystrach, A., Cheng, K., Sosa, S. & Beugnon, G. Geometry, features, and panoramic views: Ants in rectangular arenas. J. Exp. Psychol. 37(4), 420–435 (2011).
Google Scholar 
Baddeley, B., Graham, P., Husbands, P. & Philippides, A. A model of ant route navigation driven by scene familiarity. PLoS Comp. Biol. 8(1), e1002336 (2012).ADS 
CAS 

Google Scholar 
Kodzhabashev, A. & Mangan, M. Route Following Without Scanning In Biomimetic and Biohybrid Systems 199–210 (Springer, 2015).
Google Scholar 
Möller, R. A model of ant navigation based on visual prediction. J. Theo. Biol. 305, 118–130 (2012).ADS 
MathSciNet 
MATH 

Google Scholar 
Le Möel, F. & Wystrach, A. Opponent processes in visual memories: A model of attraction and repulsion in navigating insects’ mushroom bodies. PLoS Comp. Biol. 16, e1007631 (2020).
Google Scholar 
Murray, T. et al. The role of attractive and repellent scene memories in ant homing (Myrmecia croslandi). J. Exp. Biol. 223, 21002 (2020).
Google Scholar 
Jayatilaka, P., Murray, T., Narendra, A. & Zeil, J. The choreography of learning walks in the Australian jack jumper ant Myrmecia croslandi. J. Exp. Biol. 221(20), 185306 (2018).
Google Scholar 
Schwarz, S., Mangan, M., Webb, B. & Wystrach, A. Route-following ants respond to alterations of the view sequence. J. Exp. Biol. 223, 218701 (2020).
Google Scholar 
Wystrach, A., Buehlmann, C., Schwarz, S., Cheng, K. & Graham, P. Rapid aversive and memory trace learning during route navigation in desert ants. Curr. Biol. 30(100), 1927–1933 (2020).CAS 
PubMed 

Google Scholar 
Wystrach, A., Philippides, A., Aurejac, A., Cheng, K. & Graham, P. Visual scanning behaviours and their role in the navigation of the Australian desert ant Melophorus bagoti. J. Comp. Physiol. A 200(7), 615–626 (2014).
Google Scholar 
Wystrach, A., Schwarz, S., Graham, P. & Cheng, K. Running paths to nowhere: Repetition of routes shows how navigating ants modulate online the weights accorded to cues. Anim. Cogn. 2, 213–222 (2019).
Google Scholar 
MacArthur, R. H. & Pianka, E. R. On optimal use of a patchy environment. Am. Nat. 100(916), 603–609 (1966).
Google Scholar 
Krebs, J. R. Foraging Theory (Princeton University Press, 1986).
Google Scholar 
Kacelnik, A. & Bateson, M. Risky theories: The effects of variance on foraging decisions. Am. Zool. 36(4), 402–434 (1996).
Google Scholar 
Kacelnik, A. & Abreu, F. B. Risky choice and Weber’s law. J. Theor. Biol. 194(2), 289–298 (1998).ADS 
CAS 
PubMed 

Google Scholar 
Fechner, G. T. Elemente der Psychophysik Vol. 2 (Breitkopf u Härtel, 1860).
Google Scholar 
Bruce, A. C. & Johnson, J. E. V. Decision-making under risk: Effect of complexity on performance. Psychol. Rep. 79(1), 67–76 (1996).
Google Scholar 
Stevens, S. S. & Marks, L. E. Psychophysics: Introduction to its Perceptual, Neural, and Social Prospects (Routledge, 2017).
Google Scholar 
Kacelnik, A. & El Mouden, C. Triumphs and trials of the risk paradigm. Anim. Behav. 86(6), 1117–1129 (2013).
Google Scholar 
Hübner, C. & Czaczkes, T. J. Risk preference during collective decision making: Ant colonies make risk-indifferent collective choices. Anim. Behav. 132, 21–28 (2017).
Google Scholar 
De Agrò, M., Grimwade, D., Bach, R. & Czaczkes, T. J. Irrational risk aversion in an ant. Anim. Cogn. 1, 1–9 (2021).
Google Scholar 
Waddington, K. D., Allen, T. & Heinrich, B. Floral preferences of bumblebees (Bombus edwardsii) in relation to intermittent versus continuous rewards. Anim. Behav. 29(3), 779–784 (1981).
Google Scholar 
Cartar, R. V. A test of risk-sensitive foraging in wild bumble bees. Ecology 72(3), 888–895 (1991).
Google Scholar 
Perez, S. M. & Waddington, K. D. Carpenter bee (Xylocopa micans) risk indifference and a review of nectarivore risk-sensitivity studies. Am. Zool. 36(4), 435–446 (1996).
Google Scholar 
Fülöp, A. & Menzel, R. Risk-indifferent foraging behaviour in honeybees. Anim. Behav. 60(5), 657–666 (2000).PubMed 

Google Scholar 
Burns, D. D., Sendova-Franks, A. B. & Franks, N. R. The effect of social information on the collective choices of ant colonies. Behav. Ecol. 27(4), 1033–1040 (2016).
Google Scholar 
Sasaki, T., Pratt, S. C. & Kacelnik, A. Parallel vs. comparative evaluation of alternative options by colonies and individuals of the ant Temnothorax rugatulus. Sci. Rep. 8(1), 1–8 (2018).
Google Scholar 
Sasaki, T., Stott, B. & Pratt, S. C. Rational time investment during collective decision making in Temnothorax ants. Biol. Lett. 15(10), 20190542 (2019).PubMed 
PubMed Central 

Google Scholar 
Freas, C. A., Fleischmann, P. N. & Cheng, K. Experimental ethology of learning in desert ants: Becoming expert navigators. Behav. Proc. 158, 181–191 (2019).
Google Scholar 
Le Moël, F. & Wystrach, A. Towards a multi-level understanding in insect navigation. Curr. Opin. Inst. Sci. 42, 110–117 (2020).
Google Scholar 
Heinze, S. Visual navigation: Ants lose track without mushroom bodies. Curr. Biol. 30(17), R984–R986 (2020).CAS 
PubMed 

Google Scholar 
Ardin, P., Peng, F., Mangan, M., Lagogiannis, K. & Webb, B. Using an insect mushroom body circuit to encode route memory in complex natural environments. PLOS Comp. Biol. 12(2), e1004683 (2016).ADS 

Google Scholar 
Buehlmann, C. et al. Mushroom bodies are required for learned visual navigation, but not for innate visual behavior, in ants. Curr. Biol. 30(17), 3438–3443 (2020).CAS 
PubMed 

Google Scholar 
Kamhi, J. F., Barron, A. B. & Narendra, A. Vertical lobes of the mushroom bodies are essential for view-based navigation in Australian Myrmecia ants. Curr. Biol. 30(17), 3432–3437 (2020).CAS 
PubMed 

Google Scholar 
Heisenberg, M. Mushroom body memoir: From maps to models. Nat. Rev. Neurosci. 4(4), 266–275 (2003).CAS 
PubMed 

Google Scholar 
Webb, B. & Wystrach, A. Neural mechanisms of insect navigation. Curr. Opin. Inst. Sci. 15, 27–39 (2016).
Google Scholar 
Habenstein, J., Amini, E., Grübel, K., El Jundi, B. & Rössler, W. The brain of Cataglyphis ants: Neuronal organization and visual projections. J. Comp. Neurol. 528(18), 3479–3506 (2020).PubMed 

Google Scholar 
Cohn, R., Morantte, I. & Ruta, V. Coordinated and compartmentalized neuromodulation shapes sensory processing in Drosophila. Cell 163(7), 1742–1755 (2015).CAS 
PubMed 
PubMed Central 

Google Scholar 
Aso, Y. & Rubin, G. M. Dopaminergic neurons write and update memories with cell-type-specific rules. Elife 5, e16135 (2015).
Google Scholar 
Beck, C. D. O., Schroeder, B. & Davis, R. L. Learning performance of normal and mutant Drosophila after repeated conditioning trials with discrete stimuli. J. Neurosci. 20(8), 2944–2953 (2000).CAS 
PubMed 
PubMed Central 

Google Scholar 
Boto, T. & Ramaswami, M. Learning and memory: Clashing engrams in the fly brain. Curr. Biol. 31(16), R1009–R1011 (2021).CAS 
PubMed 

Google Scholar 
Bennett, J. E. M., Philippides, A. & Nowotny, T. Learning with reinforcement prediction errors in a model of the Drosophila mushroom body. Nat. Commun. 12, 22595 (2021).
Google Scholar 
Rescorla, R. A. & Wagner, A. R. A theory of Pavlovian conditioning: Variations in the effectiveness of reinforcement and nonreinforcement. In Classical Conditioning Ii: Current Theory and Research (eds Black, A. & Prokasy, W.) (Appleton-Century-Crofts, 1972).
Google Scholar  More

Study islands and bird dataSystematic ringing in spring on Mediterranean islands has been promoted by the Piccole Isole project since 198826. Standard methods of the project involve ringing between 16th April and 15th May attempting to include the peak of the spring passage of long-distance migrants. Ringing is performed from dawn to nightfall using a constant number of nets within ringing stations placed at stable sites located at representative habitats in each island (Supplementary Table S1). The use of tape-lures is not allowed. We have compiled ringing data for all the Spanish Mediterranean islands that have been applying this methodology, with the exception of Mallorca and Menorca where the ringing stations were located in wetlands and captured a large percentage of local birds (Fig. 2, Table 1). The nine study islands are spread along a south-west to north-east gradient and, with the exception of Columbrets, they are distributed in pairs of similar longitude but different latitudes (Fig. 2). Ringing stations have been operating over a variable number of years (5–27 years), with the maximum number of ringing stations operating at the same time occurring between 2003 and 2010. To include between-year variation on islands that started ringing campaigns more recently we used data from the years 2003–2018.Figure 2Image source: Google Earth. Data SIO, NOAA, US Navy, NGA, GEBCO. Image Landsat/Copernicus.Geographical location of studied islands in the western Mediterranean.Full size imageTable 1 Period of activity of the ringing stations located on each island between the years 1992 and 2018.Full size tableThe ringing period within each spring also varied in most islands, owing to funding or logistic limitations; thus, to reduce the possible effects on migrant composition we only used data from the standard period of the Piccole Isole project and from years that included at least one week of ringing in the fortnight of each month within this interval. This procedure excluded the use of some years for several islands, and the final number of data years for islands ranged between 5 and 16 (Table 1).We used only data for trans-Saharan nocturnal migrant passerines, which form the bulk of species ringed on Mediterranean islands during the standard period. The standard ringing period only covers the tail end of the short-distance migrants’ passage; thus, these species were excluded as their contribution to composition of migrants could vary mainly due to between-year variation in migration phenology. Diurnal migrants, like hirundinids and fringillids, also represent a small fraction of birds ringed and may use different cues to select stopover islands. In addition, some of these species nest in some of the islands studied and birds ringed could include breeding birds. To avoid the distorting effect of species that are captured accidentally in very small numbers, we considered only the species that were ringed in at least five separate years, or on five different islands, which limited the species considered to 35 (Supplementary Table S2). This led to the exclusion of just two species (Ficedula semitorquata with three individuals ringed in two islands and Locustella luscinioides with one individual ringed in Aire island). In addition, we only considered the number of ringed birds, since the proportion of recaptures varies among islands, likely reflecting variation in the duration of stopovers21, which could bias the comparison of the patterns of migrant species composition.Island descriptorsWe obtained two groups of variables describing the characteristics of the study islands (Tables 2, 3): (1) Variables related to geographical location: latitude, longitude, straight distance and minimum distance to the North African coast, minimum distance to the closest large body of land (continent or large island) in any direction and to the closest large body of land situated in a southerly angle between SW and SE. (2) Variables related to the habitat characteristics of the islands: area, maximum altitude and Normalized Difference Vegetation Index (NDVI). We estimated NDVI from Landsat 8 Images taken during the standard ringing period in the years 2015 and 2016. Pixels containing shoreline were excluded and the average NDVI was calculated for the rest of the pixels.Table 2 Variables describing the characteristics of the islands that included the ringing stations studied.Full size tableTable 3 Values of the island descriptors (see Table 2) and two measures of temporal variability of migrant composition in each island: average local contribution of each island to beta diversity (LCBD) and beta diversity for each island (BDTi).Full size tableContinental abundance dataAbundance estimates for western Europe were obtained from the European Red List of Birds27. We used the mean of the minimum and maximum number of pairs estimated for the 27 EU Member States as a measure of continental abundance (Supplementary Table S2).Data analysisAll analyses were done using R 3.6.128. We built a matrix of island-year x species containing the number of individuals of each selected species ringed in the study period in each island and year. Average number of individuals of each species ringed at each island was calculated and was used (log-transformed) as a dependent variable in a linear model with continental abundance (log-transformed), island and their interaction as predictors. This model was simplified using AICc as criteria to identify the best model.To analyze variation of species composition, the matrix of island-year x species was transformed using the chord transformation29 with the function decostand in the vegan package30.Using the function beta.div of the adespatial package31 we calculated beta diversity, including temporal and between-island variability (BDI,T), as the total variance of the aforementioned transformed matrix (BDTotal in29), which varies between 0 and 1 when chord distance is used. Considering that yijk is the chord transformed abundance of the species j in the island i and year k and (overline{{y }_{j}}) is the mean for species j in all islands and years altogether, then:$${SS}_{Total}=sum_{i=1}^{n}sum_{j=1}^{p}{sum_{k=1}^{q}{({y}_{ijk}-{overline{y} }_{j})}^{2}}$$$$BD_{I,T} = , SS_{Total} /left( {N – 1} right)$$where N is the total number of samples. The function beta.div also provides an estimation of contribution of localities (LCBD) and species (SCBD) to beta diversity (Table 3). Yearly LCBD (log transformed because of skewed distribution) of each island were averaged and compared between islands using ANOVA and a post-hoc Tukey test.We partitioned the above sum of squares in several ways. First, we calculated a beta diversity that considered only between-island variability, excluding temporal variability (BDI), by averaging the chord transformed abundances of each species j in each island along study years (({overline{y} }_{ij})) and applying the same procedure, but using the number of studied islands (n):$${SS}_{I}=sum_{i=1}^{n}sum_{j=1}^{p}{{({overline{y} }_{ij}-{overline{y} }_{j})}^{2}}$$$$BD_{I} = SS_{I} /left( {n – 1} right)$$Second, we calculated a beta diversity due to inter-annual variation of migrant composition within islands (BDT) as:$${SS}_{Temp}=sum_{i=1}^{n}sum_{j=1}^{p}{sum_{k=1}^{q}{({y}_{ijk}-{overline{y} }_{ij})}^{2}}$$$$BD_{T} = SS_{Temp} /left( {Y – n} right)$$where Y is the total number of study years and n is the number of studied islands (9). We also calculated a temporal beta diversity for each island i (BDTi) as the sum of squares due to variation within the island divided by the number of the island study years (Yi) minus 1:$${SS}_{Temp,i}=sum_{j=1}^{p}sum_{k=1}^{q}{({y}_{ijk}-{overline{y} }_{ij})}^{2}$$$$BD_{Ti} = SS_{Temp,i} /left( {Y_{i} – 1} right)$$Differences in temporal variability between islands could be due to different predominance of species that are more or less variable between years. To check this, we calculated Spearman’s rank correlation between the percentage of captures of each species in the total ringed on each island and BDTi and LCDB indices, for species present on all islands.We tested for the existence of differences between islands in migrant species composition using Permutational Multivariate Analysis of Variance (PERMANOVA) using the function adonis2 in the vegan package. We performed a multivariate test of homogeneity of variances using the betadisper function (vegan package) with the adjustment for small sample bias, to test if temporal variability in species composition differed between islands. We made post-hoc comparisons between islands with False Discovery Rate (FDR) correction using the function pairwise.perm.manova of the package RVAideMemoire32.To identify gradients in migrant species composition and the island characteristics that were associated with them, we employed Redundancy Analysis using the rda function (vegan package). We used the chord transformed matrix of species x island-year as a response matrix. We used two explanatory matrices, one including variables of geographical location and the other the variables related to habitat characteristics of the islands. We evaluated the relative importance of each group of variables to explain migrant species composition by performing a variation partitioning analysis, using the varpart function (vegan package). For that analysis, we followed the steps and R scripts recommended in33.Variables describing island characteristics were transformed using natural logarithms and collinearity within each group was evaluated with variance inflation factor (VIF)34. All the habitat variables presented VIF  More

Eddy-covariance observationsWe used half-hourly or hourly GPP, air temperature, VPD, SWC and incoming shortwave radiation from the recently released ICOS (Integrated Carbon Observation System)44 and the FLUXNET2015 dataset of energy, water, and carbon fluxes and meteorological data, both of which have undergone a standardized set of quality control and gap filling19. Data were already processed following a consistent and uniform processing pipeline19. This data processing pipeline mainly included: (1) thorough data quality control checks; (2) calculation of a range of friction velocity thresholds; (3) gap-filling of meteorological and flux measurements; (4) partitioning of CO2 fluxes into respiration and photosynthesis components; and (5) calculation of a correction factor for energy fluxes19. All the corrections listed were already applied to the available product19. We used incoming shortwave radiation, temperature, VPD, and SWC that were gap-filled using the marginal distribution method21. The GPP estimates from the night-time partitioning method were used for the analysis (GPP_NT_VUT_REF). SWC was measured as volumetric SWC (percentage) at different depths, varying across sites. We mainly used the surface SWC observations but deeper SWC measurements were also used when available. Data were quality controlled so that only measured and good-quality gap filled data (QC = 0 or 1) were used.Analysis of the extreme summer drought in 2018 in Europe to prove nonlinearityTo analyze the effect of summer drought in 2018 on GPP in Europe, we selected 15 sites with measurements during 2014–2018 from the ICOS dataset, representing the major ecosystems across Europe (Supplementary Table 1). Croplands were excluded due to the effect of management on the seasonal timing of ecosystem fluxes, both from crop rotation that change from year to year and from the variable timing of planting and harvesting. In croplands, the changes of GPP anomalies across different growing season could be mainly depend on crop varieties and management activities. Information of crop varieties, growing times yearly and other management data for each cropland site should be collected in future in order to fully consider and disentangle the impacts of SWC and VPD on its photosynthesis. Wetland sites were also removed because they are influenced by upstream organic matter and nutrient input, as well as fluctuating water tables. Daytime half-hourly data (7 am to 19 pm) were aggregated to daily values. At each site, the relative changes ((triangle {{{{{rm{X}}}}}})) of summer (June–July–August) GPP, SWC and VPD during 2014–2018 refer to the summer average of 2014–2018 were calculated for each year. For example, the calculation of the relative change in 2018 is shown in Eq. (1):$$triangle {{{{{rm{X}}}}}}=frac{{X}_{2018}-,{X}_{{average};{of};2014-2018}}{{X}_{{average};{of};2014-2018}}times 100 %$$
(1)
where X2018 is the mean of the daily values of (X) (GPP, SWC, or VPD) during the summer of 2018, and Xaverage of 2014–2018 is the mean of the daily values of (X) over all the summers of the 2014–2018 period. The average (triangle {{{{{rm{X}}}}}}) across a certain number of sites at each bin were used for the results in Fig. 1a.Daily time series of GPP, SWC and VPD during summer for each site were normalized (z-scores) to derive the standardized sensitivity of GPP to SWC and VPD. For each variable, the mean value across the summer of 2014–2018 was subtracted for each day at each site and then normalized by its standard deviation. At each site, we used a multiple linear regression (Eq. 2) to estimate daily GPP anomalies sensitivities to SWC and VPD anomalies across 2014–2018 and 2014–2017, respectively:$${GPP}={beta }_{1},{SWC}+{beta }_{2},{VPD}+{beta }_{3},{SWC},times {VPD}+{beta }_{4},{T}_{a}+{beta }_{5}{RAD}+b+varepsilon$$
(2)
where ({beta }_{i}) is the standardized sensitivity of GPP to each variable; ({T}_{a}) represents the air temperature; ({RAD}) represents the incoming shortwave radiation;(,b) represents the intercept; and (varepsilon) is the random error term. We compared estimated sensitivities with and without 2018 data to quantify the impacts of extreme drought in 2018 on GPP sensitivity to SWC (Fig. 1d) and VPD (Fig. 1e). The slope was calculated at each site and then the distribution of slopes across sites were plotted in Fig. 1d, e.Global analysis of the sensitivities of GPP to SWC and VPDFor the global analysis, instead of summer, we focused on the growing season and days when the SWC and VPD effects were most likely to control ecosystem fluxes and screen out days when other meteorological drivers were likely to have a larger influence on fluxes. Following previous studies5,8,45, for each site, we restrict our analyses to the days in which: (i) the daily average temperature >15 °C; (ii) sufficient evaporative demand existed to drive water fluxes, constrained as daily average VPD  > 0.5 kPa; (iii) high solar radiation, constrained as daily average incoming shortwave radiation >250 Wm−2.By combining ICOS and FLUXNET2015 data, at the global scale, we evaluated 67 sites with at least 300 days observations over the growing seasons for the years available (Supplementary Table 2). We excluded cropland and wetland sites for the above-mentioned reasons. These 67 sites were used to calculate the relative effects of low SWC and high VPD on GPP following the approach of ref. 5 (see below sections). For 8 sites, the ANN results failed performance criteria (the correlation between predicted GPP and observed GPP is {{VPD}}_{0}\ {beta }_{0},,{VPD}le {{VPD}}_{0}end{array}right.$$
(7)
where β0 and k are fitted parameters and VPD0 is 1 kPa48. Following Luo and Keenan48, we applied this method to a short time window (2–14 days) of Fc depending on the availability of flux measurements and assumed that every day in the same time window has the same daily Amax. We retrieved the daily Amax by implementing Eqs. (6) and (7) using the REddyProc R package (https://github.com/bgctw/REddyProc)20.Vcmax represents the activity of the primary carboxylating enzyme ribulose 1,5-bisphosphate carboxylase–oxygenase (Rubisco) as measured under light-saturated conditions. To evaluate the responses of Vcmax to SWC and VPD, we first calculated the daily internal leaf CO2 partial pressure (ci) in the middle of the day (11:00–14:00) via Fick’s Law (Eq. 8), excluding periods with low incoming shortwave radiation (0.7 at most sites. During the training process, weight and bias values were optimized using the Levenberg–Marquardt optimization58,59. The maximum number of epochs to train is 1000. An example to demonstrate the ANN training at one site was shown in Supplementary Fig. 3.At each site, ANN was run and sensitivities were calculated for all data within each SWC and VPD bin and the median value was used. For each of the five trained ANNs, one of the predictor variables was perturbed by one standard deviation (a value of 1 due to the initial input data normalization), and GPP was predicted again using the existing ANN with the predictors including the perturbed variable; this process was repeated for each predictor variable. The predicted values of GPP obtained with and without perturbation were then compared to determine the sensitivity values. The sample equation showing the calculation of the GPP sensitivity to VPD is shown in Eq. (10).$${{{{{{rm{Sensitivity}}}}}}}_{{VPD}}={median}left(,frac{{{GPP}}_{left({ANN};{VPD}+{stdev}left({VPD}right)right)}-{{GPP}}_{left({ANN};{all};{VAR}right)}}{{stdev}left({VPD}right)}right)$$
(10)
We repeated the ANN and sensitivity analyses five times and the median of these were used at each site. Across all sites, significances of the sensitivities for each bin were tested using t-tests (p  More

Mack, R. N. et al. Biotic invasions: causes, epidemiology, global consequences, and control. Ecol. Appl. 10, 689–710. https://doi.org/10.1890/1051-0761 (2000).Article 

Google Scholar 
Dukes, J. S. & Mooney, H. A. Disruption of ecosystem processes in western North America by invasive species. Rev. Chil. Hist. Nat. 77, 411–437 (2004).Article 

Google Scholar 
Vitousek, P. M. Biological invasions and ecosystem processes: towards an integration of population biology and ecosystem studies. Oikos 57, 7–13. https://doi.org/10.2307/3565731 (1990).Article 

Google Scholar 
Richardson, D. M. et al. Naturalization and invasion of alien plants: concepts and definitions. Diver. Distrib. 6, 93–107 (2000).Article 

Google Scholar 
Theoharides, K. A. & Dukes, J. S. Plant invasion across space and time: factors affecting nonindigenous species success during four stages of invasion. New Phytol. 176, 256–273 (2007).Article 

Google Scholar 
Pyšek, P. et al. Naturalization of central European plants in North America: species traits, habitats, propagule pressure, residence time. Ecology 96, 762–774. https://doi.org/10.1890/14-1005.1 (2015).Article 
PubMed 

Google Scholar 
Estrada, J. A., Wilson, C. H. & Flory, S. L. Clonal integration enhances performance of an invasive grass. Oikos https://doi.org/10.1111/oik.07016 (2020).Article 

Google Scholar 
Otfinowski, R. & Kenkel, N. C. Clonal integration facilitates the proliferation of smooth brome clones invading northern fescue prairies. Plant Ecol. 199, 235–242. https://doi.org/10.1007/s11258-008-9428-8 (2008).Article 

Google Scholar 
Pyšek, P. & Richardson, D. M. in Biological Invasions (ed N. Nentwig) pp. 97–125 (Springer, New York, 2007).Klimešová, J. & Klimeš, L. Clonal growth diversity and bud banks of plants in the Czech flora: an evaluation using the CLO-PLA3 database. Preslia 80, 255–275 (2008).
Google Scholar 
Klimešová, J. et al. Handbook of standardized protocols for collecting plant modularity traits. Persp. Plant Ecol. https://doi.org/10.1016/j.ppees.2019.125485 (2019).Article 

Google Scholar 
Wang, Y. J. et al. Invasive alien plants benefit more from clonal integration in heterogeneous environments than natives. New Phytol. 216, 1072–1078 (2017).Article 

Google Scholar 
Klimešová, J. in Encyclopedia of Invasive Introduced Species (eds D. Simberloff & M. Reimanek) pp. 678–679 (University of California Press, California, 2011).Ott, J. P., Klimešová, J. & Hartnett, D. C. The ecology and significance of below-ground bud banks in plants. Ann. Bot. Lond. 123, 1099–1118. https://doi.org/10.1093/aob/mcz051 (2019).Article 

Google Scholar 
Sanchez, J. M., Sanchez, C. & Navarro, L. Can asexual reproduction by plant fragments help to understand the invasion of the NW Iberian coast by Spartina patens? Flora 257, 151410. https://doi.org/10.1016/j.flora.2019.05.009 (2019).Speek, T. A. A. et al. Factors relating to regional and local success of exotic plant species in their new range. Diver. Distrib. 17, 542–551 (2011).Article 

Google Scholar 
Wang, J. Y. et al. A meta-analysis of effects of physiological integration in clonal plants under homogeneous vs heterogeneous environments. Funct. Ecol. https://doi.org/10.1111/1365-2435.13732 (2020).Article 

Google Scholar 
Maurer, D. A. & Zedler, J. B. Differential invasion of a wetland grass explained by tests of nutrients and light availability on establishment and clonal growth. Oecologia 131, 279–288. https://doi.org/10.1007/s00442-002-0886-8 (2002).ADS 
Article 
PubMed 

Google Scholar 
Mueller, I. M. & Weaver, J. E. Relative drought resistance of seedlings of dominant prairie grasses. Ecology 23, 387–398 (1942).Article 

Google Scholar 
Vetter, V. M. S. et al. Invasion windows for a global legume invader are revealed after joint examination of abiotic and biotic filters. Plant Biol. 21, 832–843. https://doi.org/10.1111/plb.12987 (2019).CAS 
Article 
PubMed 

Google Scholar 
Ibanez, I. et al. Integrated assessment of biological invasions. Ecol. Appl. 24, 25–37. https://doi.org/10.1890/13-0776.1 (2014).Article 
PubMed 

Google Scholar 
Diez, J. M. et al. Will extreme climatic events facilitate biological invasions?. Front. Ecol. Environ. 10, 249–257. https://doi.org/10.1890/110137 (2012).Article 

Google Scholar 
Davis, M. A., Grime, J. P. & Thompson, K. Fluctuating resources in plant communities: a general theory of invasibility. J. Ecol. 88, 528–534. https://doi.org/10.1046/j.1365-2745.2000.00473.x (2000).Article 

Google Scholar 
Li, W. & Stevens, M. H. H. Fluctuating resource availability increases invasibility in microbial microcosms. Oikos 121, 435–441. https://doi.org/10.1111/j.1600-0706.2011.19762.x (2012).Article 

Google Scholar 
Koerner, S. E. et al. Invasibility of a mesic grassland depends on the time-scale of fluctuating resources. J. Ecol. 103, 1538–1546. https://doi.org/10.1111/1365-2745.12479 (2015).Article 

Google Scholar 
Hendrickson, J. R. & Lund, C. Plant community and target species affect responses to restoration strategies. Rangel. Ecol. Manag. 63, 435–442 (2010).Article 

Google Scholar 
Bennett, J., Smart, A. & Perkins, L. Using phenological niche separation to improve management in a Northern Glaciated Plains grassland. Restor. Ecol. 27, 745–749. https://doi.org/10.1111/rec.12932 (2019).Article 

Google Scholar 
Jordan, N. R., Larson, D. L. & Huerd, S. C. Soil modification by invasive plants: effects on native and invasive species of mixed-grass prairies. Biol. Invas. 10, 177–190. https://doi.org/10.1007/s10530-007-9121-1 (2008).Article 

Google Scholar 
Piper, C. L., Lamb, E. G. & Siciliano, S. D. Smooth brome changes gross soil nitrogen cycling processes during invasion of a rough fescue grassland. Plant Ecol. 216, 235–246. https://doi.org/10.1007/s11258-014-0431-y (2015).Article 

Google Scholar 
Stotz, G. C., Gianoli, E. & Cahill, J. F. Biotic homogenization within and across eight widely distributed grasslands following invasion by Bromus inermis. Ecology https://doi.org/10.1002/ecy.2717 (2019).Article 
PubMed 

Google Scholar 
Dillemuth, F. P., Rietschier, E. A. & Cronin, J. T. Patch dynamics of a native grass in relation to the spread of invasive smooth brome (Bromus inermis). Biol. Invas. 11, 1381–1391. https://doi.org/10.1007/s10530-008-9346-7 (2009).Article 

Google Scholar 
Trammell, M. A. & Butler, J. L. Effects of exotic plants on native ungulate use of habitat. J. Wildlife Manag. 59, 808–816. https://doi.org/10.2307/3801961 (1995).Article 

Google Scholar 
Gibson, D. J. Grasses and Grassland Ecology (Oxford Univ. Press, 2009).
Google Scholar 
Knapp, A. K. & Smith, M. D. Variation among biomes in temporal dynamics of aboveground primary production. Science 291, 481–484. https://doi.org/10.1126/science.291.5503.481 (2001).ADS 
CAS 
Article 
PubMed 

Google Scholar 
Easterling, D. R. et al. Precipitation change in the United States. pp. 207–230 (Washington, D.C. USA, 2017).Gutschick, V. P. & BassiriRad, H. Extreme events as shaping physiology, ecology, and evolution of plants: toward a unified definition and evaluation of their consequences. New Phytol. 160, 21–42. https://doi.org/10.1046/j.1469-8137.2003.00866.x (2003).Article 
PubMed 

Google Scholar 
Briske, D. D. in Grazing management: An ecological perspective (eds R.K. Heitschmidt & J.W. Stuth) pp. 85–108 (Timber Press, Inc., 1991).Liu, F., Liu, J. & Dong, M. Ecological consequences of clonal integration in plants. Front. Plant Sci. 217, 277–287 (2016).
Google Scholar 
Hoover, D. L., Knapp, A. K. & Smith, M. D. Resistance and resilience of a grassland ecosystem to climate extremes. Ecology 95, 2646–2656. https://doi.org/10.1890/13-2186.1 (2014).Article 

Google Scholar 
VanderWeide, B. L., Hartnett, D. C. & Carter, D. L. Belowground bud banks of tallgrass prairie are insensitive to multi-year, growing-season drought. Ecosphere. https://doi.org/10.1890/Es14-00058.1 (2014).Article 

Google Scholar 
VanderWeide, B. L. & Hartnett, D. C. Belowground bud bank response to grazing under severe, short-term drought. Oecologia 178, 795–806. https://doi.org/10.1007/s00442-015-3249-y (2015).ADS 
Article 
PubMed 

Google Scholar 
Ott, J. P., Butler, J. L., Rong, Y. P. & Xu, L. Greater bud outgrowth of Bromus inermis than Pascopyrum smithii under multiple environmental conditions. J. Plant Ecol. 10, 518–527. https://doi.org/10.1093/jpe/rtw045 (2017).Article 

Google Scholar 
Oesterheld, M., Loreti, J., Semmartin, M. & Sala, O. E. Inter-annual variation in primary production of a semi-arid grassland related to previous-year production. J. Veg. Sci. 12, 137–142. https://doi.org/10.1111/j.1654-1103.2001.tb02624.x (2001).Article 

Google Scholar 
Ott, J. P. & Hartnett, D. C. Bud bank dynamics and clonal growth strategy in the rhizomatous grass, Pascopyrum smithii. Plant Ecol. 216, 395–405. https://doi.org/10.1007/s11258-014-0444-6 (2015).Article 

Google Scholar 
Carlsson, B. A. & Callaghan, T. V. Programmed tiller differentiation, intraclonal density regulation and nutrient dynamics in Carex bigelowii. Oikos 58, 219–230. https://doi.org/10.2307/3545429 (1990).Article 

Google Scholar 
Ye, X. H., Yu, F. H. & Dong, M. A trade-off between guerrilla and phalanx growth forms in Leymus secalinus under different nutrient supplies. Ann. Bot. Lond. 98, 187–191. https://doi.org/10.1093/aob/mcl086 (2006).Article 

Google Scholar 
Dibbern, J. C. Vegetative responses of Bromus inermis to certain variations in environment. Bot. Gazette 109, 44–58 (1947).Article 

Google Scholar 
Dong, X., Patton, J., Wang, G., Nyren, P. & Peterson, P. Effect of drought on biomass allocation in two invasive and two native grass species dominating the mixed-grass prairie. Grass Forage Sci. 69, 160–166. https://doi.org/10.1111/gfs.12020 (2014).Article 

Google Scholar 
Saeidnia, F., Majidi, M. M., Mirlohi, A. & Soltan, S. Physiological and tolerance indices useful for drought tolerance selection in smooth bromegrass. Crop Sci. 57, 282–289. https://doi.org/10.2135/cropsci2016.07.0636 (2017).CAS 
Article 

Google Scholar 
Vinton, M. A. & Hartnett, D. C. Effects of bison grazing on Andropogon gerardii and Panicum virgatum in burned and unbruned tallgrass prairie. Oecologia 90, 374–382. https://doi.org/10.1007/bf00317694 (1992).ADS 
CAS 
Article 
PubMed 

Google Scholar 
Eneboe, E. J., Sowell, B. F., Heitschmidt, R. K., Karl, M. G. & Haferkamp, M. R. Drought and grazing: IV. Blue grama and western wheatgrass. J. Range Manag. 55, 197–203. https://doi.org/10.2307/4003357 (2002).Article 

Google Scholar 
Broadbent, T. S., Bork, E. W. & Willms, W. D. Divergent effects of defoliation intensity and frequency on tiller growth and production dynamics of Pascopyrum smithii and Hesperostipa comata. Grass Forage Sci. 73, 532–543. https://doi.org/10.1111/gfs.12318 (2018).Article 

Google Scholar 
Donkor, N. T., Bork, E. W. & Hudson, R. J. Bromus-Poa response to defoliation intensity and frequency under three soil moisture levels. Can. J. Plant Sci. 82, 365–370. https://doi.org/10.4141/p01-040 (2002).Article 

Google Scholar 
Reynolds, J. H. & Smith, D. Trend of carbohydrate reserves in alfalfa, smooth bromegrass, and timothy grown under various cutting schedules. Crop Sci. 2, 333–336 (1962).CAS 
Article 

Google Scholar 
Lamp, H. F. Reproductive activity in Bromus inermis in relation to phases of tiller development. Bot. Gazette 113, 413–438 (1952).Article 

Google Scholar 
Paulsen, G. M. & Smith, D. Organic reserves, axillary bud activity, and herbage yields of smooth bromegrass as influenced by time of cutting, nitrogen fertilization, and shading. Crop Sci. 9, 529–534 (1969).Article 

Google Scholar 
Ott, J. P. & Hartnett, D. C. Contrasting bud bank dynamics of two co-occurring grasses in tallgrass prairie: implications for grassland dynamics. Plant Ecol. 213, 1437–1448. https://doi.org/10.1007/s11258-012-0102-9 (2012).Article 

Google Scholar 
Busso, C. A., Mueller, R. J. & Richards, J. H. Effects of drought and defoliation on bud viability in 2 caespitose grasses. Ann. Bot. Lond. 63, 477–485. https://doi.org/10.1093/oxfordjournals.aob.a087768 (1989).Article 

Google Scholar 
Tuomi, J., Nilsson, P. & Astrom, M. Plant compensatory responses-bud dormancy as an adaptation to herbivory. Ecology 75, 1429–1436. https://doi.org/10.2307/1937466 (1994).Article 

Google Scholar 
US Department of Agriculture. The PLANTS Database, (2006).Gong, K. et al. Analysis on the distribution, breeding and utilization of Bromus inermis germplasm resource in China. Heilongjiang Anim. Sci. Vet. Med. 21, 33–36 (2019).
Google Scholar 
Coupland, R. T. & Johnson, R. E. Rooting characteristics of native grassland species in Saskatchewan. J. Ecol. 53, 475–507 (1965).Article 

Google Scholar 
Gist, G. R. & Smith, R. M. Root development of several common forage grasses to a depth of eighteen inches. Agron. J. 1036–1042 (1948).Okamoto, H., Ishii, K. & An, P. Effects of soil moisture deficit and subsequent watering on the growth of four temperate grasses. Grassl. Sci. 57, 192–197. https://doi.org/10.1111/j.1744-697X.2011.00232.x (2011).Article 

Google Scholar 
Morrow, L. A. & Power, J. F. Effect of soil temperature on development of perennial forage grasses. Agron. J. 71, 7–10 (1979).Article 

Google Scholar 
Duell, E. B., Wilson, G. W. T. & Hickman, K. R. Above- and below-ground responses of native and invasive prairie grasses to future climate scenarios. Botany 94, 471–479. https://doi.org/10.1139/cjb-2015-0238 (2016).Article 

Google Scholar 
Duell, E. B., Londe, D. W., Hickman, K. R., Greer, M. J. & Wilson, G. W. T. Superior performance of invasive grasses over native counterparts will remain problematic under warmer and drier conditions. Plant Ecol. 222, 993–1006 (2021).Article 

Google Scholar 
Cully, A. C., Cully, J. F. & Hiebert, R. D. Invasion of exotic plant species in tallgrass prairie fragments. Conser. Biol. 17, 990–998. https://doi.org/10.1046/j.1523-1739.2003.02107.x (2003).Article 

Google Scholar 
DeKeyser, E. S., Meehan, M., Clambey, G. & Krabbenhoft, K. Cool season invasive grasses in northern great plains natural areas. Nat. Areas J. 33, 81–90. https://doi.org/10.3375/043.033.0110 (2013).Article 

Google Scholar 
Grant, T. A., Shaffer, T. L. & Flanders, B. Resiliency of native prairies to invasion by kentucky bluegrass, smooth brome, and woody vegetation. Rangeland Ecol. Manag. 73, 321–328. https://doi.org/10.1016/j.rama.2019.10.013 (2020).Article 

Google Scholar 
Otfinowski, R., Kenkel, N. C. & Catling, P. M. The biology of Canadian weeds. 134. Bromus inermis Leyss. Can. J. Plant Sci. 87, 183–198. https://doi.org/10.4141/p06-071 (2007).Article 

Google Scholar 
Moore, K. J. et al. Describing and quantifying growth stages of perennial forage grasses. Agron. J. 83, 1073–1077 (1991).Article 

Google Scholar 
SAS Institute. SAS 9.4. (SAS Institute Inc, 2017). More

Laurance, W. F. et al. A global strategy for road building. Nature 513, 229–232 (2014).ADS 
CAS 
PubMed 

Google Scholar 
Weng, L. et al. Mineral industries, growth corridors and agricultural development in Africa. Glob. Food Sec. 2, 195–202 (2013).
Google Scholar 
Laurance, W. F., Goosem, M. & Laurance, S. G. W. Impacts of roads and linear clearings on tropical forests. Trends Ecol. Evol. 24, 659–669 (2009).PubMed 

Google Scholar 
Trombulak, S. C. & Frissell, C. A. Review of ecological effects of roads on terrestrial and aquatic communities. Conserv. Biol. 14, 18–30 (2000).
Google Scholar 
van der Ree, R., Smith, D. J. & Grilo, C. The ecological effects of linear infrastructure and traffic. in Handbook of road ecology 1–9 (John Wiley and Sons, Ltd., 2015). https://doi.org/10.1002/9781118568170.ch1.Grilo, C., Smith, D. J. & Klar, N. Carnivores: Struggling for survival in roaded landscapes. in Handbook of road ecology 300–312 (John Wiley and Sons, Ltd., 2015). doi:https://doi.org/10.1002/9781118568170.ch35.Wallach, A. D., Izhaki, I., Toms, J. D., Ripple, W. J. & Shanas, U. What is an apex predator?. Oikos 124, 1453–1461 (2015).
Google Scholar 
Ripple, W. J. et al. Status and ecological effects of the world’s largest carnivores. Science 343, (2014).Estes, J. A. et al. Trophic downgrading of planet earth. Science 333, 301–306 (2011).ADS 
CAS 
PubMed 

Google Scholar 
Stolton, S. & Dudley, N. The New Lion Economy. Unlocking the value of lions and their landscapes. http://lionrecoveryfund.org/newlioneconomy (2019).Meijer, J. R., Huijbregts, M. A. J., Schotten, K. C. G. J. & Schipper, A. M. Global patterns of current and future road infrastructure. Environ. Res. Lett. 13, 064006 (2018). Data is available at http://www.globio.infoAscensão, F. et al. Environmental challenges for the belt and road initiative. Nat. Sustain. 1, 206–209 (2018).
Google Scholar 
Dulac, J. Global land transport infrastructure requirements – Estimating road and railway infrastructure capacity and costs to 2050. (International Energy Agency, 2013).Laurance, W. F. et al. Reducing the global environmental impacts of rapid infrastructure expansion. Curr. Biol. 25, R259–R262 (2015).CAS 
PubMed 

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

Google Scholar 
Laurance, W. F., Sloan, S., Weng, L. & Sayer, J. A. Estimating the environmental costs of Africa’s massive “development corridors”. Curr. Biol. 25, 3202–3208 (2015).CAS 
PubMed 

Google Scholar 
Sharma, R., Rimal, B., Stork, N., Baral, H. & Dhakal, M. Spatial assessment of the potential impact of infrastructure development on biodiversity conservation in lowland Nepal. ISPRS Int. J. Geo Inf. 7, 365 (2018).
Google Scholar 
IUCN. The IUCN red list of threatened species. Version 2018-1. http://www.iucnredlist.org (2018).Garrote, G. et al. Prediction of Iberian lynx road-mortality in southern Spain: A new approach using the MaxEnt algorithm. Anim. Biodivers. Conserv. 41, 217–225 (2018).
Google Scholar 
Parchizadeh, J. et al. Roads threaten Asiatic cheetahs in Iran. Curr. Biol. 28, R1141–R1142 (2018).CAS 
PubMed 

Google Scholar 
Crooks, K. R., Burdett, C. L., Theobald, D. M., Rondinini, C. & Boitani, L. Global patterns of fragmentation and connectivity of mammalian carnivore habitat. Philos. Trans. R. Soc. B Biol. Sci. 366, 2642–2651 (2011).
Google Scholar 
Kattan, G. et al. Range fragmentation in the spectacled bear Tremarctos ornatus in the northern Andes. Oryx 38, 155–163 (2004).
Google Scholar 
Valeix, M., Loveridge, A. J. & Macdonald, D. W. Influence of prey dispersion on territory and group size of African lions: A test of the resource dispersion hypothesis. Ecology 93, 2490–2496 (2012).PubMed 

Google Scholar 
Holderegger, R. & Giulio, M. D. The genetic effects of roads: a review of empirical evidence. Basic Appl. Ecol. 11, 522–531 (2010).
Google Scholar 
Riley, S. P. D. et al. A southern California freeway is a physical and social barrier to gene flow in carnivores. Mol. Ecol. 15, 1733–1741 (2006).CAS 
PubMed 

Google Scholar 
Proctor, M. F., McLellan, B. N., Strobeck, C. & Barclay, R. M. R. Genetic analysis reveals demographic fragmentation of grizzly bears yielding vulnerably small populations. Proc. R. Soc. B Biol. Sci. 272, 2409–2416 (2005).
Google Scholar 
Riley, S. P. D. et al. Individual behaviors dominate the dynamics of an urban mountain lion population isolated by roads. Curr. Biol. 24, 1989–1994 (2014).CAS 
PubMed 

Google Scholar 
Janečka, J. E. et al. Reduced genetic diversity and isolation of remnant ocelot populations occupying a severely fragmented landscape in southern Texas. Anim. Conserv. 14, 608–619 (2011).
Google Scholar 
Thatte, P., Joshi, A., Vaidyanathan, S., Landguth, E. & Ramakrishnan, U. Maintaining tiger connectivity and minimizing extinction into the next century: Insights from landscape genetics and spatially-explicit simulations. Biol. Cons. 218, 181–191 (2018).
Google Scholar 
Vaeokhaw, S. et al. Effects of a highway on the genetic diversity of Asiatic black bears. Ursus 2020, 1–15 (2020).
Google Scholar 
Benı́tez-López, A. et al. The impact of hunting on tropical mammal and bird populations. Science 356, 180–183 (2017).ADS 
PubMed 

Google Scholar 
Clements, G. R. et al. Where and how are roads endangering mammals in Southeast Asia’s forests?. PLoS ONE 9, e115376 (2014).ADS 
PubMed 
PubMed Central 

Google Scholar 
Sharma, K., Wright, B., Joseph, T. & Desai, N. Tiger poaching and trafficking in India: Estimating rates of occurrence and detection over four decades. Biol. Cons. 179, 33–39 (2014).
Google Scholar 
Espinosa, S., Branch, L. C. & Cueva, R. Road development and the geography of hunting by an Amazonian indigenous group: Consequences for wildlife conservation. PLoS ONE 9, e114916 (2014).ADS 
PubMed 
PubMed Central 

Google Scholar 
Wato, Y. A., Wahungu, G. M. & Okello, M. M. Correlates of wildlife snaring patterns in Tsavo west national park Kenya. Biol. Conserv. 132, 500–509 (2006).
Google Scholar 
Watson, F., Becker, M. S., McRobb, R. & Kanyembo, B. Spatial patterns of wire-snare poaching: Implications for community conservation in buffer zones around national parks. Biol. Cons. 168, 1–9 (2013).
Google Scholar 
Henschel, P., Hunter, L. T. B., Coad, L., Abernethy, K. A. & Mühlenberg, M. Leopard prey choice in the Congo Basin rainforest suggests exploitative competition with human bushmeat hunters. J. Zool. 285, 11–20 (2011).
Google Scholar 
Espinosa, S., Celis, G. & Branch, L. C. When roads appear jaguars decline: Increased access to an Amazonian wilderness area reduces potential for jaguar conservation. PLoS ONE 13, e0189740 (2018).PubMed 
PubMed Central 

Google Scholar 
Parsons, M. A., Newsome, T. M. & Young, J. K. The consequences of predators without prey. Front. Ecol. Environ. https://doi.org/10.1002/fee.2419 (2021).Article 

Google Scholar 
Caro, T., Dobson, A., Marshall, A. J. & Peres, C. A. Compromise solutions between conservation and road building in the tropics. Curr. Biol. 24, R722–R725 (2014).CAS 
PubMed 

Google Scholar 
Grilo, C. et al. Conservation threats from roadkill in the global road network. Glob. Ecol. Biogeogr. 30, 2200–2210 (2021).
Google Scholar 
Carter, N., Killion, A., Easter, T., Brandt, J. & Ford, A. Road development in Asia: assessing the range-wide risks to tigers. Sci. Adv. 6(18), eaaz9619 (2020).ADS 
PubMed 
PubMed Central 

Google Scholar 
Ceia-Hasse, A., Borda-de-Água, L., Grilo, C. & Pereira, H. M. Global exposure of carnivores to roads. Glob. Ecol. Biogeogr. 26, 592–600 (2017).
Google Scholar 
Gaveau, D. L. A. et al. Four decades of forest persistence, clearance and logging on Borneo. PLoS ONE 9, e101654 (2014).ADS 
PubMed 
PubMed Central 

Google Scholar 
Gerber, B. D., Karpanty, S. M. & Randrianantenaina, J. The impact of forest logging and fragmentation on carnivore species composition, density and occupancy in Madagascar’s rainforests. Oryx 46, 414–422 (2012).
Google Scholar 
Cullen, L. et al. Implications of fine-grained habitat fragmentation and road mortality for jaguar conservation in the Atlantic forest, Brazil. PLoS ONE 11, e0167372 (2016).PubMed 
PubMed Central 

Google Scholar 
Kirby, K. R. et al. The future of deforestation in the Brazilian Amazon. Futures 38, 432–453 (2006).
Google Scholar 
UNEP-WCMC & IUCN. Protected planet: The world database on protected areas. http://www.protectedplanet.net (2019).de la Torre, J. A., Gonzalez-Maya, J. F., Zarza, H., Ceballos, G. & Medellin, R. A. The jaguars spots are darker than they appear: assessing the global conservation status of the jaguar Panthera onca. Oryx 52, 300–315 (2017).
Google Scholar 
Coelho, L., Romero, D., Queirolo, D. & Guerrero, J. C. Understanding factors affecting the distribution of the maned wolf (Chrysocyon brachyurus) in South America: spatial dynamics and environmental drivers. Mamm. Biol. 92, 54–61 (2018).
Google Scholar 
Fearnside, P. M. Brazil’s Cuiabá- Santarém (BR-163) highway: The environmental cost of paving a soybean corridor through the Amazon. Environ. Manage. 39, 601–614 (2007).ADS 
PubMed 

Google Scholar 
Vetter, D., Hansbauer, M. M., Végvári, Z. & Storch, I. Predictors of forest fragmentation sensitivity in neotropical vertebrates: A quantitative review. Ecography 34, 1–8 (2011).
Google Scholar 
Morcatty, T. Q. et al. Illegal trade in wild cats and its link to Chinese-led development in central and South America. Conserv. Biol. 34, 1525–1535 (2020).PubMed 

Google Scholar 
Ramsar. Ngiri-Tumba-Maindombe. Ramsar Sites Information Service https://rsis.ramsar.org/ris/1784 (2017).Dobson, A. P. et al. Road will ruin Serengeti. Nature 467, 272–273 (2010).ADS 
CAS 
PubMed 

Google Scholar 
Riggio, J. et al. The size of savannah Africa: A lion’s (Panthera leo) view. Biodivers. Conserv. 22, 17–35 (2012).
Google Scholar 
Government of Nepal. Economic survey 2019/20. (Ministry of Finance, 2020).Jnawali, S. R. et al. The status of Nepal mammals: The national red list series. (Department of National Parks and Wildlife Conservation, 2011).Joshi, A. R. Nepal court blocks road construction in rhino stronghold of Chitwan Park. https://news.mongabay.com/2019/02/nepal-court-blocks-road-construction-in-rhino-stronghold-of-chitwan-park/ (2019).Government of Nepal. Conservation Landscapes of Nepal. (Ministry of Forest and Soil Conservation, 2016).Poudel, A. et al. Biological and socio-economic study in corridors of Terai Arc Landscape, Nepal. (Center for Policy Analysis; Development, 2013).Arlidge, W. N. S. et al. A Global Mitigation Hierarchy for Nature Conservation. Bioscience 68, 336–347 (2018).PubMed 
PubMed Central 

Google Scholar 
Ekstrom, J., Bennun, L. & Mitchell, R. A Cross-Sector Guide for Implementing the Mitigation Hierarchy. (The Biodiversity Consultancy, 2015).Malo, J. E., Suárez, F. & Díez, A. Can we mitigate animal-vehicle accidents using predictive models?. J. Appl. Ecol. 41, 701–710 (2004).
Google Scholar 
R Core Team. R: A language and environment for statistical computing (Version 4.0.3). https://www.R-project.org/ (R Foundation for Statistical Computing, 2020).Tucker, M. A., Ord, T. J. & Rogers, T. L. Evolutionary predictors of mammalian home range size: body mass, diet and the environment. Glob. Ecol. Biogeogr. 23, 1105–1114 (2014).
Google Scholar 
Santini, L., Boitani, L., Maiorano, L. & Rondinini, C. Effectiveness of protected areas in conserving large carnivores in Europe. in Protected areas 122–133 (John Wiley and Sons, Ltd., 2016). https://doi.org/10.1002/9781118338117.ch7.Rodrigues, A. S. L., Pilgrim, J. D., Hoffmann, M. & Lamoreux, J. F. The value of the IUCN red list for conservation. Trends Ecol. Evol. 21, 71–76 (2006).PubMed 

Google Scholar 
Government of Brazil. Mapas multimodais. Ministério da Infraestrutura http://www.infraestrutura.gov.br/ (2018).Assis, L. F. F. G. et al. TerraBrasilis: A spatial data analytics infrastructure for large-scale thematic mapping. ISPRS Int. J. Geo Inf. 8, 513 (2019).
Google Scholar  More