Philippa Kaur

Ecological networks: Linking structure to dynamics in food webs. (Oxford University Press, 2006).Adaptive food webs: Stability and transitions of real and model ecosystems. (Cambridge University Press, 2018).Pimm, S. L. Food Webs (Springer, 1982).Book 

Google Scholar 
Adaptive Food Webs: Stability and Transitions of Real and Model Ecosystems. (Cambridge University Press, 2017). doi:https://doi.org/10.1017/9781316871867.da Mata, A. S. Complex Networks: A Mini-review. Braz. J. Phys. 50, 658–672 (2020).ADS 
Article 

Google Scholar 
Zhang, W. Fundamentals of Network Biology. (World Scientific (Europe), 2018). https://doi.org/10.1142/q0149.Reichman, O. J., Jones, M. B. & Schildhauer, M. P. Challenges and opportunities of open data in ecology. Science 331, 703–705 (2011).ADS 
CAS 
PubMed 
Article 

Google Scholar 
Farley, S. S., Dawson, A., Goring, S. J. & Williams, J. W. situating ecology as a big-data science: Current advances, challenges, and solutions. Bioscience 68, 563–576 (2018).Article 

Google Scholar 
Osawa, T. Perspectives on biodiversity informatics for ecology. Ecol. Res. 34, 446–456 (2019).Article 

Google Scholar 
Shin, N. et al. Toward more data publication of long-term ecological observations. Ecol. Res. 35, 700–707 (2020).Article 

Google Scholar 
Pringle, R. M. & Hutchinson, M. C. Resolving food-web structure. Annu. Rev. Ecol. Evol. Syst. 51, 55–80 (2020).Article 

Google Scholar 
Derocles, S. A. P. et al. Biomonitoring for the 21st Century: Integrating Next-Generation Sequencing Into Ecological Network Analysis. in Advances in Ecological Research vol. 58 1–62 (Elsevier, 2018).Vacher, C. et al. Learning ecological networks from next-generation sequencing data. in Advances in Ecological Research vol. 54, 1–39 (Elsevier, 2016).Evans, D. M., Kitson, J. J. N., Lunt, D. H., Straw, N. A. & Pocock, M. J. O. Merging DNA metabarcoding and ecological network analysis to understand and build resilient terrestrial ecosystems. Funct. Ecol. 30, 1904–1916 (2016).Article 

Google Scholar 
Pocock, M. J. O. et al. A vision for global biodiversity monitoring with citizen science. in Advances in Ecological Research vol. 59, 169–223 (Elsevier, 2018).Sultana, M. & Storch, I. Suitability of open digital species records for assessing biodiversity patterns in cities: A case study using avian records. J. Urban Ecol. 7, juab014 (2021).Article 

Google Scholar 
Amano, T., Lamming, J. D. L. & Sutherland, W. J. Spatial gaps in global biodiversity information and the role of citizen science. Bioscience 66, 393–400 (2016).Article 

Google Scholar 
Chandler, M. et al. Contribution of citizen science towards international biodiversity monitoring. Biol. Conserv. 213, 280–294 (2017).Article 

Google Scholar 
Fontaine, C. et al. The ecological and evolutionary implications of merging different types of networks: Merging networks with different interaction types. Ecol. Lett. 14, 1170–1181 (2011).PubMed 
Article 

Google Scholar 
Martinson, H. M. & Fagan, W. F. Trophic disruption: A meta-analysis of how habitat fragmentation affects resource consumption in terrestrial arthropod systems. Ecol. Lett. 17, 1178–1189 (2014).PubMed 
Article 

Google Scholar 
Marczak, L. B., Thompson, R. M. & Richardson, J. S. Meta-analysis: Trophic level, Habitat, and productivity shape the food web effects of resource subsidies. Ecology 88, 140–148 (2007).PubMed 
Article 

Google Scholar 
McCary, M. A., Mores, R., Farfan, M. A. & Wise, D. H. Invasive plants have different effects on trophic structure of green and brown food webs in terrestrial ecosystems: A meta-analysis. Ecol. Lett. 19, 328–335 (2016).PubMed 
Article 

Google Scholar 
Cirtwill, A. R., Stouffer, D. B. & Romanuk, T. N. Latitudinal gradients in biotic niche breadth vary across ecosystem types. Proc. R. Soc. B Biol. Sci. 282, 20151589 (2015).Article 
CAS 

Google Scholar 
Fortuna, M. A., Ortega, R. & Bascompte, J. The Web of Life. ArXiv14032575 Q-Bio (2014).Brose, U. et al. Predator traits determine food-web architecture across ecosystems. Nat. Ecol. Evol. 3, 919–927 (2019).PubMed 
Article 

Google Scholar 
Mace, G. M., Norris, K. & Fitter, A. H. Biodiversity and ecosystem services: A multilayered relationship. Trends Ecol. Evol. 27, 19–26 (2012).PubMed 
Article 

Google Scholar 
Keyes, A. A., McLaughlin, J. P., Barner, A. K. & Dee, L. E. An ecological network approach to predict ecosystem service vulnerability to species losses. Nat. Commun. 12, 1586 (2021).ADS 
CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Peng, J. et al. Linking ecosystem services and circuit theory to identify ecological security patterns. Sci. Total Environ. 644, 781–790 (2018).ADS 
CAS 
PubMed 
Article 

Google Scholar 
Su, Y. et al. Modeling the optimal ecological security pattern for guiding the urban constructed land expansions. Urban For. Urban Green. 19, 35–46 (2016).Article 

Google Scholar 
Kowarik, I. Novel urban ecosystems, biodiversity, and conservation. Environ. Pollut. 159, 1974–1983 (2011).CAS 
PubMed 
Article 

Google Scholar 
Di Marco, M., Watson, J. E. M., Venter, O. & Possingham, H. P. Global biodiversity targets require both sufficiency and efficiency. Conserv. Lett. 9, 395–397 (2016).Article 

Google Scholar 
Kim, K.-H. & Pauleit, S. Landscape character, biodiversity and land use planning: The case of Kwangju City Region, South Korea. Land Use Policy 24, 264–274 (2007).Article 

Google Scholar 
Young, J. et al. Towards sustainable land use: Identifying and managing the conflicts between human activities and biodiversity conservation in Europe. Biodivers. Conserv. 14, 1641–1661 (2005).Article 

Google Scholar 
Dardonville, M., Urruty, N., Bockstaller, C. & Therond, O. Influence of diversity and intensification level on vulnerability, resilience and robustness of agricultural systems. Agric. Syst. 184, 102913 (2020).Article 

Google Scholar 
Oliver, T. H. et al. Biodiversity and resilience of ecosystem functions. Trends Ecol. Evol. 30, 673–684 (2015).PubMed 
Article 

Google Scholar 
Lau, M. K., Borrett, S. R., Baiser, B., Gotelli, N. J. & Ellison, A. M. Ecological network metrics: Opportunities for synthesis. Ecosphere 8, e01900 (2017).Article 

Google Scholar 
Newman, M. E. J. Networks. (Oxford University Press, 2018).Levine, S. Several measures of trophic structure applicable to complex food webs. J. Theor. Biol. 83, 195–207 (1980).ADS 
Article 

Google Scholar 
Guimarães, P. R. The structure of ecological networks across levels of organization. Annu. Rev. Ecol. Evol. Syst. 51, 433–460 (2020).Article 

Google Scholar 
Dormann, C. F., Frund, J., Bluthgen, N. & Gruber, B. Indices, graphs and null models: Analyzing bipartite ecological networks. Open Ecol. J. 2, 7–24 (2009).Article 

Google Scholar 
Jordán, F., Benedek, Z. & Podani, J. Quantifying positional importance in food webs: A comparison of centrality indices. Ecol. Model. 205, 270–275 (2007).Article 

Google Scholar 
Jordán, F., Liu, W. & Davis, A. J. Topological keystone species: Measures of positional importance in food webs. Oikos 112, 535–546 (2006).Article 

Google Scholar 
Jordán, F., Okey, T. A., Bauer, B. & Libralato, S. Identifying important species: Linking structure and function in ecological networks. Ecol. Model. 216, 75–80 (2008).Article 

Google Scholar 
Jiang, L. Determination of keystone species in CSM food web: A topological analysis of network structure. Netw. Biol. 5, 13 (2015).
Google Scholar 
Abarca-Arenas, L. G., Franco-Lopez, J., Peterson, M. S., Brown-Peterson, N. J. & Valero-Pacheco, E. Sociometric analysis of the role of penaeids in the continental shelf food web off Veracruz. Mexico Based By-catch Fish. Res. 87, 46–57 (2007).
Google Scholar 
Abascal-Monroy, I. M. et al. Functional and structural food web comparison of Terminos Lagoon, Mexico in Three Periods (1980, 1998, and 2011). Estuaries Coasts 39, 1282–1293 (2016).Article 

Google Scholar 
McDonald-Madden, E. et al. Using food-web theory to conserve ecosystems. Nat. Commun. 7, 10245 (2016).ADS 
CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Windsor, F. M. et al. Identifying plant mixes for multiple ecosystem service provision in agricultural systems using ecological networks. J. Appl. Ecol. 58, 2770–2782 (2021).Article 

Google Scholar 
Klaise, J. & Johnson, S. The origin of motif families in food webs. Sci. Rep. 7, 16197 (2017).ADS 
PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Estrada, E. Characterization of topological keystone species. Ecol. Complex. 4, 48–57 (2007).Article 

Google Scholar 
Thompson, R. M. & Townsend, C. R. Impacts on stream food webs of native and exotic forest: An intercontinental comparison. Ecology 84, 145–161 (2003).Article 

Google Scholar 
Bascompte, J., Melian, C. J. & Sala, E. Interaction strength combinations and the overfishing of a marine food web. Proc. Natl. Acad. Sci. 102, 5443–5447 (2005).ADS 
CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Dunne, J. A. et al. The roles and impacts of human hunter-gatherers in North Pacific marine food webs. Sci. Rep. 6, 21179 (2016).ADS 
CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Gauzens, B., Legendre, S., Lazzaro, X. & Lacroix, G. Food-web aggregation, methodological and functional issues. Oikos 122, 1606–1615 (2013).Article 

Google Scholar 
Patonai, K. & Jordán, F. Aggregation of incomplete food web data may help to suggest sampling strategies. Ecol. Model. 352, 77–89 (2017).Article 

Google Scholar 
Thompson, R. M. & Townsend, C. R. Is resolution the solution?: The effect of taxonomic resolution on the calculated properties of three stream food webs. Freshw. Biol. 44, 413–422 (2000).Article 

Google Scholar 
Abarca-Arenas, L. G. & Ulanowicz, R. E. The effects of taxonomic aggregation on network analysis. Ecol. Model. 149, 285–296 (2002).Article 

Google Scholar 
Jordán, F. & Osváth, G. The sensitivity of food web topology to temporal data aggregation. Ecol. Model. 220, 3141–3146 (2009).Article 

Google Scholar 
European Commission. Communication from the commission to the european parliament, the council, the european economic and social committee and the committee of the regions: EU Biodiversity Strategy for 2030 Bringing nature back into our lives. Preprint at https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:52020DC0380 (2020).European Parliament. European Parliament resolution of 9 June 2021 on the EU Biodiversity Strategy for 2030: Bringing nature back into our lives (P9_TA(2021)0277). Preprint at https://www.europarl.europa.eu/doceo/document/TA-9-2021-0277_EN.html (2021).Felson, A. J. & Ellison, A. M. Designing (for) Urban Food Webs. Front. Ecol. Evol. 9, 582041 (2021).Article 

Google Scholar 
Warren, P. et al. Urban food webs: Predators, prey, and the people who feed them. Bull. Ecol. Soc. Am. 87, 387–393 (2006).Article 

Google Scholar 
De Montis, A., Ganciu, A., Cabras, M., Bardi, A. & Mulas, M. Comparative ecological network analysis: An application to Italy. Land Use Policy 81, 714–724 (2019).Article 

Google Scholar 
Poisot, T. et al. Mangal—making ecological network analysis simple. Ecography 39, 384–390 (2016).Article 

Google Scholar 
Morris, Z. B., Weissburg, M. & Bras, B. Ecological network analysis of urban–industrial ecosystems. J. Ind. Ecol. 25, 193–204 (2021).Article 

Google Scholar 
Chamberlain, S. A. & Szöcs, E. taxize: Taxonomic search and retrieval in R. F1000 Research 2, 191 (2013).PubMed 
PubMed Central 
Article 

Google Scholar 
Hagberg, A. A., Schult, D. A. & Swart, P. J. Exploring network structure, dynamics, and function using networkX. in Proceedings of the 7th Python in Science Conference (eds. Varoquaux, G., Vaught, T. & Millman, J.) 11–15 (2008).Scotti, M. & Jordán, F. Relationships between centrality indices and trophic levels in food webs. Community Ecol. 11, 59–67 (2010).Article 

Google Scholar 
Gouveia, C., Móréh, Á. & Jordán, F. Combining centrality indices: Maximizing the predictability of keystone species in food webs. Ecol. Indic. 126, 107617 (2021).Article 

Google Scholar 
Allesina, S. & Pascual, M. Googling Food Webs: Can an Eigenvector Measure Species’ Importance for Coextinctions?. PLoS Comput. Biol. 5, e1000494 (2009).ADS 
MathSciNet 
PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Patro, S. G. K. & Sahu, K. K. Normalization: A preprocessing stage. https://doi.org/10.48550/ARXIV.1503.06462(2015).Reback, J. et al. pandas-dev/pandas: Pandas 1.2.3. (Zenodo, 2021). 10.5281/ZENODO.4572994.Hunter, J. D. Matplotlib: A 2D graphics environment. Comput. Sci. Eng. 9, 90–95 (2007).Article 

Google Scholar 
Waskom, M. et al. mwaskom/seaborn: v0.11.1 (December 2020). (Zenodo, 2020). 10.5281/ZENODO.4379347.Girvan, M. & Newman, M. E. J. Community structure in social and biological networks. Proc. Natl. Acad. Sci. 99, 7821–7826 (2002).ADS 
MathSciNet 
CAS 
PubMed 
PubMed Central 
MATH 
Article 

Google Scholar 
Rosvall, M., Axelsson, D. & Bergstrom, C. T. The map equation. Eur. Phys. J. Spec. Top. 178, 13–23 (2009).Article 

Google Scholar 
Gao, P. & Kupfer, J. A. Uncovering food web structure using a novel trophic similarity measure. Ecol. Inform. 30, 110–118 (2015).Article 

Google Scholar 
Gauzens, B., Thébault, E., Lacroix, G. & Legendre, S. Trophic groups and modules: Two levels of group detection in food webs. J. R. Soc. Interface 12, 20141176 (2015).PubMed 
PubMed Central 
Article 

Google Scholar 
Rudiger, P. et al. holoviz/holoviews: Version 1.14.2. (Zenodo, 2021). 10.5281/ZENODO.4581995.Pedregosa, F. et al. Scikit-learn: Machine learning in Python. J. Mach. Learn. Res. 12, 2825–2830 (2011).MathSciNet 
MATH 

Google Scholar  More

Statistical characteristics of rill cross-sectional asymmetry (RCA)The rill cross-sectional asymmetry (RCA) is a key parameter in describing rill morphology and includes the asymmetry ratio of the width (Aw) and the asymmetry ratio of the area (Aa). It reflects the differences in certain aspects of natural conditions resulting in inconsistent development speeds on both sides of a rill cross-section. The cross-section was classified as left-biased if Aw, Aa < 0, quasi-symmetrical if Aw, Aa = 0, and right skewed if Aw, Aa > 0. The left/right deflection reflects that erosion on the right happened faster than on the left, so the slope on the left is not as steep as on the right. The results of this study show that asymmetry is a common phenomenon in the cross-section of a rill. The Aw ranged from − 1.77 to 1.97, with an average value of − 0.034. There were 374 cross-sections whose RCA was less than or equal to 0, meaning that 53% of the cross-sections were right-biased. The Aa ranged from − 1.81 to 1.71, with an average of − 0.046. There were 374 cross-sections with an RCA of less than or equal to 0, meaning that 53% of the cross-sections were right-biased (Fig. 1).Figure 1Statistical characteristics of the rill cross-sectional asymmetry (RCA).Full size imageFigure 2 shows that there are four Aw groups in the interval (− 1.7, − 1.5), 53 groups in the interval (− 1.5, − 1.0), 144 groups in the interval (− 1.0, − 0.5), 173 groups in the interval (− 0.5, 0), 174 groups in the interval (0, 0.5), 120 groups in the interval (0.5, 1.0), 39 groups in the interval (1.0, 1.5), and five groups in the interval (1.5, 2). The Aa has 15 groups in the interval (− 1.8, − 1.5), 63 groups in the interval (− 1.5, − 1.0), 130 groups in the interval (− 1.0, − 0.5), 166 groups in the interval (− 0.5, 0), 161 groups in the interval (0, 0.5), 110 groups in the interval (0.5, 1.0), 53 groups in the interval (1.0, 1.5), and 14 groups in the interval (1.5, 2). The RCA of most cross-sections is concentrated in the interval (− 0.5, 0.5). This interval of Aw contains 491 cross-sections, accounting for 68.96% of the total. There are 470 cross-sections in this interval of Aa, accounting for 66.01% of the total. This indicates that, although the rill cross-section exhibits some asymmetry, the difference between both sides of the section is small (Fig. 2).Figure 2Distribution characteristics of the RCA.Full size imageThe influence of a single topographic factor on the RCACorrelation analyses of the Aw, Aa, and the slope difference on both sides (B), rill length (L), rill slope length (I), rill head catchment area (A), difference between the catchment areas of both sides (R), rill bending coefficient (K), and location of the section angle of turning of the rill (J) were carried out. The results show that the main factors that have a significant linear correlation with the Aw and the Aa are B (p < 0.01), with correlation coefficients of 0.32 and 0.22, respectively (Fig. 3). That is, the greater the difference in slope between the two sides, the more asymmetric the rill cross-section. R also has a significant linear correlation with the Aw (p < 0.05), with a correlation coefficient of 0.07. This means that the greater the difference in the catchments between the left and right sides of the rill, the greater the asymmetry of the rill cross-section. However, other topographic factors have no significant correlation with the RCA.Figure 3Correlation between rill cross-sectional asymmetry (RCA) and topographic factors.Full size imageB is the difference in slope between the left and right sides of the rill cross-section catchment area. The closer B gets to 0, the smaller the difference in slope between the left and right sides of the rill cross-section catchment area. When the catchment area slope on the right side of the cross-section is greater than that on the left side, B < 0; and when the catchment area slope on the left side of the cross-section is greater than that on the right side, B > 0. Grouping B reveals that the average RCA increases as B increases (Fig. 4). When B is (− 30, − 20), Aw is − 0.48 and Aa is − 0.38; when B is (− 20, − 10), Aw is − 0.36 and Aa is − 0.31; when B is (− 10, 0), Aw is − 0.23 and Aa is − 0.22; when B is (0, 10), Aw is 0.21 and Aa is 0.16; when B is (10, 20), Aw is 0.47 and Aa is 0.40; and when B is (20, 40), Aw is 0.31 and Aa is 0.13. These are relatively low values because this group only has two sets of cross-sections which cannot represent the characteristics of interval B. The sign of the RCA is the same as the sign of B. The directionality of the RCA is significantly affected by B. When the slope of the left catchment area is large, RCA > 0, and the rill cross-section appears to be left-biased; when the slope of the right catchment area is large, RCA < 0, and the cross-section appears to the righ-biased.Figure 4The asymmetry of different B values.Full size imageThe influence of multiple topographic factors on the RCAIn order to explore the influence of multiple topographic factors on the RCA, principal component analysis (PCA) was used to extract the main feature components of the topographic data. The PCA results show that the nine topographic factors can be reflected by two principal components at 61.84% (characteristic value: 3.117+1.211=4.328 variables) (Table 1). Therefore, the analysis of the first two principal components could reflect most of the information from all the data.Table 1 Calculation results of topographic factor principal component analysis (PCA).Full size tableThe contribution rate of the first principal component is 44.534%. The characteristic is that the factor variables have high positive loads for the four factors L, I, A, and K. L has the largest contribution rate at 88.5%, followed by A, I, and K, at 87.5%, 81.1%, and 60.2%, respectively. Therefore, the first component represents the rill slope and rill shape.The contribution rate of the second principal component is 17.303%. The characteristic is that the factor variables have high positive loads for the three factors B, J, and R. B has the largest contribution rate at 83.5%, followed by J and R, at 57.4% and 55.7%, respectively. Therefore, the second component represents the effect of the difference between the two sides of the rill.Based on the correlation between the topographic factors and the RCA of a rill cross-section in the Yuanmou dry-hot valley area, the following was observed: asymmetry in rill cross-sections is ubiquitous. The distribution range of Aw is − 1.77 to 1.97, the average value is − 0.034, and the cross-section that is right-biased accounts for 53%. A correlation analysis of the RCA and seven topographic factors shows that B has a significant positive correlation with the Aw and Aa (p < 0.01), the average RCA increases as B increases, and the directionality of the RCA is affected by B. When B > 0, RCA > 0, and the rill cross-section appears to the left; when B < 0, RCA < 0, and the cross-section appears to the right. The difference in catchment area between the sides has a significant linear correlation with the Aw (p < 0.05). Other single topographic factors have no significant correlation with the RCA. Principal component analysis and calculations show that the first principal component represents the influence of the rill slope surface and rill shape on the rill cross-sectional asymmetry. The contribution rate is 44.534%, which is characterized by a high positive load on the L, I, A, and K factors. The second principal component represents the effect of the difference between the two sides of the rill. The contribution rate is 44.534%, which is characterized by a high positive load on the B, J, and R factors. More

Hwang, H. Y. et al. Effect of cover cropping on the net global warming potential of rice paddy soil. Geoderma 292, 49–58 (2017).ADS 
CAS 
Article 

Google Scholar 
IPCC. Climate change 2013: The physical science basis. The Working Group I contribution to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC, 2013).
Google Scholar 
Cardoso, R. M., Soares, P. M. M., Lima, D. C. A. & Miranda, P. M. A. Mean and extreme temperatures in warming climate: EURO CORDEX and WRF regional climate high-resolution projection for Portugal. Clim. Dyn. 52, 129–157 (2019).Article 

Google Scholar 
Ding, T., Gao, H. & Li, W. J. Extreme high-temperature event in southern China in 2016 and the possible role of cross-equatorial flows. Int. J. Climatol. 38, 3579–3594 (2018).Article 

Google Scholar 
Escalas, A. et al. Functional diversity and redundancy across fish gut, sediment, and water bacterial communities. Environ. Microbiol. 19, 3268–3282 (2017).Article 

Google Scholar 
Philippot, L. et al. Loss in microbial diversity affects nitrogen cycling in soil. ISME J. 7, 1609–1619 (2013).CAS 
Article 

Google Scholar 
Li, Y. B. et al. Serratia spp. Are responsible for nitrogen fixation fueled by As(III) oxidation, a novel biogeochemical process identified in mine tailings. Environ. Sci. Technol 56, 2033–2043 (2022).ADS 
Article 

Google Scholar 
Jia, M., Gao, Z. W., Gu, H. J., Zhao, C. Y. & Han, G. D. Effects of precipitation change and nitrogen addition on the composition, diversity, and molecular ecological network of soil bacterial communities in a desert steppe. PLoS ONE 16, e0248194. https://doi.org/10.1371/journal.pone.0248194 (2021).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Waghmode, T. R. et al. Response of nitrifier and denitrifier abundance and microbial community structure to experimental warming in an agricultural ecosystem. Front. Microbiol. 9, 474. https://doi.org/10.3389/fmicb.2018.00474 (2018).Article 
PubMed 
PubMed Central 

Google Scholar 
Hu, Y. L., Wang, S., Niu, B., Chen, Q. & Zhang, G. Effect of increasing precipitation and warming on microbial community in Tibetan alpine steppe. Environ. Res. 189, 109917. https://doi.org/10.1016/j.envres.2020.109917 (2020).CAS 
Article 
PubMed 

Google Scholar 
Li, H. et al. Responses of soil bacterial communities to nitrogen deposition and precipitation increment are closely linked with aboveground community variation. Microb. Ecol. 71, 974–989 (2016).CAS 
Article 

Google Scholar 
Wang, H. et al. Experimental warming reduced topsoil carbon content and increased soil bacterial diversity in a subtropical planted forest. Soil Biol. Biochem. 133, 155–164 (2019).CAS 
Article 

Google Scholar 
Haumann, F. A., Gruber, N. & Münnich, M. Sea-Ice Induced Southern Ocean Subsurface Warming and Surface Cooling in a Warming Climate. AGU Advances 1, e2019AV000132. https://doi.org/10.1029/2019AV000132 (2020).ADS 
Article 

Google Scholar 
Ji, F., Wu, Z. H., Huang, J. P. & Chassignet, E. P. Evolution of land surface air temperature trend. Nat. Clim. Chang. 4, 462–466 (2014).ADS 
Article 

Google Scholar 
Sabri, N. S. A., Zakaria, Z., Mohamad, S. E., Jaafar, A. B. & Hara, H. Importance of soil temperature for the growth of temperate crops under a tropical climate and functional role of soil microbial diversity. Microbes Environ. 33, 144–150 (2018).Article 

Google Scholar 
McGrady-Steed, J. & Morin, P. T. Biodiversity, density compensation, and the dynamics of populations and functional groups. Ecology 81, 361–373 (2000).Article 

Google Scholar 
Jiang, L. Density compensation can cause no effect of biodiversity on ecosystem function. Oikos 116, 324–334 (2007).Article 

Google Scholar 
Faust, K. & Raes, J. Microbial interactions: From networks to models. Nat. Rev. Microbiol. 10, 538. https://doi.org/10.1038/nrmicro283 (2012).CAS 
Article 
PubMed 

Google Scholar 
Gao, X. X. et al. Revegetation significantly increased the bacterial-fungal interactions in different successional stages of alpine grasslands on the Qinghai-Tibetan Plateau. CATENA 205, 105385. https://doi.org/10.1016/j.catena.2021.105385 (2021).CAS 
Article 

Google Scholar 
Morriën, E. et al. Soil networks become more connected and take up more carbon as nature restoration progresses. Nat. Commun. 8, 14349. https://doi.org/10.1038/ncomms14349 (2017).ADS 
CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Banerjee, S. et al. Agricultural intensification reduces microbial network complexity and the abundance of keystone taxa in roots. ISME J. 13, 1722–1736 (2019).Article 

Google Scholar 
Pržulj, N. & Malod-Dognin, N. Network analytics in the age of big data. Science 353, 123–124 (2016).ADS 
Article 

Google Scholar 
Ratzke, C., Barrere, J. M. R. & Gore, J. Strength of species interactions determines biodiversity and stability in microbial communities. Nat. Ecol. Evol. 4, 376–383 (2020).Article 

Google Scholar 
Fuhrman, J. A. Microbial community structure and its functional implications. Nature 45, 193–199 (2009).ADS 
Article 

Google Scholar 
Zhao, M. X., Cong, J., Cheng, J. M., Qi, Q. & Zhang, Y. G. Soil microbial community assembly and interactions are constrained by nitrogen and phosphorus in broadleaf forests of southern China. Forest 11, 285. https://doi.org/10.3390/f11030285 (2020).Article 

Google Scholar 
Wan, X. L. et al. Biogeographic patterns of microbial association networks in paddy soil within Eastern China. Soil Biol. Biochem. 142, 07696. https://doi.org/10.1016/j.soilbio.2019.107696 (2020).CAS 
Article 

Google Scholar 
Yuan, M. M., Guo, X., Wu, L., Zhang, Y. & Zhou, J. Climate warming enhances microbial network complexity and stability. Nat. Clim. Change 11, 343–348 (2021).ADS 
Article 

Google Scholar 
Lassaletta, L. et al. Food and feed trade as a driver in the global nitrogen cycle: 50-year trends. Biogeochemistry 11, 225–241 (2014).Article 

Google Scholar 
Phoenix, G. K. et al. Impacts of atmospheric nitrogen deposition: Responses of multiple plant and soil parameters across contrasting ecosystems in long-term field experiments. Glob. Change Biol. 18, 1197–1215 (2012).ADS 
Article 

Google Scholar 
Nakaji, T., Fukami, M., Dokiya, Y. & Izuta, T. Effects of high nitrogen load on growth, photosynthesis and nutrient status of Cryptomeria japonica and Pinus densiflora seedlings. Trees-Struct. Funct. 15, 453–461 (2001).CAS 
Article 

Google Scholar 
Wang, H. Y. et al. Reduction in nitrogen fertilizer use results in increased rice yields and improved environmental protection. Int. J. Agric. Sustain. 15, 681–692 (2017).Article 

Google Scholar 
Zhou, X. G. & Wu, F. Z. Land-use conversion from open field to greenhouse cultivation differently affected the diversities and assembly processes of soil abundant and rare fungal communities. Sci. Total Environ. 788, 147751. https://doi.org/10.1016/j.scitotenv.2021.147751 (2021).ADS 
CAS 
Article 
PubMed 

Google Scholar 
Guo, H. et al. Long-term nitrogen & phosphorus additions reduce soil microbial respiration but increase its temperature sensitivity in a Tibetan alpine meadow. Soil Biol. Biochem. 113, 26–34 (2017).CAS 
Article 

Google Scholar 
Zhang, C. et al. Effects of simulated nitrogen deposition on soil respiration components and their temperature sensitivities in a semiarid grassland. Soil Biol. Biochem. 75, 113–123 (2014).CAS 
Article 

Google Scholar 
Zhang, J. J. et al. Different responses of soil respiration and its components to nitrogen and phosphorus addition in a subtropical secondary forest. For. Ecosyst. 8, 37. https://doi.org/10.1186/s40663-021-00313-z (2021).Article 

Google Scholar 
Norse, D. & Ju, X. T. Environmental costs of China’s food security. Agric. Ecosyst. Environ. 209, 5–14 (2015).Article 

Google Scholar 
Xu, H. F., Du, H., Zeng, F. P., Song, T. Q. & Peng, W. X. Diminished rhizosphere and bulk soil microbial abundance and diversity across succession stages in Karst area, southwest China. Appl. Soil Ecol. 158, 103799. https://doi.org/10.1016/j.apsoil.2020.103799 (2020).Article 

Google Scholar 
Li, Y. B. et al. Arsenic and antimony co-contamination influences on soil microbial community composition and functions: Relevance to arsenic resistance and carbon, nitrogen, and sulfur cycling. Environ. Int. 153, 106522. https://doi.org/10.1016/j.envint.2021.106522 (2021).CAS 
Article 
PubMed 

Google Scholar 
Zhou, J. & Fong, J. J. Strong agricultural management effects on soil microbial community in a non-experimental agroecosystem. Appl. Soil Ecol. 165, 103970. https://doi.org/10.1016/j.apsoil.2021.103970 (2021).Article 

Google Scholar 
Bárcenas-Moreno, G., Gómez-Brandón, M., Rousk, J. & Bååth, E. Adaptation of soil microbial communities to temperature: Comparison of fungi and bacteria in a laboratory experiment. Glob. Chang. Biol. 15, 2950–2957 (2009).ADS 
Article 

Google Scholar 
Tan, E. H., Zou, W., Zheng, Z., Yan, X. & Kao, S. J. Warming stimulates sediment denitrification at the expense of anaerobic ammonium oxidation. Nat. Clim. Change 10, 349–355 (2020).ADS 
CAS 
Article 

Google Scholar 
Supramaniam, Y., Chong, C. W., Silvaraj, S. & Tan, K. P. Effect of short term variation in temperature and water content on the bacterial community in a tropical soil. Appl Soil Ecol. 107, 279–289 (2016).Article 

Google Scholar 
Zhu, Y. Z., Li, Y. Y., Zheng, N. G., Chapman, S. J. & Yao, H. Y. Similar but not identical resuscitation trajectories of the soil microbial community based on either DNA or RNA after flooding. Agronomy 10, 502. https://doi.org/10.3390/agronomy10040502 (2020).CAS 
Article 

Google Scholar 
Donhauser, J., Qi, W., Bergk-Pinto, B. & Frey, B. High temperatures enhance the microbial genetic potential to recycle C and N from necromass in high-mountain soils. Glob. Chang. Biol. 27, 1365–1386 (2021).ADS 
Article 

Google Scholar 
Santoyo, G., Hernandez-Pacheco, C., Hernandez-Salmeron, J. & Hernandez-Leon, R. The role of abiotic factors modulating the plant-microbe-soil interactions: Toward sustainable agriculture. A review. Span. J. Agric. Res. 15, e03R01-e11. https://doi.org/10.5424/sjar/2017151-9990 (2017).Article 

Google Scholar 
Lefcheck, J. S. et al. Biodiversity enhances ecosystem multifunctionality across trophic levels and habitats. Nat. Commun. 6, 6936. https://doi.org/10.1038/ncomms7936 (2015).ADS 
CAS 
Article 
PubMed 

Google Scholar 
Cardinale, B. J. et al. Corrigendum: Biodiversity loss and its impact on humanity. Nature 486, 59–67 (2012).ADS 
CAS 
Article 

Google Scholar 
Ma, B., Wang, H., Dsouza, M., Lou, J. & Xu, J. Geographic patterns of co-occurrence network topological features for soil microbiota at continental scale in eastern China. ISME J. 10, 1891–1901 (2016).CAS 
Article 

Google Scholar 
Trivedi, C. et al. Losses in microbial functional diversity reduce the rate of key soil processes. Soil Biol. Biochem. 135, 267–274 (2019).CAS 
Article 

Google Scholar 
Melanie, K. et al. Effects of season and experimental warming on the bacterial community in a temperate mountain forest soil assessed by 16S rRNA gene pyrosequencing. FEMS Microbiol. Ecol. 82, 551–562 (2012).Article 

Google Scholar 
Zheng, H. F., Liu, Y., Chen, Y., Zhang, J. & Chen, Q. Short-term warming shifts microbial nutrient limitation without changing the bacterial community structure in an alpine timberline of the eastern Tibetan Plateau. Geoderma 360, 113985. https://doi.org/10.1016/j.geoderma.2019.113985 (2020).ADS 
CAS 
Article 

Google Scholar 
Finlay, B. J. & Cooper, J. L. Microbial diversity and ecosystem function. CEH Integrating Fund second progress report to the Director, Centre for Ecology and Hydrology Nov 1996–Sept (1997).Xing, X. Y. et al. Warming shapes nirS- and nosZ-type denitrifier communities and stimulates N2O emission in acidic paddy soil. Appl. Environ. Microbiol. 87, e02965-e3020. https://doi.org/10.1128/AEM.0296520 (2021).CAS 
Article 
PubMed Central 

Google Scholar 
Lin, Y. T., Whitman, W. B., Coleman, D. C., Jien, S. H. & Chiu, C. Y. Soil bacterial communities at the treeline in subtropical alpine areas. CATENA 201, 105205. https://doi.org/10.1016/j.catena.2021.105205 (2021).CAS 
Article 

Google Scholar 
Wang, J. C. et al. Impacts of inorganic and organic fertilization treatments on bacterial and fungal communities in a paddy soil. Appl. Soil Ecol. 112, 42–50 (2017).Article 

Google Scholar 
Chacón, J. M., Shaw, A. K. & Harcombe, W. R. Increasing growth rate slows adaptation when genotypes compete for diffusing resources. PLoS Comput. Biol. 16, e1007585. https://doi.org/10.1371/journal.pcbi.1007585 (2020).ADS 
CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Hartley, I. P., Hopkins, D. W., Garnett, M. H., Sommerkorn, M. & Wookey, P. A. Soil microbial respiration in arctic soil does not acclimate to temperature. Ecol. Lett. 11, 1092–1100 (2008).Article 

Google Scholar 
Baath, E. Growth rates of bacterial communities in soils at varying pH: A comparison of the thymidine and leucine incorporation techniques. Microb. Ecol. 36, 316–327 (1998).CAS 
Article 

Google Scholar 
Qin, H. L. et al. Soil moisture and activity of nitrite- and nitrous oxide-reducing microbes enhanced nitrous oxide emissions in fallow paddy soils. Biol. Fertil. Soils 56, 53–67 (2020).CAS 
Article 

Google Scholar 
Chen, Z. et al. Impact of long term fertilization on the composition of denitrifier communities based on nitrite reductase analyses in a paddy soil. Microb. Ecol. 60, 850–861 (2010).CAS 
Article 

Google Scholar 
Wei, G. S. et al. Similar drivers but different effects lead to distinct ecological patterns of soil bacterial and archaeal communities. Soil Biol. Biochem. 144, 107759. https://doi.org/10.1016/j.soilbio.2020.107759 (2020).CAS 
Article 

Google Scholar 
Bastian, F., Bouziri, L., Nicolardot, B. & Ranjard, A. L. Impact of wheat straw decomposition on successional patterns of soil microbial community structure. Soil Biol. Biochem. 41, 262–275 (2009).CAS 
Article 

Google Scholar 
Levins, R. Evolution in Changing Environments: Some Theoretical Explorations (Princeton University Press, 1968).Book 

Google Scholar  More

Darwin, C. On the Origin of Species. Facsimile of the First Edition (Harvard University Press, 1859).
Google Scholar 
Grafen, A. The phylogenetic regression. Philos. Trans. R. Soc. Lond. B Biol. Sci. 326, 119–157 (1989).ADS 
CAS 
PubMed 

Google Scholar 
Sillero, N., Reis, M., Vieira, C. P., Vieira, J. & Morales-Hojas, R. Niche evolution and thermal adaptation in the temperate species Drosophila americana. J. Evol. Biol. 27, 1549–1561 (2014).CAS 
PubMed 

Google Scholar 
Ramos, R. et al. Global spatial ecology of three closely-related gadfly petrels. Sci. Rep. 6, 23447 (2016).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Kumar, B., Cheng, J., Ge, D., Xia, L. & Yang, Q. Phylogeography and ecological niche modeling unravel the evolutionary history of the Yarkand hare, Lepus yarkandensis (Mammalia: Leporidae), through the Quaternary. BMC Evol. Biol. 19, 113 (2019).PubMed 
PubMed Central 

Google Scholar 
Wiens, J. J. & Graham, C. H. Niche conservatism: Integrating evolution, ecology, and conservation biology. Annu. Rev. Ecol. Evol. 36, 519–539 (2005).
Google Scholar 
Losos, J. B. Phylogenetic niche conservatism, phylogenetic signal and the relationship between phylogenetic relatedness and ecological similarity among species. Ecol. Lett. 11, 995–1003 (2008).PubMed 

Google Scholar 
Crisp, M. D. & Cook, L. G. Phylogenetic niche conservatism: What are the underlying evolutionary and ecological causes?. New Phytol. 196, 681–694 (2012).PubMed 

Google Scholar 
Qian, H. & Ricklefs, R. E. Geographical distribution and ecological conservatism of disjunct genera of vascular plants in eastern Asia and eastern North America. J. Ecol. 92, 253–265 (2004).
Google Scholar 
Vitt, L. J., Zani, P. A. & Espósito, M. C. Historical ecology of Amazonian lizards: Implications for community ecology. Oikos 87, 286–294 (1999).
Google Scholar 
Rice, N. H., Martínez-Meyer, E. & Peterson, A. T. Ecological niche differentiation in the Aphelocoma jays: A phylogenetic perspective. Biol. J. Linn. Soc. 80, 369–383 (2003).
Google Scholar 
Jost, L. Explosive local radiation of the genus Teagueia (Orchidaceae) in the Upper Pastaza Watershed of Ecuador. Lyonia 7, 42–47 (2004).
Google Scholar 
Antonelli, A., Verola, C. F., Parisod, C. & Gustafsson, A. L. S. Climate cooling promoted the expansion and radiation of a threatened group of South American orchids (Epidendroideae: Laeliinae). Biol. J. Linn. Soc. 100, 597–607 (2010).
Google Scholar 
Johnson, S. D., Linder, H. P. & Steiner, K. E. Phylogeny and radiation of pollination systems in Disa (Orchidaceae). Am. J. Bot. 85, 402–411 (1998).CAS 
PubMed 

Google Scholar 
Kolanowska, M., Grochocka, E. & Konowalik, K. Phylogenetic climatic niche conservatism and evolution of climatic suitability in Neotropical Angraecinae (Vandeae, Orchidaceae) and their closest African relatives. PeerJ 5, e3328 (2017).PubMed 
PubMed Central 

Google Scholar 
Dressler, R. L., Blanco, M. A., Pupulin, F. & Neubig, K. M. Proposal to conserve the name Sobralia (Orchidaceae) with a conserved type. Taxon 60, 907–908 (2011).
Google Scholar 
Baranow, P., Dudek, M. & Szlachetko, D. L. Brasolia, a new genus highlighted from Sobralia (Orchidaceae). Plant Syst. Evol. 303, 853–871 (2017).CAS 

Google Scholar 
Dressler, R. L. The major sections or groups within Sobralia, with four new species from Panama and Costa Rica, S. crispissima, S. gloriana, S. mariannae and S. nutans. Lankesteriana 5, 9–15 (2002).
Google Scholar 
Pridgeon, A. M., Cribb, P. J., Chase, M. W. & Rasmussen, F. N. Genera Orchidacearum Vol. 4: Epidendroideae Part 1 (Oxford University Press, 2005).
Google Scholar 
Van der Cingel, N. A. An Atlas of Orchid Pollination: America, Africa, Asia and Australia (Balkema, 2001).
Google Scholar 
Dodson, C. H. Why are there so many orchid species. Lankesteriana 7, 99–103 (2003).
Google Scholar 
Van Der Pijl, L. & Dodson, C. H. Orchid Flowers: Their Pollination and Evolution (University of Miami Press, 1966).
Google Scholar 
Neubig, K. M. Systematics of Tribe Sobralieae (Orchidaceae): Phylogenetics, Pollination, Anatomy, and Biogeography of a Group of Neotropical Orchids (University of Florida, 2012).
Google Scholar 
Neubig, K. M. et al. Preliminary molecular phylogenetics of Sobralia and relatives (Orchidaceae; Sobralieae). Lankesteriana 11, 307–317 (2011).
Google Scholar 
Ramírez, S. R., Roubik, D. W., Skov, C. & Pierce, N. E. Phylogeny, diversification patterns and historical biogeography of euglossine orchid bees (Hymenoptera: Apidae). Biol. J. Linn. Soc. 100, 552–572 (2010).
Google Scholar 
Gregory-Wodzicki, K. M. Uplift history of the Central and Northern Andes: A review. Geol. Soc. Am. Bull. 112, 1091–1105 (2000).ADS 

Google Scholar 
Sundell, K. E., Saylor, J. E., Lapen, T. J. & Horton, B. K. Implications of variable late Cenozoic surface uplift across the Peruvian central Andes. Sci. Rep. 9, 4877 (2019).ADS 
PubMed 
PubMed Central 

Google Scholar 
Mescua, J. F. et al. Middle to late miocene contractional deformation in Costa Rica triggered by plate geodynamics. Tectonics 36, 2936–2949 (2017).ADS 

Google Scholar 
Kolanowska, M., Mystkowska, K., Kras, M., Dudek, M. & Konowalik, K. Evolution of the climatic tolerance and postglacial ranges of the most primitive orchids (Apostasioideae) within Sunduland, Wallacea and Sahul. PeerJ 4, e2384 (2016).PubMed 
PubMed Central 

Google Scholar 
Arnal, P. et al. The evolution of climate tolerance in conifer-feeding aphids in relation to their host’s climatic niche. Ecol. Evol. 9, 11657–11671 (2019).PubMed 
PubMed Central 

Google Scholar 
Zangiabadi, S., Zaremaivan, H., Brotons, L., Mostafavi, H. & Ranjbar, H. Using climatic variables alone overestimate climate change impacts on predicting distribution of an endemic species. PLoS ONE 16, e0256918. https://doi.org/10.1371/journal.pone.0256918 (2021).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Soberón, J. & Peterson, A. Interpretation of models of fundamental ecological niches and species’ distributional areas. Biodivers. Inform. https://doi.org/10.17161/bi.v2i0.4 (2005).Article 

Google Scholar 
Jiménez-Valverde, A., Lobo, J. & Hortal, J. Not as good as they seem: The importance of concepts in species distribution modelling. Divers. Distrib. 14, 885–890. https://doi.org/10.1111/j.1472-4642.2008.00496.x (2008).Article 

Google Scholar 
Bonetti, M. F. & Wiens, J. J. Evolution of climatic niche specialization: a phylogenetic analysis in amphibians. Proc. Biol. Sci. 281, 20133229. https://doi.org/10.1098/rspb.2013.3229 (2014).Article 
PubMed 
PubMed Central 

Google Scholar 
George, P. M., Walter, E. W. & Yeuh-Lih, Y. Realized versus fundamental niche functions in a model of chaparral response to climatic change. Ecol. Modell. 7, 261–277 (1992).
Google Scholar 
Hijmans, R. J., Schreuder, M., Cruz, J. & Guarino, L. Using GIS to check co-ordinates of genebank accessions. Genet. Resour. Crop Evol. 46, 291–296 (1999).
Google Scholar 
Phillips, S. J., Dudík, M. & Schapire, R. E. A maximum entropy approach to species distribution modeling. In ICML ’04. Proceedings of the Twenty-First International Conference on MACHINE LEARNing, 655–662 (ACM, New York, 2004).Phillips, S. J., Anderson, R. & Schapire, R. E. Maximum entropy modeling of species geographic distributions. Ecol. Modell. 190, 231–259 (2006).
Google Scholar 
Elith, J. et al. A statistical explanation of MaxEnt for ecologists. Divers. Distrib. 17, 43–57 (2011).
Google Scholar 
Barve, N. et al. The crucial role of the accessible area in ecological niche modeling and species distribution modeling. Ecol. Modell. 222, 1810–1819 (2011).
Google Scholar 
Hijmans, R. J., Cameron, S. E., Parra, J. L., Jones, P. G. & Jarvis, A. Very high resolution interpolated climate surfaces for global land areas. Int. J. Climatol. 25, 1965–1978 (2005).
Google Scholar 
Fick, S. E. & Hijmans, R. J. WorldClim 2: new 1km spatial resolution climate surfaces for global land areas. Int. J. Climatol. 37, 4302–4315 (2017).
Google Scholar 
Brown, J. L. SDMtoolbox: A python-based GIS toolkit for landscape genetic, biogeographic and species distribution model analyses. Methods Ecol. Evol. 5, 694–700 (2014).
Google Scholar 
Feng, X., Park, D. S., Liang, Y., Pandey, R. & Papeş, M. Collinearity in ecological niche modeling: Confusions and challenges. Ecol. Evol. https://doi.org/10.1002/ece3.5555 (2019).Article 
PubMed 
PubMed Central 

Google Scholar 
Hosmer, D. W. & Lemeshow, S. Applied Logistic Regression (Wiley, 2000).MATH 

Google Scholar 
Mason, S. J. & Graham, N. E. Areas beneath the relative operating characteristics (ROC) and relative operating levels (ROL) curves statistical significance and interpretation. Q. J. R. Meteorol. Soc. 128, 2145–2166 (2002).ADS 

Google Scholar 
Evangelista, P. H. et al. Modelling invasion for a habitat generalist and a specialist plant species. Divers. Distrib. 14, 808–817 (2008).
Google Scholar 
R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/ (2022).Warren, D. L. et al. ENMTools 1.0: An R package for comparative ecological biogeography. Ecography 44, 504–511 (2021).
Google Scholar 
Schoener, T. W. The Anolis lizards of Bimini: Resource partitioning in a complex fauna. Ecology 49, 704–726 (1968).
Google Scholar 
Warren, D. L., Glor, R. E. & Turelli, M. Environmental niche equivalency versus conservatism: Quantitative approaches to niche evolution. Evolution 62, 2868–2883 (2008).PubMed 

Google Scholar 
Broennimann, O. et al. Measuring ecological niche overlap from occurrence and spatial environmental data. Glob. Ecol. Biogeogr. 21, 481–497 (2012).
Google Scholar 
Heibl, C. & Calenge, C. Phyloclim: integrating phylogenetics and climatic niche modeling. R package version 0.9-4. http://CRAN.R-project.org/package=phyloclim (2013).Evans, M. E., Smith, S. A., Flynn, R. S. & Donoghue, M. J. Climate, niche evolution, and diversification of the ‘“bird-cage”’ evening primroses (Oenothera, sections Anogra and Kleinia). Am. Nat. 173, 225–240 (2009).PubMed 

Google Scholar 
Paradis, E., Claude, J. & Strimmer, K. APE: Analyses of phylogenetics and evolution in R language. Bioinformatics 20, 289–290 (2004).CAS 
PubMed 

Google Scholar 
Galtier, N., Gouy, M. & Gautier, C. SeaView and Phylo_win, two graphic tools for sequence alignment and molecular phylogeny. Comput. Appl. Biosci. 12, 543–548 (1996).CAS 
PubMed 

Google Scholar 
Edgar, R. MUSCLE: Mulitiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 32, 1792–1797 (2004).CAS 
PubMed 
PubMed Central 

Google Scholar 
Nylander, J. A. A. MrModeltest v2 (Uppsala University, 2004).
Google Scholar 
Ronquist, F. & Huelsenbeck, J. P. MRBAYES: Bayesian phylogenetic inference under mixed models. Bioinformatics 19, 1572–1574 (2003).CAS 
PubMed 

Google Scholar 
Drummond, A. J., Suchard, M. A., Xie, D. & Rambaut, A. Bayesian phylogenetics with BEAUti and the BEAST 1.7. Mol. Biol. Evol. 29, 1969–1973 (2012).CAS 
PubMed 
PubMed Central 

Google Scholar 
Givnish, T. et al. Orchid phylogenomics and multiple drivers of their extraordinary diversification. Proc. Biol. Sci. https://doi.org/10.1098/rspb.2015.1553 (2015).Article 
PubMed 
PubMed Central 

Google Scholar  More

IPBES. Global assessment report on biodiversity and ecosystem services of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services. Zenodo https://doi.org/10.5281/zenodo.5657041 (2019).Laurance, W. F. et al. A global strategy for road building. Nature 513, 229–232 (2014).ADS 
CAS 
Article 

Google Scholar 
Ibisch, P. L. et al. A global map of roadless areas and their conservation status. Science 354, 1423–1427 (2016).ADS 
CAS 
Article 

Google Scholar 
Forman, R. T. T., Sperling, D. & Bissonette, J. A. Road Ecology: Science and Solutions. (Island Pr, 2003).van der Ree, R., Smith, D. J. & Grilo, C. Handbook of Road Ecology. (John Wiley & Sons, 2015).Laender. Hunting Statistics. Game casualties 2017/2018: furred game (red deer, roe deer, chamois, moufflon) https://www.statistik.at/web_en/statistics/Economy/agriculture_and_forestry/livestock_animal_production/hunting/index.html (2018).Steiner, W., Leisch, F. & Hacklander, K. A review on the temporal pattern of deer-vehicle accidents: Impact of seasonal, diurnal and lunar effects in cervids. Accident; analysis and prevention 66, (2014).Kioko, J. et al. Driver knowledge and attitudes on animal vehicle collisions in Northern Tanzania. TROPICAL CONSERVATION SCIENCE 8, 352–366 (2015).Article 

Google Scholar 
Bíl, M., Andrášik, R. & Janoška, Z. Identification of hazardous road locations of traffic accidents by means of kernel density estimation and cluster significance evaluation. Accident Analysis & Prevention 55, 265–273 (2013).Article 

Google Scholar 
Page, Y. A statistical model to compare road mortality in OECD countries. Accident Analysis and Prevention 33, 371–385 (2001).CAS 
Article 

Google Scholar 
Teixeira, F. Z. et al. Are Road-kill Hotspots Coincident among Different Vertebrate Groups? Oecologia Australis 17, 36–47 (2017).Article 

Google Scholar 
Canova, L. & Balestrieri, A. Long-term monitoring by roadkill counts of mammal populations living in intensively cultivated landscapes. Biodivers Conserv https://doi.org/10.1007/s10531-018-1638-3 (2018).Brehme, C. S., Hathaway, S. A. & Fisher, R. N. An objective road risk assessment method for multiple species: ranking 166 reptiles and amphibians in California. Landscape Ecol 33, 911–935 (2018).Article 

Google Scholar 
Heigl, F. et al. Comparing Road-Kill Datasets from Hunters and Citizen Scientists in a Landscape Context. Remote Sensing 8, (2016).Heigl, F., Horvath, K., Laaha, G. & Zaller, J. G. Amphibian and reptile road-kills on tertiary roads in relation to landscape structure: using a citizen science approach with open-access land cover data. BMC Ecol 17, 24 (2017).Article 

Google Scholar 
Dörler, D. & Heigl, F. A decrease in reports on road-killed animals based on citizen science during COVID-19 lockdown. PeerJ 9, e12464 (2021).Article 

Google Scholar 
Peer, M. et al. Predicting spring migration of two European amphibian species with plant phenology using citizen science data. Sci Rep 11, 21611 (2021).ADS 
CAS 
Article 

Google Scholar 
Schwartz, A. L. W. UK Roadkill Records. The Global Biodiversity Information Facility https://doi.org/10.15468/r3xakd (2018).Lin, T. The Taiwan Roadkill Observation Network Data Set. Version 1.3. The Global Biodiversity Information Facility https://doi.org/10.15468/cidkqi (2018).Chandler, M. et al. Contribution of citizen science towards international biodiversity monitoring. Biological Conservation 213, 280–294 (2017).Article 

Google Scholar 
Périquet, S., Roxburgh, L., le Roux, A. & Collinson, W. J. Testing the Value of Citizen Science for Roadkill Studies: A Case Study from South Africa. Front. Ecol. Evol. 6, (2018).Abra, F. D., Huijser, M. P., Pereira, C. S. & Ferraz, K. M. P. M. B. How reliable are your data? Verifying species identification of road-killed mammals recorded by road maintenance personnel in São Paulo State, Brazil. Biological Conservation 225, 42–52 (2018).Article 

Google Scholar 
Bíl, M., Kubeček, J., Sedoník, J. & Andrášik, R. Srazenazver.cz: A system for evidence of animal-vehicle collisions along transportation networks. Biological Conservation 213, 167–174 (2017). Part A.Article 

Google Scholar 
Vercayie, D. & Herremans, M. Citizen science and smartphones take roadkill monitoring to the next level. Nature Conservation 11, 29–40 (2015).Article 

Google Scholar 
Waetjen, D. P. & Shilling, F. M. Large Extent Volunteer Roadkill and Wildlife Observation Systems as Sources of Reliable Data. Front. Ecol. Evol. 5, (2017).Shilling, F. M., Perkins, S. E. & Collinson, W. Wildlife/Roadkill Observation and Reporting Systems. in Handbook of Road Ecology 492–501 (John Wiley & Sons, 2015).Eitzel, M. V. et al. Citizen Science Terminology Matters: Exploring Key Terms. Citizen Science: Theory and Practice 2, 1–20 (2017).
Google Scholar 
Haklay, M. et al. Contours of citizen science: a vignette study. Royal Society Open Science 8, 202108 (2021).ADS 
Article 

Google Scholar 
Heigl, F., Kieslinger, B., Paul, K. T., Uhlik, J. & Dörler, D. Opinion: Toward an international definition of citizen science. PNAS 116, 8089–8092 (2019).CAS 
Article 

Google Scholar 
Heigl, F. et al. Quality Criteria for Citizen Science Projects on Österreich forscht | Version 1.1. Open Science Framework https://doi.org/10.17605/OSF.IO/48J27 (2018).Heigl, F. et al. Co-Creating and Implementing Quality Criteria for Citizen Science. Citizen Science: Theory and Practice 5, 23 (2020).
Google Scholar 
Heigl, F. & Zaller, J. G. Using a Citizen Science Approach in Higher Education: a Case Study reporting Roadkills in Austria. Human Computation 1, (2014).University of Natural Resources and Life Sciences, Vienna. Roadkill, The Global Biodiversity Information Facility, https://doi.org/10.15468/ejb47y (2021).Heigl, F. & Roadkill Community. Roadkill Dataset 2014-2020 Quality level 2, Zenodo, https://doi.org/10.5281/zenodo.5878813 (2022).August, T. A. et al. Citizen meets social science: predicting volunteer involvement in a global freshwater monitoring experiment. Freshwater Science 38, 321–331 (2019).Article 

Google Scholar 
IUCN. The IUCN Red List of Threatened Species. Version 2021-3. IUCN Red List of Threatened Species https://www.iucnredlist.org/en (2021). More

Newbold, T. et al. Global effects of land use on local terrestrial biodiversity. Nature 520, 45–50 (2015).ADS 
CAS 
Article 

Google Scholar 
Díaz, S. et al. Pervasive human-driven decline of life on earth points to the need for transformative change. Science 366, eaax3100 (2019).Article 

Google Scholar 
United Nations Food and Agriculture Organization (FAO). State of the World’s Forests 2022: Forest Pathways for Green Recovery and Building Inclusive, Resilient and Sustainable Economies (FAO, 2022).Williams, M. Deforesting the Earth: From Prehistory to Global Crisis (University of Chicago Press, 2003).Laurence, W. F. et al. Averting biodiversity collapse in tropical forest protected areas. Nature 489, 290–294 (2012).ADS 
Article 

Google Scholar 
UNEP-WCMC and IUCN. Protected Planet: The World Database on Protected Areas (WDPA) http://www.protectedplanet.net (2020).Convention on Biological Diversity. First Draft of the Post-2020 Global Biodiversity Framework https://www.cbd.int/meetings/WG2020-03 (2021).Wilson, E. O. Half-Earth: Our Planet’s Fight for Life (Liveright Publishing Company, 2016).Joppa, L. N. & Pfaff, A. High and far: biases in the location of protected areas. PLoS One 4, e8273 (2009).ADS 
Article 

Google Scholar 
Sarkar, S. et al. Biodiversity conservation planning tools: present status and challenges for the future. Annu. Rev. Environ. Resour. 31, 123–159 (2006).Article 

Google Scholar 
Naidoo, R. et al. Integrating economic costs into conservation planning. Trends Ecol. Evol. 21, 681–687 (2006).Article 

Google Scholar 
Ando, A., Camm, J., Polasky, S. & Solow, A. Species distributions, land values, and efficient conservation. Science 279, 2126–2128 (1998).ADS 
CAS 
Article 

Google Scholar 
Meir, E., Andelman, S. & Possingham, H. P. Does conservation planning matter in a dynamic and uncertain world? Ecol. Lett. 7, 615–622 (2004).Article 

Google Scholar 
Costello, C. & Polasky, S. Dynamic reserve site selection. Resour. Energy Econ. 26, 157–174 (2004).Article 

Google Scholar 
The Nature Conservancy. Terrestrial Ecoregions https://geospatial.tnc.org/datasets/7b7fb9d945544d41b3e7a91494c42930_0 (2008).Hansen, M. C. et al. High-resolution global maps of 21st-century forest cover change. Science 342, 850–853 (2013).ADS 
CAS 
Article 

Google Scholar 
Kummu, M., Taka, M. & Guillaume, J. Gridded global datasets for gross domestic product and human development index over 1990–2015. Sci. Data 5, 180004 (2018).Article 

Google Scholar 
Kier, G. et al. Global patterns of plant diversity and floristic knowledge. J. Biogeogr. 32, 1107–1116 (2005).Article 

Google Scholar 
Tilman, D., Reich, P. B. & Isbell, F. Biodiversity impacts ecosystem productivity as much as resources, disturbance or herbivory. Proc. Natl Acad. Sci. USA 109, 10394–10397 (2012).ADS 
CAS 
Article 

Google Scholar 
Thompson, I., Mackey, B., McNulty, S. & Mosseler, A. Forest Resilience, Biodiversity, and Climate Change CBD Technical Series No. 43 (Secretariat of the Convention on Biological Diversity, 2009).Butsic, V., Lewis, D. J. & Radeloff, V. C. Reserve selection with land market feedbacks. J. Environ. Manag. 114, 276–284 (2013).Article 

Google Scholar 
Wilson, K. A., McBride, M. F., Bode, M. & Possingham, H. P. Prioritizing global conservation efforts. Nature 440, 337–340 (2006).ADS 
CAS 
Article 

Google Scholar 
McBride, M. F., Wilson, K. A., Bode, M. & Possingham, H. P. Incorporating the effects of socioeconomic uncertainty into priority setting for conservation investment. Conserv. Biol. 21, 1463–1474 (2007).Article 

Google Scholar 
Bode, M., Wilson, K., McBride, M. & Possingham, H. P. Optimal dynamic allocation of conservation funding among priority regions. Bull. Math. Biol. 70, 2039–2054 (2008).MathSciNet 
Article 

Google Scholar 
Pouzols, F. M. et al. Global protected area expansion is compromised by projected land-use and parochialism. Nature 516, 383–386 (2014).ADS 
Article 

Google Scholar 
Pollock, L. J., Thuiller, W. & Jetz, W. Large conservation gains possible for global biodiversity facets. Nature 546, 141–144 (2017).ADS 
CAS 
Article 

Google Scholar 
Dinerstein, E. et al. A ‘Global Safety Net’ to reverse biodiversity loss and stabilize Earth’s climate. Sci. Adv. 6, eabb2824 (2020).ADS 
Article 

Google Scholar 
Jung, M. et al. Areas of global importance for conserving terrestrial biodiversity, carbon and water. Nat. Ecol. Evol. 5, 1499–1509 (2021).Article 

Google Scholar 
Polasky, S. et al. Where to put things? Spatial land management to sustain biodiversity and economic returns. Biol. Conserv. 141, 1505–1524 (2008).Article 

Google Scholar 
Lennox, G. D., Fargione, J., Spector, S., Williams, G. & Armsworth, P. The value of flexibility in conservation financing. Conserv. Biol. 31, 666–674 (2016).Article 

Google Scholar 
Drechsler, M. & Wätzold, F. Biodiversity conservation in a dynamic world may lead to inefficiencies due to lock-in effects and path dependence. Ecol. Econ. 173, 106652 (2020).Article 

Google Scholar 
Boulton, C. A., Lenton, T. M. & Boers, N. Pronounced loss of Amazon rainforest resilience since the early 2000s. Nat. Clim. Chang. 12, 271–278 (2022).ADS 
Article 

Google Scholar 
Lomolino, M. V. Ecology’s most general, yet protean pattern: the species‐area relationship. J. Biogeogr. 27, 17–26 (2000).Article 

Google Scholar 
Drake, J. & Griffen, B. Early warning signals of extinction in deteriorating environments. Nature 467, 456–459 (2010).ADS 
CAS 
Article 

Google Scholar 
Grantham, H. et al. Anthropogenic modification of forests means only 40% of remaining forests have high ecosystem integrity. Nat. Commun. 1, 5978 (2020).ADS 
Article 

Google Scholar 
Naidoo, R. & Iwamura, T. Global-scale mapping of economic benefits from agricultural land: Implications for conservation priorities. Biol. Conserv. 140, 40–49 (2007).Article 

Google Scholar 
UNEP-FAO. United Nations Decade on Ecosystem Restoration 2021–2030 https://www.decadeonrestoration.org/ (2021).UN Climate Change Conference. Glasgow Leaders’ Declaration on Forests and Land Use https://ukcop26.org/glasgow-leaders-declaration-on-forests-and-land-use/ (2021)UN Climate Change Conference. The Global Forest Finance Pledge https://ukcop26.org/the-global-forest-finance-pledge/ (2021).Convention on Biological Diversity. Preparations for the Post-2020 Biodiversity Framework https://www.cbd.int/conferences/post2020 (2021)Deutz, A. et al. Financing Nature: Closing the Global Biodiversity Financing Gap (The Paulson Institute, The Nature Conservancy, and the Cornell Atkinson Center for Sustainability, 2020).Griscom, B. W. et al. Natural climate solutions. Proc. Natl Acad. Sci. USA 114, 11645–11650 (2017).ADS 
CAS 
Article 

Google Scholar 
Guerry, A. D. et al. Natural capital and ecosystem services informing decisions: from promise to practice. Proc. Natl Acad. Sci. USA 112, 7348–7355 (2015).CAS 

Google Scholar 
Leclère, D. et al. Bending the curve of terrestrial biodiversity needs an integrated strategy. Nature 585, 551–556 (2020).ADS 
Article 

Google Scholar 
Wintle, B. A. et al. Spending to save: what will it cost to halt Australia’s extinction crisis? Conserv. Lett. 12, e12682 (2019).Article 

Google Scholar 
Ostrom, E. Governing the Commons: The Evolution of Institutions for Collective Action (Cambridge University Press, 1990)WWF et al. The State Of Indigenous Peoples’ and Local Communities’ Lands and Territories https://wwflac.awsassets.panda.org/downloads/report_the_state_of_the_indigenous_peoples_and_local_communities_lands_and_territories_1.pdf (2021) .Chaplin-Kramer, R. et al. Conservation needs to integrate knowledge across scales. Nat. Ecol. Evol. 6, 118–119 (2022).Article 

Google Scholar 
IUCN. Spatial Data Download: Terrestrial Mammals. IUCN Red List https://www.iucnredlist.org/resources/spatial-data-download (2021).Abay, K. A., Chamberlin, J. & Berhane, G. Are land rental markets responding to rising population pressures and land scarcity in sub-Saharan Africa? Land Use Policy 101, 105139 (2021).Article 

Google Scholar  More

China economy-wide material flow identification: system boundary, processes, and materialsThe first step is to define an economy, i.e., the economic (rather than geographical) territory of a country in which the activities and transactions of producer and consumer units are resident. Additionally, the period is a total of thirty-one years, from 1990 to 2020, for the following reasons: (1) statistics before 1990 are of poor quality and are insufficient to allow us to conduct analyses; and (2) so far, statistics have just recently been updated to cover the year of 2020. Furthermore, the analytical framework (hereinafter referred to as China EW-MFA) is developed to explore material utilisation and its environmental consequences within China’s economy.The general structure of China EW-MFA is depicted in Fig. 1, which comprises seven processes. (1) Input of extracted resources: domestic natural resources are extracted from the environment to the economy through human-controlled means. (2) Output of domestic processed materials: after being processed by manufacturers, materials are released from the economy into the environment in the form of by-products and residues, which can be classified by their destinations (i.e., air, land, and water) and pathways (dissipative use and losses). (3) Input and (4) output by cross-border trade: by imports and exports, materials are transported between China’s economy and the economies of the rest of the world. (5) Input and (6) output of balancing items (BI): sometimes, materials identified in the output processes are not considered by inputs, which needs to be balanced. For example, the utilisation of fossil energy materials by combustion causes the emission of carbon dioxide (CO2) into the air, which is identified as system output, but requirements of oxygen (O2) as system input are not counted. (7) Additions to the system: within the economy, materials would have been added to the economy in the form of buildings, infrastructures, durable goods, and household appliances, which are referred to as the net additions to stock (NAS).Fig. 1The general structure of China EW-MFA. To note, white data cells can be obtained directly from official statistics, whereas grey cells are estimated.Full size imageThe last step is to specify the materials concerned in each process. Four types (in blue boxes in Fig. 1) of natural materials are extracted and input into the economy in China, i.e., harvested biomass (33 items), mined metal ores (28 items), quarried non-metallic minerals (155 items), and mined fossil energy materials (6 items in 3 classes). Materials (green boxes) released into the air are greenhouse gases (e.g., CO2, methane (CH4), dinitrogen oxide (N2O)), air pollutants (e.g., particulate matter 10 (PM10), black carbon (BC)), and toxic contaminants of mercury (Hg) in divalent, gaseous elemental, and particulate forms. Those released into the water are inorganic matters (of nitrogen (N), phosphorus (P), Arsenic (As), and four heavy metals of lead (Pb), mercury (Hg), cadmium (Cd), and chromium (Cr)) and organic matters of cyanide, petroleum, and volatile phenol. Materials released into the land are waste disposal in uncontrolled landfills, which are illegal in China. Some materials are dissipated by application, for example, fertilisers, compost, sewage sludge being applied to agricultural land, and pesticides being used to cultivate crops. Some would be unintentionally dissipated from abrasion, corrosion, erosion, and leakages. Materials (in red boxes) are BI, which includes the input of O2 and output of water vapour in the fossil energy material combustion process, the input of O2 and output of water vapour and CO2 in the respiration process of human and cultivated livestock, input and output of water in imported and exported beverages, and the output of water from domestically extracting crops.There are some messages needed to be mentioned: (1) Material of water is not included since its flow volume is more substantial than others, which needs to be independently analysed; (2) Activities of foreign tourists, cross-border transfer of emissions through natural media, etc. are excluded. (3) To be clear, we refer to a data cell as a specific flow process of a specific substance in a specific year, e.g., the number of cereals domestically extracted in 2020.Data acquisition: sources and collectionBased on our China EW-MFA, we first analyse accessibility, reliability, completeness, rules of redistribution, etc., for each data source (yellow boxes in Fig. 1), including China national database, China rural statistical yearbooks, USGS mineral yearbooks, etc. The complete list of data sources and descriptions are presented in Table 1. Then, we store the originally retrieved data source files in a semi- or unstructured format (e.g., CSV, PDF). Next, we manually collect these statistics and reorganise them according to China EW-MFA material types and processes. However, only a tiny part of retrieved statistics can be applied directly, as specified in black colour in Fig. 1.Table 1 Data sources and descriptions.Full size tableData compilation: parameter localisation and data estimationA few inconsistencies in statistics were noticed, which would result in data incompleteness. For example, the domestic extraction of vegetables has been accounted for and published since 1995, before which statistics are unavailable. The domestically harvested timber has been measured in the volume unit of cubic metres, which needs to be converted into the mass unit via density conversion factor. Therefore, acquired statistics have to be estimated, which are specified in grey colour in Fig. 1. The following section elaborates on each data cell’s estimation methods, localised parameters, references, etc. In our uploaded data files, the original statistics, data sources, and compilation methods (using formulas) are all implemented, as explained in the Data Records Section.

The input of natural resources by domestic extraction

Vegetables in crops: Statistics of vegetable production (WVegetables)16 during 1990–1994 are unavailable, which is estimated based on the relationship between the production yield (PYield) and areas (AVegetables), as shown in Eq. 1. Here, PYield is assumed to remain constant at 27.04 thousand tonnes per thousand hectares from 1990 to 1995, derived by dividing vegetable production (257,267 thousand tonnes) by areas (9,515 thousand hectares) in 1995.$${W}_{Vegetables}={P}_{Yield}times {A}_{Vegetables}$$
(1)

Nuts in crops: One of them is chestnuts. The chestnut production in 2020 is unavailable, which is assumed to be the same as in 2019.

Crop residues in biomass residues: They are referred to as that harvested production of crops that do not reach the market to be sold but are instead employed as raw materials for commercial purposes such as energy generation and livestock husbandry. This number (Wcrop residues) can be calculated by first determining the number of crop residues available from primary crop production (Wcrop) and the harvest factor (Pharvest factor), and then using the recovery rate (Precovery rate) to determine the number of crop residues used by the economy, as shown in Eq. 2. These parameters have been localized by previous studies17,18, which are adopted in this study, i.e., wheat (1.1 for Pharvest factor and 0.463 for Precovery rate), maize (1.2, 0.463), rice (0.9, 0.463), sugar cane (0.5, 0.9), beetroots (0.7, 0.9), tuber (0.5, 0.463), pulse (1.2, 0.7), cotton (3.4, 0.463), fibre crops (1.8, 0.463), silkworm cocoons (1.8, 0.463), and oil-bearing crops (1.8, 0.463).$${W}_{cropresidues}={W}_{crop}times {P}_{harvestfactor}times {P}_{recoveryrate}$$
(2)

Roughage of grazed biomass and fodder crops in biomass residues: In China, the grazed biomass for roughage includes annual forage and perennial forage, whereas fodder crops comprise straw feed, processed straw feed, and all other fodder crops. However, information19 on grazed biomass production is only accessible from 2006 to 2018, whereas fodder crop statistics are only available from 2015 to 2017. Equation 3 and Eq. 4 can be used to estimate unavailable statistics. To note, we assume that China’s domestic roughage supply structure has remained unaltered, which has two meanings. The proportion of total domestic roughage production (WDomestic production) in requirement (WRoughage requirement) has remained constant, while the proportion (PSupply fraction) of grazed biomass and fodder crop in domestic roughage production has been unchanged. The requirement (WRoughage requirement) is determined by the quantity of livestock (QLivestock) and their annual feeding amount (PAnnual intake). PAnnual intake (in tonnes per head per year) has been localised for each type of livestock4, with 4.5 for live cattle and buffaloes, 0.5 for sheep and goats, 3.7 for horses, and 2.2 for mules and asses.$${W}_{Roughagerequirement}={Q}_{Livestock}times {P}_{Annualintake}$$
(3)
$${W}_{Domesticproduction}={W}_{Roughagerequirement}times {P}_{Supplyfraction}$$
(4)

Timber in wood: As illustrated in Eq. 5, wood production16 is reported in volume units of cubic metres (VTimber), which need to be converted into mass units (WTimber) via density (PDensity). The parameter PDensity is assumed to be 0.58 tonnes per cubic metre, calculated by averaging 0.52 for coniferous types and 0.64 for non-coniferous ones4.$${W}_{Timber}={V}_{Timber}times {P}_{Density}$$
(5)

Non-ferrous metals in metal ores: Non-ferrous metal statistics are derived from two sources. China statistics20 are measured in gross ore (WMetal ores in gross ore) but are only available from 1999 to 2017, whereas the USGS statistics21 cover the period of 1990 to 2020 but they are measured in metal or concentrate content (WMetal ores in other units). Therefore, USGS statistics need to be converted with an empirical unit conversion factor (PUnit conversion factor) before being applied to estimate unavailable statistics reported by China, as shown in Eq. 6. Conversion factors are localised for each non-ferrous metal in each year from 2000 to 2017 by using USGS statistics divided by China statistics and then averaged after removing the highest value and the lowest value (i.e., trimmed mean). This factor could capture the general relationship between statistics from two separate sources, which can be used in other long time-series studies on resource management on a particular element in China.$${W}_{Metaloresingrossore}={W}_{Metaloresinotherunits}/{P}_{Unitconversionfactor}$$
(6)

Non-metallic minerals: The official China-specific information on non-metallic mineral domestic production is available between 1999 and 201720, the rest of which could be estimated from USGS statistics (1990–2020)21. Also, two differences in reporting standards are observed resulting from the material coverages and reporting units. China statistics contain eighty-eight materials in mineral ores, whereas the USGS only includes twenty in the concentrate unit. Therefore, a conversion factor is developed in this estimation, as shown in Eq. 7. This conversion factor is applied to the total amount of non-metallic mineral production, which is assumed to have been constant from 1990 to 1999 at 11.38% (1999) and 12.56% (2017) from 2017 to 2020.$${W}_{Mineralsingrossore}={W}_{Mineralsinotherunits}/{P}_{Conversionfactor}$$
(7)

Coal in fossil energy materials: Coal, mined in China, includes raw coal, peat, stone coal, and oil shale. Except for raw coal, statistics for the rest are only available from 1999 to 201720. The unavailable data (WOther coals) is estimated using Eq. 8 under the assumption that the structure of the coal supply in China barely changes. That is, the proportion (PSupply fraction) of peat, stone coal, and oil shale in raw coal production (WRaw coal) remains constant, so the 1999 proportion is applied to all years before that (earlier years of 1990–1998), while the 2017 proportion is used to the recent years between 2018 and 2020. For example, PSupply fraction for oil shale production was assumed to be 0.014% during 1990–1999, calculated by dividing raw coal production (1,250,000) by oil shale production (179) in 1999. PSupply fraction in the earlier and the recent years are 0.007% and 0.001% for peat, 0.203% and 0.031% for stone coal, and 0.014% and 0.067% for oil shale.

$${W}_{Othercoals}={W}_{Rawcoal}/{P}_{Supplyfraction}$$
(8)

The output of processed materials by release

Materials released into the air: In China, thirteen materials are released into the air, as shown in Fig. 1. The emission of sulphur dioxide (SO2) is reported in China environmental statistical yearbooks22,23, while the rest is specified in the EDGAR24. However, in EDGAR, statistics for recent years have not yet been updated, which are estimated with the value in the most recent year in our database. For example, nitrous oxide (NOx) records are only available for the years prior to 2016, with 26,365 thousand tonnes in 2015 and 26,837 in 2014. As a result of the observed decreasing trend in NOx emissions, NOx emission data for 2016–2020 is estimated to be 26,000 thousand tonnes. This estimate may be subjective due to constraints, but it would be aligned with European statistics, allowing for international comparisons. Data can be updated after the EDGAR statistics have been updated.

Materials released into the water: Ten principal materials have been found in China wastewater (both industrial and municipal) that are nitrogen (N), phosphorus (P), organic pollutants of petroleum, volatile phenol and cyanide, heavy metals of mercury (Hg), lead (Pb), cadmium (C·d), and the hexavalent chromium (Cr6+), and arsenic (As). Many statistics22,23 have been of poor quality (e.g., inconsistent material coverages between years). Given that the statistics of pollutants in industrial wastewater cover more periods and contain fewer abnormal observations, the total material emissions can be approximated from those of industrial wastewater. Equations 9 and 10 show the estimation processes. The materials in industrial wastewater (WIndustrial materials) are first identified using material mass concentration (PConcentration) and the weight of industrial wastewater (WIndustrial wastewater), and then the materials in total wastewater (WTotal materials) are identified using the proportion (PContribution) of materials in industrial wastewaters (WIndustrial materials) to the total. The assumption is that PConcentration and PContribution change gradually between years, which enables to use linear interpolation method to estimate unavailable parameters. Consider cyanide: its PConcentration was 23.61 (1‰ ppm) in 2005 and 37.31 in 2002, which was assumed to be 28.18 in 2004 and 32.74 in 2003. PConcentration was assumed to be 100% throughout the years for cyanide because all cyanide emissions in China are driven by industrial wastewater discharges. Later, the total material emissions can be derived by dividing the industrial wastewater mass by PConcentration.$${W}_{Industrialmaterials}={W}_{Industrialwastewater}times {P}_{Concentration}$$
(9)
$${W}_{Totalmaterials}={W}_{Industrialmaterials},/,{P}_{Contribution}$$
(10)

Materials released to the land: This is zero because uncontrolled landfills are illegal in China.

Materials dissipated by organic fertiliser use: In China, manure is the primary organic fertiliser, which is excreted by pigs, dairy cows, calves, sheep, horses, asses, mules, camels, chickens, and other animals. As shown in Eq. 11, the manure production (WManure) is estimated through the amounts of raised livestock (QLivestock, heads), the weight of daily manure production (PManure production, kilograms per head per day), the number of days they are raised (PFeeding period, in days per year), and the moisture content of their manure (PDry matter, %) for each type of animal. These parameters are region-specific, which have been localised by Chinese scholars25,26,27 and listed in Table 2.$${W}_{Manure}={Q}_{Livestock}times {P}_{Manureproduction}times {P}_{Feedingperiod}times {P}_{Drymatter}$$
(11)
Table 2 Localised parameters for animal manure production.Full size table

Materials dissipated by mineral fertiliser use: The mineral fertilisers used in China are four types, i.e., nitrogen (N), phosphorus (P), potash (K), and compound. Their usage (WFertiliser usage) is measured in nutrient mass (WNutrient materials), which needs to be converted into the gross mass by dividing their nutrient content (PNutrient content). Equation 12 shows the estimation. This parameter of PNutrient content is localised by the Ministry of Agriculture and Rural Affairs of China28 as 29%, 22%, 35%, and 44% for N- bearing, P- bearing, K-bearing, and compound fertilisers, respectively.$${W}_{Fertiliserusage}={W}_{Nutrientmaterials}/{P}_{Nutrientcontent}$$
(12)

Materials dissipated by sewage sludge: Sewage sludge is the residue generated by municipal wastewater treatment. As demonstrated in Eq. 13, its dissipative use (Wss, dissipation) is the untreated amount of production (Wss, production), represented by the parameter of Pss, dissipation rate. Sewage sludge production (Wss, production) statistics are only available for the years 2006–202029, and data for the remaining years can be estimated using Eq. 14 and Eq. 15. In Eq. 14, Pss, production rate represents the relationship between sewage sludge production (Wss, production, 2006–2020) and wastewater treatment (Www, treatment, 2002–2020), and in Eq. 15, Pww, treatment efficiency represents the relationship between the quantity of treated wastewater (Www, treatment, 2002–2020) and the treatment capacity (Www, treatment capacity, 1990–2020). In this estimation, three assumptions are made. The first is to estimate Www, treatment, Pww, treatment efficiency is assumed to be unchanged at 63% during 1990–2001, given it has been increasing from 63% in 2002 to ~80% in recent years. The second is that, in order to estimate Wss, production, Pss, production rate is assumed to be unchanged at 3.5 between 1990 and 2005, suggesting 3.5 tonnes of sewage sludge are generated by processing 10,000 cubic metres of wastewater. This assumption is determined by that Pss, production rate is approximately 3.5 during 2006–2010 while declines sharply and stabilises at around two during 2011–2020. The last is, to estimate the Wss,dissipation, Pss,dissipation rate is assumed to be 5% between 1990 and 2005, given it has been around 5% during 2006–2020.$${W}_{ss,dissipation}={W}_{ss,production}times {P}_{ss,dissipationrate}$$
(13)
$${W}_{ss,production}={W}_{ww,treatment}times {P}_{ss,productionrate}$$
(14)
$${W}_{ww,treatment}={W}_{ww,treatmentcapacity}times {P}_{ww,treatmentefficiency}$$
(15)

Materials dissipated by composting: Composting is a natural process that uses microbes to turn organic materials into other products, which are then used for fertilising and entering the environment. In China, composting has been used to treat two materials: feces and municipal waste, whose quantities (WComposting) were only available from 2003 to 201029. The unavailable data can be estimated using Eq. 16. The dry weight of materials treated by composting (WComposting) is proportionally related to the fresh weight of all treated materials (WTotal), the proportion treated by composting (PComposting rate), and the dry content (PDry matter). Considering that China’s composting capacity has been declining since 2001 due to the implementation of waste incineration power generation technologies30, Pcomposting rate is assumed to be the same as it was in 2003 (9.5%) between 1990 and 2002, and 1.5% in 2010 between 2011 and 2020. The parameter of PDry matter is 50%4.$${W}_{Composting}={W}_{Total}times {P}_{Compostingrate}times {P}_{Drymatter}$$
(16)

The input and output by cross-border trade. Statistics of imports and exports have been gathered since 1962 and stored in the UN Comtrade database31. However, the data quality issue of outliers, and missing values, especially in weight, is reportedly identified. In our previous work, we addressed these issues, and an improved database32 is provided. Details about our estimation methods can be found in publications33,34,35. As UN Comtrade lists 5,039 different commodity types (in 6-digit HS0 commodity code), yet only 18 material types are specified in the China EW-MFA, UN Comtrade statistics need to be aligned to the China EW-MFA framework. Therefore, we compared each commodity and each material type between them and established a correspondence table to map UN Comtrade commodity types onto our EW-MFA material types. For example, non-ferrous metal materials of China EW-MFA include commodities, such as copper ores and concentrates (260300 HS0 code), silver powder (710610), manganese, articles thereof, and waste or scrap (811100), etc., whereas biomass residues include cereal straw and husks (121300), lucerne meal and pellets (121410), and other fodder and forage products (121410). This correspondence table between HS0 and EW-MFA classification for imports and exports is provided in Supplementary File 1.

The input of balancing items

O2 required for combustion: In BI, requirements for materials can be abstracted as equalling exogenous demands minus intrinsic supplies (Eq. 17). Three parts (two demands and one supply) are considered for O2 requirements by the combustion process: (1) demanding exogenous oxygen to oxidise elements (e.g., carbon, sulphur, nitrogen, etc., except for hydrogen) released into the air, (2) demanding exogenous oxygen to oxidise the hydrogen embedded in fossil energy materials, and (3) providing intrinsic oxygen embedded in fossil energy materials. The first part can be estimated via Eq. 18 by multiplying air emissions (WEmissions) of CO2, N2O, NOx, CO, and SO2 by their oxygen content (POxygen content). For the second (Eq. 19), the oxygen demand is estimated based on the principle of mass balance by converting the hydrogen amount of domestically utilised fossil energy materials (WFossil fuel materials × PHydrogen content) via molar mass conversion factor (PMass conversion factor). PMass conversion factor equals 7.92, derived by the molar mass of one oxygen (16 g/mol) divided by that of two hydrogen atoms (2 × 1.01 g/mol). The last is the intrinsic supplies from fossil fuel materials, which is identified via Eq. 20 by multiplying the domestically utilised amount of fossil fuel materials (WFossil fuel materials) by their oxygen content (POxygen content). The parameters in this estimation are presented in Table 3. As a footnote here, the domestically utilised amount is referred to as the domestic material consumption (DMC), which equals domestic extraction (DE) plus imports (IM) and minus exports (EX).$${W}_{Requirements}={W}_{Demands}-{W}_{Supplies}$$
(17)
$${W}_{Demands}={W}_{Emissions}times {P}_{Oxygencontent}$$
(18)
$${W}_{Demands}={W}_{Fossilfuelmaterials}times {P}_{Hydrogencontent}times {P}_{Massconversionfactor}$$
(19)
$${W}_{Supplies}={W}_{Fossilfuelmaterials}times {P}_{Oxygencontent}$$
(20)
Table 3 Parameters related to combustion processes4.Full size table

O2 required for respiration: O2 is required by the metabolic activities of living organisms, the majority of which are humans and livestock. Bacteria are another sort of organism, which are not included in this estimation because their O2 requirements are too small to be quantified. The respiration-required O2 is related to the total quantity (QOrganisms) and their respiration activity by organism types, as shown in Eq. 21. The respiration activity is represented by the respiration requirement coefficient (PRespiration requirement coefficient), which is the average quantity of O2 that each organism utilises to maintain the metabolic activity, as listed in Table 4.$${W}_{Demands}={Q}_{Organisms}times {P}_{Respirationrequirementcoefficient}$$
(21)
Table 4 Parameters related to respiration processes4.Full size table

Water required for the domestic production of exported beverages: The exported beverages are produced domestically using domestically extracted materials, especially a large amount of water. The weight of water is considered in the output by cross-border trade but is not included in the domestic extraction input. The resulted imbalance can be identified by specifying the water weight in beverages, i.e., multiplying the traded beverage weight (WMaterials) by a parameter of the water content (PWater content), as given in Eq. 22. Fruit and vegetable juices (2009 in HS0 code) and beverages (code 22) are covered in the improved UN Comtrade database32, with PWater content of 85% for the first and 90% for the latter4.

$${W}_{Water}={W}_{Materials}times {P}_{Watercontent}$$
(22)

The output of balancing items.

Water vapour from combustion: Water vapour emissions by domestically combusting fossil fuel materials are contributed by two paths. The direct evaporation of embedded water is the first path (Eq. 23), which can be derived by multiplying the DMC of fossil fuel materials by their moisture content (PMoisture content). The PMoisture content for each type of fossil fuel material is listed in Table 3. The other is the generation of water vapour during hydrogen oxidation, which can be calculated by converting the oxidised weight of hydrogen to the water weight using the molar mass conversion factor (PMass conversion factor), as given in Eq. 24. PMass conversion factor equals 8.92 by dividing the molar mass of water (18.02 g/mol) by that of two hydrogen atoms (2 × 1.01 g/mol).$${W}_{Water}={W}_{Fossilfuelmaterials}times {P}_{Moisturecontent}$$
(23)
$${W}_{Water}={W}_{Fossilfuelmaterials}times {P}_{Hydrogencontent}times {P}_{Massconversionfactor}$$
(24)

Water vapour and CO2 from respiration: Respiration activities of organisms will produce water vapour and CO2, whose estimation is similar to that of O2 requirements. As shown in Eq. 25, the respiration-caused gas emissions are related to the number of organisms (QOrganisms) and the respiration activity by organism types. The latter is represented by the parameter of respiration emission coefficient (PRespiration emission coefficient), which is specified in Table 4 for water vapour and CO2 for each type of organism.$${W}_{Emissions}={Q}_{Organisms}times {P}_{Respirationemissioncoefficient}$$
(25)

Water from imported beverages: The estimation approach is the same as water by the domestic production of exported beverages, as described in Eq. 16.

Water in biomass products: Usually, the input of biomass products by domestic extraction16 has been measured in fresh weight, but their corresponding output29 by sewage sludge, composting, etc., are in dry weight, leading to an imbalance in water weight. The water weight in biomass products is calculated by multiplying their domestic extraction amount in fresh weight (WBiomass) by a parameter of moisture content at harvest (PMoisture content), as shown in Eq. 26. The values of PMoisture content by biomass products are presented in Table 5.Table 5 The moisture content at harvest for each biomass product4.Full size table

$${W}_{Water}={W}_{Biomass}times {P}_{Moisturecontent}$$
(26)
Material flow quantificationThe above attempts have quantified material inputs and outputs by flows and presented a detailed profile of material utilisation for each material in China’s economy. In order to depict the economy in a more general way, EW-MFA indicators are assessed by aggregating flows by materials or periods as below.

Domestic extraction (DE): is referred to as natural materials that are extracted from the domestic environment and are used in the domestic economy, i.e., the total input of natural materials by extraction.

Domestic processed output (DPO): is referred to as materials that are released to the domestic environment after being processed in the domestic economy, i.e., the total output of processed materials by release.

Import (IM): is referred to as all goods (in the form of raw materials, semi-finished materials, and final products) that originated from other economies and are further used in the domestic economy. It is calculated as the sum of all imported goods.

Export (EX): is referred to as all goods that originated from the domestic economy and are transported to other economies to be used. It is calculated as the sum of all exported goods.

Domestic material input (DMI): is referred to as materials that originated from the domestic environment by extraction and other economies and are available (to be used or to be stored) for the domestic economy. It is calculated as the sum of DE plus IM, as shown in Eq. 27.$$DMI=DE+IM$$
(27)

Domestic material consumption (DMC): is referred to as materials that are directly used in the domestic economy after parts of them are exported to other economies. It is calculated as the difference between DMI and EX.

Physical trade balance (PTB): is referred to as a surplus or deficit of materials for the domestic economy. It is calculated as the difference between IM and EX.

Net additions to stock (NAS): is referred to as materials that remain in the domestic economy. It is calculated by taking BI items into account, as shown in Eq. 28.

$$NAS=DMC+B{I}_{in}-DPO-B{I}_{out}$$
(28) More

Experimental set-upFour vegetation species were selected: Spartina maritima, Salicornia europaea, Halimione portulacoides and Juncus maritimus. These species were chosen for a broad representation of the biomechanical properties and morphological characteristics of saltmarsh species42,43. Plants were collected in Cantabrian estuaries in late summer and early autumn (from early September to late October) during low tide (please refer to the “Methods” section). A total of 105 boxes were collected, of which 94 boxes were used to build a 9.05 m long and 0.58 m wide meadow in a flume (Fig. 1). Five boxes were used to directly estimate the meadow standing biomass in the field (Sample 1 in Table 1), leaving 6 extra boxes for possible contingencies.Figure 1(A) Shows a sketch of the experimental flume, where the vegetation box distribution in the 100% and 50% density cases is displayed in the two upper panels and a lateral view in the bottom panel. The green boxes indicate the vegetated area in each case. Free surface sensors are displayed by blue lines and numbers. (B) Shows the four species within the flume. From left to right: view of the Spartina sp. frontal edge, aerial view of Salicornia sp., frontal view of Juncus sp. and top view of the Halimione sp. rear edge.Full size imageTable 1 Standing biomass (g/m2) and plant height (m) for the four species.Full size tableExperiments were conducted in a flume 20.71 m long and 0.58 m wide at the University of Cantabria. The flume is equipped with a piston wave maker at its left end and a dissipation beach at the rear end. The 94 vegetation boxes used to create a meadow were introduced into the flume following the pattern shown in panel A of Fig. 1 to minimize any edge effects along the edges of the boxes. To ensure a smooth transition from the bottom of the channel to the vegetated area, two false bottoms were constructed with wood, and a thin sediment layer was glued to the wood to mimic the field roughness.Three meadow densities per species were considered. The meadow density directly determined in the field was chosen under the 100% density scenario. To consider a second meadow density, and therefore a second standing biomass value, plants were removed from half of the boxes following the pattern shown in Panel A of Fig. 1 to prevent creating preferential flow channels along the meadow. This case was considered the 50% density scenario. The study of these two biomass scenarios for each vegetation species is carried out with the aim of covering a wide range of standing biomass values, including low values that may be more representative of meadow winter conditions, thus facilitating the applicability of obtained results. Finally, a second cut was made, in which all plants were removed, resulting in the final scenario with a zero density. Plants were cut from above to avoid any damage along the meadow surface (as shown in Supplementary Fig. S2). In each cut, plants in 5 boxes along the leading edge and in 5 boxes at the center of the meadow were collected to quantify the standing biomass (Samples 2 and 3 for the first cut and Sample 4 and 5 for the second cut in Table 1). Therefore, the standing biomass could be monitored throughout the entire duration of the experiments, from the field until the second cut, when all plants were removed.Once located in the flume, the meadow was evaluated under regular and random wave conditions considering three water depths, i.e., h = 0.20, 0.30 and 0.40 m. Regular waves were generated using Stokes II-, III- and V-order and Cnoidal theories when applicable. Wave heights ranging from 0.05 to 0.15 m and wave periods varying between 1.5 and 4 s were considered. Random waves were generated using a Jonswap spectrum with a peak enhancement factor of 3.3, a significant wave height varying between 0.05 and 0.15 m and a peak wave period ranging from 1.8 to 4.8 s (please refer to Supplementary Table S1). Additionally, all wave conditions were considered under the zero-density scenario with bare soil for each species. The wave height evolution along the flume was recorded using 15 capacitive free surface gauges, as shown in Fig. 1 (please refer to Supplementary Table S2 for detailed coordinates).Meadow characteristics analysisThe characteristics of the vegetation meadows were analyzed by measuring the standing biomass throughout the full duration of the experiments and by measuring the individual plant height (please refer to the “Methods” section). The mean standing biomass value obtained for each species was considered the value associated with the 100% density scenario. Then, half of the standing biomass value was considered under the 50% density scenarios since half of the boxes was randomly cut, and the standing biomass values obtained after the second cut agreed with those obtained after the first cut and in the field, as indicated in Table 1. The plant height for each species was also measured (please refer to the “Methods” section), and the resultant mean value detailed in Table 1 was considered.Wave height attenuation analysisWave height attenuation analysis was performed following previous studies reported in the literature assessing the capacity by fitting a damping coefficient6,7,35,44. The18 formulation was used for regular waves, and that of19 was used for random waves (please refer to the “Methods” section). Cases with a zero density were also considered in this analysis to quantify the influence of bare soil friction by determining the corresponding damping coefficient, ({beta }_{B}). Consequently, β was obtained in the 100% and 50% density cases and the cases without vegetation (please refer to Supplementary Tables S3, S4 and S5 to find the obtained coefficients for all cases). This allowed the determination of a new damping coefficient isolating the effect of the standing biomass, ({beta }_{SB}), following24 (please refer to the “Methods” section). Figure 2 shows an example of wave height attenuation analysis for the four species and the different densities under wave condition JS07 (Supplementary Table S1).Figure 2Analysis of wave attenuation under wave condition JS07 for Spartina sp. 100% (S100), 50% (S050) and zero density (S000); Salicornia sp. 100% (L100), 50% (L050) and zero density (L000); Juncus sp. 100% (J100), 50% (J050) and zero density (J000); and Halimione sp. 100% (H100), 50% (H050) and zero density (H000). The damping coefficients for the bare soil cases, ({beta }_{B}), are displayed in blue. The damping coefficients for the 100% and 50% density cases, (beta ), are displayed in dark and light green, respectively. The damping coefficients obtained after subtracting the dissipation obtained in the bare soil cases, ({beta }_{SB}), are displayed in black and dark gray. 95% confidence interval is shown in brackets and correlation coefficient (({rho }^{2})) for each fit is also displayed.Full size imageThe damping coefficients for the bare soil cases shown in Fig. 2, ({beta }_{B}), are consistent with the soil properties observed in the field. Spartina sp. was collected in a muddy area, whereas the other three species were collected in areas with coarser sediments and exhibited a mixture of sand and mud. For all species, wave dissipation was significantly higher under the 100% density scenario than that under the 50% density cases, as expected, highlighting the importance of the standing biomass in wave energy dissipation. It was also observed that bottom friction-induced dissipation plays a more important role for the pioneer species, i.e., Spartina sp. and Salicornia sp., than for the upper marsh species, i.e., Juncus sp. and Halimione sp., which can dissipate wave energy to a greater extent.The importance of wave parameters in the resultant wave attenuation has been highlighted by several works in the literature. Therefore, not only vegetation characteristics but also incident wave conditions determine the coastal protection capacity. Figure 3 shows a comparison of the obtained wave height attenuation due to Halimione sp. under the different wave conditions.Figure 3Analysis of wave attenuation under the different irregular wave conditions for the Halimione sp. 100% (H100) and zero-density (H000) cases. The top panel shows two cases with different h but equal Hs and Tp values (JS01 and JS08), the middle panel shows two cases with different Tp but equal h and Hs values (JS10 and JS11), and the bottom panel shows two cases with different Hs but equal h and Tp values (JS09 and JS12). 95% confidence interval is shown in brackets and correlation coefficient (({rho }^{2})) for each fit is also displayed.Full size imageThe top panel in Fig. 3 shows two cases where Hs and Tp are equal, i.e., JS01 and JS08 in Supplementary Table S1, and two water depths are considered, namely, h = 0.2 and 0.3 m. As can be observed, wave damping is higher for the smallest water depth, where most of the water column is covered by vegetation since the mean vegetation height for Halimione sp. reaches 0.187 m (Table 1). The importance of the water depth with respect to the plant height in terms of wave height attenuation has been reported by several authors44,45,46 who have highlighted this aspect based on the submergence ratio, i.e., the plant height divided by the water depth, revealing higher attenuation at lower submergence ratios on a consistent basis. Bottom friction attenuation is also higher for the smallest water depth, as expected.The middle panel of Fig. 3 shows two cases with equal h and Hs but different Tp values, namely, JS10 and JS11 in Supplementary Table S1. Wave height attenuation is higher for the shortest wave period, as well as the damping produced by bottom friction. This is in line with previous studies, such as35 and44, who conducted experiments involving simulated and real saltmarshes, respectively. Finally, the bottom panel of Fig. 3 shows two cases with different Hs but equal h and Tp values, i.e., JS09 and JS12 in Supplementary Table S1. As widely reported in the literature, e.g.,7,47,48, wave height attenuation increases with the wave height, as shown in the bottom panel of Fig. 3. Bottom friction also increases with the wave height, as expected.A set of damping coefficients was obtained via the 288 tests conducted in the laboratory, 144 tests involving regular waves and 144 tests involving random waves. Additionally, in all cases, the damping coefficient considering the isolated effect of the standing biomass, ({beta }_{SB}), was determined. The relationship of these damping coefficients to the measured standing biomass is explored in the next section with the aim of establishing a new relationship to estimate the wave damping effect of the different saltmarsh species based on the standing biomass, without the need for data fitting.Wave damping coefficient as a function of the standing biomassThe mean standing biomass obtained for the different species, Table 1, is considered here to analyze the relationship with the wave damping coefficients obtained by fitting18 formulation to wave heights measured along the meadow for regular waves and19 formulation for random waves. The plant height was highly variable among the different species (Table 1), ranging from 0.170 m for Spartina sp. to 0.714 m for Juncus sp. Then, some species were submerged at all tested water depths, while other species remained above water in all tests. In the latter cases, there remained a portion of each plant above the water level, thus not contributing to wave attenuation. To consider the actual interaction between the standing biomass and flow conditions and assuming a uniform vertical distribution, the effective standing biomass, (ESB), can be defined as follows:$$ESB=DryWeight*frac{minleft{{h}_{v},hright}}{{h}_{v}}$$
(1)
where (DryWeight) denotes the measured dry weight for each species (g/m2), ({h}_{v}) is the mean plant height and (h) is the water depth. Additionally, in the submerged cases, the same (ESB) value will impact flow differently depending on the submergence ratio, (SR), as defined in Eq. (2). To consider this effect, the standing biomass ratio, (SBR) in Eq. (3), can be defined as follows:$$SR=frac{{h}_{v}}{h}, ;;where ;; SR=1 ;;for ;;{h}_{v} >h$$
(2)
$$SBR=ESB*SR$$
(3)
Figure 4 shows the relationship between (SBR) and the measured wave damping coefficient, (beta ). The results for regular and random waves are displayed for each water depth, and a linear fit was found under each condition.Figure 4Wave damping coefficient, (beta ), as a function of the standing biomass ratio, (SBR), under all regular (left panels) and random (right panels) wave conditions. Each panel shows the wave trains assessed at each water depth, h = 0.20, 0.30 and 0.40 m. The results for the 100% density case are marked with circles and those for the 50% density case are marked with squares. The linear fitting results obtained under each wave condition are also displayed.Full size imageUnder each wave condition, a linear fitting relationship between (beta ) and (SBR) was obtained for the eight (SBR) values, as shown in Fig. 4. For similar (SBR) values, the highest (beta ) values were consistently obtained at the smallest water depth, highlighting the notable influence of this parameter on the obtained wave attenuation. Following previous works, such as those of24 and25, who considered the vegetation submerged solid volume fraction to estimate the resulting wave attenuation and established a common relationship for different water depths, the volumetric standing biomass, (VSB), can be defined as follows:$$VSB= SBR*frac{1}{h}$$
(4)
(VSB) is expressed in units of g/m3, which is the weight per unit volume. Exploring the relationship of (beta ) with this new parameter, it was found that the results for the three water depths could be fitted with a single linear relationship, as shown in Fig. 5. However, despite the linear trend observed in Fig. 5, notable data scatter was observed for each (VSB) value. Each of these groups corresponds to a certain water depth and (SBR) value, which were determined under different wave heights and wave periods.Figure 5Wave damping coefficient, (beta ), as a function of the volumetric standing biomass, (VSB), under all regular (top panel) and random (bottom panel) wave conditions. The obtained linear fitting results are displayed in both panels. 95% confidence interval is shown in brackets and correlation coefficient (({rho }^{2})) for each fit is also displayed.Full size imageFinally, to account for the characteristics of the incident wave conditions, including the wave height and period, two nondimensional parameters were considered. The first parameter, considering the wave height, is the relative wave height, defined as the ratio of the incident wave height to the water depth, (H/h). Previous studies have highlighted the importance of this parameter in the resultant wave attenuation (e.g.24,44). Under random wave conditions, the considered wave height is ({H}_{rms}), according to wave attenuation analysis. The second parameter, considering the effect of the different wave periods and the importance of the number of wave lengths inside the vegetation length49, is the relative meadow length, defined as the ratio of the meadow length to the wave length, ({L}_{v}/L). To ensure consistency with the above wave attenuation analysis, in which the wave damping amount per unit length was obtained, the unit meadow length was considered here. Thus, the hydraulic standing biomass, (HSB), can be defined as:$$HSB=VSB*frac{H}{h}*frac{{L}_{v}}{L}$$
(5)
Figure 6 shows the relationship obtained between (beta ) and this new variable under all regular and random conditions following the linear fitting relationship of (beta =A*HSB+B), where (A) and (B) are fitting constants with units of (g/m2)−1 and m−1, respectively.Figure 6Wave damping coefficient, (beta ), as a function of the hydraulic standing biomass, (HSB), under all regular (top panel) and random (bottom panel) wave conditions. Both panels show linear fitting results obtained without considering the saturation point, indicated by the black solid line, and those obtained considering the saturation point, indicated by the gray solid line. The black dashed line indicates the saturation point. 95% confidence interval is shown in brackets and correlation coefficient (({rho }^{2})) for each fit is also displayed.Full size imageThe linear fitting results obtained between (beta ) and (HSB) under regular and random wave conditions are shown in Fig. 6 as solid black lines and expressed as Eqs. (6) and (7), respectively, where values between brackets are the 95% confidence interval for each coefficient.$$beta =9.206cdot {10}^{-4} left(9.006cdot {10}^{-5}right)*HSB+0.103 (0.021)$$
(6)
$$beta =1.192 cdot {10}^{-3} left(9.124 cdot {10}^{-5}right)*HSB+0.071 (0.016)$$
(7)
The inclusion of incident wave condition characteristics reduces the resulting data scatter, highlighting the role of the wave height and period in the obtained wave attenuation, as described in the previous section. An interesting aspect observed in Fig. 6 is that the four cases with the highest wave damping coefficients yielded similar values for the different (HSB) values. Under regular wave conditions, the mean (beta ) value for these four cases is 0.76, and under random wave conditions, the value reaches 0.68. This may indicate that the damping coefficient has reached its maximum value and no longer increases with increasing (HSB) value. To analyze this aspect in more detail, the wave height evolution measured for the four tests in which (beta ) reaches its maximum value are plotted (as shown in Supplementary Fig. S3). These tests correspond to Halimione sp. with a density of 100% and the shallowest water depth, h = 0.20 m. This species achieved the highest standing biomass value among the species considered in these experiments, and for h = 0.20 m, almost the entire water column was covered by vegetation. For these tests, a notable wave height attenuation was observed, where the wave height strongly decayed along the first 5 m of vegetation, and the wave height entirely dissipated along the last 4 m (as shown in Supplementary Fig. S3). The wave damping equation cannot suitably reproduce the strong wave decay within this few meters. Then, an almost constant wave damping coefficient value is reached under the different considered wave conditions, and a saturation regime is observed, in which the wave height beyond the meadow can be assumed to be negligible. To consider this phenomenon, a two-section fitting relationship is proposed, as shown in Fig. 6. The value of the saturation damping coefficient, chosen as the mean value of the four cases analyzed, is plotted as a dashed gray line, and a linear fit is obtained for the remaining data. The two-section fitting relationship is expressed in Eqs. (8) and (9) for both regular and random waves, respectively, where values between brackets are the 95% confidence interval for each coefficient.$$beta =left{begin{array}{ll}1.020 cdot {10}^{-3}left(1.112 cdot {10}^{-4}right)*HSB+0.088 ; (0.020) \ 0.758; (0.027)end{array}right. begin{array}{l} ;;0 < HSB < 659\ ;; HSB > 659end{array}$$
(8)
$$beta =left{begin{array}{l}1.310cdot {10}^{-3}left(1.232cdot {10}^{-4}right)*HSB+0.059; (0.017) \ 0.684 ;(0.066)end{array}right. begin{array}{l};;0474end{array}$$
(9)
All damping coefficients considered in the previous analysis were obtained without subtracting any additional source of dissipation such as bottom and wall friction. Previous works, such as24, highlighted the high importance of considering any other sources of wave dissipation besides the effect of vegetation elements when quantifying the wave height attenuation capacity. In this case, the flume walls were made of glass, and the friction induced by these walls could be considered negligible. However, bottom friction could be significant, as observed in tests run after removing all vegetation stems. Then, the wave damping coefficient obtained after subtracting the bottom friction contribution, ({beta }_{SB}), is studied here. Figure 7 shows the relationship obtained between this damping coefficient, ({beta }_{SB}), and hydraulic standing biomass, (HSB).Figure 7Wave damping coefficient, ({beta }_{SB}), as a function of the hydraulic standing biomass, (HSB), under all regular (top panel) and random (bottom panel) wave conditions. Both panels show linear fitting results obtained without considering the saturation point, indicated by the black solid line, and those obtained considering the saturation point, indicated by the gray solid line. The black dashed line indicates the saturation point. 95% confidence interval is shown in brackets and correlation coefficient (({rho }^{2})) for each fit is also displayed.Full size imageA linear relationship was also obtained for ({beta }_{SB}), revealing correlation coefficients similar to those obtained when analyzing (beta ). The obtained linear relationships under regular and random wave conditions are expressed as Eqs. (10) and (11), respectively, where values between brackets are the 95% confidence interval for each coefficient. A two-section fitting relationship, Eqs. (12) and (13), was also included considering the saturation regime obtained in the Halimione sp. 100% density and h = 0.20 m cases with a ({beta }_{SB}=) 0.69 and 0.63 under regular and random wave conditions, respectively.$${beta }_{SB}=1.051*{10}^{-3} left(7.063cdot {10}^{-5}right)*HSB$$
(10)
$${beta }_{SB}=1.296*{10}^{-3} left(6.894cdot {10}^{-5}right)*HSB$$
(11)
$${beta }_{SB}=left{begin{array}{l}1.151cdot {10}^{-3} left(7.445cdot {10}^{-5}right)*HSB \ 0.685 ;(0.047)end{array}right. begin{array}{l} ;; 0599end{array}$$
(12)
$${beta }_{SB}=left{begin{array}{l}1.396cdot {10}^{-3}left(7.919cdot {10}^{-5}right)*HSB \ 0.631 ;left(0.055right)end{array}right. begin{array}{l};; 0451end{array}$$
(13)
As can be noted, the ({beta }_{SB}) values are significantly lower than those obtained for (beta ), especially in the shallowest water depth cases where bottom friction is the highest, as discussed above. The estimation of (beta ) and ({beta }_{SB}) allows two possible approaches to determine the wave damping effect of a saltmarsh. The first approach, based on (beta ), includes wave damping induced by the combined effect of vegetation and bottom friction. Therefore, the consideration of (beta ) in analytical or numerical analysis could provide the total dissipation induced by the species under study, and sediment characteristics are not necessary for analysis. Considering that saltmarsh species grow in muddy to sandy environments and that the major contribution to the obtained wave attenuation is associated with vegetation, this approach may be the best option if soil properties are not thoroughly characterized.The second approach relies on the definition of ({beta }_{SB}). In this case, the wave damping contributions of vegetation drag and bottom friction are separated. Then, ({beta }_{SB}) can be used in cases where the effect of both momentum sinks can be separately evaluated. To quantify the wave damping contribution of vegetation drag only, ({beta }_{SB}) can be used, and then, the additional friction due to the bottom effect can be added considering the soil properties in each case. This second approach assumes a linear sum of both momentum sinks and could be applicable when soil properties are thoroughly characterized. More