Ecology

Huijbers, P. M. C., Flach, C.-F. & Larsson, D. G. J. A conceptual framework for the environmental surveillance of antibiotics and antibiotic resistance. Environ. Int. 130, 104880 (2019).Article 

Google Scholar 
Aarestrup, F. M. & Woolhouse, M. E. J. Using sewage for surveillance of antimicrobial resistance. Science 367, 630–632 (2020).Article 

Google Scholar 
European Commission. Proposal for a revised Urban Wastewater Treatment Directive. European Commission https://environment.ec.europa.eu/publications/proposal-revised-urban-wastewater-treatment-directive_en (2022).US Centres for Disease Control and Prevention. COVID-19 impacts on environment (e.g., water, soil) and sanitation: addressing antimicrobials and antimicrobial resistant threats in the environment. US Centres for Disease Control and Prevention https://www.cdc.gov/drugresistance/pdf/covid19/COVID19-Impacts-AR-Environment-Sanitation-508.pdf (2021).Flach, C.-F., Hutinel, M., Razavi, M., Åhrén, C. & Larsson, D. G. J. Monitoring of hospital sewage shows both promise and limitations as an early-warning system for carbapenemase-producing Enterobacterales in a low-prevalence setting. Water Res. 200, 117261 (2021).Article 

Google Scholar 
Larsson, D. G. J. & Flach, C.-F. Antibiotic resistance in the environment. Nat. Rev. Microbiol. 20, 257–269 (2022).Article 

Google Scholar 
Newton, R. J. et al. Sewage reflects the microbiomes of human populations. mBio 6, e02574 (2015).Article 

Google Scholar 
Huijbers, P. M. C., Larsson, D. G. J. & Flach, C. F. Surveillance of antibiotic resistant Escherichia coli in human populations through urban wastewater in ten European countries. Environ. Pollut. 261, 114200 (2020).Article 

Google Scholar 
Laxminarayan, R. & Macauley, M. K. The Value of Infromation: Methodological Frontiers and New Applications in Environment and Health 1st edn (Springer Dordrecht, 2012).Munk, P. et al. Genomic analysis of sewage from 101 countries reveals global landscape of antimicrobial resistance. Nat. Commun. 13, 7251 (2022).Article 

Google Scholar  More

Johnson, C. N. et al. Biodiversity losses and conservation responses in the Anthropocene. Science 356, 270–275 (2017).Article 
CAS 
PubMed 

Google Scholar 
Rockström, J. et al. A safe operating space for humanity. Nature 461, 472–475 (2009).Article 
PubMed 

Google Scholar 
Maxwell, S. L. et al. Area-based conservation in the twenty-first century. Nature 586, 217–227 (2020).Article 
CAS 
PubMed 

Google Scholar 
Schulze, K. et al. An assessment of threats to terrestrial protected areas. Conserv. Lett. 11, e12435 (2018).Article 

Google Scholar 
Bingham, H. C. et al. (eds). Protected Planet Report 2020 (UNEP-WCMC & IUCN, 2021); https://livereport.protectedplanet.net/Buchanan, G. M., Butchart, S. H., Chandler, G. & Gregory, R. D. Assessment of national-level progress towards elements of the Aichi Biodiversity Targets. Ecol. Indic. 116, 106497 (2020).Article 

Google Scholar 
Xu, H. et al. Ensuring effective implementation of the post-2020 global biodiversity targets. Nat. Ecol. Evol. 5, 411–418 (2021).Article 
PubMed 

Google Scholar 
Report of the Open-ended Working Group on the Post-2020 Global Biodiversity Framework on Its Third Meeting (CBD Secretariat, 2022); https://www.cbd.int/conferences/post2020/wg2020-03/documentsRodrigues, A. S. & Cazalis, V. The multifaceted challenge of evaluating protected area effectiveness. Nat. Commun. 11, 5147 (2020).Article 
PubMed Central 
CAS 
PubMed 

Google Scholar 
Geldmann, J., Manica, A., Burgess, N. D., Coad, L. & Balmford, A. A global-level assessment of the effectiveness of protected areas at resisting anthropogenic pressures. Proc. Natl Acad. Sci. USA 116, 23209–23215 (2019).Article 
PubMed Central 
CAS 
PubMed 

Google Scholar 
Starnes, T. et al. The extent and effectiveness of protected areas in the UK. Glob. Ecol. Conserv. 30, e01745 (2021).Article 

Google Scholar 
Kremen, C. et al. Aligning conservation priorities across taxa in Madagascar with high-resolution planning tools. Science 320, 222–226 (2008).Article 
CAS 
PubMed 

Google Scholar 
Cazalis, V. et al. Mismatch between bird species sensitivity and the protection of intact habitats across the Americas. Ecol. Lett. 24, 2394–2405 (2021).Article 
PubMed 

Google Scholar 
Venter, O. et al. Targeting global protected area expansion for imperiled biodiversity. PLoS Biol. 12, e1001891 (2014).Article 
PubMed Central 
PubMed 

Google Scholar 
Gamero, A. et al. Tracking progress toward EU biodiversity strategy targets: EU policy effects in preserving its common farmland birds. Conserv. Lett. 10, 395–402 (2017).Article 

Google Scholar 
Pellissier, V. et al. Effects of Natura 2000 on nontarget bird and butterfly species based on citizen science data. Conserv. Biol. 34, 666–676 (2020).Article 
CAS 
PubMed 

Google Scholar 
Princé, K., Rouveyrol, P., Pellissier, V., Touroult, J. & Jiguet, F. Long-term effectiveness of Natura 2000 network to protect biodiversity: a hint of optimism for common birds. Biol. Conserv. 253, 108871 (2021).Article 

Google Scholar 
Cunningham, C. A., Thomas, C. D., Morecroft, M. D., Crick, H. Q. P. & Beale, C. M. The effectiveness of the protected area network of Great Britain. Biol. Conserv. 257, 109146 (2021).Article 

Google Scholar 
Duckworth, G. D. & Altwegg, R. Effectiveness of protected areas for bird conservation depends on guild. Divers. Distrib. 24, 1083–1091 (2018).Article 
PubMed Central 
PubMed 

Google Scholar 
Rada, S. et al. Protected areas do not mitigate biodiversity declines: a case study on butterflies. Divers. Distrib. 25, 217–224 (2019).Article 

Google Scholar 
Terraube, J., Van Doninck, J., Helle, P., & Cabeza, M. Assessing the effectiveness of a national protected area network for carnivore conservation. Nat. Commun. 11, 2957 (2020).Article 
PubMed Central 
CAS 
PubMed 

Google Scholar 
Lenoir, J. et al. Species better track the shifting isotherms in the oceans than on land. Nat. Ecol. Evol. 4, 1044–1059 (2020).Article 
PubMed 

Google Scholar 
van Teeffelen, A., Meller, L., van Minnen, J., Vermaat, J. & Cabeza, M. How climate proof is the European Union’s biodiversity policy? Regional Environ. Change 15, 997–1010 (2015).Article 

Google Scholar 
Thomas, C. D. & Gillingham, P. K. The performance of protected areas for biodiversity under climate change. Biol. J. Linn. Soc. Lond. 115, 718–730 (2015).Article 

Google Scholar 
Gillingham, P. K. et al. The effectiveness of protected areas in the conservation of species with changing geographical ranges. Biol. J. Linn. Soc. Lond. 115, 707–717 (2015).Article 

Google Scholar 
Geldmann, J. et al. Effectiveness of terrestrial protected areas in reducing habitat loss and population declines. Biol. Conserv. 161, 230–238 (2013).Article 

Google Scholar 
Stokstad, E. Species? Climate? Cost? Ambitious goal means trade-offs. Science 371, 555 (2021).Article 
CAS 
PubMed 

Google Scholar 
Brlík, V. et al. Long-term and large-scale multispecies dataset tracking population changes of common European breeding birds. Sci. Data 8, 21 (2021).Article 
PubMed Central 
PubMed 

Google Scholar 
Stanbury, A. et al. The status of bird populations: the fifth Birds of Conservation Concern in the United Kingdom, Channel Islands and Isle of Man and second IUCN Red List assessment of extinction risk for Great Britain. Br. Birds 114, 723–747 (2021).
Google Scholar 
Dudley, N. (ed). Guidelines for Applying Protected Area Management Categories (IUCN, 2008).Deguignet, M. et al. Measuring the extent of overlaps in protected area designations. PLoS ONE 12, e0188681 (2017).Article 
PubMed Central 
PubMed 

Google Scholar 
JNCC. Common Standards Monitoring: Introduction to the Guidance Manual (JNCC Resource Hub, 2004).Hayhow, D. B. et al. State of Nature 2019 (RSPB, 2019).Schleicher, J. et al. Statistical matching for conservation science. Conserv. Biol. 34, 538–549 (2019).Article 
PubMed Central 
PubMed 

Google Scholar 
Waldron, A. et al. Protecting 30% of the Planet for Nature: Costs, Benefits and Economic Implications (Campaign for Nature, 2020); https://helda.helsinki.fi/handle/10138/326470Franks, S. E., Roodbergen, M., Teunissen, W., Carrington Cotton, A. & Pearce‐Higgins, J. W. Evaluating the effectiveness of conservation measures for European grassland‐breeding waders. Ecol. Evol. 8, 10555–10568 (2018).Article 
PubMed Central 
PubMed 

Google Scholar 
Pearce-Higgins, J. W. et al. Site-based adaptation reduces the negative effects of weather upon a southern range margin Welsh black grouse Tetrao tetrix population that is vulnerable to climate change. Clim. Change 153, 253–265 (2019).Article 

Google Scholar 
Jellesmark, S. et al. A counterfactual approach to measure the impact of wet grassland conservation on U.K. breeding bird populations. Conserv. Biol. 35, 1575–1585 (2021).Article 
PubMed 

Google Scholar 
Morrison, C. A. et al. Covariation in population trends and demography reveals targets for conservation action. Proc. Biol. Sci. 288, 20202955 (2021).PubMed Central 
PubMed 

Google Scholar 
Donald, P. F. et al. International conservation policy delivers benefits for birds in Europe. Science 317, 810–813 (2007).Article 
CAS 
PubMed 

Google Scholar 
Martay, B. et al. Monitoring landscape-scale environmental changes with citizen scientists: Twenty years of land use change in Great Britain. J. Nat. Conserv. 44, 33–42 (2018).Article 

Google Scholar 
Sullivan, M. J. P., Newson, S. E. & Pearce‐Higgins, J. W. Changing densities of generalist species underlie apparent homogenization of UK bird communities. Ibis 158, 645–655 (2016).Article 

Google Scholar 
Wauchope, H. S. et al. Evaluating impact using time-series data. Trends Ecol. Evol. 36, 196–205 (2021).Article 
PubMed 

Google Scholar 
Devictor, V. et al. Differences in the climatic debts of birds and butterflies at a continental scale. Nat. Clim. Change 2, 121–124 (2012).Article 

Google Scholar 
Lehikoinen, P., Santangeli, A., Jaatinen, K., Rajasärkkä, A. & Lehikoinen, A. Protected areas act as a buffer against detrimental effects of climate change—evidence from large‐scale, long‐term abundance data. Glob. Change Biol. 25, 304–313 (2019).Article 

Google Scholar 
Gaüzère, P., Jiguet, F. & Devictor, V. Can protected areas mitigate the impacts of climate change on bird’s species and communities? Diversity Distrib. 22, 625–637 (2016).Article 

Google Scholar 
Neate‐Clegg, M. H. C., Jones, S. E. I., Burdekin, O., Jocque, M. & Şekercioğlu, Ç. H. Elevational changes in the avian community of a Mesoamerican cloud forest park. Biotropica 50, 805–815 (2018).Article 

Google Scholar 
Oliver, T. H. et al. Large extents of intensive land use limit community reorganization during climate warming. Glob. Change Biol. 23, 2272–2283 (2017).Article 

Google Scholar 
Hiley, J. R., Bradbury, R. B., Holling, M. & Thomas, C. D. Protected areas act as establishment centres for species colonizing the UK. Proc. Biol. Sci. 280, 20122310 (2013).PubMed Central 
PubMed 

Google Scholar 
Thomas, C. D. et al. Protected areas facilitate species’ range expansions. Proc. Natl Acad. Sci. USA 109, 14063–14068 (2012).Article 
PubMed Central 
CAS 
PubMed 

Google Scholar 
Grace, M. K. et al. Testing a global standard for quantifying species recovery and assessing conservation impact. Conserv. Biol. 35, 1833–1849 (2021).Article 
PubMed 

Google Scholar 
Gibbons, D. W., Reid, J. B. & Chapman, R. A. The New Atlas of Breeding Birds in Britain & Ireland 1988–1991 (T. & A. D. Poyser, 1993).Balmer, D. E. et al. Bird Atlas 2007–11: the Breeding and Wintering Birds of Britain and Ireland (BTO, 2013).Gillings, S. et al. Breeding and wintering bird distributions in Britain and Ireland from citizen science bird atlases. Glob. Ecol. Biogeogr. 28, 866–874 (2019).Article 

Google Scholar 
Freeman, S. N., Noble, D. G., Newson, S. E. & Baillie, S. R. Modelling population changes using data from different surveys: the Common Birds Census and the Breeding Bird Survey. Bird Study 54, 61–72 (2007).Article 

Google Scholar 
Robinson, R. A., Julliard, R. & Saracco, J. F. Constant effort: studying avian population processes using standardised ringing. Ring. Migr. 24, 199–204 (2009).Article 

Google Scholar 
Cave, V. M., Freeman, S. N., Brooks, S. P., King, R. & Balmer, D. E. in Modeling Demographic Processes in Marked Populations, 949–963 (Springer, 2009).Rowland, C. S. et al. Land Cover Map 2015 (1km Percentage Aggregate Class, GB) (eds Thomson, D. L. et al) (Environmental Information Data Centre, 2017); https://doi.org/10.5285/7115bc48-3ab0-475d-84ae-fd3126c20984Rowland, C. S. et al. Land Cover Map 2015 (1km Percentage Aggregate Class, N. Ireland) (Environmental Information Data Centre, 2017); https://doi.org/10.5285/362feaea-0ccf-4a45-b11f-980c6b89a858ASTER Global Digital Elevation Model V003 (dataset). NASA EOSDIS Land Processes DAAC (NASA/METI/AIST/Japan Space Systems and U.S./Japan ASTER Science Team, 2019); https://doi.org/10.5067/ASTER/ASTGTM.003Schiavina, M., Freire, S. & MacManus, K. GHS-SMOD R2019A – GHS Settlement Layers, Updated and Refined REGIO Model 2014 in Application to GHS-BUILT R2018A and GHS-POP R2019A, Multitemporal (1975-1990-2000-2015) (European Commission Joint Research Centre, 2019); https://doi.org/10.2905/42E8BE89-54FF-464E-BE7B-BF9E64DA5218Robinson, R. A. BirdFacts: Profiles of Birds Occurring in Britain & Ireland (BTO, 2005).Gibbons, D. W. et al. Bird species of conservation concern in the United Kingdom, Channel Islands and Isle of Man: revising the Red Data List. RSPB Conserv. Rev. 10, 7–18 (1996).
Google Scholar 
Stone, B. H. et al. Population estimates of birds in Britain and in the United Kingdom. Br. Birds 90, 1–22 (1997).
Google Scholar 
Woodward, I. et al. Population estimates of birds in Great Britain and the United Kingdom. Br. Birds 113, 69–104 (2020).
Google Scholar 
R Core Team (2020). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/Joppa, L. N. & Pfaff, A. High and far: biases in the location of protected areas. PLoS ONE 4, e8273 (2009).Article 
PubMed Central 
PubMed 

Google Scholar 
Bull, J. W., Strange, N., Smith, R. J. & Gordon, A. Reconciling multiple counterfactuals when evaluating biodiversity conservation impact in social‐ecological systems. Conserv. Biol. 35, 510–521 (2020).Article 
PubMed 

Google Scholar 
Jellesmark, S. et al. Assessing the global impact of targeted conservation actions on species abundance. Preprint at bioRxiv https://doi.org/10.1101/2022.01.14.476374 (2022).Wauchope, H. S. et al. Protected areas have a mixed impact on waterbirds but management helps. Nature 605, 103–107 (2022).Article 
CAS 
PubMed 

Google Scholar 
Ho, D. E., Imai, K., King, G. & Stuart, E. A. MatchIt: nonparametric preprocessing for parametric causal inference. J. Stat. Softw. 42, 1–28 (2011).Article 

Google Scholar 
Wood, S. N. Generalized Additive Models: an Introduction with R 2nd edn (Chapman and Hall/CRC, 2017).Hartig, F. DHARMa: Residual diagnostics for hierarchical (multi-level/mixed) regression models. R package v.0.4.4 (2021); https://CRAN.R-project.org/package=DHARMaJetz, W., Thomas, G. H., Joy, J. B., Hartmann, K. & Mooers, A. O. The global diversity of birds in space and time. Nature 491, 444–448 (2012).Article 
CAS 
PubMed 

Google Scholar 
Hadfield, J. D. MCMC methods for multi-response generalized linear mixed models: the MCMCglmm R package. J. Stat. Softw. 33, 1–22 (2010).Article 

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

Google Scholar 
Johnston, A. et al. Species traits explain variation in detectability of UK birds. Bird Study 61, 340–350 (2014).Article 

Google Scholar 
Hill, M. O. Diversity and evenness: a unifying notation and its consequences. Ecology 54, 427–432 (1973).Article 

Google Scholar 
Ho, D. E., Imai, K., King, G. & Stuart, E. A. Matching as nonparametric preprocessing for reducing model dependence in parametric causal inference. Political Anal. 15, 199–236 (2007).Article 

Google Scholar 
Julliard, R., Clavel, J., Devictor, V., Jiguet, F. & Couvet, D. Spatial segregation of specialists and generalists in bird communities. Ecol. Lett. 9, 1237–1244 (2006).Article 
PubMed 

Google Scholar 
Devictor, V., Julliard, R., Couvet, D. & Jiguet, F. Birds are tracking climate warming, but not fast enough. Proc. Biol. Sci. 275, 2743–2748 (2008).PubMed Central 
PubMed 

Google Scholar  More

Global statistics on declining biodiversity can give the impression that every population of every species is in a downward spiral. In fact, many populations are stable or growing, while a small number of species faces truly existential challenges. These photos capture some specific crises. They are images of threats unfolding, of desperate attempts at species defence and of the beautiful living world that is at stake.
The 15th United Nations Biodiversity Conference, COP15, opens in Montreal, Canada, on 7 December. At the meeting, delegates will attempt to agree on goals for stabilizing species’ declines by 2030 and reverse them by mid-century. The current draft framework agreement promises nothing less than a “transformation in society’s relationship with biodiversity”.
Help for the kelp. Tasmania’s forests of giant kelp (Macrocystis pyrifera) are dying as climate change shifts ocean currents, bringing warm water to the east coast of the temperate Australian island. The kelp forests host an entire ecosystem, including abalone and crayfish — both economically important species and part of local food culture. Now, researchers at the Institute for Marine and Antarctic Studies in Hobart are breeding kelp plants that can tolerate warmer conditions, and replanting them along the coast — a trial for what they hope will become a landscape-scale restoration. More

Machado, D. et al. Polarization of microbial communities between competitive and cooperative metabolism. Nat. Ecol. Evol. 5, 195–203 (2021).Article 
PubMed 
PubMed Central 

Google Scholar 
Thompson, L. R. et al. A communal catalogue reveals Earth’s multiscale microbial diversity. Nature 551, 457–463 (2017).Article 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Price, M. Erroneous predictions of auxotrophies by CarveMe. https://doi.org/10.1038/s41559-022-01936-3 (2022).Machado, D., Andrejev, S., Tramontano, M. & Patil, K. R. Fast automated reconstruction of genome-scale metabolic models for microbial species and communities. Nucleic Acids Res. 46, 7542–7553 (2018).Article 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Price, M. N., Deutschbauer, A. M. & Arkin, A. P. GapMind: automated annotation of amino acid biosynthesis. mSystems 5, e00291-20 (2020).Article 
PubMed 
PubMed Central 

Google Scholar 
Mee, M. T., Collins, J. J., Church, G. M. & Wang, H. H. Syntrophic exchange in synthetic microbial communities. Proc. Natl. Acad. Sci. USA 111, E2149–E2156 (2014).Article 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Ponomarova, O. et al. Yeast creates a niche for symbiotic lactic acid bacteria through nitrogen overflow. Cell Syst. 5, 345–357.e6 (2017).Article 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Zengler, K. & Zaramela, L. S. The social network of microorganisms—how auxotrophies shape complex communities. Nat. Rev. Microbiol. 16, 383–390 (2018).Article 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Giri, S. et al. Metabolic dissimilarity determines the establishment of cross-feeding interactions in bacteria. Curr. Biol. 31, 5547–5557.e6 (2021).Article 
CAS 
PubMed 

Google Scholar 
Morris, J. J., Lenski, R. E. & Zinser, E. R. The black queen hypothesis: evolution of dependencies through adaptive gene loss. mBio 3, e00036-12 (2012).Article 
PubMed 
PubMed Central 

Google Scholar 
Campbell, K. et al. Self-establishing communities enable cooperative metabolite exchange in a eukaryote. eLife 4, e09943 (2015).Article 
PubMed 
PubMed Central 

Google Scholar 
D’Souza, G. & Kost, C. Experimental evolution of metabolic dependency in bacteria. PLOS Genet. 12, e1006364 (2016).Article 
PubMed 
PubMed Central 

Google Scholar 
Ziesack, M. et al. Engineered interspecies amino acid cross-feeding increases population evenness in a synthetic bacterial consortium. mSystems 4, e00352-19 (2019).Article 
PubMed 
PubMed Central 

Google Scholar 
Ryback, B., Bortfeld-Miller, M. & Vorholt, J. A. Metabolic adaptation to vitamin auxotrophy by leaf-associated bacteria. ISME J. https://doi.org/10.1038/s41396-022-01303-x (2022). More

Ah-King, M. Flexible mate choice in Encyclopedia of Animal Behavior, 2nd edn Vol. 4 (ed Jae Chun Choe) 421–431 (Academic Press, 2019).Harestad, A. S. & Bunnell, F. L. Home range and body weight—A reevaluation. Ecology 60, 389–402 (1979).Article 

Google Scholar 
O’Neill, R. V., Milne, B. T., Turner, M. G. & Gardner, R. H. Resource utilization scales and landscape pattern. Landsc. Ecol. 2, 63–69 (1988).Article 

Google Scholar 
Tricas, T. C. Determinants of feeding territory size in the corallivorous butterflyfish, Chaetodon multicinctus. Anim. Behav. 37, 830–841. https://doi.org/10.1016/0003-3472(89)90067-5 (1989).Article 

Google Scholar 
Tremblay, I., Thomas, D., Blondel, J., Perret, P. & Lambrechts, M. M. The effect of habitat quality on foraging patterns, provisioning rate and nestling growth in Corsican Blue Tits Parus caeruleus. Ibis 147, 17–24. https://doi.org/10.1111/j.1474-919x.2004.00312.x (2005).Article 

Google Scholar 
Watts, D. P. The influence of male mating tactics on habitat use in Mountain Gorillas (Gorilla gorilla beringei). Primates 35, 35–47. https://doi.org/10.1007/BF02381484 (1994).Article 

Google Scholar 
Lescroël, A. et al. Working less to gain more: when breeding quality relates to foraging efficiency. Ecology 91, 2044–2055. https://doi.org/10.1890/09-0766.1 (2010).Article 
PubMed 

Google Scholar 
Tufto, J., Anderson, R. & Linnell, J. Habitat use and ecological correlates of home range size in a small cervid: the roe deer. J. Anim. Ecol. 65, 715–724. https://doi.org/10.2307/5670 (1996).Article 

Google Scholar 
Morellet, N. et al. Seasonality, weather and climate affect home range size in roe deer across a wide latitudinal gradient within Europe. J. Anim. Ecol. 82, 1326–1339. https://doi.org/10.1111/1365-2656.12105 (2013).Article 
PubMed 

Google Scholar 
Anderson, D. P. et al. Scale-dependent summer resource selection by reintroduced elk in Wisconsin, USA. J. Wildl. Manag. 69, 298–310. https://doi.org/10.2193/0022-541X(2005)069%3c0298:SSRSBR%3e2.0.CO;2 (2005).Article 

Google Scholar 
Olsson, P. M. O. et al. Movement and activity patterns of translocated elk (Cervus elaphus nelsoni) on an active coal mine in Kentucky. Wildl. Biol. Pract. 3, 1–8. https://doi.org/10.2461/wbp.2007.3.1 (2007).Article 

Google Scholar 
Porter, W. P., Sabo, J. L., Tracy, C. R., Reichman, O. J. & Ramankutty, N. Physiology on a landscape scale: plant–animal interactions. Integr. Comp. Biol. 42, 431–453. https://doi.org/10.1093/icb/42.3.431 (2002).Article 
PubMed 

Google Scholar 
Berg, J. E. et al. Mothers’ movements: shifts in calving area selection by partially migratory elk. J. Wildl. Manag. 85, 1476–1489. https://doi.org/10.1002/jwmg.22099 (2021).Article 

Google Scholar 
Lehman, C. P. et al. Elk resource selection at parturition sites, Black Hills, South Dakota. J. Wildl. Manag. 80, 465–478. https://doi.org/10.1002/jwmg.1017 (2016).Article 

Google Scholar 
Johnson, B. K., Kern, J. W., Wisdom, M. J., Findholt, S. L. & Kie, J. G. Resource selection and spatial separation of mule deer and elk during spring. J. Wildl. Manag. 64, 685–697. https://doi.org/10.2307/3802738 (2000).Article 

Google Scholar 
Grace, J. & Easterbee, N. The natural shelter for red deer (Cervus elaphus) in a Scottish glen. J. Appl. Ecol. 16, 37–48. https://doi.org/10.2307/2402726 (1979).Article 

Google Scholar 
Demarchi, M. W. & Bunnell, F. L. Estimating forest canopy effects on summer thermal cover for Cervidae (deer family). Can. J. For. Res. 23, 2419–2426. https://doi.org/10.1139/x93-299 (1993).Article 

Google Scholar 
Parker, K. L. & Gillingham, M. P. Estimates of critical thermal environments for mule deer. J. Range. Manag. 43, 73–81 (1990).Article 

Google Scholar 
Proffitt, K. M. et al. Changes in elk resource selection and distributions associated with a late-season elk hunt. J. Wildl. Manag. 74, 210–218. https://doi.org/10.2193/2008-593 (2010).Article 

Google Scholar 
Webb, S. L., Dzialak, M. R., Harju, S. M., Hayden-Wing, L. D. & Winstead, J. B. Influence of land development on home range use dynamics of female elk. Wildl. Res. 38, 163–167. https://doi.org/10.1071/WR10101 (2011).Article 

Google Scholar 
Rumble, M. A., Benkobi, L. & Gamo, R. S. Elk responses to humans in a densely roaded area. Intermt. J. Sci. 11, 10–24 (2005).
Google Scholar 
McCorquodale, S. M. Sex-specific movements and habitat use by elk in the Cascade Range of Washington. J. Wildl. Manag. 67, 729–741. https://doi.org/10.1890/15-1607.1 (2003).Article 

Google Scholar 
Saïd, S. & Servanty, S. The influence of landscape structure on female roe deer home-range size. Landsc. Ecol. 20, 1003–1012. https://doi.org/10.1007/s10980-005-7518-8 (2005).Article 

Google Scholar 
Seddon, P. J., Armstrong, D. P. & Maloney, R. F. Developing the science of reintroduction biology. Conserv. Biol. 21, 303–312. https://doi.org/10.1111/j.1523-1739.2006.00627.x (2007).Article 
PubMed 

Google Scholar 
Hale, S. L. & Koprowski, J. L. Ecosystem-level effects of keystone species reintroduction: a literature review. Restor. Ecol. 26, 439–445. https://doi.org/10.1111/rec.12684 (2018).Article 

Google Scholar 
Cheyne, S. M. Wildlife reintroduction: considerations of habitat quality at the release site. BMC Ecol. 6, 5. https://doi.org/10.1186/1472-6785-6-5 (2006).Article 
PubMed 
PubMed Central 

Google Scholar 
Hegel, T. M., Gates, C. C. & Eslinger, D. The geography of conflict between elk and agricultural values in the Cypress Hills, Canada. J. Eniron. Manag. 90, 222–235. https://doi.org/10.1016/j.jenvman.2007.09.005 (2009).Article 

Google Scholar 
Walter, W. D. et al. Management of damage by elk (Cervus elaphus) in North America: a review. Wildl. Res. 37, 630–646. https://doi.org/10.1071/WR10021 (2010).Article 

Google Scholar 
Jung, T. S. Extralimital movements of reintroduced bison (Bison bison): implications for potential range expansion and human–wildlife conflict. Eur. J. Wildl. Res. 63, 35. https://doi.org/10.1007/s10344-017-1094-5 (2017).Article 

Google Scholar 
Buchholtz, E. K., Stronza, A., Songhurst, A., McCulloch, G. & Fitzgerald, L. A. Using landscape connectivity to predict human-wildlife conflict. Biol. Conserv. 248, 108677. https://doi.org/10.1016/j.biocon.2020.108677 (2020).Article 

Google Scholar 
Hodgson, J. A., Moilanen, A., Wintle, B. A. & Thomas, C. D. Habitat area, quality and connectivity: striking the balance for efficient conservation. J. Appl. Ecol. 48, 148–152. https://doi.org/10.1111/j.1365-2664.2010.01919.x (2011).Article 

Google Scholar 
Murie, O. The Elk of North America (Stackpole Co., 1951).
Google Scholar 
VDWR. Virginia elk management plan 2019–2028 (ed Virginia Department of Wildlife Resources) (Virginia Department of Wildlife Resources, 2019).Lituma, C. M. et al. Terrestrial wildlife in the post-mined Appalachian landscape: status and opportunities in Appalachia’s Coal-Mined Landscapes (eds Carl E. Zipper & Jeff Skousen) 135–166 (Springer, 2021).Lupardus, J. L., Muller, L. I. & Kindall, J. L. Seasonal forage availability and diet for reintroduced elk in the Cumberland Mountains, Tennessee. Southeast. Nat. 10, 53–74. https://doi.org/10.1656/058.010.0105 (2011).Article 

Google Scholar 
Schneider, J. et al. Food habits of reintroduced elk in southeastern Kentucky. Southeast. Nat. 5, 535–546. https://doi.org/10.1656/1528-7092(2006)5[535:Fhorei]2.0.Co;2 (2006).Article 

Google Scholar 
Smith, T. N., Keller, B. J., Chitwood, M. C., Hansen, L. P. & Millspaugh, J. J. Diet composition and selection of recently reintroduced elk in Missouri. Am. Midl. Nat. 180, 143–159. https://doi.org/10.1674/0003-0031-180.1.143 (2018).Article 

Google Scholar 
Franklin, J. A., Zipper, C. E., Burger, J. A., Skousen, J. G. & Jacobs, D. F. Influence of herbaceous ground cover on forest restoration of eastern US coal surface mines. New. For. 43, 905–924. https://doi.org/10.1007/s11056-012-9342-8 (2012).Article 

Google Scholar 
Popp, J. N., Toman, T., Mallory, F. F. & Hamr, J. A century of elk restoration in eastern North America. Restor. Ecol. 22, 723–730. https://doi.org/10.1111/rec.12150 (2014).Article 

Google Scholar 
Cook, J. G., Irwin, L. L., Bryant, L. D., Riggs, R. A. & Thomas, J. W. Relations of forest cover and condition of elk: a test of the thermal cover hypothesis in the summer and winter. Wildl. Monogr. 141, 3–61 (1998).
Google Scholar 
Parker, K. L. & Robbins, C. T. Thermoregulation in mule deer and elk. Can. J. Zool. 62, 1409–1422. https://doi.org/10.1139/z84-202 (1984).Article 

Google Scholar 
Mao, J. S. et al. Habitat selection by elk before and after wolf reintroduction in Yellowstone National Park. J. Wildl. Manag. 69, 1691–1707. https://doi.org/10.2193/0022-541X (2005).Article 

Google Scholar 
Wolff, J. O. & Van Horn, T. Vigilance and foraging patterns of American elk during the rut in habitats with and without predators. Can. J. Zool. 81, 266–271. https://doi.org/10.1139/z03-011 (2003).Article 

Google Scholar 
Beck, J. L. & Peek, J. M. Diet composition, forage selection, and potential for forage competition among elk, deer, and livestock on aspen–sagebrush summer range. Rangel. Ecol. Manag. 58, 135–147. https://doi.org/10.2111/03-13.1 (2005).Article 

Google Scholar 
Ford, W. M., Johnson, A. S. & Hale, P. E. Nutritional quality of deer browse in southern Appalachian clearcuts and mature forests. For. Ecol. Manag. 67, 149–157. https://doi.org/10.1016/0378-1127(94)90013-2 (1994).Article 

Google Scholar 
Sikes, R. S., Gannon, W. L. & The American Care and Use Committee of the American Society of Mammalogists. Guidelines of the American Society of Mammalogists for the use of wild mammals in research. J. Mammal. 92, 235–253. https://doi.org/10.1644/10-mamm-f-355.1 (2011).Percie du Sert, N. et al. The ARRIVE guidelines 2.0: updated guidelines for reporting animal research. PLoS Biol. 18, e3000410. https://doi.org/10.1371/journal.pbio.3000410 (2020).Article 
PubMed 
PubMed Central 

Google Scholar 
Powell, J. W. Physiographic Regions of the United States. (American Book Company, 1895).Braun, E. L. Forests of the Cumberland Mountains. Ecol. Monogr. 12, 413–447. https://doi.org/10.2307/1943039 (1942).Article 

Google Scholar 
Clark, J. B. The Vascular Flora of Breaks Interstate Park, Pike County, Kentucky, and Dickenson County, Virginia Master of Science thesis, Eastern Kentucky University (2012).Pericak, A. A. et al. Mapping the yearly extent of surface coal mining in Central Appalachia using Landsat and Google Earth Engine. PLoS ONE 13, e0197758. https://doi.org/10.1371/journal.pone.0197758 (2018).Article 
PubMed 
PubMed Central 

Google Scholar 
Boettner, F. et al. An assessment of the natural assets in the Appalachian Region: forest resources (ed Appalachian Regional Commission Report) 97 (Washington, DC, 2014).NOAA. Summary of monthly normals Grundy, VA 1991 – 2020 data (National Oceanic and Atmospheric Administration (2022).U.S. Geological Survey (USGS) Gap Analysis Project (GAP). GAP/LANDFIRE national terrestrial ecosystems 2011: U.S. Geological Survey data release (2016).Clark, M. The Nature Conservancy Eastern Division & North Atlantic Landscape Conservation Cooperative. Terrestrial habitat, Northeast data (2017).ESRI. ArcGIS desktop version 10.8.1 (Environmental Systems Research Institute, 2020).Ford, W. M. et al. Influence of elevation and forest type on community assemblage and species distribution of shrews in the central and southern Appalachians in Advances in the Biology of the Shrews II Vol. 1(eds. J.F. Merritt, S. Churchfield, R. Hutterer and B.A. Sheftel) 303–315(Special Publication of the International Society of Shrew Biologists, 2006).Kniowski, A. B. & Ford, W. M. Predicting intensity of white-tailed deer herbivory in the Central Appalachian Mountains. J. For. Res. 29, 841–850. https://doi.org/10.1007/s11676-017-0476-6 (2018).Article 

Google Scholar 
Fleming, C. H. & Calabrese, J. M. ctmm: continuous-time movement modeling. R package version 0.6.0 (2021).R Core Team. R: a language and environment for statistical computing (R Foundation for Statistical Computing, 2020).Fleming, C. H. et al. Estimating where and how animals travel: an optimal framework for path reconstruction from autocorrelated tracking data. Ecology 97, 576–582. https://doi.org/10.1890/15-1607.1 (2016).Article 
PubMed 

Google Scholar 
Hijmans, R. J. raster: geographic data analysis and modeling. R package version 3.4-5 (2020).Becker, R. A., Chambers, J. M. & Wilks, A. R. The New S Language (Wadsworth and Brooks/Cole, 1988).Kuznetsova, A., Brockhoff, P. B. & Christensen, R. B. H. lmerTest package: tests in linear mixed effects models. J. Stat. Softw. 82, 1–26. https://doi.org/10.18637/jss.v082.i13 (2017).Article 

Google Scholar 
Burnham, K. P. & Anderson, D. R. Model Selection and Inference: A Practical Use of the Information-Theoretic Approach (Springer, 1998).Turner, M. G., Wu, Y., Romme, W. H. & Wallace, L. L. A landscape simulation model of winter foraging by large ungulates. Ecol. Modell. 69, 163–184. https://doi.org/10.1016/0304-3800(93)90026-O (1993).Article 

Google Scholar 
Taper, M. L. & Gogan, P. J. P. The northern Yellowstone elk: density dependence and climatic conditions. J. Wildl. Manag. 66, 106–122. https://doi.org/10.2307/3802877 (2002).Article 

Google Scholar 
Green, R. A. & Bear, G. D. Seasonal cycles and daily activity patterns of Rocky Mountain elk. J. Wildl. Manag. 54, 272–279. https://doi.org/10.2307/3809041 (1990).Article 

Google Scholar 
Craighead, J. J., Craighead, F. C. J., Ruff, R. L. & O’Gara, B. W. Home ranges and activity patterns of nonmigratory elk of the Madison Drainage herd as determined by biotelemetry. Wildl. Monogr. 33, 3–50 (1973).
Google Scholar 
Gittleman, J. L. & Thompson, S. D. Energy allocation in mammalian reproduction. Am. Zool. 28, 863–875. https://doi.org/10.1093/icb/28.3.863 (1988).Article 

Google Scholar 
Beier, P. & McCullough, D. R. Factors influencing white-tailed deer activity patterns and habitat use. Wildl. Monogr. 109, 3–51 (1990).
Google Scholar 
Ciuti, S., Davini, S., Luccarini, S. & Apollonio, M. Variation in home range size of female fallow deer inhabiting a sub-Mediterranean habitat. Rev. Ecol. 58, 381–395 (2003).
Google Scholar 
Vore, J. M. & Schmidt, E. M. Movements of female elk during calving season in northwest Montana. Wildl. Soc. Bull. 29, 720–725 (2001).
Google Scholar 
Wickstrom, M. L., Robbins, C. T., Hanley, T. A., Spalinger, D. E. & Parish, S. M. Food intake and foraging energetics of elk and mule deer. J. Wildl. Manag. 48, 1285–1301. https://doi.org/10.2307/3801789 (1984).Article 

Google Scholar 
Van Soest, P. J. Allometry and ecology of feeding behavior and digestive capacity in herbivores: a review. Zoo. Biol. 15, 455–479 (1996). https://doi.org/10.1002/(SICI)1098-2361(1996)15:53.0.CO;2-AEsmaeili, S. et al. Body size and digestive system shape resource selection by ungulates: a cross-taxa test of the forage maturation hypothesis. Ecol. Lett. 24, 2178–2191. https://doi.org/10.1111/ele.13848 (2021).Article 
PubMed 

Google Scholar 
Demment, M. W. & Van Soest, P. J. A nutritional explanation for body-size patterns of ruminant and nonruminant herbivores. Am. Nat. 125, 641–672 (1985).Article 

Google Scholar 
Anderson, D. P. et al. Factors influencing female home range sizes in elk (Cervus elaphus) in North American landscapes. Landsc. Ecol. 20, 257–271. https://doi.org/10.1007/s10980-005-0062-8 (2005).Article 

Google Scholar 
Maigret, T. A., Cox, J. J. & Yang, J. Persistent geophysical effects of mining threaten ridgetop biota of Appalachian forests. Front. Ecol. Environ. 17, 85–91. https://doi.org/10.1002/fee.1992 (2019).Article 

Google Scholar 
Beier, P. Sex differences in quality of white-tailed deer diets. J. Mammal. 68, 323–329. https://doi.org/10.2307/1381471 (1987).Article 

Google Scholar 
Parker, K. L., Barboza, P. S. & Gillingham, M. P. Nutrition integrates environmental responses of ungulates. Funct. Ecol. 23, 57–69. https://doi.org/10.1111/j.1365-2435.2008.01528.x (2009).Article 

Google Scholar 
Wichrowski, M. W., Maehr, D. S., Larkin, J. L., Cox, J. J. & Olsson, M. P. O. Activity and movements of reintroduced elk in southeastern Kentucky. Southeast. Nat. 4, 365–374. https://doi.org/10.1656/1528-7092(2005)004[0365:Aamore]2.0.Co;2 (2005).Article 

Google Scholar 
Relyea, R. A., Lawrence, R. K. & Demarais, S. Home range of desert mule deer: testing the body-size and habitat-productivity hypotheses. J. Wildl. Manag. 64, 146–153. https://doi.org/10.2307/3802984 (2000).Article 

Google Scholar  More

Whether land acts as a carbon sink or source depends largely on two opposite fluxes: carbon uptake through photosynthesis and carbon release through turnover. Turnover occurs through multiple processes, including but not limited to, leaf senescence, tree mortality, and respiration by plants, microbes, and animals. Each of these processes is sensitive to climate, and ecologists and climatologists have been working to figure out how temperature regulates biological activities and to what extent the carbon cycle responds to global warming. Previous theoretical and experimental studies have yielded conflicting relationships between temperature and carbon turnover, with large variations across ecosystems, climate and time-scale1,2,3,4. Writing in Nature Geoscience, Fan et al.5 find that hydrometeorological factors have an important influence on how the turnover time of land carbon responds to changes in temperature. More

Vo, Q. T., Oppelt, N., Leinenkugel, P. & Kuenzer, C. Remote sensing in mapping mangrove ecosystems: An object-based approach. Remote Sens. 5, 183–201. https://doi.org/10.3390/rs5010183 (2013).Article 

Google Scholar 
Kertész, Á. & Křeček, J. Landscape degradation in the world and in Hungary. Hung. Geogr. Bull. 68, 201–221. https://doi.org/10.15201/hungeobull.68.3.1 (2019).Article 

Google Scholar 
Vorster, A. G., Evangelista, P. H., Stovall, A. E. L. & Ex, S. Variability and uncertainty in forest biomass estimates from the tree to landscape scale: The role of allometric equations. Carbon Balance Manag. 15, 8. https://doi.org/10.1186/s13021-020-00143-6 (2020).Article 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Blackman, A. Evaluating forest conservation policies in developing countries using remote sensing data: An introduction and practical guide. For. Policy Econ. 34, 1–16. https://doi.org/10.1016/j.forpol.2013.04.006 (2013).Article 

Google Scholar 
Wilfong, B. N., Gorchov, D. L. & Henry, M. C. Detecting an invasive shrub in deciduous forest understories using remote sensing. Weed Sci. 57, 512–520. https://doi.org/10.1614/WS-09-012.1 (2009).Article 
CAS 

Google Scholar 
Dyderski, M. K. & Pawlik, Ł. Spatial distribution of tree species in mountain national parks depends on geomorphology and climate. For. Ecol. Manag. 474, 118366. https://doi.org/10.1016/j.foreco.2020.118366 (2020).Article 

Google Scholar 
Milosevic, D., Dunjić, J. & Stojanović, V. Investigating micrometeorological differences between saline steppe, forest-steppe and forest environments in northern Serbia during a clear and sunny autumn day. Geogr. Pannonica 24(3), 176–186. https://doi.org/10.5937/gp24-25885 (2020).Article 

Google Scholar 
Modzelewska, A., Fassnacht, F. E. & Stereńczak, K. Tree species identification within an extensive forest area with diverse management regimes using airborne hyperspectral data. Int. J. Appl. Earth Obs. Geoinf. 84, 101960. https://doi.org/10.1016/j.jag.2019.101960 (2020).Article 

Google Scholar 
Wulder, M. Optical remote-sensing techniques for the assessment of forest inventory and biophysical parameters. Prog. Phys. Geogr. Earth Environ. 22, 449–476. https://doi.org/10.1177/030913339802200402 (1998).Article 

Google Scholar 
Tang, L., Shao, G. & Dai, L. Roles of digital technology in China’s sustainable forestry development. Int. J. Sustain. Dev. World Ecol. 16, 94–101. https://doi.org/10.1080/13504500902794000 (2009).Article 

Google Scholar 
Richter, R., Reu, B., Wirth, C., Doktor, D. & Vohland, M. The use of airborne hyperspectral data for tree species classification in a species-rich Central European forest area. Int. J. Appl. Earth Obs. Geoinform. 52, 464–474. https://doi.org/10.1016/j.jag.2016.07.018 (2016).Article 

Google Scholar 
Thenkabail, P., Gumma, M., Teluguntla, P. & Ahmed, M. I. Hyperspectral remote sensing of vegetation and agricultural crops. Photogramm. Eng. Remote Sens. 80, 695–723 (2014).
Google Scholar 
Fassnacht, F. E. et al. Review of studies on tree species classification from remotely sensed data. Remote Sens. Environ. 186, 64–87. https://doi.org/10.1016/j.rse.2016.08.013 (2016).Article 

Google Scholar 
Vangi, E. et al. The new hyperspectral satellite PRISMA: Imagery for forest types discrimination. Sensors 21, 1182. https://doi.org/10.3390/s21041182 (2021).Article 
PubMed 
PubMed Central 

Google Scholar 
Burai, P., Beko, L., Lenart, C., Tomor, T. & Kovacs, Z. Individual tree species classification using airborne hyperspectral imagery and lidar data. In 2019 10th Workshop on Hyperspectral Imaging and Signal Processing: Evolution in Remote Sensing (WHISPERS) 1–4. https://doi.org/10.1109/WHISPERS.2019.8921016 (2019).Kumar, B., Dikshit, O., Gupta, A. & Singh, M. K. Feature extraction for hyperspectral image classification: A review. Int. J. Remote Sens. 41, 6248–6287. https://doi.org/10.1080/01431161.2020.1736732 (2020).Article 

Google Scholar 
Li, X., Li, Z., Qiu, H., Hou, G. & Fan, P. An overview of hyperspectral image feature extraction, classification methods and the methods based on small samples. Appl. Spectrosc. Rev. https://doi.org/10.1080/05704928.2021.1999252 (2021).Article 

Google Scholar 
Wang, J. & Chang, C.-I. Independent component analysis-based dimensionality reduction with applications in hyperspectral image analysis. IEEE Trans. Geosci. Remote Sens. 44, 1586–1600. https://doi.org/10.1109/TGRS.2005.863297 (2006).Article 

Google Scholar 
Hamada, Y., Stow, D. A., Coulter, L. L., Jafolla, J. C. & Hendricks, L. W. Detecting Tamarisk species (Tamarix spp.) in riparian habitats of Southern California using high spatial resolution hyperspectral imagery. Remote Sens. Environ. 109, 237–248. https://doi.org/10.1016/j.rse.2007.01.003 (2007).Article 

Google Scholar 
Ibarrola-Ulzurrun, E., Marcello, J. & Gonzalo-Martin, C. Assessment of component selection strategies in hyperspectral imagery. Entropy 19, 666. https://doi.org/10.3390/e19120666 (2017).Article 
MathSciNet 

Google Scholar 
Dabiri, Z. & Lang, S. Comparison of independent component analysis, principal component analysis, and minimum noise fraction transformation for tree species classification using APEX hyperspectral imagery. ISPRS Int. J. Geo-Inf. 7, 488. https://doi.org/10.3390/ijgi7120488 (2018).Article 

Google Scholar 
Priyadarshini, K. N., Sivashankari, V., Shekhar, S. & Balasubramani, K. Comparison and evaluation of dimensionality reduction techniques for hyperspectral data analysis. Proceedings 24, 6. https://doi.org/10.3390/IECG2019-06209 (2019).Article 

Google Scholar 
Arslan, O., Akyürek, Ö., Kaya, Ş & Şeker, D. Z. Dimension reduction methods applied to coastline extraction on hyperspectral imagery. Geocarto Int. 35, 376–390. https://doi.org/10.1080/10106049.2018.1520920 (2020).Article 

Google Scholar 
Kadavi, P. R., Lee, W.-J. & Lee, C.-W. Analysis of the pyroclastic flow deposits of mount sinabung and Merapi using landsat imagery and the artificial neural networks approach. Appl. Sci. 7, 935. https://doi.org/10.3390/app7090935 (2017).Article 

Google Scholar 
Schlosser, A. D. et al. Building extraction using orthophotos and dense point cloud derived from visual band aerial imagery based on machine learning and segmentation. Remote Sens. 12, 2397. https://doi.org/10.3390/rs12152397 (2020).Article 

Google Scholar 
Latifi, H., Fassnacht, F. & Koch, B. Forest structure modeling with combined airborne hyperspectral and LiDAR data. Remote Sens. Environ. 121, 10–25. https://doi.org/10.1016/j.rse.2012.01.015 (2012).Article 

Google Scholar 
Clark, M. L., Roberts, D. A. & Clark, D. B. Hyperspectral discrimination of tropical rain forest tree species at leaf to crown scales. Remote Sens. Environ. 96, 375–398. https://doi.org/10.1016/j.rse.2005.03.009 (2005).Article 

Google Scholar 
Melgani, F. & Bruzzone, L. Classification of hyperspectral remote sensing images with support vector machines. IEEE Trans. Geosci. Remote Sens. 42, 1778–1790. https://doi.org/10.1109/ICIECS.2009.5363456 (2004).Article 

Google Scholar 
Belgiu, M. & Drăguţ, L. Random forest in remote sensing: A review of applications and future directions. ISPRS J. Photogramm. Remote Sens. 114, 24–31. https://doi.org/10.1016/j.isprsjprs.2016.01.011 (2016).Article 

Google Scholar 
Manandhar, R., Odeh, I. O. A. & Ancev, T. Improving the accuracy of land use and land cover classification of landsat data using post-classification enhancement. Remote Sens. 1, 330–344. https://doi.org/10.3390/rs1030330 (2009).Article 

Google Scholar 
Thakkar, A. K., Desai, V. R., Patel, A. & Potdar, M. B. Post-classification corrections in improving the classification of Land Use/Land Cover of arid region using RS and GIS: The case of Arjuni watershed, Gujarat, India. Egypt. J. Remote Sens. Space Sci. 20, 79–89. https://doi.org/10.1016/j.ejrs.2016.11.006 (2017).Article 

Google Scholar 
El-Hattab, M. M. Applying post classification change detection technique to monitor an Egyptian coastal zone (Abu Qir Bay), Egypt. J. Remote Sens. Space Sci. 19, 23–36. https://doi.org/10.1016/j.ejrs.2016.02.002 (2016).Article 

Google Scholar 
Bhosale, N., Manza, R., Kale, K., Scholar, R. & Professor, A. Analysis of effect of gaussian, salt and pepper noise removal from noisy remote sensing images. Pceedings of teh Second International Conference on ERCICA 386–390. http://rameshmanza.in/Publication/Narayan_Bhosle/Analysis%20of%20Effect%20of%20Gaussian.pdf (2014).Schöll, K., Kiss, A., Dinka, M. & Berczik, Á. Flood-pulse effects on zooplankton assemblages in a river-floodplain system (Gemenc Floodplain of the Danube, Hungary). Int. Rev. Hydrobiol. 97, 41–54. https://doi.org/10.1002/iroh.201111427 (2012).Article 

Google Scholar 
Ágoston-Szabó, E., Schöll, K., Kiss, A. & Dinka, M. The effects of tree species richness and composition on leaf litter decomposition in a Danube oxbow lake (Gemenc, Hungary). Fundam. Appl. Limnol. https://doi.org/10.1127/fal/2017/0675 (2017).Article 

Google Scholar 
Guti, G. Water bodies in the Gemenc floodplain of the Danube, Hungary: (A theoretical basis for their typology). Opusc Zool. 33, 49–60 (2001).
Google Scholar 
Berczik, Á. & Dinka, M. Bibliography of hydrobiological research on the Gemenc and Béda: Karapancsa floodplains of the River Danube (1498–1436 rkm) including the publications of the Danube Research Institute of the Hungarian Academy of Sciences between 1968 and 2017. Opusc. Zool. 49, 191–197. https://doi.org/10.18348/opzool.2018.2.191 (2018).Article 

Google Scholar 
Ceulemans, R., McDonald, A. J. S. & Pereira, J. S. A comparison among eucalypt, poplar and willow characteristics with particular reference to a coppice, growth-modelling approach. Biomass Bioenergy 11, 215–231. https://doi.org/10.1016/0961-9534(96)00035-9 (1996).Article 

Google Scholar 
Haneca, K., Katarina, Č & Beeckman, H. Oaks, tree-rings and wooden cultural heritage: A review of the main characteristics and applications of oak dendrochronology in Europe. J. Archaeol. Sci. 36, 1–11. https://doi.org/10.1016/j.jas.2008.07.005 (2009).Article 

Google Scholar 
Jones, T. G., Coops, N. C. & Sharma, T. Assessing the utility of airborne hyperspectral and LiDAR data for species distribution mapping in the coastal Pacific Northwest, Canada. Remote Sens. Environ. 114, 2841–2852. https://doi.org/10.1016/j.rse.2010.07.002 (2010).Article 

Google Scholar 
Sothe, C. et al. Tree species classification in a highly diverse subtropical forest integrating UAV-based photogrammetric point cloud and hyperspectral data. Remote Sens. 11, 1338. https://doi.org/10.3390/rs11111338 (2019).Article 

Google Scholar 
Nambiar, E. K. S. & Sands, R. Competition for water and nutrients in forests. Can. J. For. Res. 23, 1955–1968. https://doi.org/10.1139/x93-247 (1993).Article 

Google Scholar 
Mayoral, C., Calama, R., Sánchez-González, M. & Pardos, M. Modelling the influence of light, water and temperature on photosynthesis in young trees of mixed Mediterranean forests. New For. 46, 485–506. https://doi.org/10.1007/s11056-015-9471-y (2015).Article 

Google Scholar 
Stojanović, D. B., Levanič, T., Matović, B. & Orlović, S. Growth decrease and mortality of oak floodplain forests as a response to change of water regime and climate. Eur. J. For. Res. 134, 555–567. https://doi.org/10.1007/s10342-015-0871-5 (2015).Article 

Google Scholar 
Dyderski, M. K. & Jagodziński, A. M. Impact of invasive tree species on natural regeneration species composition, diversity, and density. Forests 11, 456. https://doi.org/10.3390/f11040456 (2020).Article 

Google Scholar 
Jia, S., Ji, Z., Qian, Y. & Shen, L. Unsupervised band selection for hyperspectral imagery classification without manual band removal. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 5, 531–543. https://doi.org/10.1109/JSTARS.2012.2187434 (2012).Article 

Google Scholar 
Karpouzli, E. & Malthus, T. The empirical line method for the atmospheric correction of IKONOS imagery. Int. J. Remote Sens. 24, 1143–1150. https://doi.org/10.1080/0143116021000026779 (2003).Article 

Google Scholar 
Richards, J. A. Remote Sensing Digital Image Analysis (Springer, 2013). https://doi.org/10.1007/978-3-642-30062-2.Book 

Google Scholar 
Sharifi Hashjin, S. & Khazai, S. A new method to detect targets in hyperspectral images based on principal component analysis. Geocarto Int. 37, 2679–2697. https://doi.org/10.1080/10106049.2020.1831625 (2022).Article 

Google Scholar 
Kaiser, H. F. The varimax criterion for analytic rotation in factor analysis. Psychometrika 23, 187–200 (1958).Article 
MATH 

Google Scholar 
Shah, C. A., Arora, M. K. & Varshney, P. K. Unsupervised classification of hyperspectral data: An ICA mixture model based approach. Int. J. Remote Sens. 25, 481–487. https://doi.org/10.1080/01431160310001618040 (2004).Article 

Google Scholar 
Tharwat, A. Independent component analysis: An introduction. Appl. Comput. Inform. 17, 222–249. https://doi.org/10.1016/S1364-6613(00)01813-1 (2020).Article 

Google Scholar 
Villa, A., Chanussot, J., Jutten, C., Benediktsson, J. A. & Moussaoui, S. On the use of ICA for hyperspectral image analysis. In 2009 IEEE International Geoscience and Remote Sensing Symposium vol. 4 IV-97-IV–100. https://doi.org/10.1109/IGARSS.2009.5417363 (2009).Hyvärinen, A. & Oja, E. Independent component analysis: Algorithms and applications. Neural Netw. 13, 411–430. https://doi.org/10.1016/s0893-6080(00)00026-5 (2000).Article 
PubMed 

Google Scholar 
Otukei, J. R. & Blaschke, T. Land cover change assessment using decision trees, support vector machines and maximum likelihood classification algorithms. Int. J. Appl. Earth Obs. Geoinf. 12, S27–S31. https://doi.org/10.1016/j.jag.2009.11.002 (2010).Article 

Google Scholar 
Murty, M. N. & Raghava, R. Kernel-based SVM. In Support Vector Machines and Perceptrons: Learning, Optimization, Classification, and Application to Social Networks (eds Murty, M. N. & Raghava, R.) 57–67 (Springer, 2016). https://doi.org/10.1007/978-3-319-41063-0_5.Chapter 
MATH 

Google Scholar 
Seidl, D., Ružiak, I., Koštialová Jančíková, Z. & Koštial, P. Sensitivity analysis: A tool for tailoring environmentally friendly materials. Expert Syst. Appl. 208, 118039. https://doi.org/10.1016/j.eswa.2022.118039 (2022).Article 

Google Scholar 
Zhao, D., Pang, Y., Liu, L. & Li, Z. Individual tree classification using airborne LiDAR and hyperspectral data in a natural mixed forest of Northeast China. Forests 11, 303. https://doi.org/10.3390/f11030303 (2020).Article 

Google Scholar 
Aksoy, S. & Akcay, H. G. Multi-resolution segmentation and shape analysis for remote sensing image classification. In Proceedings of 2nd International Conference on Recent Advances in Space Technologies, 2005. RAST 2005. 599–604 (2005). https://doi.org/10.1109/RAST.2005.1512638.Dalponte, M., Ørka, H. O., Ene, L. T., Gobakken, T. & Næsset, E. Tree crown delineation and tree species classification in boreal forests using hyperspectral and ALS data. Remote Sens. Environ. 140, 306–317. https://doi.org/10.1016/j.rse.2013.09.006 (2014).Article 

Google Scholar 
Amini, S., Homayouni, S., Safari, A. & Darvishsefat, A. A. Object-based classification of hyperspectral data using Random Forest algorithm. Geo-Spat. Inf. Sci. 21, 127–138. https://doi.org/10.1080/10095020.2017.1399674 (2018).Article 

Google Scholar 
Congalton, R. G. A review of assessing the accuracy of classifications of remotely sensed data. Remote Sens. Environ. 37, 35–46. https://doi.org/10.1016/0034-4257(91)90048-B (1991).Article 

Google Scholar 
Foody, G. M. Status of land cover classification accuracy assessment. Remote Sens. Environ. 80, 185–201. https://doi.org/10.1016/S0034-4257(01)00295-4 (2002).Article 

Google Scholar 
Tharwat, A. Classification assessment methods. Appl. Comput. Inform. 17, 168–192. https://doi.org/10.1016/j.aci.2018.08.003 (2020).Article 

Google Scholar 
Field, F. Discovering Statistics Using IBM SPSS Statistics. SAGE Publications Ltd https://uk.sagepub.com/en-gb/eur/discovering-statistics-using-ibm-spss-statistics/book257672 (2022).R Core Team. R: A language and environment for statistical computing. https://www.gbif.org/tool/81287/r-a-language-and-environment-for-statistical-computing (2022).Galucci, M. Generalized Mixed Models module. R package version 2.0.5. https://gamlj.github.io/gzlmmixed.html More

Study areaFigure 2 shows the location of the study area on a map of China generated by ArcGIS software. This study’s field experiments were carried out in the Shuanghe Town agricultural comprehensive water-saving demonstration area (40°42′ N; 107°24′ E), which is located in the middle reaches of the Hetao Irrigation Area of Inner Mongolia. The duration of the experimental process ranged from April in 2018 to October in 2020. The experimental area was characterized by a mid-temperate semi-arid continental climate. The average annual precipitation was determined to be 138 mm and the average evaporation was approximately 2332 mm. The majority of the rainfall was concentrated during summer and autumn seasons, and the accumulation of salt in the surface soil was considered to be serious in the spring and winter months. The average rainfall during maize growth period was 75.3 mm. The 0 to 40 cm soil layers in the experimental area were categorized as silty loam soil, with an average bulk density ranging from 1.42 to 1.53 g cm−3. A maize straw layer with a thickness of 5 cm was buried at a depth of 40 cm, and then the land was leveled. Also, in addition to autumn watering and spring irrigation procedures, water from the Yellow River was used three times for irrigation during the entire growth period of the maize crops. The adopted irrigation method belonged to border irrigation. Urea (46% N) were used as the fertilizer types.Figure 2The location of the study area.Full size imageField trials design and data collectionWe carried out experiment 1 from 2018 to 2019, and the data obtained were used for model training and to determine the hyper-parameters. The experimental design is shown in Table 1. The PNN model trained from the data obtained in experiment 1 predicted the optimal range of irrigation amount and nitrogen application rate (N rate) for each growth period of maize. In these ranges, the soil organic matter and total nitrogen could be kept above 20 g/kg and 1.6 g/kg, respectively, the soil salt content was less than 2 g/kg, and the pH value was between 6.5 and 7.5. In order to verify the accuracy and feasibility of the range of irrigation and nitrogen application simulated by PNN, the field experiment 2 was set in 2020 based on the range simulated by PNN and to evaluate the fitting degree between measured and simulated values of soil indicators under the same amount of irrigation and nitrogen application. The experimental design is shown in Table 2.Table 1 Experimental 1 design scheme.Full size tableTable 2 Experimental 2 design scheme.Full size tableThe experimental design were repeated for three times. The plot area of each treatment measuring 8 × 9 = 72 m2. The surrounding area was separated using 1.2 m buried polyethylene plastic film, and 30 cm was left at the top to prevent fertilizer and water from flowing into each other. The field management process was consistent with that used by the local farmers. The film width of maize was 1.1 m, with each film covering two rows. The plant spacing was approximately 45 cm, and the row spacing was 35 cm. In addition, the planting density of the maize was 60,000 plants/hm2.During the entire growth period of the maize crops, soil samples were collected from the 0 to 20 cm, 20 to 40 cm, 40 to 60 cm, 60 to 80 cm, and 80 to 100 cm soil layers using a soil drill and a three-point method was adopted. The soil samples were stored at 4 °C for the determination of total nitrogen, organic matter, total salt content, and pH values. The total nitrogen, organic matter, total salt content, and pH were determined using a KDN-AA double tube azotometer, MWD-2 microwave universal digestion device, TU1810PC ultraviolet–visible spectrophotometer, and a TU18950 double beam ultraviolet–visible spectrophotometer, respectively.Soil parameters measured include organic matter (SOM), total nitrogen (TN), Salt and pH. The data set includes pre-irrigation and post-irrigation reports from 2018 to 2020. Statistical parameters regarding the soil data are shown in Table 3.Table 3 Various meteorological variables and their descriptive statistics.Full size tableThe dataset obtained in Experiment 1 in 2018 to 2019 was 2490 rows in size, the 80/20 principle was used to data into training, and testing sets were required for ML modeling; 80% of data were employed for model training, while the remaining 20% were used for testing. Specifically, the data corresponding to the treatments with the nitrogen application rate (N rate) of 75 kg/hm2 (N3) in all the treatments (W1N3, W2N3, W3N3) were used as the test set, and the data of the other treatments were used as the training set. The training set was used to initiate ML parameter training. Subsequently, The test set was employed to assess the model. The dataset size in 2020 was 1080 rows, which was used to verify ML modeling.Figure 3 shows the changes of soil indexes over time for each treatment in the field test (take the 0–40 cm soil in the main distribution area of maize roots as an example). There are differences under the influence of different irrigation amounts. When irrigation is 90 mm, soil SOM is 13.25% and 7.00% higher than 60 mm and 120 mm, and soil TN is 4.59% and 6.50% higher than 60 mm and 120 mm, respectively. The soil Salt was 23.30% lower than 60 mm, and the pH was 4.16% and 4.36% lower than that of 60 mm and 120 mm, respectively. It can be seen that irrigation of 90 mm is more favorable for increasing soil SOM and TN contents and reducing soil salinity and alkalinity. Soil SOM and TN contents were the highest at n 75 kg/hm2, which were 4.38% and 8.34% higher than those at N = 93.3 kg/hm2, respectively. Soil Salt was the lowest at N = 60 kg/hm2, which was 3.02% lower than those at N = 75 kg/hm2, with a small gap with other levels. In conclusion, nitrogen application of 75 kg/hm2 was beneficial to increase soil organic matter and nitrogen content, and nitrogen application of 60 kg/hm2 was beneficial to controlling soil salt content.Figure 3Changes in soil organic matter, total nitrogen, salinity, and pH under different treatments over time (a case study of 2019).Full size imageMachine learning (ML) models used for irrigation and nitrogen application strategiesFive ML frames were used to estimate the irrigation and N rate. These models are preference Neural Network (PNN), Support Vector Regression (SVR), Linear Regression (LR), Logistic Regression (LOR), and traditional BP Neural Networks (BPNN). Among them, the prediction effects of linear, Poly, and rbf kernel functions are respectively tried in SVR framework. The torch framework was used to train and test machine learning models in Python.Development of preference neural networkModel frameworkThe preference neural network (PNN) which was proposed for the first time in this study was a typical deep learning model. PNN can be regarded as an approximate natural function in order to describe the complete dependence of the soil fertility indexes, including the effects of soil total nitrogen, organic matter, total salt content, and pH values on irrigation and nitrogen applications. More specifically, PNN has the ability to optimize the function by constructing the mapping y = f (x, θ) and learning parameter θ.First, the input end of PNN model was defined as matrix X ∈ ℝn×d (in which n is the sample size, n = 2490; and d is the dimension of each input vector, d = 6), where {xi} i=1, …, n ∈ X represents the vectorized set of total nitrogen, organic matter, salt content, and pH used for measuring the soil fertility, as well as the nitrogen application and irrigation durations (expressed by days after sowing). At the same time, the output end of the model was defined as the matrix Y ∈ ℝn×2, which represented the levels of the irrigation and nitrogen fertilizer applications. The goal of the proposed PNN model was to learn the fixed mapping Y′ = f (X; θ) ⇒Y through the given input matrix X, where θ is the well optimized learnable parameters which can be obtained via PNN training. Meanwhile, the predicted value Y′ will infinitely approach the measured value Y. The structure and the algorithm of this study’s PNN model is shown in Fig. 4 and Table. 4.Figure 4Schematic diagram for the PNN structural connections. In the figure, it can be seen that when each input vector passed through each layer of the PNN, it is first multiplied by the Hadamard product of the weight matrix and preference value matrix for the purpose of obtaining a weight matrix with preference properties. After the matrix was activated by the Relu Function, Batch Normalization Module Methods and the Dropout Module were used for random suspension and normalization processing, and the input of the next layer was obtained.Full size imageTable 4 Algorithm of Preference neural network.Full size tableLayer-by-layer affine transformationA good definition of the affine transformation of the information flow between layers is considered to be the key to neural network model training. Generally speaking, the learnable parameter θ of each layer of a model includes the weight parameter w and the preference parameter b. The hidden representation hl of the l-th layer in PNN is defined as follows:$${h}_{l}({h}_{l-1};{W}_{l},{b}_{l})={h}_{l-1}^{mathrm{T}}{W}_{l}+{b}_{l}$$
(1)

where Wl and bl represent the learnable weight and bias variables of the l layer, respectively, and hl-1 is the hidden representation of the upper layer. Therefore, when l = 1, then h0 = X.In the present study, using the hierarchical update rules, a given input data stream was allowed to pass through each hidden layer with intermediate operations, and then finally reached the output end.Preference structureThe correlation between different production behavior factors (e.g., irrigation levels) and different natural factors (e.g., soil organic matter) differs in agricultural production. However, the traditional fully connected neural network has the characteristic that nodes of one layer are fully connected with all nodes of subsequent layers, resulting in the neurons between production behavior factors and natural factors with very weak correlation still all being connected. Conversely, connections between neurons corresponding to factors with solid correlations are not strengthened.Therefore, in this study the preference value module was specially developed. By first calculating the correlation and significance between different production behavior factors (irrigation amount, N rate) and different soil fertility factors (organic matter, total nitrogen, total salt and pH), the preference value between the above two types of variables was calculated, and the preference matrix was constructed. Then the Hadamard product of the weight matrix and preference matrix was used to realize the artificial intervention and guidance to the neural network’s learning process.In order to reduce the adverse impact of non-normality of data on correlation analysis as much as possible, this study rank-based inverse normal (RIN) transformations (i.e., conversion to rank score) methods were used to normally process the data28. The RIN transformation function used here is as follows:$$f(x)={Phi }^{-1}left(frac{{x}_{r}-frac{1}{2}}{n}right)$$
(2)

where Φ–1 is the inverse normal cumulative distribution function, and n is the sample size.The normal cumulative distribution function is represented as follows: for discrete variables, the sum of probabilities of all values less than or equal to a, and its formula is as shown below:$${F}_{X}(a)=P(Xle a)$$
(3)
The RIN normalized conversion values meet the requirements of normal distribution, Pearson correlation analysis and t-test can be directly performed, and the formula used was as follows:$$r(X,Y)=frac{mathrm{Cov}(X,Y)}{sqrt{left(mathrm{Var}left[Xright]mathrm{Var}left[mathrm{Y}right]right)}}$$
(4)

where r (X, Y) is the Pearson Correlation Coefficient, Var [X] is the variance of X, and Var [Y] is the variance of Y, Cov (X, Y) is the covariance of X and Y, which represents the overall error of the two variables. The t-test is performed on the normalized data after rank-based inverse normal (RIN) transformation method, and the formula is as follows:$$t=sqrt{frac{n-2}{1-{r}^{2}}}$$
(5)

where n is the number of samples, and r represents the Pearson Correlation Coefficient. Preference value is the concentrated embodiment of correlation and significance between variables, and the calculation formula is as follows:$${PV}_{ij}=frac{r({X}_{i},{Y}_{j})}{{P}_{ij}+e}$$
(6)

where PVij represents the preference values between the variables Xi and Yj, Xi represents the ith production behavior factor (e.g., irrigation amount), and Yj represents the jth soil fertility factor (e.g., soil organic matter content), ({P}_{ij}) is obtained by looking up the table based on the t, and e is a constant, taking 0.001 in order to prevent the denominator of the formula from being 0.In order to make the preference values of the various indicators in the same order of magnitude more stable, the preference values were normalized:$${PV}_{normal}=pm frac{left|{PV}_{i}-{PV}_{avg}right|}{sqrt{frac{sum_{i=1}^{N}{({PV}_{i}-{PV}_{avg})}^{2}}{N-1}}}$$
(7)

where N represents the number of variables related to the experimental treatments, PVi -PVavg takes the absolute value, while the positive or negative values of the PVnormal were determined by the positive or negative values of the correlation r.The PNN integrated the preference matrixes into the neural network structures by identifying the Hadamard products of the learnable weights between the preference matrixes and the input and output data. By referring to Eq. (1) in the hierarchical affine transformation, the preference constraint of PNN could be expressed as follows:$${h}_{l}({h}_{l-1};{W}_{l},{b}_{l})={h}_{l-1}^{T}{W}_{l}odot P+{b}_{l}$$
(8)

where P is the preference matrix calculated by Eq. (8), and ⊙ represents the Hadamard product of the corresponding elements of the matrix. The structure of preference neural network and preference value are shown in Figs. 5 and 6.Figure 5Schematic diagram of the preference connection structures of the preference neural networks. The depth of the network detailed in the figure only illustrates the preference connection structure (for a better demonstration), and does not indicate the depth of the PNN used in the experiment.Full size imageFigure 6PVnormal between production behavior factors and natural factors. Since soil depth, days, irrigation amount and N rate were all artificially set variables, and there was no objective correlation in the data set. Therefore, the preference values among these variables were default e = 0.001.Full size imageHyper-parameters of PNNWe conducted experiments on the datasets with varying the hyper-parameters (such as the number of PNN layers and hidden layers, the number of nodes in each layer, learning rate, dropout rate and batch size) to understand that how the Hyper-parameters impact on the performance of PNN.We select the activation function and learning rate by referring to the neural network structure commonly used in similar fields (1 hidden layer and 64 hidden nodes)29,30. It is found that ReLU has better performance than other activation functions (sigmoid, tanh). The performance is best when the learning rate is around 0.005. It is generally believed that neural networks with more hidden layers are able, with the same number of resources, to address more complex problems31, but excessively increasing network depth will easily lead to overfitting32. Since there is no direct method to select the optimal number of hidden layers and nodes33, this study first calculated the structure of one hidden layer and 64 nodes in each layer, and found that the combined effect was poor (R2 of irrigation and nitrogen application were 0.3971 and 0.4124, respectively). Therefore, the trial-and-error method is adopted. The number of hidden layers starts from 1 and is incremented by 1 to test the maximum number of 10 hidden layers. The number of nodes in each layer were tested with a maximum number of 100 hidden neurons, starting with 5 and increasing by 5.We found that when the number of hidden layers of PNN exceeds 6, and the number of nodes in each layer exceeds 65, the performance will drop significantly. The reason behind this phenomenon could be the current dataset size is insufficient for larger scale of the PNN model. In the consideration of that the size of new dataset we can obtain very year is similar to the current dataset size, we believe that current hyper-paramter settings of PNN is in a reasonable condition.After that, the number of layers was fixed as 6, and the number of nodes in each layer were tested 10 times with 60 as the starting point and 1 as the increment, we found that when the number of nodes was 64, the improvement of the fit degree was no longer noticeable. On this basis, we changed different activation functions and learning rate again, and found that PNN still has the best performance when the activation function is ReLU and the learning rate is 0.005. Then, different batch sizes and dropout rates were tried. The two parameters had weaker effects on the performance than the other parameters, and the performance was optimal at 256 and 0.1, respectively.The hyper-parameters include:

1.

number of PNN layers;

2.

number of hidden layers;

3.

types of activation function;

4.

percentage of dropout;

5.

learning rate;

6.

loss function;

7.

optimizer;

8.

batch size;

9.

number of epochs;

10.

number of workers.

The ideal PNN structure for the study comprises these layers:

1.

number of PNN layers is 8;

2.

number of hidden layers is 6;

3.

Fully connected layers with 64 nodes and ReLU activation function

4.

dropout with 0.1.

5.

the learning rate is 0.005;

6.

loss function is Huber Loss Methods (HLM);

7.

optimizer: ADAM;

8.

epochs is 500;

9.

the batch size is 256;

10.

number of workers is 6.

Hyper-parameters of other modelsLR algorithms and LOR do not have hyper-parameters that need to be adjusted. A part of the hyper-parameters of the SVR model was determined by referring to Guan Xiaoyan’s research34, and a part of the hyper-parameters of the BPNN model was determined by referring to Gu Jian’s research27. RMLP takes the same hyperparameters as PNN. The hyperparameters of SVR and BPNN models are shown in Table 5.Table 5 Hyper-parameters of other model.Full size tableModel performance evaluationThe proposed PNN model was trained and validated using the field measured data from 2020 and the performance achievements of PNN were evaluated by the root mean square errors, mean square errors, and mean absolute errors as follows:$$RMSE=sqrt{frac{{sum }_{i=1}^{n}{({y}_{ipre}-{y}_{imea})}^{2}}{n}}$$
(9)
$${R}^{2}=1-frac{{sum }_{i=1}^{n}{({y}_{ipre}-{y}_{imea})}^{2}}{{sum }_{i=1}^{n}{({y}_{ipre}-{y}_{iavg})}^{2}}$$
(10)
$$MAE=frac{{sum }_{i=1}^{n}left|{y}_{ipre}-{y}_{iavg}right|}{n}$$
(11)
Model multidimensional fertility targetsThe soil fertility grade classification of soil organic matter, soil total nitrogen content and salt content in this study was based on the soil fertility grade classification results by the Agriculture and Animal Husbandry Bureau of Bayannur City, along with the local standard Technical Specifications for the Assessment and Rating Criteria of Cultivated Land Quality (DB 15/T 1086, 2016), as the shown in Tables 6 and 7.Table 6 Soil organic matter and Soil total nitrogen degrees.Full size tableTable 7 Grading of the salinization degrees.Full size tableIn the evaluation system of soil fertility referencing the Technical Specifications for Assessment and Rating Criteria of Cultivated Land Quality (DB 15/T 1086, 2016), the pH was divided into four grades according to the membership degrees of the land productivity evaluations, as detailed in Table 8.Table 8 pH grading degrees of the cultivated land.Full size tableBased on the classification standard of soil fertility obtained by the Bureau of Agriculture and Animal Husbandry of Bayannur City, when the farmland soil is at the high fertility level, the soil organic matter and total nitrogen content should be more than 20 g/kg and 1.6 g/kg, respectively. Soil salt content was less than 2 g/kg. Meanwhile, the pH value is kept between 6.5 and 7.5. More