Ecology

1.Helm, B. et al. Annual rhythms that underlie phenology: Biological time-keeping meets environmental change. Proc. R. Soc. B Biol. Sci. 280, 20130016 (2013).
Google Scholar 
2.Bradshaw, W. E. & Holzapfel, C. M. Evolution of animal photoperiodism. Annu. Rev. Ecol. Evol. Syst. 38, 1–25 (2007).
Google Scholar 
3.Dawson, A. Control of the annual cycle in birds: Endocrine constraints and plasticity in response to ecological variability. Philos. Trans. R. Soc. B Biol. Sci. 363, 1621–1633 (2008).
Google Scholar 
4.Dawson, A., King, V. M., Bentley, G. E. & Ball, G. F. Photoperiodic control of seasonality in birds. J. Biol. Rhythms 16, 365–380 (2001).CAS 
PubMed 

Google Scholar 
5.Wingfield, J. C. & Kenagy, G. J. Natural regulation of reproductive cycles. Vertebr. Endocrinol. Fundam. Biomed. Implic. 4, 181–241 (1991).
Google Scholar 
6.Hahn, T. P., Pereyra, M. E., Sharbaugh, S. M. & Bentley, G. E. Physiological responses to photoperiod in three cardueline finch species. Gen. Comp. Endocrinol. 137, 99–108 (2004).CAS 
PubMed 

Google Scholar 
7.Perfito, N., Meddle, S. L., Tramontin, A. D., Sharp, P. J. & Wingfield, J. C. Seasonal gonadal recrudescence in song sparrows: Response to temperature cues. Gen. Comp. Endocrinol. 143, 121–128 (2005).CAS 
PubMed 

Google Scholar 
8.Shutt, J. D. et al. The environmental predictors of spatio-temporal variation in the breeding phenology of a passerine bird. Proc. R. Soc. B Biol. Sci. 286, 20190952 (2019).
Google Scholar 
9.Drake, A. & Martin, K. Rainfall and nest site competition delay mountain bluebird and tree swallow breeding but do not impact productivity. Auk 137, 1–18 (2020).
Google Scholar 
10.Bison, M. et al. Best environmental predictors of breeding phenology differ with elevation in a common woodland bird species. Ecol. Evolut. https://doi.org/10.1002/ece3.6684 (2020).Article 

Google Scholar 
11.McNamara, J. M., Barta, Z., Klaassen, M. & Bauer, S. Cues and the optimal timing of activities under environmental changes. Ecol. Lett. 14, 1183–1190 (2011).PubMed 
PubMed Central 

Google Scholar 
12.Thackeray, S. J. et al. Phenological sensitivity to climate across taxa and trophic levels. Nature 535, 241–245 (2016).ADS 
CAS 
PubMed 

Google Scholar 
13.Moussus, J.-P., Clavel, J., Jiguet, F. & Julliard, R. Which are the phenologically flexible species? A case study with common passerine birds. Oikos 120, 991–998 (2011).
Google Scholar 
14.Chamberlain, D. et al. The altitudinal frontier in avian climate impact research. Ibis 154, 205–209 (2012).
Google Scholar 
15.Wipf, S., Stoeckli, V. & Bebi, P. Winter climate change in alpine tundra: Plant responses to changes in snow depth and snowmelt timing. Clim. Change 94, 105–121 (2009).ADS 

Google Scholar 
16.Jonas, T., Rixen, C., Sturm, M. & Stoeckli, V. How alpine plant growth is linked to snow cover and climate variability. J. Geophys. Res. Biogeosci. 113, G03013 (2008).ADS 

Google Scholar 
17.Kudo, G. & Hirao, A. S. Habitat-specific responses in the flowering phenology and seed set of alpine plants to climate variation: Implications for global-change impacts. Popul. Ecol. 48, 49–58 (2006).
Google Scholar 
18.Trant, A., Higgs, E. & Starzomski, B. M. A century of high elevation ecosystem change in the Canadian Rocky Mountains. Sci. Rep. 10, 9698 (2020).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
19.Ceppi, P., Scherrer, S. C., Fischer, A. M. & Appenzeller, C. Revisiting Swiss temperature trends 1959–2008. Int. J. Climatol. 32, 203–213 (2012).
Google Scholar 
20.Pepin, N. et al. Elevation-dependent warming in mountain regions of the world. Nat. Clim. Chang. 5, 424–430 (2015).ADS 

Google Scholar 
21.Rosenzweig, C. et al. Attributing physical and biological impacts to anthropogenic climate change. Nature 453, 353–357 (2008).ADS 
CAS 
PubMed 

Google Scholar 
22.Brunetti, M. et al. Precipitation variability and changes in the greater Alpine region over the 1800–2003 period. J. Geophys. Res. Atmos. 111, D11107 (2006).ADS 

Google Scholar 
23.Napoli, A., Crespi, A., Ragone, F., Maugeri, M. & Pasquero, C. Variability of orographic enhancement of precipitation in the Alpine region. Sci. Rep. 9, 13352 (2019).ADS 
PubMed 
PubMed Central 

Google Scholar 
24.Diffenbaugh, N. S., Scherer, M. & Ashfaq, M. Response of snow-dependent hydrologic extremes to continued global warming. Nat. Clim. Chang. 3, 379–384 (2013).ADS 
PubMed 

Google Scholar 
25.Beniston, M., Keller, F. & Goyette, S. Snow pack in the Swiss Alps under changing climatic conditions: An empirical approach for climate impacts studies. Theoret. Appl. Climatol. 74, 19–31 (2003).ADS 

Google Scholar 
26.Parmesan, C. Ecological and evolutionary responses to recent climate change. Annu. Rev. Ecol. Evol. Syst. 37, 637–669 (2006).
Google Scholar 
27.Saalfeld, S. T. et al. Phenological mismatch in Arctic-breeding shorebirds: Impact of snowmelt and unpredictable weather conditions on food availability and chick growth. Ecol. Evol. 9, 6693–6707 (2019).PubMed 
PubMed Central 

Google Scholar 
28.Tulp, I. & Schekkerman, H. Has prey availability for arctic birds advanced with climate change? Hindcasting the abundance of tundra arthropods using weather and seasonal variation. Arctic 61, 48–60 (2008).
Google Scholar 
29.Leung, M.C.-Y. et al. Phenology of hatching and food in low Arctic passerines and shorebirds: Is there a mismatch?. Arctic Sci. 4, 538–556 (2018).
Google Scholar 
30.Grabowski, M. M., Doyle, F. I., Reid, D. G., Mossop, D. & Talarico, D. Do Arctic-nesting birds respond to earlier snowmelt? A multi-species study in north Yukon, Canada. Polar Biol. 36, 1097–1105 (2013).
Google Scholar 
31.Liebezeit, J. R., Gurney, K. E. B., Budde, M., Zack, S. & Ward, D. Phenological advancement in arctic bird species: Relative importance of snow melt and ecological factors. Polar Biol. 37, 1309–1320 (2014).
Google Scholar 
32.Hendricks, P. Spring snow conditions, laying date, and clutch size in an alpine population of American Pipits. J. Field Ornithol. 74, 423–429 (2003).
Google Scholar 
33.Pereyra, M. E. Effects of snow-related environmental variation on breeding schedules and productivity of a high-altitude population of dusky flycatchers (Empidonax oberholseri). Auk 128, 746–758 (2011).
Google Scholar 
34.Resano-Mayor, J. et al. Snow cover phenology is the main driver of foraging habitat selection for a high-alpine passerine during breeding: implications for species persistence in the face of climate change. Biodivers. Conserv. 28, 2669–2685 (2019).
Google Scholar 
35.Venables, W. N. & Ripley, B. D. Modern Applied Statistics with S (Springer, 2002).MATH 

Google Scholar 
36.Bears, H., Martin, K. & White, G. C. Breeding in high-elevation habitat results in shift to slower life-history strategy within a single species. J. Anim. Ecol. 78, 365–375 (2009).CAS 
PubMed 

Google Scholar 
37.García-González, R., Aldezabal, A., Laskurain, N. A., Margalida, A. & Novoa, C. Influence of snowmelt timing on the diet quality of pyrenean rock ptarmigan (Lagopus muta pyrenaica): Implications for reproductive success. PLoS ONE 11, e0148632 (2016).PubMed 
PubMed Central 

Google Scholar 
38.Antor, R. J. Arthropod fallout on high alpine snow patches of the Central Pyrenees, northeastern Spain. Arct. Alp. Res. 26, 72–76 (1994).
Google Scholar 
39.Brambilla, M. et al. Foraging habitat selection by alpine white-winged snowfinches Montifringilla nivalis during the nestling rearing period. J. Ornithol. 158, 277–286 (2017).
Google Scholar 
40.Heiniger, P. H. Anpassungsstrategien des Schneefinken (Montifringilla nivalis) an die extremen Umweltbedingungen des Hochgebirges. Der Ornithol. Beobachter 88, 193–207 (1991).
Google Scholar 
41.MacDonald, E. C., Camfield, A. F., Jankowski, J. E. & Martin, K. An alpine-breeding songbird can adjust dawn incubation rhythms to annual thermal regimes. Auk 131, 495–506 (2014).
Google Scholar 
42.Mortensen, L. O., Schmidt, N. M., Høye, T. T., Damgaard, C. & Forchhammer, M. C. Analysis of trophic interactions reveals highly plastic response to climate change in a tri-trophic high-arctic ecosystem. Polar Biol. 39, 1467–1478 (2016).
Google Scholar 
43.Grangé, J. L. Biologie de la reproduction de la Niverolle alpine Montifringilla nivalis dans les Pyrénnées occidentales françaises. Nos Oiseaux 55, 67–82 (2008).
Google Scholar 
44.Strinella, E., Vianale, P., Pirrello, S. & Artese, C. Biologia riproduttiva del Fringuello Alpino Montifringilla nivalis a Campo Imperatore nel Parco Nazionale del Gran Sasso e Monti della Laga (AQ). Alula 18, 95–100 (2011).
Google Scholar 
45.Visser, M. E. et al. Variable responses to large-scale climate change in European Parus populations. Proc. R. Soc. Lond. Ser. B Biol. Sci. 270, 367–372 (2003).
Google Scholar 
46.Knaus, P. et al. Schweizer Brutvogelatlas 2013–2016. Verbreitung und Bestandsentwicklung der Vögel in der Schweiz und im Fürstentum Liechtenstein. (Schweizerische Vogelwarte, 2018).47.Basist, A., Bell, G. D. & Meentemeyer, V. Statistical relationships between topography and precipitation patterns. J. Clim. 7, 1305–1315 (1994).ADS 

Google Scholar 
48.Hock, R. et al. High mountain areas. in IPCC Special Report on the Ocean and Cryosphere in a Changing Climate (eds. Pörtner, H. O. et al.). 131–202. (IPCC-Intergovernmental Panel on Climate Change, 2019).49.Schmidt, N. M., Reneerkens, J., Christensen, J. H., Olesen, M. & Roslin, T. An ecosystem-wide reproductive failure with more snow in the Arctic. PLOS Biol. 17, e3000392 (2019).CAS 
PubMed 
PubMed Central 

Google Scholar 
50.Martin, K. & Wiebe, K. L. Coping mechanisms of alpine and arctic breeding birds: extreme weather and limitations to reproductive resilience. Integr. Comp. Biol. 44, 177–185 (2004).PubMed 

Google Scholar 
51.Williams, C. T. et al. Seasonal reproductive tactics: Annual timing and the capital-to-income breeder continuum. Philos. Trans. R. Soc. B Biol. Sci. 372, 20160250 (2017).
Google Scholar 
52.Barlow, K. E. et al. Citizen science reveals trends in bat populations: The National Bat Monitoring Programme in Great Britain. Biol. Cons. 182, 14–26 (2015).
Google Scholar 
53.Strebel, N., Kéry, M., Schaub, M. & Schmid, H. Studying phenology by flexible modelling of seasonal detectability peaks. Methods Ecol. Evol. 5, 483–490 (2014).
Google Scholar 
54.Maggini, R. et al. Are Swiss birds tracking climate change?: Detecting elevational shifts using response curve shapes. Ecol. Model. 222, 21–32 (2011).
Google Scholar 
55.Gilg, O. et al. Climate change and the ecology and evolution of Arctic vertebrates. Ann. N. Y. Acad. Sci. 1249, 166–190 (2012).ADS 
PubMed 

Google Scholar 
56.Gossmann, T. I. et al. Ice-age climate adaptations trap the alpine marmot in a state of low genetic diversity. Curr. Biol. 29, 1712–1720 (2019).CAS 
PubMed 
PubMed Central 

Google Scholar 
57.Charmantier, A. & Gienapp, P. Climate change and timing of avian breeding and migration: Evolutionary versus plastic changes. Evol. Appl. 7, 15–28 (2014).PubMed 

Google Scholar 
58.Klein, G., Vitasse, Y., Rixen, C., Marty, C. & Rebetez, M. Shorter snow cover duration since 1970 in the Swiss Alps due to earlier snowmelt more than to later snow onset. Clim. Change 139, 637–649 (2016).
Google Scholar 
59.Scridel, D. et al. A review and meta-analysis of the effects of climate change on Holarctic mountain and upland bird populations. Ibis 160, 489–515 (2018).
Google Scholar 
60.Strinella, E., Scridel, D., Brambilla, M., Schano, C. & Korner-Nievergelt, F. Potential sex-dependent effects of weather on apparent survival of a high-elevation specialist. Sci. Rep. 10, 8386 (2020).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
61.Gottfried, M. et al. Continent-wide response of mountain vegetation to climate change. Nat. Clim. Chang. 2, 111–115 (2012).ADS 

Google Scholar 
62.Kharouba, H. M. & Wolkovich, E. M. Disconnects between ecological theory and data in phenological mismatch research. Nat. Clim. Chang. 10, 406–415 (2020).ADS 

Google Scholar 
63.Summers-Smith, J. Handbook of the Birds of the World, Volume 14: Bush-Shrikes to Old World Sparrows. (2009).64.Glutz von Blotzheim, U., Bauer, K. & Bezzel, E. I: Passeridae. in Handbuch der Vögel Mitteleuropas. Vol. 12 (Akademische Verlagsgesellschaft, 1997).65.Antor, R. J. The importance of arthropod fallout on snow patches for the foraging of high-alpine birds. J. Avian Biol. 26, 81–85 (1995).
Google Scholar 
66.Gonseth, Y., Wohlgemuth, T., Sansonnens, B. & Buttler, A. Die Biogeographischen Regionen der Schweiz. Erläuterungen und Einteilungsstandard. Umwelt Materialien. Vol. 137 (2001).67.Thornton, P. E., Running, S. W. & White, M. A. Generating surfaces of daily meteorological variables over large regions of complex terrain. J. Hydrol. 190, 214–251 (1997).ADS 

Google Scholar 
68.Magnusson, J., Gustafsson, D., Hüsler, F. & Jonas, T. Assimilation of point SWE data into a distributed snow cover model comparing two contrasting methods. Water Resour. Res. 50, 7816–7835 (2014).ADS 

Google Scholar 
69.Helbig, N., van Herwijnen, A., Magnusson, J. & Jonas, T. Fractional snow-covered area parameterization over complex topography. Hydrol. Earth Syst. Sci. 19, 1339–1351 (2015).ADS 

Google Scholar 
70.Begert, M. & Frei, C. Long-term area-mean temperature series for Switzerland—Combining homogenized station data and high resolution grid data. Int. J. Climatol. 38, 2792–2807 (2018).
Google Scholar 
71.Bates, D., Mächler, M., Bolker, B. & Walker, S. Fitting linear mixed-effects models using lme4. ArXiv e-prints 1406 (2015).72.R Core Team. R: A Language and Environment for Statistical Computing. (2020).73.Gelman, A. & Su, Y.-S. Arm: Data analysis using regression and multilevel/hierarchical models. (2020).74.Carpenter, B. et al. Stan: A probabilistic programming language. J. Stat. Softw. 76 (2017).75.Stan Development Team. RStan: The R interface to Stan. (2020).76.Gabry, J. shinystan: Interactive Visual and Numerical Diagnostics and Posterior Analysis for Bayesian Models. (2018).77.Pebesma, E. J. Multivariable geostatistics in S: The gstat package. Comput. Geosci. 30, 683–691 (2004).ADS 

Google Scholar 
78.Pebesma, E. & Bivand, R. S. S classes and methods for spatial data: the sp package. R News 5, 9–13 (2005).
Google Scholar 
79.Gelman, A. & Greenland, S. Are confidence intervals better termed “uncertainty intervals”?. BMJ 366, I5381 (2019).
Google Scholar  More

Study area and samplingSmall mammals (murid rodents and shrews) were captured using mouse-type snap traps in Tatarstan, Russian Federation (Fig. 1, Table S1). Area type (urban or rural), vegetation (forest or field) and distance from trapping points to the nearest human settlement were recorded. The distinction between forest and field was made based on the UN Food and Agriculture Organization’s criteria23,24. Each administrative division in the Tatarstan was defined to be urban or rural by the Federal Service of State Statistics of Russian Federation25. Based on these criteria, Kazan city and Naberezhnye Chelny city were classified as urban districts and Vysokogorsky district, Yelabuzhsky district, Laishevsky district, Mamadyshsky district, Nizhnekamsky district, Pestrechinsky district and Tukayevsky district were classified as rural districts. Small mammals were captured during the spring and fall periods of 2016 and 2017. Fifty traps were placed in a line every 5 m in one place. Traps were baited and left for one night. Animal suffering was minimized as snap traps cause rapid death in murid rodents and shrews. Each captured small mammal’s species, age, and sex were morphologically identified using a reference guide26, and the animals were then stored at − 20 °C until their brains were isolated.EthicsAll experiments were performed in compliance with relevant Russian and Japanese and institutional laws and guidelines and were approved by the Ministry of Health of the Russian Federation and the Animal Research Committee of Gifu University (Permit Nos. MU 3.1.1029-01, and 17060, respectively). Study was carried out in compliance with the ARRIVE guidelines (https://arriveguidelines.org).DNA extraction and PCRBrain tissue samples were prepared as described previously12. Brain samples stored at − 20 °C were transferred to a − 86 °C deep freezer. Each deep-frozen whole brain sample was homogenized in 1 ml of a 0.9% saline solution. Total DNA was extracted from the brain tissues of each small mammal using a Genomic DNA Purification Kit (Promega, Madison, WI, USA), following the manufacturer’s instructions. Nested PCR was performed with the Takara PCR Amplification Kit (Takara Bio Inc., Foster City, California, USA) according to the manufacturer’s instructions. The primer sets and PCR conditions used to detect the B1 gene from T. gondii were those described previously12.MappingSpatial referencing of the sampling sites was conducted using global positioning system navigation with a Garmin eTrex 10 device. Visualization of cartographic data and measurements of the distances from the trapping points to the nearest human settlements were performed using QGIS 3.12 software27. Geodetic coordinates were projected into planar rectangular coordinates in the Universal Transverse Mercator projection on the WGS-84 ellipsoid (Universal Transverse Mercator, zone 39N). The overview map of the European part of Russia was made in the Lambert Conformal Conic Projection. Map coordinates are represented as geodetic coordinates (WGS-84, degrees and minutes north latitude and east longitude). To visualize thematic objects (administrative boundaries, forests, agricultural lands, and water bodies), a set of vector data layers, NextGIS (Russia), was purchased from OpenStreetMap and contributors, 2021 (https://data.nextgis.com). Data license: ODbL.Dataset and statistical analysesMultivariate logistic regression was performed using the R statistical software package (version 3.6.3)28 to assess the trapping point area (urban or rural), vegetation (forest or field), small mammal species type (alien or non-alien species), age (0–2 months-old juveniles, 3–6 months-old adults or ≧ 6 months old), sex (male or female) and distance from trapping points to the nearest human settlements as risk factors for PCR positivity. According to previous reports2,13,16,17,18, four species, Mi. arvalis, A. flavicollis, A. agrarius, A. uralensis, and three species, My. glareolus, S. araneus and D. nitedula are considered alien and non-alien species, respectively. Quantitative data were replaced with 0 or 1 dummy variables, and age data were replaced by 0, 1 and 2 for juveniles, adults and elders, respectively. Multicollinearity of the explanatory variables was evaluated using Spearman’s coefficient29 calculated using dplyr, FSA and psych packages30,31,32. None of the Spearman’s coefficients were  > 0.6. To find the best fit model, a forward selection procedure was used. Predictive performance and model fitting were assessed using the area under the receiver operating characteristic (ROC) curve, area under the curve (AUC) and corrected Akaike’s information criterion (AICc) with Akaike weight (Wi). AICc and Wi were calculated using the MuMIN package33, and the AUC was calculated using the R pROC package34. P-values of  More

The micro-CT results from Arthrorhynchus agree perfectly with the previously known light microscope and transmission electron microscope images2. This emphasizes that microtomography is a good technique to visualize the type of fungal attachment to the host and especially the penetration of the cuticle, apart from the study of thallus in amber fossils17. As Jensen et al. (2019) demonstrated the presence of a haustorium in Arthrorhynchus using scanning electron microscopy, we are confident that the lack of penetration and haustorium in Rickia found by micro-CT is real. This is also in agreement with results from the scanning electron microscopical investigation of the attachment sites of R. gigas, which exhibits no indication of penetration and are very similar to those of R. wasmannii previously shown18.Despite the absence of a haustorium, and hence without any obvious means of obtaining nutrition, Rickia gigas is quite a successful fungus, being often abundant on several species of Afrotropical millipedes of the family Spirostreptidae10. It was originally described from Archispirostreptus gigas, and Tropostreptus (= ‘Spirostreptus’) hamatus20, and was subsequently reported from several other Tropostreptus species19.A further challenge for Laboulbeniales growing on millipedes is that infected millipedes, in some species even adults, may moult, shedding the exuviae with the fungus, as has been observed by us on an undescribed Rickia species on a millipede of the genus Spirobolus (family Spirobolidae).The question of how non-haustoriate Laboulbeniales obtain nutrients has been discussed by several authors18, including staining experiments using fungi of the non-haustoriate genus Laboulbenia on various beetles21. Whereas the surface of the main thallus was almost impenetrable to the dye applied (Nile Blue), the smaller appendages could sometimes be penetrated21. The dye injection into the beetle elytra upon which the fungi were sitting, actually spread from the elytron into the fungus, thus indicating that in spite of the lack of a haustorium, the fungus is able to extract nutrients from the interior of its host21.Such experiments have not been performed on Rickia species, but the possibility that nutrients may pass from the host into the basis of the fungus cannot be excluded. For this genus, or at least R. gigas, there may, however, be an alternative way to obtain nutrients: the small opening in the circular wall by which the thallus is attached to the host may allow nutrients from the surface of the millipede or from the environment to seep into the foot of the fungus. However, further experiments are needed in order to evaluate this hypothesis. Moreover, we should not exclude a potential role of primary and secondary appendages in Laboulbeniales nutrition, as we still do not understand exactly their functional role on the fungus life cycle11.The predominant position of the Laboulbeniales on the host might be related to the absence or presence of a haustorium. Thus, the haustoriate species of the genus Arthrorhynchus are most frequently encountered in large numbers on the arthrodial membranes of the host’s abdomen, although some thalli are found on legs2,22. At the arthrodial membranes the cuticle is more flexible and therefore might be easier to penetrate by a parasite. Furthermore, most tissues providing/storing nutrition (e.g., fat body) are located within the abdomen. In contrast, non-haustoriate fungi as are often located on more stiff and sclerotized body-parts like the genus Rickia on the legs or body-rings of millipedes7,20,23 or the genus Laboulbenia on the elytra of beetles21,24. A reason for this might be that the non-haustoriate forms, which are only superficially attached to the host need a more or less smooth surface for adherence and can easily become detached from a flexible surface, which is movable in itself, like the arthrodial membrane, while the haustoriate forms are firmly anchored within the hosts’ cuticle.Whereas the vast majority of the more than 2000 described species of Laboulbeniales show no sign of host penetration, haustoria have been reported from some other genera18, including Trenomyces parasitizing bird lice25,26, Hesperomyces growing on coccinellid beetles and Herpomyces on cockroaches (formerly a Laboulbeniales and now in the order Herpomycetales10), with pernicious consequences on the hosts’ fitness18,27. Micro-CT studies on these genera could help to understand the host penetration. In order to fully understand how Laboulbeniales obtain nourishment, although other approaches are, also needed—for the time being it remains a mystery how the non-haustoriate Laboulbeniales sustain themselves. More

Collection and domestication of the wild populationsThe academic permission for collections and research on medicinal plants was obtained from the Head of Biotechnology Department, Research Institute of Modern Biological Techniques, University of Zanjan, Zanjan, Iran. The study complies with all relevant guidelines. Some populations of wild spinaches were harvested during spring season 2013 from the mountain habitat of this wild plant in the Tarom region of Zanjan province from an altitude of 2500–3000 m and were transferred to the greenhouses conditions. The domestication and cultivation experiments were conducted at Research Institute of Modern Biological Techniques, University of Zanjan, 1579° m above sea level, with 48° 28′ longitude and 36° 40′ latitude, from April 2013 to August 2020. The resulted seeds were cultured on pots to produce adequate seeds. The seedlings were transferred to the field with rows spaced 50 cm apart and also 50 cm between plants within the rows. Two seeds per hill were planted in an area of approximately 50 m2. Based on the organic conditions, no fertilization was performed. Thinning was done 25 days after emergence, leaving one plant per hill. The other cultural practices were those normally adopted for cultivation in the region.Mass selection of populationsIn the first year, phenotypic studies were performed during the growing season and weak, diseased and underdeveloped plants were removed from the field before the flowering stage. Then plants with the same phenotype and the desired traits were selected and after harvesting, their seeds were mixed. This election cycle was repeated for 5 years. In the final year, the new mass selected population was compared in a pilot project with cultivated spinach in traits such as yield, resistance to wilt, cold and pests, diseases, and mineral contents. This variety before the certification in the related national organization is a candida cultivar. It is a developed population that will be evaluated in the session of the Iranian variety of introduction committee.The seeds of cultivated spinach (Spinacia oleracea L. |Varamin 88|) were prepared from the Research Institute of Modern Biological Techniques, University of Zanjan, Zanjan, Iran.Performing tests of stability, uniformity and differentiationTo assess morphologically and differentiate advanced uniformity in the studied population (Candida cultivar), the population was managed as a randomized complete block design with three replications over 2 years according to the instructions for spinach differentiation, uniformity, and stability (DUS Testing) of the International Union New Plant Cultivation (UPOV) and some morphological traits on plants or parts of plants. The studied traits included: cotyledon length, presence or absence of anthocyanin in petiole and veins, green color intensity, shrinkage, presence of lobes in the petiole, petiole state, petiole length, foil shape, foil edge shape, tip shape, and part of the length of the petiole, the time of flowering and the color of the seeds.Mineral analysesTo compare the mineral content of mass-selected population-medicinal spinach (MSP) with cultivated spinach (Spinacia oleracea L. var. Varamin 88), both plants were planted in pots and fields on similar conditions. In five leaves stage, plant samples were taken from both leaf and crown sections. The sampling method was such that after removing half a meter from the beginning and end of each plot (to remove the marginal effect) and also removing the two sidelines, five plants were harvested randomly for plant mineral analysis. Atomic absorption spectroscopy was used to determine the mineral content including iron (Fe), zinc (Z), manganese (Mn), and copper (Cu).The dried samples of root-crown and leave were stored, and later grounded and analyzed for iron (Fe), zinc (Z), manganese (Mn), and copper (Cu) in mass-selected variety (MSP) and cultivated spinach (CSP). Studied minerals were measured using atomic absorption spectrometry in the model of GBC AVANTA (GBC scientific equipment Ltd., Melbourne, Vic., Australia).Calibration of AAS was done using the working standard prepared from commercially available metal/mineral standard solutions (1000 μg/mL, Merck, Germany). The most appropriate wavelength, hollow cathode lamp current, gas mixture flow rate, slit width, and other AAS instrument parameters for metals/minerals were selected as given in the instrument user’s manual, and background correction was used during the determination of metals/minerals. Measurements were made within the linear range of working standards used for calibration15,16.The concentrations of all the minerals were expressed as mg/1000 g (ppm) dry weight of the sample. Each value is the mean of three replicate determination ± standard deviation.Scanning electron microscopy (SEM)For SEM studies, the seeds enveloping were removed and were acetolyzed in a 1:9 sulfuric acid-acetic anhydride solution. The seeds were vigorously shaken for 5 min. Then, they were left for 24–48 h in the solution. After this time, seeds were again shaken for 5 min and then washed.in distilled water by shaking for a further 5 min. The seeds were dried overnight and then were mounted on stubs and covered with Au–Pd by sputter coater model SC 7620. After coating, coated seeds were photographed with an LEO 1450 VP Scanning Electron Microscope. All photographs were taken in the Taban laboratory (Tehran, Iran).Statistical analysisThe statistical evaluation including: data transformation, analysis of variance and comparison of means were performed (SPSS software, Version 11.0). The experiment was structured following a randomized complete block design (RCBD) with three replications. Means comparisons were conducted using an ANOVA protected the least significant difference (LSD) test, with the ANOVA confidence levels of 0.95. Data were presented with their standard deviations (SD). More

Dispersal modelsIn this study, the dispersal model consists of two parts, namely, kernel and observation model (Fig. 1). The main purpose of the kernel was employed to estimate the proportion of pollen dispersed from location s′ to location s and calculate the expected number of CP grains. The observation model used the expected number of CP grains as a parameter and described the number of CP grains at location s (Ys) by a specific distribution in the following:$${Y}_{s}sim fleft(left.{y}_{s}right|{{varvec{theta}}}_{s}right),$$
(1)
where f indicates the probability density function (PDF) of the specific distribution. The θs is the parameter vector of the distribution. This study constructed eight different dispersal models combined with two observation models, two kernels, and two conditions of the field border (FB) effect (Table 1). The details of the kernels and observation models were described in the following subsections.Figure 1Graphical summary of the establishment of the dispersal model using ZIP distribution observation model as an example.Full size imageTable 1 List of dispersal models constructed in this study.Full size tableKernelsThe kernel indicates the probability when the pollen emitted at location s′ and would fall down at location s. It can be expressed as γ(s, s′), where s′ is the source location closest to location s. Numerous kernels have been used to describe various dispersal phenomena24. The output of the kernel represents the donor pollen density of location s. In order to calculate the expected number of CP grains, the donor pollen density is multiplied by the average total grain number described as follows:$${lambda }_{s}=Ktimes gamma left(s,{s}^{^{prime}}right),$$
(2)
where λs and K indicate the expected number of CP grains at location s and the average number of grains per cob, respectively. The effect of the FB was introduced into the kernel to suit to the small-scale farming system in Asia. This study assumed that the relation between the pollen density at the first recipient row and the width of the FB displayed an exponential decrease25,26. To evaluate the improvement of the kernel with the FB effect, the kernels without the FB effect were also established in this study.The compound exponential kernel (γExpo) has been used in the previous pollen dispersal study27. Our study introduced the FB effect into this kernel. Therefore, the form of the compound exponential kernel can be expressed as follows:$$gamma_{{{text{Expo}}}} left( {s,s^{prime}} right) = left{ {begin{array}{*{20}l} {K_{e} exp left( { – a_{1} d^{*} left( {s,s^{prime}} right)} right)exp left( { – ksqrt {FB} } right),} \ {K_{e} exp left( { – a_{1} D – a_{2} left( {d^{*} left( {s,s^{prime}} right) – D} right)} right)exp left( { – ksqrt {FB} } right),} \ end{array} } right.begin{array}{*{20}l} {{text{if}},, d^{*} left( {s,s^{prime}} right) le D} \ {{text{if}} ,,d^{*} left( {s,s^{prime}} right) > D,} \ end{array}$$
(3)
where Ke, a1, a2, k, D are the parameters of the kernel. d*(s, s′) indicates the shortest distance between locations s′ and s in which the width of the FB has been subtracted. In the compound exponential kernel without the FB effect, the exponential term of the FB effect was removed and the d*(s, s′) was replaced directly by the shortest distance between s′ and s.The second kernel applied in this study was the modified Cauchy kernel (γCauchy) which was based on the PDF of the Cauchy distribution and the concept of compound distribution. The modified Cauchy kernel is represented as follows:$$gamma_{Cauchy} left( {s,s^{prime}} right) = left{ {begin{array}{*{20}l} {frac{2beta }{{pi left[ {beta^{2} + d^{*} left( {s,s^{prime}} right)^{2} } right]}}{text{exp}}left( { – ksqrt {FB} } right),} \ {frac{2beta }{{pi left[ {beta^{2} + D^{2} + c_{1} left( {d^{*} left( {s,s^{prime}} right) – D} right)^{2} } right]}}{text{exp}}left( { – ksqrt {FB} } right),} \ end{array} } right.begin{array}{*{20}l} {{text{if}} ,,d^{*} left( {s,s^{prime}} right) le D} \ {{text{if}} ,,d^{*} left( {s,s^{prime}} right) > D,} \ end{array}$$
(4)
where the β indicates the decline rate of the curve. Parameters of k and D are same as the compound exponential kernel. c1 indicates the relative slow decrease of pollen density at further distances. Similarly, in the modified Cauchy kernel without the FB effect, the term of the FB effect was removed and the d*(s, s′) was replaced directly by the shortest distance between s′ and s in which the row spacing (0.75 m) had been subtracted.Observation modelsBecause of the high proportions of zero value observations, the present study assumed that the CP grain count followed the zero-inflated Poisson (ZIP) distribution to account for zero-excess condition28. The ZIP distribution was first proposed by Lambert29, and several studies had applied the ZIP distribution to deal with the CP data27,30. The ZIP distribution consists of a Dirac distribution in zero and a Poisson distribution. Therefore, the distribution of CP grain count at location s (Ys) can be expressed as follows:$${Y}_{s}sim mathrm{ZIP}left(1-{q}_{s},{uplambda }_{s}right),$$
(5)
where qs indicates the probability of an observation following a Poisson distribution, and λs is the parameter of Poisson distribution calculated by Eq. (2). Furthermore, the parameter qs can be assumed to depend on the shortest distance between the recipient and donor plants. The border effect is also included in the estimation of qs because it is related to the distance effect. The relationship among distance, border, and the qs can be described using the following logistic function:$${q}_{s}=frac{1}{1+mathrm{exp}({b}_{1}-{b}_{2}{d}^{*}left(s,{s}^{^{prime}}right))},$$
(6)
where b1 and b2 are the parameters of the logistic function. The d*(s, s′) was the shortest distance between s′ and s in the version of dispersal models without the FB effect. The Poisson distribution was also used as an observation model for comparison with the ZIP observation model.Experimental and meteorological data collectionThe pollen dispersal data were collected from experiments performed in 2009 and 2010 at the geographic coordinates 23° 47′ N, 120° 26′ E, and an altitude of 20 m. These experiments were coded as 2009-1, 2009-2, and 2010-1, respectively. The experiment 2009-2 was divided into 2009-2A (without the FB) and 2009-2B (with the FB) based on the presence of the FB. The different layouts of the field experiments were designed to investigate the effect of the FB. Two commercial glutinous maize varieties, black pearl (purple grain) and Tainan No. 23 (white grain), were selected as the pollen donor and pollen recipient, respectively. The distance between the plants in a row was 25 cm, whereas the distance between the rows was 75 cm. The recipient plots consisted of 82 and 91 rows in 2009 and 2010 experiments, respectively.The CP rate was determined based on the differences in grain color on recipient cobs as a result of the xenia effect31. In the sampling framework, the whole field was divided into many grids and corn samples were collected from each grid in the whole field. The CP rate of each grid was calculated using the method presented in a previous study32 and defined as:$$mathrm{CP}left(%right)=left[sum_{i=1}^{n}{Cob}_{i}/left(ntimes Kright)right],$$
(7)

where Cobi and n indicate ith cob and total number of cobs in the grid, respectively. K is the average grain number per cob. Meteorological data were collected from the meteorological station at geographic coordinates 23° 35′ N, 120° 27′ E, and an altitude of 20 m. The detailed experimental setup was described in our previous study33. The study complies with relevant institutional, national, and international guidelines and legislation.Statistical analysesAll statistical analyses were performed using SAS (Statistical Analysis System, version 9.4). The dispersal model parameters were estimated by two methods. First, the nonlinear model estimation was conducted by PROC NLMIXED to evaluate the fitting and predictive abilities of dispersal models. Then the dispersal models with the observation model performed better fitting ability were re-estimated using the Bayesian estimation method to assess the uncertainty by PROC MCMC. In the Bayesian method, the noninformative prior distribution was used to estimate all parameters (Supplementary Table S1). The iteration of Markov Chain was 500,000 times and the burn-in was set to 450,000 iterations. In order to reduce the autocorrelations in the chain, the thinned value was set to 25.The validation method used in this study was the threefold cross-validation for the results of both estimation methods. The data from three experiments were combined and randomly partitioned into three sub-datasets. To avoid the heterogeneity of the different field designs and distances among sub-datasets, the observations from the same field design and same distance were considered as a group, and then partitioned into three parts. Each sub-dataset contained one part of all groups. At each validation run, two sub-datasets were selected as the training set, and the remaining one was used for validation.The fitting ability of the dispersal models was evaluated based on two criteria, namely, Akaike information criterion (AIC), Deviance, and coefficient of determination (R2). The smaller values of AIC or deviance indicate a better fitting. The higher R2 value represents a better fitting performance. The correlation coefficient (r) between the predicted and actual CP rates was used to assess the predictive ability. The deviance information criterion (DIC) was used to evaluate the performance of dispersal model fitting for the Bayesian estimation. The criterion values calculated from three training and validation sets were averaged to assess the overall results. The uncertainty of the model parameter was quantified by the standard deviation (SD) of parameter posterior distribution. The 95% credible intervals of posterior predictive distribution constructed by the 2.5th and 97.5th percentiles of 200,000 samples generated from the posterior predictive distribution were used to assess the predictive uncertainty. Furthermore, to assess the zero-excess condition, the percentage of observed zero CP grain events was compared with the Poisson probability of the zero CP grain event. A zero-excess condition occurred if the observed percentage was higher than the Poisson probability34. More

1.Maynard Smith, J., Feil, E. J. & Smith, N. H. Population structure and evolutionary dynamics of pathogenic bacteria. Bioessays 22, 1115–1122 (2000).
Google Scholar 
2.Garud, N. R., Good, B. H., Hallatschek, O. & Pollard, K. S. Evolutionary dynamics of bacteria in the gut microbiome within and across hosts. PLoS Biol. 17, e3000102 (2019). Using metagenomic samples form the human gut microbiome, the authors infer lineage structure from within-host polymorphisms in more than 40 species to show adaptation on short timescales can be seeded by HGT.PubMed 
PubMed Central 

Google Scholar 
3.Frazão, N., Sousa, A., Lässig, M. & Gordo, I. Horizontal gene transfer overrides mutation in Escherichia coli colonizing the mammalian gut. Proc. Natl Acad. Sci. USA 116, 17906–17915 (2019). Using the mouse microbiome as a study system, the authors show that rapid, phage-mediated HGT can transfer beneficial genes — already present in a resident strain — to an invading strain.PubMed 
PubMed Central 

Google Scholar 
4.Smith, J. M., Smith, N. H., O’Rourke, M. & Spratt, B. G. How clonal are bacteria? Proc. Natl Acad. Sci. USA 90, 4384–4388 (1993).PubMed 
PubMed Central 

Google Scholar 
5.Dykhuizen, D. E. & Green, L. Recombination in Escherichia coli and the definition of biological species. J. Bacteriol. 173, 7257–7268 (1991).PubMed 
PubMed Central 

Google Scholar 
6.Feil, E. J. et al. Recombination within natural populations of pathogenic bacteria: short-term empirical estimates and long-term phylogenetic consequences. Proc. Natl Acad. Sci. USA 98, 182–187 (2001).PubMed 
PubMed Central 

Google Scholar 
7.Suerbaum, S. et al. Free recombination within Helicobacter pylori. PNAS 95, 12619–12624 (1998).PubMed 
PubMed Central 

Google Scholar 
8.Smillie, C. S. et al. Ecology drives a global network of gene exchange connecting the human microbiome. Nature 480, 241–244 (2011).PubMed 

Google Scholar 
9.Lozupone, C. A. et al. The convergence of carbohydrate active gene repertoires in human gut microbes. Proc. Natl Acad. Sci. USA 105, 15076–15081 (2008).PubMed 
PubMed Central 

Google Scholar 
10.Bradley, P. H., Nayfach, S. & Pollard, K. S. Phylogeny-corrected identification of microbial gene families relevant to human gut colonization. PLoS Computational Biol. 14, e1006242 (2018). The authors use phylogenetic linear regression to control for important confounders and identify genes potentially involved in adaptation to the human gut.
Google Scholar 
11.Andreani, N. A., Hesse, E. & Vos, M. Prokaryote genome fluidity is dependent on effective population size. ISME J. 11, 1719–1721 (2017).PubMed 
PubMed Central 

Google Scholar 
12.Mcinerney, J. O., Mcnally, A. & Connell, M. J. O. Why prokaryotes have pangenomes. Nat. Publ. Gr. 2, 1–5 (2017).
Google Scholar 
13.Shapiro, B. J. The population genetics of pangenomes. Nat. Microbiol. 2, 1005860 (2017).
Google Scholar 
14.Vos, M. & Eyre-walker, A. Are pangenomes adaptive or not? Nat. Microbiol. https://doi.org/10.1038/s41564-017-0067-5 (2017).Article 
PubMed 
PubMed Central 

Google Scholar 
15.Johnsborg, O., Eldholm, V. & Håvarstein, L. S. Natural genetic transformation: prevalence, mechanisms and function. Res. Microbiol. 158, 767–778 (2007).PubMed 

Google Scholar 
16.Johnston, C., Martin, B., Fichant, G., Polard, P. & Claverys, J. P. Bacterial transformation: distribution, shared mechanisms and divergent control. Nat. Rev. Microbiol. 12, 181–196 (2014).PubMed 

Google Scholar 
17.Pimentel, Z. T. & Zhang, Y. Evolution of the natural transformation protein, ComEC, in Bacteria. Front. Microbiol. 9, 1–10 (2018).
Google Scholar 
18.Roux, S., Hallam, S. J., Woyke, T. & Sullivan, M. B. Viral dark matter and virus–host interactions resolved from publicly available microbial genomes. eLife 4, 1–20 (2015).
Google Scholar 
19.Camarillo-Guerrero, L. F. et al. Massive expansion of human gut bacteriophage diversity. Cell 184, 1098–1109.e9 (2021).PubMed 
PubMed Central 

Google Scholar 
20.Guglielmini, J., Quintais, L., Garcillán-Barcia, M. P., de la Cruz, F. & Rocha, E. P. C. The repertoire of ice in prokaryotes underscores the unity, diversity, and ubiquity of conjugation. PLoS Genet. 7, e1002222 (2011).PubMed 
PubMed Central 

Google Scholar 
21.Dubey, G. P. & Ben-Yehuda, S. Intercellular nanotubes mediate bacterial communication. Cell 144, 590–600 (2011).PubMed 

Google Scholar 
22.Abe, K., Nomura, N. & Suzuki, S. Biofilms: hot spots of horizontal gene transfer (HGT) in aquatic environments, with a focus on a new HGT mechanism. FEMS Microbiol. Ecol. 96, 1–12 (2020).
Google Scholar 
23.Bárdy, P. et al. Structure and mechanism of DNA delivery of a gene transfer agent. Nat. Commun. 11, 3034 (2020).PubMed 
PubMed Central 

Google Scholar 
24.Hasegawa, H., Suzuki, E. & Maeda, S. Horizontal plasmid transfer by transformation in Escherichia coli: environmental factors and possible mechanisms. Front. Microbiol. 9, 1–6 (2018).
Google Scholar 
25.Seitz, P. & Blokesch, M. Cues and regulatory pathways involved in natural competence and transformation in pathogenic and environmental Gram-negative bacteria. FEMS Microbiol. Rev. 37, 336–363 (2013).PubMed 

Google Scholar 
26.Wall, D. Kin recognition in bacteria. Annu. Rev. Microbiol. 70, 143–160 (2016).PubMed 
PubMed Central 

Google Scholar 
27.Frye, S. A., Nilsen, M., Tønjum, T. & Ambur, O. H. Dialects of the DNA uptake sequence in Neisseriaceae. PLoS Genet. https://doi.org/10.1371/journal.pgen.1003458 (2013).Article 
PubMed 
PubMed Central 

Google Scholar 
28.Redfield, R. J. et al. Evolution of competence and DNA uptake specificity in the Pasteurellaceae. BMC Evol. Biol. 6, 1–15 (2006).
Google Scholar 
29.Dion, M. B., Oechslin, F. & Moineau, S. Phage diversity, genomics and phylogeny. Nat. Rev. Microbiol. https://doi.org/10.1038/s41579-019-0311-5 (2020).Article 
PubMed 

Google Scholar 
30.Siguier, P., Gourbeyre, E. & Chandler, M. Bacterial insertion sequences: their genomic impact and diversity. FEMS Microbiol. Rev. 38, 865–891 (2014).PubMed 

Google Scholar 
31.Vulić, M., Dionisio, F., Taddei, F. & Radman, M. Molecular keys to speciation: DNA polymorphism and the control of genetic exchange in enterobacteria. Proc. Natl Acad. Sci. USA 94, 9763–9767 (1997).PubMed 
PubMed Central 

Google Scholar 
32.Majewski, J. et al. Barriers to genetic exchange between bacterial species: Streptococcus pneumoniae transformation. J. Bacteriol. 182, 1016–1023 (2000).PubMed 
PubMed Central 

Google Scholar 
33.Wyres, K. L. et al. Pneumococcal capsular switching: a historical perspective. J. Infect. Dis. 207, 439–449 (2013).PubMed 

Google Scholar 
34.Hallet, B. & Sherratt, D. J. Transposition and site-specific recombination: adapting DNA cut-and-paste mechanisms to a variety of genetic rearrangements. FEMS Microbiol. Rev. 21, 157–178 (1997).PubMed 

Google Scholar 
35.Durrant, M. G., Li, M. M., Siranosian, B. A., Montgomery, S. B. & Bhatt, A. S. A bioinformatic analysis of integrative mobile genetic elements highlights their role in bacterial adaptation. Cell Host Microbe 27, 140–153.e9 (2020).PubMed 

Google Scholar 
36.Rajeev, L., Malanowska, K. & Gardner, J. F. Challenging a paradigm: the role of DNA homology in tyrosine recombinase reactions. Microbiol. Mol. Biol. Rev. 73, 300–309 (2009).PubMed 
PubMed Central 

Google Scholar 
37.Hickman, A. B., Chandler, M. & Dyda, F. Integrating prokaryotes and eukaryotes: DNA transposases in light of structure. Crit. Rev. Biochem. Mol. Biol. 45, 50–69 (2010).PubMed 
PubMed Central 

Google Scholar 
38.Oliveira, P. H., Touchon, M., Cury, J. & Rocha, E. P. C. The chromosomal organization of horizontal gene transfer in bacteria. Nat. Commun. 8, 1–10 (2017).
Google Scholar 
39.Wadsworth, C. B., Arnold, B. J., Sater, M. R. A. & Grad, Y. Azithromycin resistance through interspecific acquisition of an epistasis-dependent efflux pump component and transcriptional regulator in Neisseria gonorrhoeae. mBio 9, 1–17 (2018).
Google Scholar 
40.Arevalo, P., VanInsberghe, D., Elsherbini, J., Gore, J. & Polz, M. F. A reverse ecology approach based on a biological definition of microbial populations. Cell 178, 820–834.e14 (2019). The authors create a metric of recent gene flow to define ecological populations and discover genes that have experienced positive selection across populations.PubMed 

Google Scholar 
41.Croucher, N. J. et al. Horizontal DNA transfer mechanisms of bacteria as weapons of intragenomic conflict. PLoS Biol. 14, 1–42 (2016). A model of transformation with known bias towards the acquisition of shorter alleles suggests HGT may effectively purge bacterial genomes of parasitic MGEs.
Google Scholar 
42.Apagyi, K. J., Fraser, C. & Croucher, N. J. Transformation asymmetry and the evolution of the bacterial accessory genome. Mol. Biol. Evol. 35, 575–581 (2018).PubMed 

Google Scholar 
43.Mira, A., Ochman, H. & Moran, N. A. Deletional bias and the evolution of bacterial genomes. Trends Genet. 17, 589–596 (2001).PubMed 

Google Scholar 
44.Kuo, C.-H. & Ochman, H. Deletional bias across the three domains of life. Genome Biol. Evol. 1, 145–152 (2009).PubMed 
PubMed Central 

Google Scholar 
45.Lawrence, J. G. & Roth, J. R. Selfish operons: horizontal transfer may drive the evolution of gene clusters. Genetics 143, 1843–1860 (1996).PubMed 
PubMed Central 

Google Scholar 
46.Hehemann, J. H. et al. Transfer of carbohydrate-active enzymes from marine bacteria to Japanese gut microbiota. Nature 464, 908–912 (2010).PubMed 

Google Scholar 
47.Campbell, A. Prophage insertion sites. Res. Microbiol. 154, 277–282 (2003).PubMed 

Google Scholar 
48.Chu, N. D. et al. A mobile element in mutS drives hypermutation in a marine Vibrio. mBio 8, 1–13 (2017).
Google Scholar 
49.Bobay, L. M., Rocha, E. P. C. & Touchon, M. The adaptation of temperate bacteriophages to their host genomes. Mol. Biol. Evol. 30, 737–751 (2013).PubMed 

Google Scholar 
50.Lee, H., Doak, T. G., Popodi, E., Foster, P. L. & Tang, H. Insertion sequence-caused large-scale rearrangements in the genome of Escherichia coli. Nucleic Acids Res. 44, 7109–7119 (2016).PubMed 
PubMed Central 

Google Scholar 
51.Parkhill, J. et al. Comparative analysis of the genome sequences of Bordetella pertussis, Bordetella parapertussis and Bordetella bronchiseptica. Nat. Genet. 35, 32–40 (2003).PubMed 

Google Scholar 
52.Moran, N. A. & Plague, G. R. Genomic changes following host restriction in bacteria. Curr. Opin. Genet. Dev. 14, 627–633 (2004).PubMed 

Google Scholar 
53.Hendry, T. et al. Ongoing transposon-mediated genome reduction in the luminous bacterial symbionts of deep-sea ceratioid anglerfishes. mBio https://doi.org/10.1128/mBio.01033-18 (2018).Article 
PubMed 
PubMed Central 

Google Scholar 
54.Waterworth, S. C. et al. Horizontal gene transfer to a defensive symbiont with a reduced genome in a multipartite beetle microbiome. mBio https://doi.org/10.1128/mBio.02430-19 (2020).Article 
PubMed 
PubMed Central 

Google Scholar 
55.Vos, M. et al. Rates of lateral gene transfer in prokaryotes: high but why? Trends Microbiol. 23, 598–605 (2015).PubMed 

Google Scholar 
56.Cohen, E., Kessler, D. A. & Levine, H. Recombination dramatically speeds up evolution of finite populations. Phys. Rev. Lett. 94, 1–4 (2005).
Google Scholar 
57.Levin, B. R. & Cornejo, O. E. The population and evolutionary dynamics of homologous gene recombination in bacteria. PLoS Genet. https://doi.org/10.1371/journal.pgen.1000601 (2009).Article 
PubMed 
PubMed Central 

Google Scholar 
58.Arnold, B. J. et al. Weak epistasis may drive adaptation in recombining bacteria. Genetics 208, 1247–1260 (2018).PubMed 
PubMed Central 

Google Scholar 
59.Moradigaravand, D. & Engelstädter, J. The effect of bacterial recombination on adaptation on fitness landscapes with limited peak accessibility. PLoS Comput. Biol. 8, 35–37 (2012).
Google Scholar 
60.Cooper, T. F. Recombination speeds adaptation by reducing competition between beneficial mutations in populations of Escherichia coli. PLoS Biol. 5, 1899–1905 (2007).
Google Scholar 
61.Winkler, J. & Kao, K. C. Harnessing recombination to speed adaptive evolution in Escherichia coli. Metab. Eng. 14, 487–495 (2012).PubMed 

Google Scholar 
62.Chu, H. Y., Sprouffske, K. & Wagner, A. The role of recombination in evolutionary adaptation of Escherichia coli to a novel nutrient. J. Evol. Biol. 30, 1692–1711 (2017).PubMed 

Google Scholar 
63.Arnold, B. et al. Fine-scale haplotype structure reveals strong signatures of positive selection in a recombining bacterial pathogen. Mol. Biol. Evol. https://doi.org/10.1093/molbev/msz225 (2019).Article 
PubMed Central 

Google Scholar 
64.Yahara, K. et al. The landscape of realized homologous recombination in pathogenic bacteria. Mol. Biol. Evol. 33, 456–471 (2016).PubMed 

Google Scholar 
65.Engelstädter, J. & Moradigaravand, D. Adaptation through genetic time travel? Fluctuating selection can drive the evolution of bacterial transformation. Proc. R. Soc. B Biol. Sci. 281, 20132609 (2014).
Google Scholar 
66.Cohan, F. M. Periodic selection and ecological diversity in bacteria. Selective Sweep https://doi.org/10.1007/0-387-27651-3_7 (2007).Article 

Google Scholar 
67.Shapiro, B. J., David, L. A., Friedman, J. & Alm, E. J. Looking for Darwin’s footprints in the microbial world. Trends Microbiol. 17, 196–204 (2009).PubMed 

Google Scholar 
68.Shapiro, B. J. et al. Population genomics of early events in the ecological differentiation of bacteria. Science 336, 48–51 (2012).PubMed 
PubMed Central 

Google Scholar 
69.Rosen, M., Davison, M., Bhaya, D. & Fisher, D. S. Fine-scale diversity and extensive recombination in a quasisexual bacterial population occupying a broad niche. Science 348, 1019–1024 (2015).PubMed 

Google Scholar 
70.Bendall, M. L. et al. Genome-wide selective sweeps and gene-specific sweeps in natural bacterial populations. ISME J. 10, 1589–1601 (2016).PubMed 
PubMed Central 

Google Scholar 
71.Porter, S. S., Chang, P. L., Conow, C. A., Dunham, J. P. & Friesen, M. L. Association mapping reveals novel serpentine adaptation gene clusters in a population of symbiotic Mesorhizobium. ISME J. 11, 248–262 (2017).PubMed 

Google Scholar 
72.Crits-Christoph, A., Olm, M. R., Diamond, S., Bouma-Gregson, K. & Banfield, J. F. Soil bacterial populations are shaped by recombination and gene-specific selection across a grassland meadow. ISME J. https://doi.org/10.1038/s41396-020-0655-x (2020).Article 
PubMed 
PubMed Central 

Google Scholar 
73.Woods, L. C. et al. Horizontal gene transfer potentiates adaptation by reducing selective constraints on the spread of genetic variation. Proc. Natl Acad. Sci. USA 117, 26868–26875 (2020).PubMed 
PubMed Central 

Google Scholar 
74.Miralles, R., Gerrish, P. J., Moya, A. & Elena, S. F. Clonal interference and the evolution of RNA viruses. Science 285, 1745–1747 (1999).PubMed 

Google Scholar 
75.De Visser, J. A. G. M., Zeyl, C. W., Gerrish, P. J., Blanchard, J. L. & Lenski, R. E. Diminishing returns from mutation supply rate in asexual populations. Science 283, 404–406 (1999).PubMed 

Google Scholar 
76.Good, B. H., Rouzine, I. M., Balick, D. J., Hallatschek, O. & Desai, M. M. Distribution of fixed beneficial mutations and the rate of adaptation in asexual populations. Proc. Natl Acad. Sci. USA 109, 4950–4955 (2012).PubMed 
PubMed Central 

Google Scholar 
77.Takeuchi, N., Cordero, O. X., Koonin, E. V. & Kaneko, K. Gene-specific selective sweeps in bacteria and archaea caused by negative frequency-dependent selection. BMC Biol. 13, 1–11 (2015). The authors show that in the presence of NFDS, genes or mutations that are unconditionally beneficial can spread through populations only via HGT, giving rise to gene-specific sweeps.
Google Scholar 
78.Corander, J. et al. Frequency-dependent selection in vaccine-associated pneumococcal population dynamics. Nat. Ecol. Evol. 2017, 1950–1960 (2018).
Google Scholar 
79.Rodriguez-Valera, F. et al. Explaining microbial population genomics through phage predation. Nat. Rev. Microbiol. 7, 828–836 (2009).PubMed 

Google Scholar 
80.Good, B. H., McDonald, M. J., Barrick, J. E., Lenski, R. E. & Desai, M. M. The dynamics of molecular evolution over 60,000 generations. Nature 551, 45–50 (2017).PubMed 
PubMed Central 

Google Scholar 
81.Ramiro, R. S., Durão, P., Bank, C. & Gordo, I. Low mutational load allows for high mutation rate variation in gut commensal bacteria. PLoS Biol. https://doi.org/10.1101/568709 (2020).Article 
PubMed 
PubMed Central 

Google Scholar 
82.Holt, R. D. Bringing the Hutchinsonian niche into the 21st century: ecological and evolutionary perspectives. Proc. Natl Acad. Sci. USA 106, 19659–19665 (2009).PubMed 
PubMed Central 

Google Scholar 
83.Cohan, F. M. Transmission in the origins of bacterial diversity, from ecotypes to phyla. Microbiol. Spectr. https://doi.org/10.1128/9781555819743.ch18 (2017).Article 
PubMed 

Google Scholar 
84.Fondi, M. et al. “Every gene is everywhere but the environment selects”: global geolocalization of gene sharing in environmental samples through network analysis. Genome Biol. Evol. 8, 1388–1400 (2016).PubMed 
PubMed Central 

Google Scholar 
85.Cohan, F. M. The effects of rare but promiscuous genetic exchange on evolutionary divergence in prokaryotes. Am. Nat. 143, 965–986 (1994).
Google Scholar 
86.Majewski, J. & Cohan, F. M. Adapt globally, act locally: the effect of selective sweeps on bacterial sequence diversity. Genetics 152, 1459–1474 (1999).PubMed 
PubMed Central 

Google Scholar 
87.Messer, P. W. & Petrov, D. A. Population genomics of rapid adaptation by soft selective sweeps. Trends Ecol. Evol. https://doi.org/10.1016/j.tree.2013.08.003 (2013).Article 
PubMed 
PubMed Central 

Google Scholar 
88.Cui, Y. et al. Epidemic clones, oceanic gene pools, and Eco-LD in the free living marine pathogen Vibrio parahaemolyticus. Mol. Biol. Evol. 32, 1396–1410 (2015).PubMed 

Google Scholar 
89.Skwark, M. et al. Interacting networks of resistance, virulence and core machinery genes identified by genome-wide epistasis analysis. PLoS Genet. https://doi.org/10.1371/journal.pgen.1006508 (2016).Article 

Google Scholar 
90.Pensar, J. et al. Genome-wide epistasis and co-selection study using mutual information. Nucleic Acids Res. 47, e112–e112 (2019).PubMed 
PubMed Central 

Google Scholar 
91.Puranen, S. et al. SuperDCA for genome-wide epistasis analysis. Microb. Genomics 4, e000184 (2018).
Google Scholar 
92.Whelan, F. J., Rusilowicz, M. & McInerney, J. O. Coinfinder: detecting significant associations and dissociations in pangenomes. Microb. Genomics 6, e000338 (2020).
Google Scholar 
93.Slomka, S. et al. Experimental evolution of bacillus subtilis reveals the evolutionary dynamics of horizontal gene transfer and suggests adaptive and neutral effects. Genetics 216, 543–558 (2020).PubMed 
PubMed Central 

Google Scholar 
94.Maddamsetti, R. & Lenski, R. E. Analysis of bacterial genomes from an evolution experiment with horizontal gene transfer shows that recombination can sometimes overwhelm selection. PLoS Genet. 14, 1–30 (2018).
Google Scholar 
95.Knöppel, A., Lind, P. A., Lustig, U., Näsvall, J. & Andersson, D. I. Minor fitness costs in an experimental model of horizontal gene transfer in bacteria. Mol. Biol. Evol. 31, 1220–1227 (2014).PubMed 

Google Scholar 
96.Collins, R. E. & Higgs, P. G. Testing the infinitely many genes model for the evolution of the bacterial core genome and pangenome. Mol. Biol. Evol. 29, 3413–3425 (2012).PubMed 

Google Scholar 
97.Baumdicker, F., Hess, W. R. & Pfaffelhuber, P. The infinitely many genes model for the distributed genome of bacteria. Genome Biol. Evol. 4, 443–456 (2012).PubMed 
PubMed Central 

Google Scholar 
98.Haegeman, B. & Weitz, J. S. A neutral theory of genome evolution and the frequency distribution of genes. BMC Genomics 13, 196 (2012).PubMed 
PubMed Central 

Google Scholar 
99.Hughes, A. L. Evidence for abundant slightly deleterious polymorphisms in bacterial populations. Genetics 169, 533–538 (2005).PubMed 
PubMed Central 

Google Scholar 
100.Van Passel, M. W. J., Marri, P. R. & Ochman, H. The emergence and fate of horizontally acquired genes in Escherichia coli. PLoS Comput. Biol. 4, e1000059 (2008).PubMed 
PubMed Central 

Google Scholar 
101.Hao, W. & Golding, G. B. The fate of laterally transferred genes: life in the fast lane to adaptation or death. Genome Res. 16, 636–643 (2006).PubMed 
PubMed Central 

Google Scholar 
102.Lerat, E., Daubin, V., Ochman, H. & Moran, N. A. Evolutionary origins of genomic repertoires in bacteria. 3, e130 (2005).103.Lobkovsky, A. E., Wolf, Y. I. & Koonin, E. V. Gene frequency distributions reject a neutral model of genome evolution. Genome Biol. Evol. 5, 233–242 (2013).PubMed 
PubMed Central 

Google Scholar 
104.Sela, I., Wolf, Y. I. & Koonin, E. V. Theory of prokaryotic genome evolution. Proc. Natl Acad. Sci. USA 113, 11399–11407 (2016).PubMed 
PubMed Central 

Google Scholar 
105.Charlesworth, B. Effective population size and patterns of molecular evolution and variation. Nat. Rev. Genet. https://doi.org/10.1038/nrg2526 (2009).Article 
PubMed 

Google Scholar 
106.Cohan, F. M. & Perry, E. B. A systematics for discovering the fundamental units of bacterial diversity. Curr. Biol. 17, 373–386 (2007).
Google Scholar 
107.Domingo-Sananes, M. R. & McInerney, J. O. Selection-based model of prokaryote pangenomes. bioRxiv https://doi.org/10.1101/782573 (2019).Article 

Google Scholar 
108.Azarian, T. et al. Frequency-dependent selection can forecast evolution in Streptococcus pneumoniae. PLoS Biol. 18, e3000878 (2020). The authors provide evidence that NFDS is a pervasive evolutionary force that shapes the accessory genome of S. pneumoniae.PubMed 
PubMed Central 

Google Scholar 
109.Bobay, L. M., Touchon, M. & Rocha, E. P. C. Pervasive domestication of defective prophages by bacteria. Proc. Natl Acad. Sci. USA 111, 12127–12132 (2014). Although prophages can be considered parasitic, the authors show evidence of purifying selection within prophage genes, suggesting that they serve a beneficial purpose within their bacterial hosts.PubMed 
PubMed Central 

Google Scholar 
110.Puigbò, P., Lobkovsky, A. E., Kristensen, D. M., Wolf, Y. I. & Koonin, E. V. Genomes in turmoil: quantification of genome dynamics in prokaryote supergenomes. BMC Med. 12, 1–19 (2014).
Google Scholar 
111.Lynch, M. Streamlining and simplification of microbial genome architecture. Annu.Rev.Microbiol. 60, 327–349 (2006).PubMed 

Google Scholar 
112.Bobay, L. & Ochman, H. Factors driving effective population size and pan-genome evolution in bacteria. BMC Evol. Biol. 18, 15 (2018).
Google Scholar 
113.Brito, I. L. et al. Mobile genes in the human microbiome are structured from global to individual scales. Nature 535, 435–439 (2016).PubMed 
PubMed Central 

Google Scholar 
114.Evans, T. G. Considerations for the use of transcriptomics in identifying the ‘genes that matter’ for environmental adaptation. J. Exp. Biol. 218, 1925–1935 (2015).PubMed 

Google Scholar 
115.Cain, A. K. et al. A decade of advances in transposon-insertion sequencing. Nat. Rev. Genet. 21, 526–540 (2020).PubMed 

Google Scholar 
116.Wu, M. et al. Genetic determinants of in vivo fitness and diet responsiveness in multiple human gut Bacteroides. Science (80-.) 350, aac5992 (2015).
Google Scholar 
117.Poulsen, B. E. et al. Defining the core essential genome of Pseudomonas aeruginosa. Proc. Natl Acad. Sci. USA 116, 10072–10080 (2019).PubMed 
PubMed Central 

Google Scholar 
118.Pál, C., Papp, B. & Lercher, M. J. Adaptive evolution of bacterial metabolic networks by horizontal gene transfer. Nat. Genet. 37, 1372–1375 (2005).PubMed 

Google Scholar 
119.Ansari, A. & Didelot, X. Inference of the properties of the recombination process from whole bacterial genomes. Genetics 196, 253–265 (2014).PubMed 

Google Scholar 
120.Lin, M. & Kussell, E. Inferring bacterial recombination rates from large-scale sequencing datasets. Nat. Methods 16, 199–204 (2019). The authors develop a fast and clever method that uses linkage information to estimate recombination rates and the diversity of the gene pool that has contributed alleles to the sample via HGT.PubMed 

Google Scholar 
121.Marttinen, P. et al. Detection of recombination events in bacterial genomes from large population samples. Nucleic Acids Res. 40, 1–12 (2012).
Google Scholar 
122.Didelot, X. & Wilson, D. J. ClonalFrameML: efficient inference of recombination in whole bacterial genomes. PLoS Comput. Biol. 11, 1–18 (2015).
Google Scholar 
123.Croucher, N. J. et al. Rapid phylogenetic analysis of large samples of recombinant bacterial whole genome sequences using Gubbins. https://doi.org/10.1371/journal.pcbi.1004041 (2015).124.Mostowy, R. et al. Efficient inference of recent and ancestral recombination within bacterial populations. Mol. Biol. Evol. 34, 1167–1182 (2017).PubMed 
PubMed Central 

Google Scholar 
125.Yahara, K., Didelot, X., Ansari, M. A., Sheppard, S. K. & Falush, D. Efficient inference of recombination hot regions in bacterial genomes. Mol. Biol. Evol. 31, 1593–1605 (2014).PubMed 
PubMed Central 

Google Scholar 
126.Daubin, V., Moran, N. A. & Ochman, H. Phylogenetics and the cohesion of bacterial genomes. Science 301, 829–832 (2003).PubMed 

Google Scholar 
127.Daubin, V. & Szollosi, G. Horizontal gene transfer and the tree of life. Cold Spring Harb. Perspect. Biol. https://doi.org/10.1007/978-94-007-2941-4_37 (2016).Article 
PubMed 
PubMed Central 

Google Scholar 
128.Bertelli, C., Tilley, K. E. & Brinkman, F. S. L. Microbial genomic island discovery, visualization and analysis. Brief. Bioinform. 20, 1685–1698 (2019).PubMed 

Google Scholar 
129.Rocha, E. P. C. et al. Comparisons of dN/dS are time dependent for closely related bacterial genomes. J. Theor. Biol. 239, 226–235 (2006).PubMed 

Google Scholar 
130.Kryazhimskiy, S. & Plotkin, J. B. The population genetics of dN/dS. PLoS Genet. https://doi.org/10.1371/journal.pgen.1000304 (2008).Article 
PubMed 
PubMed Central 

Google Scholar 
131.Charlesworth, B. & Charlesworth, D. Elements of Evolutionary Genetics (Roberts and Company Publishers, 2010).132.Castillo-Ramírez, S. et al. The impact of recombination on dN/dS within recently emerged bacterial clones. PLoS Pathog. https://doi.org/10.1371/journal.ppat.1002129 (2011).Article 
PubMed 
PubMed Central 

Google Scholar 
133.David, S. et al. Dynamics and impact of homologous recombination on the evolution of Legionella pneumophila. PLoS Genet. 13, 1–21 (2017).
Google Scholar 
134.Dillon, M., Thakur, S., Almeida, R. & Guttman, D. Recombination of ecologically and evolutionarily significant loci maintains genetic cohesion in the Pseudomonas syringae species complex. Genome Biol. https://doi.org/10.1101/227413 (2019).Article 
PubMed 
PubMed Central 

Google Scholar  More

Results agreed closely with Ref.32 when wind speed-specific drag coefficients were introduced that seemed to be reasonable, as the latter lie slightly under the values published by Ref.17 for a stiffer Canary date palm (Phoenix canariensis). Notwithstanding, the model’s outcome on its own would not be a valid contribution to the field of palm biomechanics and risk assessment. Hence, and from here onwards, the following observations are deemed to be crucial:The researcher32 used data on the mechanical properties of green tissue of senile 80–100-year-old coconut palm stems harvested in 2010 in Fiji and Samoa and which were then published in Ref.34. His model predicted that the critical wind speed for failure of the stem fibres was 82.8 km/h (23 m/s). However, for instance, the cyclone “Val” hit Samoa with wind speeds up to 140 knots (259.28 km/h) in 199135. The real palms from Ref.34 that served for Ref.32 must have withstood, along their life span of 80–100 years, many times wind speeds that exceeded the theoretical “critical wind speed” of 82.8 km/h as predicted by Ref.32. And yet, those palms still stood upright when they were harvested. And no mechanically-damaged tissue in the harvested cocostems was reported in Refs.[32, 34], which suggests that those coconut palms had not suffered any failure of their stem fibres, not even when wind speeds up to 259.28 km/h hit the island they were growing on. This observation adds up to others (e.g.10,19,20,33) that suggest that the engineering approach as used by Ref.32 and used, albeit much simpler, in e.g. the “tree-statics” of Ref.3 may have a limited predictive value.Next: Lowest FI was calculated here at a height of 13.1 m up the stem, whereas32 predicted failure between 6 and 10 m at 23 m/s (while the failure area would move towards the stem base with higher wind speeds). The reason behind this disagreement is simple: the herein employed model uses simple beam theory, which assumes the deformation or curvature of the stem is null. And32, on the other hand, depicted the bending over of the slender palm with a highly-pronounced curvature in the lower half of the stem, which would, hence, cause greater stresses. And as this happens, the upper half of the stem would align itself more with the wind and thus be less deformed, while also the experienced stresses would be lower there (see the figure in Ref.32 p. 126).Furthermore, the present model “fabricated” the rising stresses, by modeling the stem as a hollow wind turbine tower with a changing t/R related to wind speed. The rationale behind this approach is the following: the beam theory neglects that, in palm stems, stresses increase exponentially from the neutral axis and that non-linear deformations and strong curvatures can be experienced by slender palms in high winds, as reported by e.g.23,32. This is thus one of the arguments against the simple beam theory in slender trees and palms, and which has been used in the present model and by e.g.1,3,4,5,16,17,18. One of the premises of that theory is that deformation of the beam should be small. And if, for instance, a slender Mexican Fan Palm bends over, the curvature of the stem could be too pronounced to be faithfully modeled with that theory. And real stresses due to that curvature would hence be higher than predicted. For instance, a company markets commercial pulling test on palms and shows it application on a 22-m-tall and leaning Mexican Fan Palms (La Aduana, Málaga)36. And theirs may thus be an untenable and risky suggestion. A crucial observation can be made here: if regular beam theory (as used in commercial pulling tests and wind load analysis software packages) had been blindly used for the coconut palm simulation (i.e. assuming a stiff and solid beam with no sharp rise in stresses due to the pronounced curvature of the stem and the bending stress increasing linearly from the neutral axis as if the cross-section were both full and isotropic), breaking safety factors would have been greatly overestimated (nearly doubled at 60 m/s) as stresses would have been underestimated. Which could lead to dangerous, and deadly, situations in real-life palm risk assessments.Next: As mentioned in Ref.20 the results of Ref.32 were not validated experimentally. Hence, and even barring structural defects such as cracks or decay, it is a theoretical model that may not allow for accurate safety predictions yet. Furthermore, palm stems were reported to break either at mid-stem or just below the crown37. Whereas32 predicted failure well below mid-stem, extending towards the bottom of the stem at higher wind speeds and so, his results do not agree with real failures as reported by Ref.37. Also, the Red Palm Weevil (Rhynchophorus ferrugineus) has been said to affect the structural stability of palm trunks by excavating tunnels which could lead to their collapse38. The loss of structural integrity due to this tunneling is a type of defect that is not taken into account in Ref.32 either. And Ref.39 suggested several factors that may influence breakage of coconut palms subject to cyclones, and not all are taken into account in the simulation: e.g. the ratio between diameter of the bole and stem height, different mechanical failure modes related to certain hybrids (e.g. the of Malayan Red Dwarf versus Tall Palms), biomechanical degradation due to Phytophthora palmivora and crown characteristics (weight and volume of fronds and crop). The latter also reported fracture of the bole at the root-soil plate level and below ground, while other palms were left leaning after partial uprooting and others had their stems broken at different heights39. This suggests that the predictive value of the approach used by Ref.32 is rather limited and that, even though his is an impressive and highly-valuable contribution, it may not be extrapolable to real-life palm risk assessments yet.Next, the commercial pulling tests and wind load analysis software of Refs.4,5 have been used on palms in Spain. Nevertheless, no clear references pointing towards scientific, peer-reviewed papers of theirs could be found in those publications, relating to scientifically sound data and scientifically contrasted procedures that would support their methods. Which, surprisingly, stands in sharp contrast with their criticisms toward a competitor in business: that the latter would offer (similar) methods without adducing supporting scientific evidence4,5. Also an influential publication asserted with a “generalised tipping curve” that the critical threshold of tilt angle for uprooting would be 2.5° (assertedly based on 400 trees), which seems to be the foundation of their commercial pulling tests3. That strong claim was later questioned as it still seemed unproven and thus hypothetical19. Moreover, the asserted results of3 stand in sharp contrast with those of recent researchers who, after two million measurements on more than 8000 trees, do not assert to have found any critical threshold of tilt angle yet40. Pulling tests have been seriously questioned of late and it is not clear to what extent their predictive value in dicotyledonous trees is reliable or not19,20. Which makes their use for monocotyledonous palms even more questionable as the latter are biomechanically very different from trees. Pulling tests were developed from pulling over dicotyledonous trees, with their corresponding characteristics (regarding material and geometry) of their root system and stem base. However, and on the other hand, the fleshy palm roots tolerate significant bending and twisting before they undergo mechanical failure41. And the roots also sprout from the stembase in a way that is similar to onions. Moreover, the effects of organic exudates from the roots on the aggregation of soil particles, which would thereby create a cement-like soil consistency, was reported too41. And these exudates have seemingly not been reported for dicots yet. And, thus, recommending pulling tests for palms seems to stand for blindly extrapolating hypothetical (as seemingly yet unproven) values for dicotyledons, into the unknown and unexplored realm of palms, and this is could thus be deemed very questionable.Next, destructive experiments were experimentally carried out with the pulling test method of Refs.3,45 on a hollow date palm (Paseo Marítimo, Mataró, Spain) that presented a thin residual wall and a large longitudinal opening42. The palm was pulled with a fixed maximum load of 1.5 kN while performing 20 consecutive measurements with a strain gauge sensor or “elastometer”. The measuring tools (elastometer, inclinometer) had been provided by Refs.3,45. The distance between the two pins of the strain gauge sensor was 2000 mm. The direction of the pull was perpendicular to the opening of the cavity. The first 18 measurements were carried out by placing the sensor aligned with the stem (longitudinally) and in the direction of the pull, conform to the classic pulling test procedure. The highest longitudinal deformation (aligned with the stem) recorded was 0.089 mm (at 2 m height) and 0.045 mm at the height of the cavity (1 m) with a static pulling load of 1.5 kN and conform to the classic pulling test procedure of Ref.3. However, two alternative measurements were made afterwards by placing the sensor first in an angle of 90° (horizontally and bridging the cavity opening) and then in an angle of approximately 30° over the open cavity (Fig. 1), to assess if there could be any shear. Then, the strain gauge sensor recorded the astonishing value of 0.321 mm at the same pulling load (1.5 kN) when placed obliquely (30°) over the open cavity. Which is more than seven-fold the maximum axial strain measured at the same height of the stem and aligned with the pull, which indicates that shear (and not longitudinal strain) was the highest. The two sides of the open cavity seemed to slide over each other. This phenomenon can be visualised by bending a softcover book: the pages slide over eachother. Palm wood is much weaker regarding shear stresses and when this is coupled with extraordinary shear deformation (due e.g. a large, open cavity such as here) failure of the hollow stem can be triggered at lower loads than predicted by the beam theory. And if this extremely hollow stem exhibited high shear deformations under a transverse pull, then logic says that also other structural behaviours (e.g. ovalization, cracking or kinking) could set in.Figure 1The strain gauge sensor was placed obliquely over an open cavity in a date palm, recording high shear deformations. The red arrows show the direction of the strain, producing shear as the two halves of the stem seemed to slide over each other, while a static pull was being applied.Full size imageTo continue: the death of a man crushed by a Canary date palm in 2020 in Barcelona (Spain), triggered the implementation of drilling with the Resistograph of Ref.1 and “oscillation tests” on 2026 date palms with a “reliability of almost a 100%”, carried out by Refs.36,43. Nevertheless, it was stated that drilling cannot predict the residual strength of a structurally damaged trunk44,45. And it was shown that boring into decayed zones would probably augment the speed at which decay spreads46. Micro-drilling would allow fungi to grow out radially due to the microenvironment the narrow channel creates47. And drilling was also questioned by Ref.48. And correct assessments would rely on comparing results with both known standards and decay-free cores taken from the same tree, which would make this method thus highly-invasive49. Hence, (micro)drilling is currently highly questionable. And it was reported that the “oscillation tests” of that company were as follows: the palm is pulled manually with a rope and if the whole stem moves the palm stem is regarded as sound50. This was the only description found of those “oscillation tests” on palms after exhaustive online literature review. That company recommends that their oscillation test be used in all palm risk assessments36. Nevertheless, no clear references can be found in Refs.36,43 that would point to peer-reviewed publications in scientific journals that would validate their claim, so it is not clear if it is scientifically tenable or not. However, their claim could be interesting, if robust and supporting scientific evidence could be adduced and if their procedure were unbiased, documentable and reproducible by a third party.Next, the famous “70%” criterion was adduced by the Municipal Manager of the City Council of Barcelona after the deadly accident with a Canary date palm and the manager was reported to have said that inner decay was considered problematic if at least 70% of the stem diameter was affected by it12. And that the risk of breaking would gain importance if the extent of decay occupied at least 70% of the stem radius13. However, if the palm collapsed with only 25% of its radius affected by decay, should one then not immediately refute the “70%” criterion? Notwithstanding, the applicability of this criterion as suggested for palms by Ref.1, can easily be refuted scientifically too: Firstly, suspicions of falsification were published regarding that highly-influential Visual Tree Assessment (VTA) rule t/R = 0.32 or 70%51. That t/R rule for risk assessments defined the (supposedly) allowed degree of hollowness of a tree trunk and is still being used world-wide, although it is allegedly the result of falsification51. Secondly, that famous “70% criterion” was, moreover, developed for dicotyledonous trees and not for monocotyledonous palms. Thirdly, tangential MOR (tension perpendicular to grain) in coconut wood was found to be as low as 0.233 kN/cm2 whereas longitudinal MOR could reach 5.22 kN/cm252. And shear strength of date palm wood could be as low as 7.14% compared with its longitudinal MOR52. Even longitudinal MOR varied greatly between coconut, oil and date palm52. For oil palm (Elaeis guineensis), the proportion of tensile strength perpendicular to grain to longitudinal MOR was reported to be only 2.08%53. And another researcher found that tangential MOR would be 36.77% approximately of longitudinal MOR in coconut palms32. Therefore, tangential cracking followed by longitudinal splitting in straight hollow stems would thus be triggered earlier in date and oil palms than in coconut palms. Palm wood is highly anisotropic compared to dicotyledonous hardwood trees and, above all, there are great differences both among different palm species and even within the same palm species, depending on e.g. age and growth. And, hence, a fixed t/R rule seems to be an untenable recommendation for palms. And this would even not be applicable in stems with side openings or those with only heart decay30. Or with irregularly-distributed pockets of decay combined with cracks or invaginations. Even if one dared to leave aside the fact that high stresses can be caused by strong curvatures of slender palms in high winds (see e.g.23,32) and those are not taken into account in the VTA t/R rule of 70%. Cross-sectional flattening or ovalization, leading to cracking of thin-walled hollow stems, is neglected by classic beam theory too, while those structural failures depend highly on the MOE of the wood54. And MOE, MOR and density vary greatly according to species, age, et cetera and thus preclude a fixed t/R rule entirely from being useful when the aim is to predict the structural collapse of (especially) palms.Next: Scientists published breaking safety factors and critical wind speeds for severely decayed date palms in a highly-frequented area in a major Spanish city16. Their suggestion that those decayed palms would withstand wind speeds of at least 135 km/h even made headlines55. They did not fully explain the underlying methodology in their online-published report. However, the manual of the acoustic tomograph (Fakopp56) they used, suggests that breaking safety factors and critical wind speeds were calculated by means of simple beam theory, which is based on longitudinal stress and strength, and assumes that wood is a homogeneous material (i.e. isotropy). However, palm wood is highly anisotropic, meaning that shear, delamination, torsional, radial and tangential tensile stresses could cause structural collapse, especially in decayed palm stems, even if the beam theory were applicable. So, their claim that decayed palm stems in a market square would withstand hurricane-type winds, is possibly based on what looks like a methodological error on their behalf. On the other hand, a method for the breaking risk assessment of Canary date palms was offered in a scientific paper through beam theory and a hypothetical static wind load17. But, fortunately, the latter acknowledged that their approach could not be valid for decayed stem areas due to shear and that failure due to progressive fatigue was not considered either. And a highly-cited researcher calculated longitudinal stress in a Mexican fan palm (Washingtonia robusta), seemingly based on the assumption that bending stress would increase linearly from the neutral axis and by means of a static pulling test combined with simple beam theory18. However, the herein offered remarks suggests that the latter’s approach for palms may be a simplification that has room for improvement too. A Spanish researcher published results from static bending tests of planks, sawn from a Canary date palm, to be used in the context of mechanistic risk assessment models57. Unfortunately, the procedure he used precludes his results from being useful in that context, as current models need MOR and MOE values obtained from compressive axial and tangential tensile tests. Nevertheless, in his review he rightly acknowledged the dubious efficiency of published risk assessment models, as they would depend too much on a wide palette of unknown variables (e.g. Cd, MOE, MOR, density) and that decay detecting tools were not efficient, as reference values did not exist (e.g. to calculate strength loss in comparison with sound palm wood)57.Next: Some readers will surely feel tempted to use the model for, and extrapolate the results to, commercial palm risk assessments. But that would clearly be premature, as can be inferred from the following observations: a wind-speed-specific drag coefficient was proposed herein to simulate the coconut palm of Ref.32. But a sturdy stem (in contrast with the studied 25-m-tall and flexible stem) would need to be modeled with a different Cd. And to make Cd transferrable to other palms, non-linear deformation would have to be taken into account. This non-linearity is the result of the slenderness, anisotropy and geometry of the stem, the flexibility of the crown and overall out-of-phase damping. A greater curvature of the stem leads to higher three-dimensional stresses (longitudinal, radial and tangential). And Poisson’s ratios also determine deformation of the fibres32 and this may differ too. And all of this clearly exceeds the capacities and predictive power of simple beam theory. Also, wind drag has commonly been estimated as being proportional to the square of the wind speed. However, it was shown that this estimation may be too high for flexible culms, as at higher wind speeds the drag would be linearly proportional58. And, hence, real loads would be lower than predicted. They also found that the risk of mechanical damage was comparatively lower at higher wind speeds, as the plants’ height was reduced by up to 45%58. It was suggested that coconut palms would resist hurricanes better than dicotyledonous trees because of the same strategy32. On the other hand, common sense and observation suggest that the mass of palms (stem, crown and crop) combined with violent gusts may lead to dynamic loading that far exceeds predictions that take into account static loading only. Two simulation studies did not consider dynamic loading, damping, looping or inertia effects20,32. Nevertheless, the results the first agreed very well with commercial softwares that, assertedly, would include dynamics and natural frequencies20. For instance, it was asserted that “and statics integrated methods that combine static pulling with dynamic wind load assessment (Wessolly 1991; Brudi and van Wassenaer 2002; Detter and Rust 2013)” (sic)59. Related authors also suggested that a natural frequency factor was incorporated in their calculation of the wind load and bending/uprooting moment of their pulling tests3. However, robust scientific evidence that would support their claim was not found and neither did the mathematical simulations find any evidence of dynamics20. Not including the influence of dynamics (e.g. the swinging of slender trees and palms) in a wind load analysis could underestimate real wind-induced loads. Which means that the palm or tree could thus fall down even if it had been assessed as “safe”. Also the weight of crop (e.g. dates or coconuts) could add inertial forces to the swinging and this could be a subject for future research on wind loads in palms20.Further: Mechanical properties (strength, stiffness and density) of green palm tissue are still a relatively unexplored field, although several palm species have been studied17,22,32,52,53,60,61,62. Properties from these publications of other palm species could be introduced in the model to explore their importance relative to other influencing factors such as slenderness, wind speeds, loads, et cetera. But, and even though this procedure has led to good agreement for FI of Ref.32, more research on the applicability of the model should be carried out.Also: The herein used approach is based on a simplified version of the theory of elasticity, which ignores stress concentrations (e.g. around knurls or defects in wood), Inglis’ potential energies, fatigue and crack propagations as described by Ref.63, which can lead to unexpected structural collapse if one relies solely on simple beam theory. The need to explore those ideas was suggested, as understanding their influence could be the key to understanding the relationship between structural failure and wind42. Fortunately, those relatively unexplored ideas were later applied to calculate critical wind speeds for failure in forest trees64. This could thus be an interesting starting point for research on the breaking prediction of palms and trees.Moreover, the beam theory as used by some of the herein mentioned authors is aimed only at predicting conventional bending failure (axial compression stress that exceeds MOR), while low t/R ratios can lead to Brazier buckling or tangential cracking followed by longitudinal splitting in hollow stems30. The formulations were offered to predict the bending moment at which cracking failure would occur in hollow trees, based on t/R, MOE and tangential tensile MOR54. So, and for instance, if one took an oil palm and a coconut palm, both hollow and with an identical t/R and wind loading, the first would crack earlier than the second due to a lower tangential MOR. And as t/R decreased, failure modes would be bending failure, cracking and Brazier buckling respectively, for oil palm. Whereas in coconut palm, and depending on t/R, bending failure would occur earlier than the other two modes due to a comparatively higher tangential MOR/longitudinal MOR proportion. Which is also evidence why fixed t/R rules (e.g. 0.32 or 70% of the radius) and beam theory (e.g. pulling tests) cannot be applicable to palms. Hence, incorporating cracking and buckling predictions in the assessment of decayed and concentrically hollow palm stems, could also be an interesting lead.Next, it was found that the Brazier calculations (based on MOE) agreed with the BS outputs (based on MOR) of the model for all heights along the stem and all wind speeds, when the stem was modeled as untapered, which is interesting. The relationships of varying MOR and MOE along the stem were based on the measured densities of Ref.32, so there seems to be a consistent mathematical relationship between the MOR, MOE and density values32. At first sight, densities of green palm wood could thus be an interesting future research subject. However, and on the other hand, it was reported that no correlation existed between density and mechanical properties in date palm wood52. Which would make future mechanistic modeling thus even more challenging.Next: None of the herein investigated models and criteria fulfill the requirements (i.e. that models should account for cell wall expansion and sclerification as a result of height growth and age) stated by Ref.60. And they could therefore be precluded from being useful for palms, as the latter both grow and age.Researchers also recorded longitudinal tensile stress on the surface of upright growing trunks, whereas compression stress was found at the bent area of leaning trunks in coconut palms due to growth stresses65. They also found compression stress in the outermost portion of the inner cylinder of the coconut stems, which they said was radically different from dicotyledonous and coniferous trees. So, this also questions the application of fixed t/R rules, pulling tests or wind load analysis combined with beam theory on palms, as those methods neglect growth stresses and their biomechanical importance. For instance, if the central cylinder were missing (e.g. due to butt rot caused by Ganoderma zonatum) then the lack of those inner and outer growth stresses and strains should be accounted for.However, now we will suppose, and for the sake of the argument, that we approached the pitfall in which some of the aforementioned companies and researchers have seemingly already fallen. At a first glance, it would be appealing to suggest the following method: consider that commercial software packages for wind load and breakage predictions were successfully simulated20. And that special software packages were also suggested to accurately measure the vertical area of e.g. a palm crown, which would thus allow to perform a wind load estimation that would meet the standards of the commercial software packages investigated20. Suppose a wind speed-specific drag factor be introduced, such as proposed for Canary date palms by Ref.17 or the one found here for coconut palm. And that the formulations for the critical bending moments for tangential cracking from Ref.54 and the ones employed in this study for pure bending failure be incorporated, together with the wood properties as published for several palm species by e.g.17,22,32,52,53,60,61,62. Furthermore, values for peripheral material properties were obtained from the ring that corresponds to the outer third of the radius32,52. And take a non-linear bending stress distribution in the cross-section of the stem, which rises exponentially from the neutral fibre to the peripheral outer ring made of the most dense, stiffest and strongest tissues21. Then, a simplified assumption would be to calculate stresses taking into account only the outer third of the radius (i.e. t/R = 0.33), as if it were a hollow wind turbine tower and as has been done in the present study. And this, to simulate (in an extremely simplified manner) non-linear bending stress and peripheral material properties (note: it should absolutely be stressed here that this is not regarded as a validation of the VTA t/R = 0.32 rule, as the rationale for its use in the model differ from the rationale of Refs.1,15, while the inapplicability of the latter’s claim has been amply evidenced in this study). In this way, theoretical safety factors could then be calculated and compared for bending versus cracking failures of the hollow palm stem, for varying wind speeds and several palm species. This would thus be similar to the widely-cited Statics Integrated Assessment (SIA) and Statics Integrated Methods (SIM) of Refs.3,45, but then for palms and slightly enhanced (as it adopts cracking failure, the varying material properties across and along the stem and a wind speed-specific drag factor). And as less advanced methods (e.g.1,3,16 have already been commercially marketed, a non-scholar could perhaps be tempted to commercialise this model in a software package or use it for their consultancy services too. However, this approach would still suffer from the same limitations as described in Refs.19,20 and in the present study. And it would still be theoretical, as the variables concerning the structural stability of hollow trees and palms may be too diverse to be assessed with current methods19. And the combination of small deviations in the real palms from e.g. the published values for MOR and MOE and theoretical drag factors (and hence predicted wind loads) could result in a global deviation that may invalidate the outcomes (the latter concerns all of the herein investigated methods too)33. Hence, and even though it is not the corresponding author’s idea to wholly negate the usefulness of the herein investigated methods, it is crucial to point out that both their validation and predictive value are seemingly problematic.A crucial rationale for presenting the utterly simplistic model in this paper was the following: supposedly complex models such as e.g.3,4,5,45 or the advanced 3DFE simulation of Ref.32 may obscure that fact that those models can be as tied to the same limitations as the simple model presented herein. And apparently complex equations (or e.g. a high number of citations of the related papers) may deviate the readers from the fact that factual empirical and scientific evidence could still be missing that would validate the models for real-life purposes. Hence, a simple model such as the herein presented one, may serve the purpose of pointing out the flaws and limitations of the seemingly more advanced models, while it even seems capable of simulating internationally-renown commercial software and methods20.So, the time seems to be ripe now to go beyond the classical procedures as trusted upon by the arboricultural community so far (and discussed before).Even in straight, thin and idealised cantilever beams, bending–torsion coupling deformations can arise due to the dissimilar bending stiffnesses when the two planes (horizontal/vertical) of the cross-section are of uneven dimensions (instead of a e.g. a perfectly circular or annular cross-section)66. Palm and tree stems are not always perfectly round due to dissimilar diameters in the horizontal versus the vertical plane (e.g. in cases of reaction wood, open cavities or uneven radial growth due to touching physical obstacles). Hence, simple beam theory (e.g. pulling tests) may thus not account for torsional (and, ultimately, catastrophic) behaviours, even in straight stems. Moreover, if improperly applied, simple beam theory may theoretically predict the strength and stiffness requirements of a structure to be satisfying, while unforeseen collapse may later occur because of the loss of stability (buckling), including intriguing phenomena such as non-linear geometric deformation and wrinkles66. Translated into arboricultural language: the tree or palm that had been assessed as “safe”, suddenly collapses unexpectedly. Therefore the need in this paper to show the arboricultural community that structural collapses, that have been studied for centuries in other fields such as mechanics and engineering, should not be ignored.The risk of buckling of a Mexican Fan Palm (Washingtonia robusta), assessed by the corresponding author in 2003 in the Atocha train station (Madrid, Spain), gave birth to a proposal to assess the risk of Euler buckling while carrying out wind load estimations in order to optimise artificial supports (e.g. cabling of the palm to nearby structures)28. Prior to the assessment of the last standing palm, several other slender Mexican fan palms in that train station had already collapsed, even though the interior of this giant greenhouse is free of wind loads (Fig. 2). The photograph is a testimony to a rather neglected fact in commercial arboricultural methods: structural collapse in absence of wind loading and pure post-buckling failure. In this case it was hypothesized that these palms had initially become elastically unstable, by exceeding their critical stem height and weight. It was hypothesized, too, that this had been caused by their unlimited growth towards the glass ceiling searching for light, the absence of external loading stimuli such as wind (the lack of which would have made the palms not to invest in stiffer and denser wood) and optimum growing conditions (permanent moisture and warmth). The weight of the crown, small horizontal displacements, watering from the ceiling (i.e. fog to keep the atmosphere moist) and resulting gravity forces would then have further influenced the failure process, leading to final collapse. This example illustrates how plants can adapt to their environment and that biomechanical failure can be possible in total absence of wind loading. In large-wave Euler buckling, the column curves and deviates laterally to escape from compressive loads (such as e.g. self-weight) before axial stresses surpass axial MOR. The column becomes elastically unstable and buckles under its own weight. The critical weight divided by self-weight gives the safety factor and only when this safety factor is higher than unity can columns, or plants, bear additional loads such as wind, snow or ice. The critical buckling height or weight is a function of stem height, diameter, tapering, MOE, density of the wood and loading conditions27. The latter also showed that buckling safety can be overestimated if the stem is improperly assumed to be untapered, cylindrical, free of imperfections and isotropic27. They also offered an overview of why predictions of structural collapse may easily differ from real-life situations27. Moreover, the bifurcation point is the sudden jumping process of a beam from a straight-line to a bent shape, causing instability or buckling66. Pre-buckling analysis has proven to be rather straightforward for a simple pole, while the post-buckling process that describes the finite deformation of the structure (which may lead to its collapse after damage and faults accumulate to a certain value) requires a large set of numerical solutions67. Strong geometric nonlinearities and large displacements of the post-buckling behavior of a slender rod were studied, leading to a quantitative calculation of the post-buckling deflections of a hollow oil sucker rod67. Translated into the world of palm biomechanics, it means that: while pre-buckling of the stem would already be a daunting task due to the varying taper and MOE, predicting its post-buckling behaviour and final collapse (including structural faults such as e.g. cracks or pockets of rot) seems to be out of reach, as palm stems are not human-made structures. And yet, it seems reasonable not to ignore this type of structural behaviour in future palm risk assessments. It was acknowledged too that Brazier buckling played a crucial role in the local instability of plant stems66. And this was also a reason to include Brazier buckling of a hollow wind turbine tower to simulate breaking safeties of the coconut stem in the present paper.Figure 2The slender Mexican fan palm anchored to the ceiling of the Atocha train station, Madrid. The cabling configuration was installed to minimise damage in case of post-buckling collapse. The other palms had collapsed before, even though there are no events of wind inside this giant greenhouse.Full size imageNow, and as a second part of this “Discussion” section, the following reasonings elucidated from literature overview, visual observation and intellectual reasoning, are presented to postulate ideas that may serve to show the way towards a future unified theory on palm risk assessment.First: The biomechanical structure of palms seems to have evolved towards highly-efficient energy dissipation and viscoelastic damping capacities under strong and dynamic wind loading. To achieve this, a triple-helical mesh of tough (high tensile strength) fibrovascular bundles is embedded in a soft parenchymatous foam, which both contribute to damping and energy dissipation32,68. The fibrovascular bundles run along the stem in a screw-like fashion and across the stem in a radial zigzag pattern (this also sets palm wood also apart from dicotyledonous wood, as in the latter the fibres are stiffly glued together and, most importantly, axially aligned). It was asserted that this screw-like pattern can hold the bending stem together under high wind loading as it lends the stem a higher stiffness and strength when the fibrovascular bundle orientations varied between 0° and 9°32. This pattern was also suggested to minimise longitudinal splitting and thus enhance the mechanical efficiency of the stem32. This structure was an inspiration for spirally-laminated hollow veneer-based composite poles32. Also high microfibril angles across the fibre cap would result in a high extensibility of the stiffening tissue, which would enable palms to cope with considerable deformations under wind loads in Mexican fan palms69. Large deformations in bending and torsion under wind loads of the petioles were said to combine efficiently with water and nutrition conduction, due to the optimized connection of their vascular bundles to the leaf traces68, which allows to suppose that also the crown is optimized regarding damping and energy dissipation brought about by dynamic winds. And the contribution of both parenchymatous and vascular tissue of palms to energy dissipation, dynamic response and flexibility, and thereby improving impact resistance, was described too70.Second: Palm wood is highly sensitive to shear, delamination and splitting in comparison with dicotyledons. For instance, when samples were taken by Ref.17 to perform longitudinal compression and tension tests, then this irregular structure of the palm tissues unwillingly led to longitudinal fractures, sliding and shear in the samples, and thus seriously limiting experimental data on axial MOR. And thick disks of coconut wood were manually torn apart, while the delamination followed the helical pattern of the fibrovascular bundles that tangentially deviated across the disc diameter32. Hence, it is thus not unreasonable to suppose that this sensitivity to delamination and shear may lead to the stem’s structural collapse, especially when this helical path of bundles is interrupted by a mechanical defect (e.g. pockets of rot, irregular decay, cracks or tunneling by Red Palm Weevil). This reasoning aligns with another researcher’s too, who likened coconut and oil palm stems to a composite material made of a matrix and reinforced elements and found that shear and tension perpendicular to grain greatly govern the bending behaviour and structural stability of the stem52. The aforementioned “spirally-laminated hollow veneer-based composite poles” suggested by Ref.32 may be very stiff and strong when undamaged (i.e. if this helical pathway of fibrovascular bundles is not interrupted by a mechanical defect and thus a completely defect-free beam). But, an interruption along this path may trigger delamination and splitting along the “veneer”. Crack propagation and splitting could thus follow the helical path of the fibrovascular bundles. And predictions based solely on simple beam theory and axial stress and strain would then be less than acceptably reliable. Observations and experiments that seem to support this hypothesis are e.g. the aforementioned Canary date palm that crushed a man in Barcelona, as a small inner crack was said to have triggered the sturdy stem’s collapse with a breeze of only 38.2 km/h10. Also pulling test experiments carried out in 2004 by the corresponding author showed that the mechanically damaged palm stems under artificial loading started splitting first, leading to full collapse afterwards42. Those experiments (partially published in 200542) had been kindly supported by Josep Selga S.L., the City Councils of Terrassa and Mataró and the Asociación Española de Arboricultura, while the instrumentation had been kindly provided too (Picus tomograph: L. Göcke Argus Electronics; Pulling tests: Brudi and Partner Tree consult and Dr. Ing. L. Wessolly; Resistograph F300, IML: the City Council Terrassa). The aim was to assess whether the pulling tests of Refs.3,45 could be adapted to palms or not and if experimental data for MOE could be obtained from standing palms. Acoustic tomography (Picus tomograph) and microdrilling (Resistograph F300) had also been carried out on several damaged palms, but had not facilitated any reliable breakage prediction either (unpublished results). An example is shown in Fig. 3 where a desert fan palm (Washingtonia filifera) collapsed under a static pull, after slanted longitudinal splitting and delamination was initiated at the border of the open cavity (upwards and downwards)42. Also Fig. 4 shows how delamination (triggered at the height of the open cavity under a static pull) led to total collapse of a date palm stem. No primary axial compression failure was observed macroscopically42. And a hollow date palm exhibited extremely high shear values in comparison with axial deformation at the height of a large, open cavity (Fig. 1)42. Moreover, it is not unreasonable to suppose that if strong, cyclic and repetitive dynamic wind loading had beaten these three palms (instead of a static pull), the risk of structural failure could have been heightened by progressive fatigue of the wood around the structural defects (and thus earlier crack formations/propagations and at lower loads than with the static pull).Figure 3When a decayed desert fan palm stem was statically pulled, collapse was initiated by splitting of the hollow stem. Cracks first appeared above and below the open cavity and initiated at its borders (red arrows) and total collapse only ensued after large longitudinal splitting and delamination.Full size imageFigure 4When a decayed date palm stem was statically pulled, splitting was initiated at the open cavity and total collapse ensued by delamination.Full size imageThird: Highly deformable and soft but elastic materials can exhibit types of structural deformation under mechanical loads that are unlike those commonly observed in elastic structures that behave linearly71. Kinking at the inner side of soft, elastic cylinders was observed after the cylinders had become elastically unstable due to Euler buckling71. The extreme localization of curvature at the compressed inner (not outer) side exceeded a critical value leading to a sharp fold. When the cylinder was kept under a bending load for several minutes, irreversible defects appeared at the location of the inner kink which, in subsequent loading cycles, progressively lowered the cylinder’s structural stability under the same amount of load71. Translated into palm stems, and assuming they are highly deformable, soft and elastic, this means that inner kinks and defects could appear and lead to structural collapse due to fatigue and cyclic loading beyond the critical curvature. Brazier buckling was also observed in soft, elastic and hollow cylinders and the occurrence of either kinking and/or ovalization was found to be dependent on the ratio between the diameter and the wall thickness71. When one envisages palm stems as has been done in the present paper (a viscoelastic cylinder), then the kinking and ovalization of the cylinder (here: the palm stem), after becoming elastically unstable, could thus lead to abrupt structural collapse while not obeying simple beam theory. Calculation of the critical curvature at which buckling sets in was said to be rather straightforward, but the posterior evolution of the kink or defect would need detailed non-linear theory71. Hence, the modeling of elastic pre-buckling (i.e. prior to these aforementioned structural failures) seems to be more within reach for palm stems than post-buckling collapses. No experiments have been performed on palms yet to either confirm or refute these extrapolated suggestions, but the latter are possibly worth considering in future research or risk assessments.Fourth: Developing a mechanical model seems currently out of reach as strength and stiffness (and thus damping) seem to evolve over time in the palm stem as a function of the location of the vascular bundles within the trunk, age (and ensueing additional cell wall layers and (secondary) growth within the trunk) and growth conditions52. Also the lack of a statistical correlation between MOR and MOE and wood density in date palms is, inexplicably, contrary to other investigated palm species, which also obstructs the path towards reliable mechanistic modelization52.Fifth: It was stated that “Reliable prediction of delamination growth is still proving to be problematic” in human-made wood products, whereas simply localising starting points for delamination would possibly be more within reach72. From which it can thus be inferred, that reliable predictions of delamination-triggered collapses of Nature-made palm and tree trunks seem currently to be out of reach. But that would still be no reason to neglect this type of structural failure).Sixth: The existence of silica in palms was mentioned by Ref.52 (p. 158) and studied by e.g.73,74. Researchers concluded from a literature review that the mechanical properties of palms could be enchanced by silica73. And the role of silica in plants was described as: “Biomineralization is a naturally occurring process by which living organisms form skeletons from inorganic minerals such as silica and calcium”75. The latter also found flexural rigidity in rice plant leaves to increase with increasing silica content. It has been suggested by practitioners and arborists in Spain that silica and biomineralization would make the palm stem stiffer and stronger around structurally defective areas, as an alleged reaction to strength loss percieved by the palm itself (i.e. a substitute for compensation or thigmomorphogenesis as studied in dicotyledons), but no scientific findings were found that would support their suggestion.Seventh: Local mechanical performance (i.e. damping and the diminution of stress discontinuities) of a Mexican fan palm stem could be controlled by the plant itself up to a certain point by adaptation69. Which would further complicate the mechanical modeling of structural stability versus (wind) loads.Eighth: The cracking formulation of Ref.54 should unfortunately be precluded from being useful in hollow palms, as their formulation assumes that the fibres are aligned along the tree axis, while palms present a mesh of triple-helical fibrovascular bundles in a screw-like pattern along and across the palm stem.Ninth: Based on visual observation, young and still flexible and soft Mexican fan and windmill (Trachycarpus fortunei) palm stems seem to exhibit a viscoelastic behaviour when manually pushed and pulled. Their moving out-of-phase with the pulls can be felt by hand and feels like a structure made of foam, but with a certain resilience. Their behaviour resembles neither that of a steel spring nor that of foam or a stiff and non-deformable beam. And this in contrast with e.g. flexible dicotyledonous saplings and tree branches that almost behave like springs or lashes when laterally loaded and released by hand. Also visual observation of the damped manner in which older, taller and stiffer Mexican fan, date and windmill (Trachycarpus fortunei) palm stems move out-of-phase in strong winds seems to confirm this. And in several palm species the woven mesh of leaf sheath palm fibres attached to the stem also exhibits a damping and viscoelastic behaviour when manually manipulated. In windmill palm for instance, the stem is wrapped in a burlap-like mesh of brown and coarse leaf sheath fibre, clasped around the trunk. Manual manipulation of that mesh suggests that friction among the fibres could contribute to damping of leaf and stem movements. A review of published findings on damping and energy dissipation in palms seems to confirm these visual obervations too (see32,68,69,70). A viscoelastic structure exhibits a non-linear response to the strain rate, in which cyclic stress is out-of-phase with strain, as some of the stored energy is recovered upon removal of the load, while the remaining energy is dissipated as heat. The modulus is represented by a complex quantity: on the one hand the stiffness is defined by elastic behaviour and, on the other hand, the energy dissipative ability of the material is defined by the material’s viscous behaviour. Hence, one could thus hypothesise that the palm stem could be neither an elastic nor a viscous structure, but a combination of both.So now, the aforementioned observations lead us to the following:The herein postulated model envisages the palm stem as a viscoelastic and hollow cylinder prone to Euler and Brazier buckling and ovalization and kinking. This hypothetical model could graphically be imagined as a hollow foam pool noodle with a triple-helicoidal embedded mesh of tough (a high tensile strength) fibre bundles. Both the foam of the pool noodle and the mesh of fibres contribute to the damping while the latter also adds flexural stiffness under bending. The cylinder exhibites a non-linear response to the strain rate, in which cyclic stress is out-of-phase with strain, which makes the whole structure viscoelastic. This envisaging was the main reason why Eq. (10) for Brazier buckling, with a constant t/R for all wind speeds, was experimentally applied to simulate FI of the cocostem of Ref.32.However, it would also be prone to delamination, splitting and shear as the bundles are glued together with “foam” along their screw-like path. The momentum the cylinder should withstand should be a result of dynamic wind loads, mass and inertia that cause a non-linear deformation and pronounced curvature of the cylinder (non-linear due to the varying material properties along the (tapered) stem and structural damping). Strains in the stem would then not be linearly proportional to the load, by which Hooke’s law (ut tensio sic vis) would not be not applicable. And stress would rise non-linearly along the stem radius from the core to the periphery. Progressive fatigue of the wood, or at structural defects (e.g. crack initiations and progressive propagation due to repeated dynamic wind loading), should be taken into account. This model would now possibly align quite well with the scientific findings cited in this paper.Nevertheless, a simple mind experiment can reveal the additional baffling challenges found in real palms: imagine a date palm trunk that has been severely tunneled by Red Palm Weevil and/or pockets of rot: the structure resembles a piece of Gruyère cheese and allows remaining bundles of sound strands to be torn off by hand, as the stiff vascular bundles are just lightly glued together by means of a foamy parenchymatous tissue. The remaining bundles and volumes of sound wood, bordering the void and decayed spaces, could then resemble irregularly shaped columns. Now, imagine the loading of this disk of “Gruyère cheese” due to a bending moment: an infinite variety of kinds of structural failure would take place within the remaining “columns”: buckling, sliding, shear, sideways kinking of the fibres, torsion, crack propagations along a triple-helical path, stress concentrations, et cetera. And as the smart reader will surely agree to, this three-dimensional failure process is totally impossible to depict, or assess, by means of drilling, tomography or simple beam theory. Doubtful readers can have a look at the Figs. 10 and 11 in Ref.17 and imagine that the wood blocks in those figures were the remaining “columns” of our imagined trunk. And, as it can be seen in those figures, the blocks structurally failed due to shear, even under pure axial compression and tension17. And now let us add the following: looping movements of a tall palm in winds has already been recorded18. These looping and circular motions of the stem, inevitably, cause a rotative loading of the cross-section of that same stem. This rotative motion thus causes compression stress (and tension stress on the opposite side) at the periphery and in a circular motion, Real wind loading of palms is thus very dissimilar to the unidirectional loadings (assumed or performed) by e.g.1,3,16,17,18,32,36. And let us add too, that shearing behaviours can be caused due to structural defects (e.g. see Fig. 1), and couple this with the rotative motion and possible progressive fatigue processes in the root system and stem. Now the abovementioned reasonings leave us with a mind-boggling panorama of infinite variables, which seemingly precludes all herein investigated methods from being reliable. However, and from a constructive point of view, these postulated ideas are possibly the best starting point for the development of a future risk assessment method. And the herein offered observations can be used by arboricultural professionals to enhance their tree and palm risk assessment consultancy reports.This is only a partial theory, which need not cancel out others per sé, but may overlap others so as to reach a more acceptable degree of predictive accuracy. This may be a step toward a more complete, fully-unified and more reliable theory that would enable us to make predictions that agree with observations to an acceptable degree of accuracy. Constructing a complete theory from scratch looks excruciatingly difficult now, so perhaps the way forward would be to overlap existing partial theories. Partial theories describe a limited variety of events while leaving others aside. Current partial theories in arboriculture do not seem to be valid on their own19,20,33. Examples are theories that neglect common mechanical behaviours of the wooden body33, simple beam theory and dubious t/R criteria for palm risk assessments. Or predictions of uprooting and breakage that are based on a static wind load analysis, if the latter does not take into account the influence of slenderness, dynamics, mass and inertia in slender and top-loaded (due to e.g. a lion-tailed crown or heavy crop) palms and trees20. A complete theory would thus contain a number of parameters which values, in real-life, cannot be predicted yet and such values may have to be chosen to fit in through experiment. A very appealing goal would be now to overcome this mind-boggling and infinite combination of behaviours and (structural and material) properties, and distill it all into one simple and generic law/model, as was elegantly done for buckling by Ref.24.Researchers have taken sound stems as a starting point (e.g.17,18,32. But, perhaps structurally-damaged trunks should be the place from where to start, as the latter are generally the aim and goal of risk assessments. Future methods could thus perhaps focus on deformations of the stem under circular (wind or artificial) loading, while three-dimensional mechanical behaviours and failures can reasonably be expected within a damaged stem. And also three-dimensional material properties should be taken into account: i.e. MOR and MOE in all anatomical directions. But, as taking those values from published tables would not be feasible (due to the high variability of those properties), different methods from the ones used by e.g.1,3,5,16,17,18,32,36 should perhaps be devised. For instance, a preliminary investigation was carried out on forced vibrations, and resulting resonance frequency values, for a Mexican fan palm, in the light of the identification of trunk decay and its level of severity76. And this could perhaps open up new leads for research. Vibration analysis could monitor repetitive motion signals, to detect abnormal vibration patterns and levels, which could allow the assessment of the overall structural condition of the trunk. But then one would still be left wondering whether that approach would reliably assess e.g. the risk of delamination and crack propagation, or ovalization and kinking.Nevertheless, it is now clear that if we stay within the limits of the theories that are the basis of methods such as e.g. the tree-statics of3, t/R rules used by Ref.1,15 or the ill-fated pulling tests as reported by Ref.77, then our mind will possibly not be able to devise the path of evolution. More

1.Kaneda, T. PRB’s 2016 World Population Data Sheet. http://www.prb.org/Publications/Datasheets/2016/2016-world-population-data-sheet.aspx (2016).2.Bedford, J. et al. A new twenty-first century science for effective epidemic response. Nature 575(7781), 130–136 (2019).CAS 
Article 
ADS 

Google Scholar 
3.Neiderud, C.-J. How urbanization affects the epidemiology of emerging infectious diseases. Infect. Ecol. Epidemiol. 5(1), 27060 (2015).PubMed 

Google Scholar 
4.Dye, C. Health and urban living. Science 319(5864), 766–769 (2008).CAS 
Article 
ADS 

Google Scholar 
5.Fang, C. & Yu, D. Urban agglomeration: An evolving concept of an emerging phenomenon. Landsc. Urban Plan. 162, 126–136 (2017).Article 

Google Scholar 
6.Lee, V. J. et al. Epidemic preparedness in urban settings: New challenges and opportunities. Lancet. Infect. Dis 20(5), 527–529 (2020).CAS 
Article 

Google Scholar 
7.Mollalo, A., Vahedi, B. & Rivera, K. M. GIS-based spatial modeling of COVID-19 incidence rate in the continental United States. Sci. Tot. Environ. 728, 138884 (2020).CAS 
Article 

Google Scholar 
8.Small, C., MacDonald, A. J., & Sousa, D. Spatial network connectivity of population and development in the USA; Implications for disease transmission. Preprint at https://arxiv.org/abs/2004.14237v1 (2020).9.Wang, Z. et al. Quantifying uncertainties in nighttime light retrievals from Suomi-NPP and NOAA-20 VIIRS day/night band data. Remote Sens. Environ. 263, 112557 (2021).Article 
ADS 

Google Scholar 
10.Liu, Q. et al. Spatiotemporal patterns of COVID-19 impact on human activities and environment in Mainland China using nighttime light and air quality data. Remote Sens. 12(10), 1576 (2020).Article 
ADS 

Google Scholar 
11.Elvidge, C., Ghosh, T., Hsu, F.-C., Zhizhin, M. & Bazilian, M. The Dimming of lights in China during the COVID-19 pandemic. Remote Sens. 12(17), 2851 (2020).Article 
ADS 

Google Scholar 
12.Venter, Z.S., Barton, D.N., Gundersen, V., Figari, H., & Nowell, M. Urban nature in a time of crisis: Recreational use of green space increases during the COVID-19 outbreak in Oslo, Norway. Preprint at https://doi.org/10.31235/osf.io/kbdum (2020).13.Chauhan, A. & Singh, R. P. Decline in PM2.5 concentrations over major cities around the world associated with COVID-19. Environ. Res. 187, 109634 (2020).CAS 
Article 

Google Scholar 
14.Kasturi, D. K., Kamarul Zaman, N. A. F., Kaskaoutis, D. G. & Latif, M. T. COVID-19’s impact on the atmospheric environment in the Southeast Asia region. Sci. Tot. Environ. 736, 139658 (2020).Article 

Google Scholar 
15.Forman, R.T.T. Land Mosaics: The Ecology of Landscapes and Regions. (Cambridge University Press, 1995)16.Turner, M. G., Gardner, R. H. & O’Neill, R. V. Landscape Ecology in Theory and Practice (Springer, 2001).
Google Scholar 
17.Wiens, J. A., Schooley, R. L. & Weeks, R. D. Patchy landscapes and animal movements: Do beetles percolate?. Oikos 78(2), 257–264 (1997).Article 

Google Scholar 
18.Dolman, P. M., Hinsley, S. A., Bellamy, P. E. & Watts, K. Woodland birds in patchy landscapes: the evidence base for strategic networks. Ibis 149, 146–160 (2007).Article 

Google Scholar 
19.Laiolo, P. & Tella, J. L. Landscape bioacoustics allow detection of the effects of habitat patchiness on population structure. Ecology 87(5), 1203–1214 (2006).Article 

Google Scholar 
20.Jia, Y., Tang, L., Xu, M. & Yang, X. Landscape pattern indices for evaluating urban spatial morphology—A case study of Chinese cities. Ecol. Ind. 99, 27–37 (2019).Article 

Google Scholar 
21.Bosch, M. & Chenal, J. Spatiotemporal patterns of urbanization in three Swiss urban agglomerations: Insights from landscape metrics, growth modes and fractal analysis. Preprint https://doi.org/10.1101/645549 (2019).Article 

Google Scholar 
22.Wang, H., Huang, Y., Wang, D. & Chen, H. Effects of urban built-up patches on native plants in subtropical landscapes with ecological thresholds—A case study of Chongqing city. Ecol. Indic. 108, 105751 (2020).Article 

Google Scholar 
23.McGarigal, K., & Cushman, S.A. The gradient concept of landscape structure. in Issues and Perspectives in Landscape Ecology (eds. Wiens, J., & Moss, M.) 112–119 (Cambridge University Press, 2005).24.Cushman, S. A., Gutzweiler, K., Evans, J., McGarigal, K. The gradient paradigm: a conceptual and analytical framework for landscape ecology. in Spatial Complexity, Informatics, and Wildlife Conservation (eds. Cushman, S. A. & Huettmann, F.) 83–108. (Springer, 2010).25.McGarigal, K., Tagil, S. & Cushman, S. A. Surface metrics: An alternative to patch metrics for the quantification of landscape structure. Landsc. Ecol. 24(3), 433–450 (2009).Article 

Google Scholar 
26.Fan, C., Myint, S. W., Rey, S. J. & Li, W. Time series evaluation of landscape dynamics using annual Landsat imagery and spatial statistical modeling: Evidence from the Phoenix metropolitan region. Int. J. Appl. Earth Obs. Geoinf. 58, 12–25 (2017).Article 
ADS 

Google Scholar 
27.Kowe, P., Mutanga, O., Odindi, J. & Dube, T. A quantitative framework for analysing long term spatial clustering and vegetation fragmentation in an urban landscape using multi-temporal landsat data. Int. J. Appl. Earth Obs. Geoinf. 88, 102057 (2020).Article 

Google Scholar 
28.Wang, J., Kuffer, M. & Pfeffer, K. The role of spatial heterogeneity in detecting urban slums. Comput. Environ. Urban Syst. 73, 95–107 (2019).Article 

Google Scholar 
29.Wu, D. Q. et al. Multi-scale identification of urban landscape structure based on two-dimensional wavelet analysis: The case of metropolitan Beijing, China. Ecol. Complex. 43, 100832 (2020).Article 

Google Scholar 
30.Rocchini, D. et al. Fourier transforms for detecting multitemporal landscape fragmentation by remote sensing. Int. J. Remote Sens. 34(24), 8907–8916 (2013).Article 

Google Scholar 
31.QGIS Development Team. QGIS Version 3.16.10. Geographic Information System. Open-Source Geospatial Foundation Project. https://www.qgis.org/en/site/ (2019).32.Bennett, M. M. & Smith, L. C. Advances in using multitemporal night-time lights satellite imagery to detect, estimate, and monitor socioeconomic dynamics. Remote Sens. Environ. 192, 176–197 (2017).Article 
ADS 

Google Scholar 
33.Zhuo, L. et al. Modelling the population density of China at the pixel level based on DMSP/OLS non-radiance-calibrated night-time light images. Int. J. Remote Sens. 30(4), 1003–1018 (2009).Article 

Google Scholar 
34.Zhang, Q. & Seto, K. C. Mapping urbanization dynamics at regional and global scales using multi-temporal DMSP/OLS nighttime light data. Remote Sens. Environ. 115(9), 2320–2329 (2011).CAS 
Article 
ADS 

Google Scholar 
35.Yang, B. et al. A spatio-temporal method for crime prediction using historical crime data and transitional zones identified from nightlight imagery. Int. J. Geogr. Inf. Sci. 34(9), 1740–1764 (2020).Article 

Google Scholar 
36.Chen, X. Nighttime lights and population migration: Revisiting classic demographic perspectives with an analysis of recent European data. Remote Sens. 12(1), 169 (2020).Article 

Google Scholar 
37.Li, X., Li, D., Xu, H. & Wu, C. Intercalibration between DMSP/OLS and VIIRS night-time light images to evaluate city light dynamics of Syria’s major human settlement during Syrian Civil War. Int. J. Remote Sens. 38(21), 5934–5951 (2017).Article 

Google Scholar 
38.Duan, X., Hu, Q., Zhao, P., Wang, S. & Ai, M. An approach of identifying and extracting urban commercial areas using the nighttime lights satellite imagery. Remote Sens. 12(6), 1029 (2020).Article 
ADS 

Google Scholar 
39.Krikigianni, E., Tsiakos, C. & Chalkias, C. Estimating the relationship between touristic activities and night light emissions. Eur. J. Remote Sens. 52(sup1), 233–246 (2019).Article 

Google Scholar 
40.Israeli Ministry of Health. Online Geodatabase on COVID-19 Exposures (In Hebrew). https://imoh.maps.arcgis.com/apps/webappviewer/index.html?id=20ded58639ff4d47a2e2e36af464c36e&locale=he&/41.Sharav, N. et al. Gush Dan Metro. The Economic, Social and Urban Impacts of the Gush Dan Metro Network Executive Report. Ministry of Finance, Ministry of Transport and Road Safety, National Economic Council Prime Minister’s Office, NATA, Ayalon Highways, July 2020. (in Hebrew). https://www.gov.il/BlobFolder/reports/metro_goshdan_jul_2020/he/Metro_Executive_Summary_14_july_2020.pdf42.MathWorks 2019. MATLAB ver. 2019b Computer Program. https://www.mathworks.com/. (The MathWorks Inc., 2019). 43.Moellering, H. & Tobler, W. Geographical variances. Geogr. Anal. 4(1), 34–50 (1972).Article 

Google Scholar 
44.Wu, J., Jelinski, D., Luck, M. & Tueller, P. T. Multiscale analysis of landscape heterogeneity: Scale variance and pattern metrics. Ann. GIS 6(1), 6–19 (2000).Article 

Google Scholar 
45.Shen, W., Darrel Jenerette, G., Wu, J. & Gardner, R. H. Evaluating empirical scaling relations of pattern metrics with simulated landscapes. Ecography 27(4), 459–469 (2004).Article 

Google Scholar 
46.Wu, J. Effects of changing scale on landscape pattern analysis: Scaling relations. Landsc. Ecol. 19(2), 125–138 (2004).Article 

Google Scholar 
47.Mahadevan, A. & Campbell, J. W. Biogeochemical patchiness at the sea surface. Geophys. Res. Lett. 29(19), 32-1-32–4 (2002).Article 
ADS 

Google Scholar 
48.Yamamoto, D. Scales of regional income disparities in the USA, 1955 2003. J. Econ. Geogr. 8(1), 79–103 (2007).Article 

Google Scholar 
49.USGS Earth Explorer. https://earthexplorer.usgs.gov.50.Román, M. O. et al. NASA’s black marble nighttime lights product suite. Remote Sens. Environ. 210, 113–143 (2018).Article 
ADS 

Google Scholar 
51.Google Earth Engine Data Catalog, Earth Observation Group, Payne Institute for Public Policy, Colorado School of Mines. https://developers.google.com/earth-engine/datasets/catalog/NOAA_VIIRS_DNB_MONTHLY_V1_VCMCFG. More