Ecology

To our knowledge, this study provides the first recorded information on the active swimming of Japanese eels and on their transport by currents in the open ocean. Specifically, the strong flow of the KC largely dominated the movements of the eels and transported them northeastward while they swam mainly southward, and active swimming contributed a little to their travel trajectories. In contrast, the swimming of eels made a relatively higher contribution to their travel trajectories in the TS area.Our in situ estimates of the mean swimming speeds of Japanese eels (26–41 cm/s) were similar or slightly lower than those of European eels. In the acoustic tracking experiment of European eels considering environmental current vectors, their swimming speeds were 35–58 cm/s in the coastal midwater26. In a laboratory experiment using stamina tunnels with stable temperatures, the optimal swimming speeds of European eels were estimated to be 61–68 cm/s (0.74–1.02 BL/s)56, which were higher than the in situ estimates. The minimum swimming speed of European eels is considered to be 40 cm/s if they will arrive at their spawning area in the Sargasso Sea (distance of 5500 km) in 6 months, and their optimal swimming speeds were sufficient to migrate over the long distance in time for the near-spawning period after escape from their growth habitats56. However, field studies using PSAT tagging also reported that in situ migration speeds (including transport by currents) were less than the optimal swimming speeds and suggested that some European eels could reach their spawning area within the near-spawning periods and that others only arrive in time for the following spawning season19.Our estimated effective swimming speed of Japanese eels, all day and all night over the tracking periods, ranged from 3 to 30 cm/s with individual variations. These estimates were consistent with the swimming speeds (excluding transport by currents) of 2.2–15.1 km/day (2–18 cm/s) estimated in the PSAT study of Japanese eels14. Silver-phase Japanese eels start migrating from their coastal growth habitats in Japan primarily in October to December57, 58, and spawning near the West Mariana Ridge occurs in April to August33, 35. Numerical models assuming that migrating eels use true navigation (readjusted compass) or a constant compass heading (fixed compass from the departure place to the spawning site) indicate that the minimal swimming speed required to arrive at the spawning area within 8 months is 10–12 cm/s37. Our estimated effective swimming speeds of five out of ten eels during the day and eight out of ten eels during the night were similar or higher than these minimal speeds. The low effective swimming speeds frequently observed during the day might be due to the relatively low values observed in the swimming speed at 10 min intervals and the swimming directions often varying during the day. When eels swim with stable orientation, as observed in three of the eels (WE2999_TS, WE3001_TS, and WE3002_TS) during the night, the effective swimming speeds exceeded 25 cm/s. If such a stable orientation is maintained and compensate the low speeds during the day, the eels that leave during autumn and winter will be able to arrive at the spawning area during the next spring to summer.It should also be noted that the swimming speeds in body length per second were significantly higher in shallow water during the night than in deep water during the day. In the open ocean, anguillid species exhibit DVMs during oceanic migration, swimming at depth during the day and in the shallows during the night9,10,11,12,13,14,15,16,17,18,19,20,21,22. These DVMs are likely related to the possible avoidance from visual predators under light conditions19 or maturation control59. Essentially, through the DVMs, the eels encounter low temperatures ( 20 °C) during the same day. Generally, the swimming speeds of fishes are restricted by the ambient water temperature60, and the water temperature encountered through DVMs might influence the horizontal-swimming speeds of Japanese eels.Other factors besides swimming speed are important for the success of eel migrations, such as adapting to mesopelagic zones that silver eels undergo during their spawning migrations. The most important and obvious morphological adaptation in mesopelagic fish is their well-developed eyes, and migrating eels also seem to use this strategy. These fish often have relatively large pupils61, high photosensitive structures, such as tubular eyes62, a pure rod multibank retina63, and maximum rhodopsin absorption to adapt to the blue-green light in the deep sea64. The eyes of catadromous eels displayed enlargement during their transformation into migrating silver-phase eels65, 66 and potentially increase their retinal surface area, which results in the possibility of increased photon capture. In addition, the rhodopsins in the eyes change from a freshwater type with a maximum absorption of ~ 500 nm to a deep-sea type with a maximum absorption of ~ 480 nm67,68,69. Their extreme sensitivity to light is evident through their DVM in mesopelagic water, where the timing of a large descent and ascent in the DVM demonstrated by migrating catadromous eels is precisely synchronized with sunrise and sunset. Furthermore, eels alter their swimming depth in response to the phase of the Moon9, 15, 20, 21, appearing to be capable of perceiving extremely low-intensity moonlight.This study showed that three eels released in the TS area (mainly 300–400-m depth) and one eel in the KC area (near surface) were found to change their swimming direction around the time of the solar culmination when the Sun’s bearing changed. The clockwise and counterclockwise trajectories of these eels corresponded to whether the Sun moved from the east to west in the southern and northern sky, suggesting that they demonstrated horizontal negative phototaxis swimming to avoid sunlight. They might move to avoid high-intensity sunlight horizontally, not vertically, as they gradually increase the swimming depths possibly due to acclimation to cold deep water after release. The daytime swimming depths of the eels became deeper day-by-day after their release (Fig. 4); a similar phenomenon was observed in European eels12, American eels17, and long fin eels13. Recently, Higuchi et al.20 observed that the daytime swimming depths of Japanese eels released in the TS area gradually became deeper until 13 days after their release. These facts indicate that they gradually acclimate to the cold water at the deep depths after release. Since this tracking study was conducted 2–8 days after their release, the daytime swimming depth of eels would not have reached a steady state yet. The relatively high intensity from sunlight at the shallow depths where eels swam immediately after release in the TS area might cause horizontal avoidance behavior from the light.In other cases, many eels, especially those released in the KC area, did not demonstrate the rotational behavior. The eels in the KC area mostly stayed deeper (500–800 m) during the day than the eels in the TS area (stayed at depths of 300–600 m) even during the periods shortly after their release. This is possibly due to higher water temperatures even at the deeper depths in the KC area (Fig. 4). The eels in the TS area did not demonstrate clear rotational behavior at depths of more than 400 m. The PSAT studies have reported that the steady swimming depths during the day were 500–800 m14, 20. Therefore, it was assumed that the rotational behavior observed in some eels was not a regular behavior during their migration. However, the rotational behavior observed in this study suggests that they surely perceive the horizontal direction of Sun’s bearing at 400 m depths at least. Generally, they exhibit DVM precisely synchronizing with sunrise and sunset and surely perceive the change in sunlight intensity at deeper depths9,10,11,12,13,14,15,16,17,18,19,20,21,22. Even though the rotational behavior were not observed below 400 m, it remains unknown whether the eels could not perceive the Sun’s bearing from the light penetrated at depth; thus, further investigation of response to underwater light is required in future.While possible negative phototaxis behaviors were observed in some eels after release around the time of solar culmination, the trajectories of ten eels during the entire period of tracking experiments implied that each eel tended to swim meridionally toward the bearing of the Sun at culmination. We observed that eels released at middle (20°–34° N) and low (12°–13°N) latitudes tended to swim southward and northward in the meridional direction, respectively (Fig. 6A, B). The tendency to move in a north–south swimming direction corresponded to whether the Sun culminated to the north or south: eels swam southward if the culmination occurred in the southern sky, but they swam northward if it occurred in the northern sky (Fig. 6). In the KC area (33°–35° N), the Sun rose in the southeast, passed celestial meridian in the southern sky, and set in the southwest (Fig. 6C). At 20° N in the summer time when the tracking study was conducted, the Sun also passed a celestial meridian in the southern sky, but rose in the northeast and set in the northwest (Fig. 6C). When Sun culmination occurred in the southern sky, the meridional swimming directions tended to be southward (Fig. 6A). However, at 12° to 13° N in the summer time, the Sun rose in the northeast, passed the celestial meridian in the northern sky, and set in the northwest (Fig. 6D). When the Sun at culmination appeared in the northern sky, the meridional swimming directions tended to be northward (Fig. 6B). Furthermore, the swimming behavior by one eel (WE4264_TS) that was released slightly south (14° 15′ N) from the latitude with the Sun passing through the zenith was also indicative of the meridional swimming traits. This eel moved in a northerly direction on the first day, but then it lost its north–south bias in swimming around 14° 30′ N, where the Sun nearly passed through the zenith (Figs. 1 and 6D). These observations imply that the eels might move toward the latitude with the Sun passing through the zenith.Figure 6Swimming trajectories of eels and solar paths in the celestial sphere viewed from east during each tracking period. Swimming trajectories of eels released at (A) 20°N in the tropical–subtropical area and the Kuroshio Current area, and (B) 12°–14°15′N in the tropical–subtropical area. Solar paths through the north (N)–south (S) axis and the zenith at the time of tracking in (C) 20°N in the tropical–subtropical area and the Kuroshio Current area, and (D) 12°–14°15′N in the tropical–subtropical area.Full size imageTheoretically, it is possible for mesopelagic animals to use solar cues for navigation at depths shallower than the asymptotic depth, below which penetrating light rays are symmetrical around the vertical axis and the polarization plane becomes horizontal. For example, the Sargasso Sea, where the two Atlantic catadromous eels spawn1, 3, has extremely transparent water70, and the major axis of radiance distribution still remains tilted in the mesopelagic zone. The angle of maximum radiance of sunlight at 475 nm was 13° at depths of 400 m when the Sun’s elevation was 60° (Fig. 7)52, 53. In highly transparent water, the asymptotic depth could be as high as 1000–1200 m, and the depths below this cannot be utilized for compass use53. Currently, it is not possible to verify whether the Sun culminating to north or south caused the meridional swimming tendencies of eels in this study. Potentially, these meridional swimming tendencies could be due to other orientation clues, such as the geomagnetic field, as discussed for temperate anguillid eels17, 45. Nevertheless, in future studies, it would be worthwhile considering solar cues as a possible candidate factor in the orientation of eels, even when under faint underwater light conditions.Figure 7Optical features of underwater sunlight. (A) Schematic diagram of sunlight penetrating the deep ocean at 90° to the solar bearing. The line of arrows indicates the major axis of the incident beam in a vertical plane perpendicular to the Sun’s bearing. Blue light (around 475 nm) reaches the lowest depths. With increasing depth, the light field alters its character into a less directed distribution and a lower energetic level through scattering and absorption processes. Penetrating light rays are symmetrical around the region below the asymptotic depth. (B) An example of spectral radiance distribution (e. g. 475 nm) at a certain depth. The radiance distribution is shown by an ellipsoid and the major axis is drawn by a line with arrow. The refracted angular deviation (a) of the major axis of underwater radiance distribution from the vertical axis equals the tilt of the electric vector (ee bar) from the horizontal axis53. When the Sun’ s elevation was 60° in the Sargasso Sea, the radiance distributions were measured at three different depths and the tilt of the electric vector were estimated to be 24° at depths of 100 m and 200 m and 13° at depth of 400 m52, 53.Full size imageGiven that eels might be able to use the Sun’s bearing at culmination to orient their meridional swimming direction, this orientation scheme could support a clockwise eel migration route following a partial subtropical gyre2, 37. Japanese eels that departed from the nursery area first transported northeastward via the strong KC. Maintaining southward swimming in the current, they eventually crossed the current and shifted to the southward migration course. When they enter the KC, movement to the left of the bearing of the Sun at culmination (i.e., south) is the typical pattern for the early migration of eels from Japan. The movements of eels observed in the KC were consistent with the expected route; however, eels released at low latitudes of the TS area often swam northward but also westward, which resulted in their traveling an unreasonable distance from the spawning area. This might be due to their behavior during early migration. In this study, eels were transported from Japan and released into the open ocean at low latitudes. They might have swum toward the expected bearing of the Sun at culmination as if they were in the north and moved to the left of the Sun’s bearing along with the North Equatorial Current, which would mimic the early migration of eels leaving Japan and moving along the KC.Among the eels tracked in this study were individuals with impaired swim bladders, yellow-phase eels in the process of hormone-treatment maturation, and silver-phase eels collected from different rivers in different years. Despite these variations, the swimming characteristics of the eels did not differ in terms of their DVM behavior16 and swimming speed. Nevertheless, confirmation of our results using samples with a uniform status in future research would be highly desirable. In this study, the tracked eel position was assumed to be identical to that of the tracking ship and the errors between these two positions could not be evaluated; thus, the positioning of tracked fish also may need to be improved in future studies. Experimental studies, such as tracking of blind, magnetically disturbed, or olfactory-blocked eels, could help obtain or eliminate alternative candidate clues and enhance our understanding of the navigational system of anguillid eels. Controlled laboratory experiments are required to directly quantify the ability of eels to perceive radiance distribution or polarization, along with any associated behaviors. In addition, the internal clock of eels required to perform celestial navigation should be investigated. Meanwhile, the results obtained from this study can enhance our knowledge of the mechanisms underlying the migratory behaviors of eels in the open ocean. More

Data collection: soil organic carbonThe process of data acquisition, selection and harmonisation is illustrated in the Supplementary Fig. S15 and in the Supplementary Tables S4–6. We conducted a systematic review for peer-reviewed journal articles, published before December 2018, from Web of Science, and Google Scholar with the search terms “(tree species OR forest) AND (soil organic carbon OR soil organic matter)”. We also used studies listed in two previously published meta-analyses16,17, or cited in already retained references (including references in English, French, Spanish, Portuguese, or Russian). For inclusion in the analysis we chose studies based on the following criteria: (1) the study reported soil organic carbon (SOC) or soil organic matter (SOM) concentrations or pools, at least in the topsoil layer and under at least two single-species forest stands; (2) the stands had to be older than 10 years81; (3) the stands had not experienced a major disturbance that differed between tree species, for at least 30 years (e.g., we rejected studies that compared natural forests with planted forests that were less than 30 years old); (4) The SOC concentration was More

Both in Polistes biglumis and P. gallicus in most of the inhabited nests all types of brood were present: eggs, larvae and pupae (Table S1), with the exception of one foundress nest of P. biglumis with only one egg. The size of thermographed nests was quite variable in both species, the number of cells ranging from 18 to 99 in P. biglumis (mean: 61.6 cells), and from 19 to 381 in P. gallicus (mean: 101.7 cells) (Table S1). The mean number of wasps on the thermographed nests was higher in P. gallicus (12.6 wasps) than in P. biglumis (7.1 wasps). All nests of Polistes biglumis we observed in this study were built on stone substrate or walls (Figs. 1c, 2a). Only recently we found one nest built on a pile of wood. The choice of the nest substrate was more diverse in P. gallicus (Figs. 1d, 2b). They chose stone, concrete, walls, window grilles, and metal of fences or doorframes.Figure 2Examples of nests and fieldwork set-up in Obergail (a) and Sesto Fiorentino (b). 1 = thermocouple wire; 2 = global radiation sensor, 3 = Peltier-element IR reference source.Full size imageDaily nest temperature coursePolistes biglumisFigure 3 shows a sequence of thermograms of a P. biglumis nest taken from dawn to dusk. Before sunrise the temperatures of the nest and of the wasps on it were quite low (mean ~ 15 °C) and uniform (~ 12 to 17.5 °C; Fig. 3a). The temperature of the stone substrate where the nest was built on was considerably higher (~ 20 °C). After sunrise (Fig. 3b,c) the nest temperature began to rise quickly. It only needed 13 min of sunshine (radiation) to heat the nest from ~ 17 to ~ 25 °C. Within one hour, temperature differences of almost 20 °C were measured within the nest. At 6:50, when the highest temperature on the nest was already at 36.2 °C, fast movements of the adults with inspections of the cells were observed (Fig. 3c). Soon afterwards the increasing temperature induced the wasps to start fanning (arrow in Fig. 3d). The wasps also began to gather water and spread it on and inside cells to cool the nest by evaporation (Fig. 3d,e). Towards late morning, some parts of the nest reached temperatures as high as 46 °C (Fig. 3e)! As soon as the nest was shaded by the substrate (~ 13:00) the nest temperature decreased according to the decrease in ambient temperature (Fig. 3f,g), reaching ~ 21 °C on average after dusk (Fig. 3h). At that time the substrate temperature (~ 25 °C) was still about 4 °C higher than the nest temperature.Figure 3Thermograms of a P. biglumis nest during a whole day (19.07.2017). (a) Before sunrise at 6:20; (b) during sunrise (06:33); (c) nest temperature increasing fast in sunshine; (d) with a fanner for convective nest cooling (arrow; see also Fig. S4); (e) with water drops for evaporative cooling when sunshine increased part of the nest to temperatures  > 45 °C; (f,g) after sunset (nest now in shade) in the afternoon; (h) at dusk with wasps sitting motionless on the nest. Time = CEST = UTC + 2 h.Full size imageThe nest and body temperatures of a complete 24 h cycle of a different nest are shown in Fig. 4a. At night the nest temperature and the wasps’ thorax temperature decreased slowly according to the decrease of the air temperature. The substrate temperature was always higher than the mean nest temperature, which surely helped to keep the nest temperature higher than the temperature of the surrounding air (Tanest). Variation of within-nest temperature (max–min) was low at night. As soon as solar radiation increased in early morning, the nest temperature and the body temperature of the wasps on it increased rapidly, and the variation of nest temperature (max–min) increased (see also Fig. 3b). Though the maximum nest temperature reached values as high as 46.9 °C, cooling measures of the wasps (fanning and spreading of water drops, see below) kept the mean nest temperature always below 38.5 °C. Cooling of the nest after sunset (at the nest) was much slower than the increase in the morning, following the decrease of ambient and substrate temperature (Fig. 4a,b).Figure 4Examples of daily temperature changes of nests and wasps of P. biglumis (a,b) and P. gallicus (c,d). Tthorax = mean thorax surface temperature of up to five adult individuals per time of measurement; gray ribbon: total range of nest temperatures (Tmax:Tmin) with mean; Tsubstrate = temperature beside the nest (see Fig. S1c,d); Tanest = ambient air temperature directly at the nest. Ta = ambient air temperature in shade 1–3 m away from nest; Radiation = global radiation hitting the nest; black bars = fanning events at the time of thermographic measurements: actually, many more fanning events were observed. (c) Fanning was never observed! See also Fig. S2 for another example of a P. gallicus nest in shade. Time = CEST = UTC + 2 h.Full size imagePolistes gallicusMost P. gallicus nests were built in locations with no or only little direct sunshine (Figs. 2b, 4c, Fig. S2). In their habitats temperatures in midsummer are often already quite high in the morning, and may increase to values higher than 40 °C during the day (Fig. 4d). Mean temperatures of the nest and of the imagines on it were usually higher than the air temperature close to the nest (Tanest). In most nests variation of within-nest temperature (max–min) remained small throughout the day. On hot days (Tanest  > 40 °C), however, maximum temperatures of empty cells in the nest margin sometimes reached values as high as 49.9 °C even in shade. Body temperature of the adults was mostly similar to the mean nest temperature (Fig. 4c, Fig. S2). At night, the nest temperature decreased according to the decrease of Tanest, similar to P. biglumis but at a higher level (Fig. 4d).The situation was different in one large nest which had been built in a location exposed to the morning sun (Figs. 4d, 5). On a hot day when Tanest increased to values higher than 42 °C, the body temperature of the adults increased to values up to 5 °C higher than the mean nest temperature. Nevertheless, though the combined effects of high air temperature and intense insolation increased part of the nest to a temperature of ~ 58 °C (Fig. 4d), mean nest temperature was kept below 41 °C. This was accomplished by cooling with many water droplets in the cells (dark spots in Fig. 5), and by the occurrence of fanning during the period when the sun was shining on the nest (Fig. 4d; see arrows in Fig. 5c). Fanning, however, was quite rare in all the other observed nests, even during the hottest time of the day! Water droplets were carried onto this nest until evening (Fig. 5h), as at that time the nest temperature was still at about 35–38 °C.Figure 5Thermograms of a large P. gallicus nest during a whole day (01.08.2017). Thermograms are rotated 90° clockwise (the upper part is on the right). (a) Before sunrise (6:36); (b) during sunrise (06:46) with the first water drops visible (dark spots); (c) with two fanners for convective nest cooling (arrows, see also Fig. 4d); (d) with more cooling drops; (e) after sunset at the nest site (nest now in shade); (f–h) after sunset in the afternoon and evening. Time = CEST = UTC + 2 h. For temperature evaluation see Fig. 4d.Full size imageBody and nest temperaturesFigure 6 shows a comparison of the dependence of body and nest temperatures on ambient air temperature and insolation between the two species. In the lower ranges of air temperature, usually at night, body temperature followed Tanest closely in both species. The exposition of the P. biglumis nests to the morning sun at ESE (Fig. 7) increased the wasp body temperature to values of often more than 15 °C higher than the surrounding air. However, body temperatures remained always below 40 °C (Fig. 6a). In P. gallicus, by contrast, the body temperature of the wasps increased considerably above 40 °C, to maximum values of about 46 °C, especially (but not exclusively) during intense insolation in the nest exposed to the morning sun (Fig. 6b).Figure 6Surface temperature of the thorax of adult wasps, of different stages of brood and of water drops of P. biglumis (left) and P. gallicus (right), in dependence on ambient air temperature close to the nest (Tanest) and global radiation (color scale). Egg f.n. = single egg on a foundress nest; diagonal lines = isolines. Regressions were calculated for shaded conditions (Radiation = 0–100 W/m2; black or gray solid lines) and sunshine (Radiation  > 100 W/m2; pink broken lines); P  More

Field collectionEndophytic fungi and insects were assessed from dormant twig samples from 155 tree species at 51 locations in 32 countries. Sampled tree species belonged to genera that are native to, and occur widely across, either the northern or southern hemisphere, since very few tree genera occur naturally in both hemispheres (e.g., in our study only Podocarpus appears in both hemispheres but has a limited distribution in the northern hemisphere). We sampled largely in botanical gardens and arboreta, which allowed us to sample native and non-native, congeneric and confamiliar, tree species at each location. At each location, one native and one to three non-native congeneric or confamiliar tree species were sampled.At each location, twenty 50-cm long asymptomatic twigs were collected from 1–5 individual trees per species, from different branches and different parts of the crown (Fig. 1). The number of individual trees per species depended on the number of trees available in the specific botanical garden or arboretum, which was often low (Table 1). All twigs per tree species and location were pooled and analysed as a single sample. On some occasions two samples of the same tree species at the same location are considered. Sampling was conducted in the month with the shortest day-length in the year (end of December 2017 in the Northern hemisphere, end of June 2018 in the Southern hemisphere). Samples originating from a tropical region (eleven samples from Tanzania) were collected in June 2018. Trees were sampled in winter to align with the timing of trade, i.e. most woody plants are traded in winter or early spring, as plants will be planted in the following spring, and to reduce the risk of introducing foliar pests in deciduous trees. Evergreen gymnosperm and angiosperm tree species, which were also considered, do not lose foliage during winter, and are thus sold with leaves/needles.Table 1 Site information for sampling locations included in this study.Full size tableFungal endophytesTo assess fungal communities, a total of 352 samples from 145 native and non-native tree species, belonging to nine families of angiosperms and gymnosperms, were collected. Sampling was done at 44 locations in 28 countries on five continents (Fig. 1, Table 1).From each twig in a sample, one bud, one needle/leaf and one 1 cm long twig segment were taken (Fig. 1). Needles from gymnosperms, and leaves from evergreen angiosperms were sampled to accurately assess the risk of trading these species. Twig segments were cut from the twig bases. The selected plant parts were surface sterilized by immersion in 75% ethanol for 1 min, 4% NaOCl for 5 min, and 75% ethanol for 30 s26. After air drying on a sterile bench, the following material from each of 20 twigs per sample was pooled: half of one bud, a 0.5 cm long piece of a needle (from gymnosperms) or a 0.25 cm2 leaf (for evergreen angiosperms) and a 0.5 cm long piece of twig.DNA extraction, PCR amplification and Illumina sequencingTotal genomic DNA was extracted from 50 mg of pooled, surface sterilized, and ground tissue (Fig. 1) using DNeasy PowerPlant Pro Kit (Qiagen, Hilden, Germany), following the manufacturer’s instructions. For a total of 31 out of 352 samples, DNA was extracted from different tissues separately, and DNA extracts were then pooled. DNA concentrations were quantified using the Qubit dsDNA BR Assay Kit (Thermo Fisher Scientific, Waltham, USA) on a Qubit 3.0 Fluorometer (Thermo Fisher Scientific) and DNA was diluted to 5 ng/μl. Samples that yielded less than 5 ng/μl were not diluted. The ITS2 region was amplified with the 5.8S-Fung and ITS4-Fung primers27. PCR amplifications were carried out in 20 μl reaction volumes containing 25 ng of DNA template, 1 mg/ml BSA, 1 mM of MgCl2, 0.4 μM of each primer, and 0.76 × JumpStart REDTaq ReadyMix Reaction Mix (Sigma-Aldrich, Steinheim, Germany). PCR was performed using Veriti 96-Well Thermal Cycler (Applied Biosystems, Foster City, CA, USA) as described in Franić et al. (2019). Each sample was amplified in triplicates and successful PCR amplification confirmed by visualization of the PCR products, before and after pooling the triplicates, on 1.5% (w/v) agarose gel with ethidium bromide staining. Pooled amplicons were sent to the Génome Québec Innovation Center at McGill University (Montréal, Quebec, Canada) for barcoding using Fluidigm Access Array technology (Fluidigm, South San Francisco, CA, USA) and paired-end sequencing on the Illumina MiSeq v3 platform (Illumina Inc., San Diego, CA, USA). Raw sequences obtained in this study are deposited at the NCBI Sequence Read Archive under BioProject accession number PRJNA70814822.Bioinformatics and taxonomical classification of ASVsQuality filtering and delineation into ASVs were done with a customized pipeline28 largely based on VSEARCH29, as described by Herzog et al.30. The output data available on Figshare show the abundances of fungal ASVs in the samples24. Taxonomic classification of ASVs was conducted using Sintax31 implemented in VSEARCH against the UNITE v.7.2 database32 with a bootstrap support of 80%. The data on the taxonomic classification of fungal ASVs is deposited in Figshare24.Quality filtering, delineation into ASVs, and taxonomical assignments were done on a larger data set (total of 474 samples), which increased the confidence in the selected centroid sequences. This data set consisted of (1) sequences obtained from 352 samples of pooled tree tissues that are presented here22, (2) sequences obtained from 33 samples of pooled tree tissues which were not included in this manuscript due to violation of the common protocol, (3) sequences from 21 contaminated samples (positive DNA extraction controls), including sequences from the two control samples (not presented here), and (4) sequences obtained from 66 samples of non-pooled tree tissues of Pinus sylvestris and Quercus robur that were collected from the subset of locations considered in this study, but for a different study, and are thus not presented here.Herbivorous insectsInsects were assessed from 227 samples of 109 tree species, collected at 31 locations and in 18 countries (Fig. 1, Table 1).The collected twigs (twenty 50 cm twigs per species per location) were brought to a laboratory close to each sampling location and inspected for the presence of insects that overwinter as adults. Twigs were kept at room temperature with the cut ends immersed in water to induce budding and to allow the development of insects that overwinter as larvae, pupae or eggs. Twigs from each sample were protected with gauze bags to prevent insects moving between samples (Fig. 1). Twigs were inspected for the presence of insects daily for 4 weeks and all collected insects were stored in 95% ethanol for further examination.Morphological and molecular identificationInsects were inspected using a stereo microscope and sorted to taxonomic orders and feeding guilds (i.e. herbivores, predators, parasitoids and other). The abundance of the different feeding guilds and taxonomic orders in the samples is presented in a file deposited on Figshare24. Herbivorous insects were further sorted into morphospecies and at least one specimen per morphospecies was stored at −20 °C for molecular analysis. The abundance of the different morphospecies in each sample is presented in a file deposited on Figshare24. Specimens for molecular analysis were photographed with a Leica DVM6 digital microscope and the Leica Application Suite X (LAS X). Depending on the size of the insects, the whole individual or parts (e.g. legs, head) were used for molecular analysis. Genomic DNA was extracted with a KingFisher (Thermo Fisher Scientific) extraction protocol suitable for insects (35 min incubation at RT, 30 min wash at RT with 3 different washing buffers, 13 min elution at 60 °C) in a 96-well plate. PCR for the COI was carried out in 25 µl reaction volume with 2 µl diluted DNA (1:10), 0.5 µM of each of the primers LCO1490 and HCO219833 and 1 x REDTaq ReadyMix Reaction Mix (Sigma-Aldrich) using a Veriti 96-Well Thermal Cycler (Applied Biosystems) with the following setting: 2 min at 94 °C, five cycles of 30 s at 94 °C, 40 s at 45 °C, and 1 min at 72 °C, 35 cycles of 30 s at 94 °C, 50 s at 51 °C, and 1 min at 72 °C, and a final extension step at 72 °C for 10 min. The success of amplification was verified by electrophoresis of the PCR products in 1.5% (w/v) agarose gel at 90 V for 30 min with ethidium bromide staining. A standard Sanger sequencing of the PCR products in both directions with the same primers was done at Macrogen Europe, Amsterdam, Netherlands. Sequences were assembled and edited with CLC Workbench (Version 7.6.2, Quiagen) and compared to reference sequences in BOLD34. If no conclusive results were found, sequences were compared to reference sequences in the National Centre for Biotechnology Information (NCBI) GenBank databases35. Specimens were assigned to species if the query sequence showed less than 1% divergence from the reference sequence. If two or more taxa matched within the same range, the assignment was ranked down to the next taxonomic level (i.e., genus). When no species match was obtained based on the above criteria, a genus was assigned with a divergence of less than 3%. For lower taxonomic groups the 100 nearest sequences were inspected on the Blast Tree (Fast Minimum Evolution Method) and the taxonomic relationship was evaluated based on that tree. If none of the approaches above revealed a conclusive taxonomic assignment, the morphological identification was taken as reference. The results of morphological and molecular identification of insect specimens are presented in a file deposited on Figshare24. Insect sequences are deposited in GenBank database under accession numbers MW441337-MW44176725.Sample metadataPairwise geographic distances (Euclidean distances) between sampling locations were calculated based on the geographic coordinates of the locations, with function “dist” in the R statistical programme36.Climate data, including mean annual temperature, mean annual precipitation, and temperature seasonality were obtained from the WorldClim database37, at a resolution of 2.5 min, and represent averages between 1970 and 2000.A host-tree phylogeny was constructed with the phylomatic function from the package brranching38 in R using the “zanne2014” reference tree39. One Eucalyptus sample collected in Argentina and two Eucalyptus samples collected in Tunisia were not identified to species. To place them in the phylogeny, we assigned them to different congeneric species that were not sampled in this study and that we considered as representative samples of phylogenetic diversity from across Eucalyptus genus (E. viminalis, E. robusta and E. radiata). Pairwise phylogenetic distances between study tree species were calculated using the “cophenetic” function in R36.The described sample metadata are available in a file on Figshare24. More

Pure coordination gameIn this section we study the Pure Coordination Game (PCG) (also known as doorway game, or driving game) in which (R=1), (S=0), (T=0), and (P=1), resulting in a symmetric payoff matrix with respect to the two strategies:$$begin{gathered} begin{array}{*{20}c} {} & {quad ; {text{A}}} &; {text{B}} \ end{array}hfill \ begin{array}{*{20}c} {text{A}} \ {text{B}} \ end{array} left( {begin{array}{*{20}c} 1 & 0 \ 0 & 1 \ end{array} } right) hfill \ end{gathered}$$
(2)
There are two equivalent equilibria for both players coordinating at the strategy A or B (a third Nash equilibrium exists for players using a mix strategy of 50% A and 50% B). As the absolute values of the payoff matrix are irrelevant and the dynamics is defined by ratios between payoffs from different strategies, the payoff matrix (2) represents all games for which the relation (R=P >S=T) is fulfilled.In the PCG the dilemma of choosing between safety and benefit does not exist, because there is no distinction between risk-dominant and payoff-dominant equilibrium. Both strategies yield equal payoffs when players coordinate on them and both have the same punishment (no payoff) when players fail to coordinate. Therefore, the PCG is the simplest framework to test when coordination is possible and which factors influence it and how. It is in every player’s interest to use the same strategy as others. Two strategies, however, are present in the system at the beginning of the simulation in equal amounts. From the symmetry of the game we can expect no difference in frequency of each strategy being played, when averaged over many realisations. Still, the problem of when the system reaches full coordination in one of the strategies is not trivial. We address this question here.Figure 1Time evolution of the coordination rate (alpha) (in MC steps) in individual realisations for different values of the degree k in a random regular network of (N=1000) nodes, using (a) the replicator dynamics, (b) the best response, and (c) the unconditional imitation update rule.Full size imageFigure 2Coordination rate (alpha) and interface density (rho) vs degree k of a random regular network for (N=1000) using (a) the replicator dynamics, (b) the best response, and (c) the unconditional imitation update rule. Each green circle represents one of 500 realisations for each value of the degree k and the average value is plotted with a solid line, separately for (alpha >0.5) and (alpha le 0.5). Results are compared to the ER random network ((alpha _{ER})) with the same average degree.Full size imageFirst, we look at single trajectories as presented in Fig. 1. Some of them quickly reach (alpha =0) or 1, or stop in a frozen state without obtaining global coordination. Other trajectories take much longer and extend beyond the time scale showed in the figure. What we can already tell is that the process of reaching coordination is slower in the replicator dynamics where it usually takes more time than in the best response and unconditional imitation to reach a frozen configuration. For all update rules the qualitative effect of the connectivity is similar—for bigger degree it is more likely to obtain full coordination and it happens faster. For the UI, however, larger values of degree than for the RD and BR are required to observe coordination. For example, in the case of (k=10) or 20 the system stops in a frozen disorder when using UI, while for the RD and BR it quickly reaches a coordinated state of (alpha =0) or 1.To confirm the conclusions from observation of trajectories, we present the average outcome of the system’s evolution in the Fig. 2. The first thing to notice is that all plots are symmetrical with respect to the horizontal line of (alpha = 0.5). It indicates that the strategies are indeed equivalent as expected. In all cases there is a minimal connectivity required to obtain global coordination. For the RD and BR update rules this minimum value is (k=4), although in the case of BR the system fails to coordinate for small odd values of k due to regular character of the graph. This oscillating behaviour does not exist in Erdős–Rényi random networks. When nodes choose their strategies following the UI rule much larger values of k are required to obtain full coordination. Single realisations can result in (alpha = 0), or 1 already for (k=15). However, even for (k=60) there is still a possibility of reaching a frozen uncoordinated configuration.The important conclusion is that there is no coordination without a sufficient level of connectivity. In order to confirm that this is not a mere artefact of the random regular graphs we compare our results with those obtained for Erdős–Rényi (ER) random networks76,77 (black dashed line in Fig. 2). The level of coordination starts to increase earlier for the three update rules, but the general trend is the same. The only qualitative difference can be found in the BR. The oscillating level of coordination disappears and it doesn’t matter if the degree is odd or even. This shows that different behaviour for odd values of k is due to topological traps in random regular graphs78. Our results for the UI update rule are also consistent with previous work reporting coordination for a complete graph but failure of global coordination in sparse networks40.Figure 3Examples of frozen configuration reached under the UI update rule for small values of the average degree k in random regular networks (top row) and Erdős–Rényi networks (bottom row) with 150 nodes. Red colour indicates a player choosing the strategy A, blue colour the strategy B. Note the topological differences between random regular and ER networks when they are sparse. For (k=1) a random regular graph consists of pairs of connected nodes, while an ER network has some slightly larger components and many loose nodes. For (k=2) a random regular graph is a chain (sometimes 2–4 separate chains), while an ER network has one large component and many disconnected nodes. For (k=3) and (k=4) a random regular graph is always composed of one component, while an ER network has still a few disconnected nodes.Full size imageSince agents using the RD and BR update rule do not achieve coordination for small values of degree, one might suspect that the network is just not sufficiently connected for these values of the degree, i.e. there are separate components. This is only partially true. In Fig. 3, we can see the structures generated by random regular graph and by ER random graph algorithms. Indeed, for (k=1) and 2 the topology is trivial and a large (infinite for (k=1)) average path length23 can be the underlying feature stopping the system to reach coordination. For (k=3), however, the network is well connected with one giant component and the system still does not reach the global coordination when using RD or BR. For the UI update rule coordination arrives even for larger values of k. Looking at the strategies used by players in Fig. 3 we can see how frozen configuration without coordination can be achieved. There are various types of topological traps where nodes with different strategies are connected, but none of them is willing to change the strategy in the given update rule.We next consider the question of how the two strategies are distributed in the situations in which full coordination is not reached. Looking at the trajectories in Fig. 1 we can see that there are only few successful strategy updates in such scenario and the value of (alpha) remains close to 0.5 until arriving at a frozen state for (k=2) (also (k=7) for UI). This suggests that there is not enough time, in the sense of the number of updates, to cluster the different strategies in the network. Therefore, one might expect that they are well mixed as at the end of each simulation. However, an analysis of the density of active links in the final state of the dynamics, presented in Fig. 2, shows a slightly more complex behaviour. When the two strategies are randomly distributed (i.e. well mixed) in a network, the interface density takes the value (rho =0.5). When the two strategies are spatially clustered in the network there are only few links connecting them and therefore the interface density takes small values. Looking at the dependence of (rho) on k, we find that for the replicator dynamics the active link density starts at 0.5 for (k=1), then drops below 0.2 for (k=2) and 3 indicating good clustering between strategies, to fall to zero for (k=4) where full coordination is already obtained. When using the best response update rule the situation is quite different. For (k=1) there are no active links, (rho =0), and hardly any for (k=2). There is a slight increase of the active link density for (k=3), to drop to zero again for (k=4) due to full coordination. Because of the oscillatory level of coordination there are still active links for odd values of (kP) (otherwise we can rename the strategies and shuffle the columns and rows). What defines the outcome of a game are the greater than and smaller than relations among the payoffs. Therefore we can add/subtract any value from all payoffs, or multiply them by a factor grater than zero, without changing the game. Thus, the payoff matrix (1) can be rewritten as:$$begin{gathered} begin{array}{*{20}c} {} & {qquad {text{A}}} & {quad quad {text{B}}} \ end{array} ;; hfill \ begin{array}{*{20}c} {text{A}} \ {text{B}} \ end{array} left( {begin{array}{*{20}c} 1 & {frac{{S – P}}{{R – P}}} \ {frac{{T – P}}{{R – P}}} & 0 \ end{array} } right) hfill \ end{gathered}$$
(3)
which, after substituting (S’=frac{S-P}{R-P}) and (T’=frac{T-P}{R-P}), is equivalent to the matrix: $$begin{gathered} begin{array}{*{20}c} {} &quad ;;{text{A}} &; {text{B}} \ end{array} ;quad quad quad quad quad quad begin{array}{*{20}c} {} & quad; {text{A}} & ;{text{B}} \ end{array} hfill \ begin{array}{*{20}c} {text{A}} \ {text{B}} \ end{array} left( {begin{array}{*{20}c} 1 & {S^{prime}} \ {T^{prime}} & 0 \ end{array} } right)xrightarrow[{{text{apostrophes}}}]{{{text{skipping}}}}begin{array}{*{20}c} {text{A}} \ {text{B}} \ end{array} left( {begin{array}{*{20}c} 1 & S \ T & 0 \ end{array} } right) hfill \ end{gathered}$$
(4)
From now on we omit the apostrophes and simply refer to parameters S and T. This payoff matrix can represent many games, including e.g. the prisoner’s dilemma14,46 (for (T >1) and (S More

IPCC Climate Change 2014: Synthesis Report. In Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change (eds Core Writing Team, Pachauri, R. K. & Meyer L. A.) 151 (IPCC, Geneva, Switzerland, 2014).Grizzetti, B., Bouraoui, F. & Aloe, A. Changes of nitrogen and phosphorus loads to European seas. Glob. Change Biol. 18, 769–782 (2012).
Google Scholar 
Hoegh-Guldberg, O. & Bruno, J. F. The impact of climate change on the world’s marine ecosystems. Science 328, 1523–1528 (2010).CAS 
PubMed 

Google Scholar 
Duarte, C. M. Global change and the future ocean: a grand challenge for marine sciences. Front. Mar. Sci. 1, 1–16 (2014).
Google Scholar 
Richardson, A. J. & Schoeman, D. S. Climate impact on plankton ecosystems in the Northeast Atlantic. Science 305, 1609–1612 (2004).CAS 
PubMed 

Google Scholar 
Rose, J. M. et al. Effects of increased pCO2 and temperature on the North Atlantic spring bloom. II. Microzooplankton abundance and grazing. Mar. Ecol. Prog. Ser. 388, 27–40 (2009).CAS 

Google Scholar 
Sommer, U., Paul, C. & Moustaka-Gouni, M. Warming and ocean acidification effects on phytoplankton—from species shifts to size shifts within species in a mesocosm experiment. PLoS ONE 10, 1–17 (2015).
Google Scholar 
Garzke, J., Hansen, T., Ismar, S. M. H. & Sommer, U. Combined effects of ocean warming and acidification on copepod abundance, body size and fatty acid content. PLoS ONE 11, 1–22 (2016).
Google Scholar 
Horn, H. G., Boersma, M., Garzke, J., Sommer, U. & Aberle, N. High CO2 and warming affect microzooplankton food web dynamics in a Baltic Sea summer plankton community. Mar. Biol. 167, 1–17 (2020).
Google Scholar 
Boyd, P. W. et al. Experimental strategies to assess the biological ramifications of multiple drivers of global ocean change—a review. Glob. Change Biol. 24, 2239–2261 (2018).
Google Scholar 
Stewart, R. I. A. et al. Mesocosm experiments as a tool for ecological provided for ecological climate-change research. In Advances in Ecological Research/Guy Woodward (ed. O’Gorman, E. J.) 71–181 (Academic Press, 2013).Rost, B. & Riebesell, U. Coccolithophores and the biological pump: responses to environmental changes. In Coccolithophores: From Molecular Processes to Global Impact (eds Thierstein, H. R. & Young, J. R.) 99–125 (Springer, 2004).Peter, K. H. & Sommer, U. Phytoplankton cell size reduction in response to warming mediated by nutrient limitation. PLoS ONE 8, 1–6 (2013).
Google Scholar 
Bermúdez, J. R., Riebesell, U., Larsen, A. & Winder, M. Ocean acidification reduces transfer of essential biomolecules in a natural plankton community. Sci. Rep. 6, 1–8 (2016).
Google Scholar 
Peter, K. H. & Sommer, U. Interactive effect of warming, nitrogen and phosphorus limitation on phytoplankton cell size. Ecol. Evolution 5, 1011–1024 (2015).
Google Scholar 
Alvarez-Fernandez, S. et al. Plankton responses to ocean acidification: the role of nutrient limitation. Prog. Oceanogr. 165, 11–18 (2018).
Google Scholar 
Stramski, D., Sciandra, A. & Claustre, H. Effects of temperature, nitrogen, and light limitation on the optical properties of the marine diatom Thalassiosira pseudonana. Limnol. Oceanogr. 47, 392–403 (2002).CAS 

Google Scholar 
Marañón, E. Cell size as a key determinant of phytoplankton metabolism and community structure. Annu. Rev. Mar. Sci. 7, 241–264 (2015).
Google Scholar 
Peñuelas, J., Sardans, J., Rivas‐Ubach, A. & Janssens, I. A. The human-induced imbalance between C, N and P in Earth’s life system. Glob. Change Biol. 18, 3–6 (2011).
Google Scholar 
Azam, F. et al. The ecological role of water-column microbes in the sea. Mar. Ecol. Prog. Ser. 10, 257–63. (1983).
Google Scholar 
Legendre, L. & Le Fèvre, J. Microbial food webs and the export of biogenic carbon in oceans. Aquat. Microb. Ecol. 9, 69–77 (1995).
Google Scholar 
Beaufort, L. et al. Sensitivity of coccolithophores to carbonate chemistry and ocean acidification. Nature 476, 80–83 (2011).CAS 
PubMed 

Google Scholar 
Langer, G., Nehrke, G., Probert, I., Ly, J. & Ziveri, P. Strain-specific responses of Emiliania huxleyi to changing seawater carbonate chemistry. Biogeosciences 6, 2637–2646 (2009).CAS 

Google Scholar 
Winter, A., Henderiks, J., Beaufort, L., Rickaby, R. E. M. & Brown, C. W. Poleward expansion of the coccolithophore Emiliania huxleyi. J. Plankton Res. 36, 316–325 (2014).CAS 

Google Scholar 
Hopkins, J., Henson, S. A., Painter, S. C., Tyrrell, T. & Poulton, A. J. Phenological characteristics of global coccolithophore blooms. Glob. Biogeochemical Cycles 29, 239–253 (2015).CAS 

Google Scholar 
León, P. et al. Seasonal variability of the carbonate system and coccolithophore Emiliania huxleyi at a Scottish Coastal Observatory monitoring site. Estuar., Coast. Shelf Sci. 202, 302–314 (2018).
Google Scholar 
Rivero-Calle, S., Gnanadesikan, A., Del Castillo, C. E., Balch, W. M. & Guikema, S. D. Multidecadal increase in North Atlantic coccolithophores and the potential role of rising CO2. Science 350, 1533–1537 (2015).CAS 
PubMed 

Google Scholar 
Purdie, D. A. & Finch, M. S. Impact of a coccolithophorid bloom on dissolved carbon dioxide in sea water enclosures in a Norwegian fjord. Sarsia 79, 379–387 (1994).
Google Scholar 
Nejstgaard, J. C., Gismervik, I. & Solberg, P. T. Feeding and reproduction by Calanus finmarchicus, and microzooplankton grazing during mesocosm blooms of diatoms and the coccolithophore Emiliania huxleyi. Mar. Ecol. Prog. Ser. 147, 197–217 (1997).
Google Scholar 
Leblanc, K. et al. Distribution of calcifying and silicifying phytoplankton in relation to environmental and biogeochemical parameters during the late stages of the 2005 North East Atlantic Spring Bloom. Biogeosciences 6, 2155–2179 (2009).CAS 

Google Scholar 
Sett, S. et al. Temperature modulates coccolithophorid sensitivity of growth, photosynthesis and calcification to increasing seawater pCO2. PLoS ONE 9, e88308 (2014).PubMed 
PubMed Central 

Google Scholar 
Benner, I. et al. Emiliania huxleyi increases calcification but not expression of calcification-related genes in long-term exposure to elevated temperature and pCO2. Philos. Trans. R. Soc. B 368, 20130049 (2013).
Google Scholar 
Borchard, C., Borges, A. V., Händel, N. & Engel, A. Biogeochemical response of Emiliania huxleyi (PML B92/11) to elevated CO2 and temperature under phosphorous limitation: a chemostat study. J. Exp. Mar. Biol. Ecol. 410, 61–71 (2011).CAS 

Google Scholar 
Harrison, P. J. et al. Geographical distribution of red and green Noctiluca scintillans. Chin. J. Oceanol. Limnol. 29, 807–831 (2011).
Google Scholar 
Johns, D. G., Edwards, M., Greve, W. & SJohn, A. W. G. Increasing prevelance of the marine cladoceran Penilia avirostris (Dana, 1852) in the North Sea. Helgol. Mar. Res. 59, 215–218 (2005).
Google Scholar 
O’Connor, M. I. O., Piehler, M. F., Leech, D. M., Anton, A. & Bruno, J. F. Warming and resource availability shift food web structure and metabolism. PLoS Biol. 7, 1–6 (2009).
Google Scholar 
Cross, W. F., Hood, J. M., Benstead, J. P., Huryn, A. D. & Nelson, D. Interactions between temperature and nutrients across levels of ecological organization. Glob. change Biol. 21, 1025–1040 (2015).
Google Scholar 
Boersma, M. et al. Temperature driven changes in the diet preference of omnivorous copepods: no more meat when it’s hot? Ecol. Lett. 19, 45–53 (2016).PubMed 

Google Scholar 
Anderson, T. R., Hessen, D. O., Boersma, M., Urabe, J. & Mayor, D. J. Will invertebrates require increasingly carbon-rich food in a warming world? Am. Naturalist 190, 725–742 (2017).
Google Scholar 
Kirchner, M., Sahling, G., Uhlig, G., Gunkel, W. & Klings, K.-W. Does the red tide-forming dinoflagellate Noctiluca scintillans feed on bacteria? Sarsia 81, 45–55 (2015).
Google Scholar 
Elbrächter, M. & Qi, Y. Aspects of Noctiluca (Dinophyceae) population dynamics. In Physiological Ecology of Harmful Algal Blooms (ed. Anderson, M. D.) 315–335 (Springer-Verlag, 1998).Atienza, D., Saiz, E. & Calbet, A. Feeding ecology of the marine cladoceran Penilia avirostris: natural diet, prey selectivity and daily ration. Mar. Ecol. Prog. Ser. 315, 211–220 (2006).
Google Scholar 
Zhang, S., Liu, H., Chen, B. & Chih-Jung, W. Effects of diet nutritional quality on the growth and grazing of Noctiluca scintillans. Sci. Rep. 527, 73–85 (2015).CAS 

Google Scholar 
Reid, P. C., Borges, M. F. & Svendsen, E. A regime shift in the North Sea circa 1988 linked to changes in the North Sea horse mackerel fishery. Fish. Res. 50, 163–171 (2001).
Google Scholar 
Beaugrand, G., Brander, K. M., Lindley, J. A., Souissi, S. & Reid, P. C. Plankton effect on cod recruitment in the North Sea. Nature 426, 661–664 (2003).CAS 
PubMed 

Google Scholar 
Payne, M. R. et al. Recruitment in a changing environment: the 2000s North Sea herring recruitment failure. ICES J. Mar. Sci. 66, 272–277 (2009).
Google Scholar 
Perälä, T., Olsen, E. M. & Hutchings, J. A. Disentangling conditional effects of multiple regime shifts on Atlantic cod productivity. PLoS ONE 15, e0237414 (2020).PubMed 
PubMed Central 

Google Scholar 
Behrenfeld, M. J., Boss, E. S. & Halsey, K. H. Phytoplankton community structuring and succession in a competition-neutral resource landscape. ISME COMMUN. 1, 1–8 (2021).Monteiro, F. M. et al. Why marine phytoplankton calcify. Sci. Adv. 2, 1–14 (2016).
Google Scholar 
Mayers, K. M. J. et al. The possession of coccoliths fails to deter microzooplankton grazers. Front. Mar. Sci. 7, 976 (2020).
Google Scholar 
Zhao, Y. et al. Grazing by microzooplankton and copepods on the microbial food web in spring in the southern Yellow Sea, China. Mar. Life Sci. Technol. 2, 442–455 (2020).
Google Scholar 
Aberle, N. et al. High tolerance of microzooplankton to ocean acidification in an Arctic coastal plankton community. Biogeosciences 10, 1471–1481 (2013).
Google Scholar 
Horn, H. G. et al. Low CO2 sensitivity of Microzooplankton communities in the Gullmar Fjord, Skagerrak: evidence from a long-term Mesocosm Study. PLoS ONE 11, 1–24 (2016).
Google Scholar 
Chen, B., Landry, M. R., Huang, B. & Liu, H. Does warming enhance the effect of microzooplankton grazing on marine phytoplankton in the ocean? Limnol. Oceanogr. 57, 519–526 (2012).CAS 

Google Scholar 
Vázquez-Domínguez, E., Vaqué, D. & Gasol, J. M. Temperature effects on the heterotrophic bacteria, heterotrophic nanoflagellates, and microbial top predators of the NW Mediterranean. Aquat. Microb. Ecol. 67, 107–121 (2012).
Google Scholar 
Lara, E. et al. Experimental evaluation of the warming effect on viral, bacterial and protistan communities in two contrasting Arctic systems. Aquat. Microb. Ecol. 70, 17–32 (2013).
Google Scholar 
Olson, M. B., Solem, K. & Love, B. Microzooplankton grazing responds to simulated ocean acidification indirectly through changes in prey cellular characteristics. Mar. Ecol. Prog. Ser. 604, 83–97 (2018).CAS 

Google Scholar 
Sherr, E. B. & Sherr, B. F. Bacterivory and herbivory: key roles of phagotrophic protists in pelagic food webs. Microb. Ecol. 28, 223–235 (1994).CAS 
PubMed 

Google Scholar 
Brander, K. & Kiørboe, T. Decreasing phytoplankton size adversely affects ocean food chains. Glob. Change Biol. 26, 5356–5357 (2020).
Google Scholar 
Irigoien, X. et al. A high frequency time series at weathership M, Norwegian Sea, during the 1997 spring bloom: feeding of adult female Calanus finmarchicus. Mar. Ecol. Prog. Ser. 172, 127–137 (1998).
Google Scholar 
Fenchel, T. The microbial loop—25 years later. J. Exp. Mar. Biol. Ecol. 366, 99–103 (2008).
Google Scholar 
Aberle, N., Malzahn, A. M., Lewandowska, A. M. & Sommer, U. Some like it hot: the protozooplankton— copepod link in a warming ocean. Mar. Ecol. Prog. Ser. 519, 103–113 (2015).
Google Scholar 
Berglund, J., Müren, U., Båmstedt, U. & Andersson, A. Efficiency of a phytoplankton-based and a bacteria-based food web in a pelagic marine system. Limnol. Oceanogr. 52, 121–131 (2007).CAS 

Google Scholar 
Sherr, E. B. & Sherr, B. F. Heterotrophic dinoflagellates: a significant component of microzooplankton biomass and major grazers of diatoms in the sea. Mar. Ecol. Prog. Ser. 352, 187–197 (2007).
Google Scholar 
Gifford, D. J. The protozoan-metazoan trophic link in pelagic ecosystems. J. Protozool. 38, 81–86 (1991).
Google Scholar 
Rollwagen-Bollens, G. & Gifford, S. The role of protistan microzooplankton in the upper San Francisco estuary planktonic food web: source or sink? Estuaries Coasts 34, 1026–1038 (2011).CAS 

Google Scholar 
Anjusha, A. et al. Trophic efficiency of plankton food webs: observations from the Gulf of Mannar and the Palk Bay, Southeast Coast of India. J. Mar. Syst. 115, 40–61 (2013).
Google Scholar 
IPCC. Global Warming of 1.5 °C. An IPCC Special Report on the impacts of global warming of 1.5 °C above pre-industrial levels and related global greenhouse gas emission pathways. In The Context of Strengthening the Global Response to The Threat of Climate Change, Sustainable Development, and Efforts to Eradicate Poverty (Masson-Delmotte, V. et al (eds.) 616 (IPCC, Geneva, Switzerland, 2018).Pansch, A., Winde, V., Asmus, R. & Asmus, H. Tidal benthic mesocosms simulating future climate change scenarios in the field of marine ecology. Limnol. Oceanogr.: Methods 14, 257–267 (2016).
Google Scholar 
van Leeuwen, S., Tett, P., Mills, D. & van der Molen, J. Stratified and nonstratified areas in the North Sea: long-term variability and biological and policy implications. J. Geophys. Res.: Oceans 120, 4670–4686 (2015).
Google Scholar 
Grasshoff, K., Kremling, K. & Ehrhardt, M. (eds). Methods of Seawater Analysis, 3rd edn. (Wiley-VCH, Weinheim, 1999).Dickson, A. G. An exact definition of total alkalinity and a procedure for the estimation of alkalinity and total inorganic carbon from titration data. Deep-Sea Res. 28, 609–623 (1981).CAS 

Google Scholar 
Pierrot, D. E., Lewis, E. & Wallace, D. W. R. MS Excel program developed for CO2 system calculations. ORNL/CDIAC-105a. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, U.S. Department of Energy, Oak Ridge, Tennessee https://doi.org/10.3334/CDIAC/otg.CO2SYS_XLS_CDIAC105a (2006).Dickson, A. G. & Millero, F. J. A comparison of the equilibrium constants for the dissociation of carbonic acid in seawater media. Deep-Sea Res. 34, 1733–1743 (1987).CAS 

Google Scholar 
Arrigo, K. R. et al. Phytoplankton community structure and the drawdown of nutrients and CO2 in the Southern Ocean. Science 283, 365–368 (1999).CAS 
PubMed 

Google Scholar 
Utermöhl, H. Zur Vervollkommnung der quantitativen Phytoplankton- Methodik. Int. Ver. für. Theoretische und Angew. Limnologie: Mitteilungen 9, 1–38 (1958).
Google Scholar 
McEwen, G. F., Johnson, M. W. & Folsom, T. R. A statistical analysis of the performance of the Folsom plankton sample splitter, based upon test observations. Archiv für. Archiv Meteorologie, Geophysik und Bioklimatologie, Ser. A 7, 502–527 (1954).
Google Scholar 
Sell, D. W. & Evans, M. S. A statistical analysis of subsampling and an evaluation of the Folsom plankton splitter. Hydrobiologia 94, 223–230 (1982).
Google Scholar 
Boersma, M., Wiltshire, K. H., Kong, S., Greve, W. & Renz, J. Long-term change in the copepod community in the southern German Bight. J. Sea Res. 101, 41–50 (2015).
Google Scholar 
Marie, D., Simon, N. & Vaulot, D. Phytoplankton cell counting by flow cytometry. Algal Culturing Tech. 1, 253–267 (2005).
Google Scholar 
Hillebrand, H., Dürselen, C., Kirschtel, D., Pollingher, U. & Zohary, T. Biovolume calculation for pelagic and benthic microalgae. J. Phycol. 35, 403–424 (1999).
Google Scholar 
Menden-Deuer, S. & Lessard, E. J. Carbon to volume relationships for dinoflagellates, diatoms, and other protist plankton. Limnol. Oceanogr. 45, 569–579 (2000).CAS 

Google Scholar 
Putt, M. & Stoecker, D. K. An experimentally determined carbon: volume ratio for marine “oligotrichous” ciliates from estuarine and coastal waters. Limnol. Oceanogr. 34, 1097–1103 (1989).
Google Scholar 
Beran, A. et al. Carbon content and biovolume of the heterotrophic dinoflagellate Noctiluca scintillans from the Northern Adriatic Sea. In Proceedings of the CESUM-BS 2003, Varna. 28 (Book of Abstracts, UNESCO, Paris, 2003).Lee, S. & Fuhrman, J. A. Relationships between biovolume and biomass of naturally derived marine bacterioplankton. Appl. Environ. Microbiol. 53, 1298–1303 (1987).CAS 
PubMed 
PubMed Central 

Google Scholar 
Kraberg, A., Baumann, M. & Dürselen, C. Coastal Phytoplankton: Photo Guide for Northern European Seas (Dr. Friedrich Pfeil, München, 2010).R Core Team. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/ (2021). More

Rutz, C. Predator fitness increases with selectivity for odd prey. Curr. Biol. 22, 820–824 (2012).CAS 
PubMed 

Google Scholar 
Santos, C. D. et al. Personality and morphological traits affect pigeon survival from raptor attacks. Sci. Rep. 5, 1–8 (2015).
Google Scholar 
Brown, M. B. & Wells, E. Skeletal dysplasia-like syndromes in wild giraffe. BMC Res. Notes 13, 569 (2020).PubMed 
PubMed Central 

Google Scholar 
van Grouw, H. What colour is that bird? The causes and recognition of common colour aberrations in birds. Br. Birds 106, 17–29 (2013).
Google Scholar 
Slagsvold, T., Rofstad, G. & Sandvik, J. Partial albinism and natural selection in the hooded crow Corvus corone cornix. J. Zool. 214, 157–166 (1988).
Google Scholar 
Stevens, M. et al. Revealed by conspicuousness: distractive markings reduce camouflage. Behav. Ecol. 24, 213–222 (2013).
Google Scholar 
van Grouw, H. What’s in a name? Nomenclature for colour aberrations in birds reviewed. Bull. Br. Ornithol. Club 141, 276–299 (2021).
Google Scholar 
Parsons, G. J. & Bonderup-Nielsen, S. Partial albinism in an island population of Meadow Voles, Microtus pennsylvanicus, from Nova Scotia. Can. Field-Nat. 109, 263–264 (1995).
Google Scholar 
Reis, A. da S., Zampaulo, R. de A. & Talamoni, S. A. Frequency of leucism in a colony of Anoura geoffroyi (Chiroptera: Phyllostomidae) roosting in a ferruginous cave in Brazil. Biota Neotropica 19(3): e20180676. https://doi.org/10.1590/1676-0611-BN-2018-0676 (2019).Jehl, J. R. Leucism in Eared Grebes in western north America. Condor 87, 439–441 (1985).
Google Scholar 
Forrest, S. & Naveen, R. Prevalence of leucism in Pygoscelid penguins of the Antarctic peninsula. Waterbirds 23, 283–285 (2000).
Google Scholar 
González-Ortegón, E., Drake, P., Quigley, D. T. G. & Cuesta, J. A. Leucism in the European sardine Sardina pilchardus (Clupeidae). Ecol. Indic. 117, 106544 (2020).
Google Scholar 
David, B. Z. First report of partial leucism in the poison frog Epipedobates anthonyi (Anura: Dendrobatidae) in El Oro, Ecuador. Neotrop. Biodivers. 7, 1–4 (2021).
Google Scholar 
Krecsák, L. Albinism and leucism among European Viperinae: a review. Russ. J. Herpetol. 15, 97–102 (2008).
Google Scholar 
Ritland, K., Newton, C. & Marshall, H. D. Inheritance and population structure of the white-phased “Kermode” black bear. Curr. Biol. 11, 1468–1472 (2001).CAS 
PubMed 

Google Scholar 
Galván, I., Bijlsma, R. G., Negro, J. J., Jarén, M. & Garrido-Fernández, J. Environmental constraints for plumage melanization in the northern goshawk Accipiter gentilis. J. Avian Biol. 41, 523–531 (2010).
Google Scholar 
Pijpe, A., Gardien, K. L. M., Meijeren-Hoogendoorn, R. E. van, Middelkoop, E. & Zuijlen, P. P. M. van. Scar Symptoms: Pigmentation Disorders in Textbook On Scar Management (eds. Téot, L., Mustoe, T. A., Middelkoop, E. & Gauglitz, G. G.) 109–115 (Springer, 2020).Edelaar, P. et al. Apparent selective advantage of leucism in a coastal population of Southern caracaras (Falconidae). Evol. Ecol. Res. 13, 187–196 (2011).
Google Scholar 
Ellegren, H., Lindgren, G., Primmer, C. R. & Møller, A. P. Fitness loss and germline mutations in barn swallows breeding in Chernobyl. Nature 389, 593–596 (1997).ADS 
CAS 
PubMed 

Google Scholar 
Benítez-López, A. & García-Egea, I. First record of an aberrantly colored Pin-tailed Sandgrouse (Pterocles alchata). Wilson J. Ornithol. 127, 755–759 (2015).
Google Scholar 
Zbyryt, A., Mikula, P., Ciach, M., Morelli, F. & Tryjanowski, P. A large-scale survey of bird plumage colour aberrations reveals a collection bias in Internet-mined photographs. Ibis 163, 566–578 (2020).
Google Scholar 
Bensch, S., Hansson, B., Hasselquist, D. & Nielsen, B. Partial albinism in a semi-isolated population of Great Reed Warblers. Hereditas 133, 167–170 (2000).CAS 
PubMed 

Google Scholar 
Izquierdo, L. et al. Factors associated with leucism in the common blackbird Turdus merula. J. Avian Biol. 49, e01778 (2018).
Google Scholar 
Møller, A. P. & Mousseau, T. A. Albinism and phenotype of barn swallows (Hirundo rustica) from Chernobyl. Evolution 55, 2097–2104 (2001).PubMed 

Google Scholar 
Troscianko, J., Wilson-Aggarwal, J., Stevens, M. & Spottiswoode, C. N. Camouflage predicts survival in ground-nesting birds. Sci. Rep. 6, 1–8 (2016).
Google Scholar 
Aragonés, J., Arias de Reyna, L. & Recuerda, P. Visual communication and sexual selection in a nocturnal bird species, Caprimulgus ruficollis, a balance between crypsis and conspicuousness. Wilson Bull. 111, 340–345 (1999).
Google Scholar 
Negro, J. J., Bortolotti, G. R. & Sarasola, J. H. Deceptive plumage signals in birds: manipulation of predators or prey? Biol. J. Linn. Soc. 90, 467–477 (2007).
Google Scholar 
Brooke, M. de L. Unexplained recurrent features of the plumage of birds. Ibis 152, 845–847 (2010).Forero, M. G., Tella, J. L. & García, L. Age related evolution of sexual dimorphism in the Red-necked Nightjar Caprimulgus ruficollis. J. Ornithol. 136, 447–451 (1995).
Google Scholar 
Camacho, C. Early age at first breeding and high natal philopatry in the Red-necked Nightjar Caprimulgus ruficollis. Ibis 156, 442–445 (2014).
Google Scholar 
Camacho, C. et al. The road to opportunities: landscape change promotes body-size divergence in a highly mobile species. Curr. Zool. 62, 7–14 (2016).PubMed 
PubMed Central 

Google Scholar 
Forero, M. G., Tella, J. L. & Oro, D. Annual survival rates of adult Red-necked Nightjars Caprimulgus ruficollis. Ibis 143, 273–277 (2001).
Google Scholar 
Henner, J. et al. Genetic mapping of the (G)-locus responsible for the coat color phenotype “Progressive Greying with Age” in horses (Equus caballus). Mamm. Genome 13, 535–537 (2002).CAS 
PubMed 

Google Scholar 
Edson, J. M. An epidemic of albinism. Auk 45, 377–378 (1928).
Google Scholar 
Camacho, C., Palacios, S., Sáez, P., Sánchez, S. & Potti, J. Human-induced changes in landscape configuration influence individual movement routines: lessons from a versatile, highly mobile species. PLoS ONE 9, e104974 (2014).ADS 
PubMed 
PubMed Central 

Google Scholar 
Enders, F. & Post, W. White-spotting in the genus Ammospiza and other grassland sparrows. Bird-Band. 42, 210–219 (1971).
Google Scholar 
Sage, B. L. Albinism and melanism in birds. Br. Birds 55, 201–225 (1962).
Google Scholar 
O’Sullivan, J. D. B. et al. The biology of human hair greying. Biol. Rev. 96, 107–128 (2021).PubMed 

Google Scholar 
Nichols, J. D., Hines, J. E. & Blums, P. Tests for senescent decline in annual survival probabilities of common pochards, Aythya ferina. Ecology 78, 1009–1018 (1997).
Google Scholar 
Owen, M. & Skimmings, P. The occurrence and performance of leucistic Barnacle Geese Branta leucopsis. Ibis 134, 22–26 (1992).
Google Scholar 
Mulder, T., Campbell, C. J. & Ruxton, G. D. Evaluation of disruptive camouflage of avian cup-nests. Ibis 163, 150–158 (2021).
Google Scholar 
Holyoak, D. Variable albinism of the flight feathers as an adaptation for recognition of individual birds in some Polynesian populations of Acrocephalus warblers. Ardea 66, 112–117 (1978).
Google Scholar 
Griffith, S. C., Parker, T. H. & Olson, V. A. Melanin- versus carotenoid-based sexual signals: is the difference really so black and red? Anim. Behav. 71, 749–763 (2006).
Google Scholar 
Galván, I., Jorge, A., Nielsen, J. T. & Møller, A. P. Pheomelanin synthesis varies with protein food abundance in developing goshawks. J. Comp. Physiol. B 189, 441–450 (2019).PubMed 

Google Scholar 
Zaragoza-Trello, C., Vilà, M., Botías, C. & Bartomeus, I. Interactions among global change pressures act in a non-additive way on bumblebee individuals and colonies. Funct. Ecol. 35, 420–434 (2021).
Google Scholar 
Rollin, N. A note on abnormally marked Song Thrushes and Blackbirds. Trans. Nat. Hist. Soc. Northumberl. Durh. Newctle upon Tyne 10, 183–184 (1953).Guerrero-Bosagna, C. et al. Transgenerational epigenetic inheritance in birds. Environ. Epigenet. 4, dvy008 (2018).Camacho, C., Negro, J. J., Redondo, I., Palacios, S. & Sáez-Gómez, P. Correlates of individual variation in the porphyrin-based fluorescence of red-necked nightjars (Caprimulgus ruficollis). Sci. Rep. 9, 1–9 (2019).
Google Scholar 
Camacho, C. Tropical phenology in temperate regions: extended breeding season in a long-distance migrant. Condor 115, 830–837 (2013).
Google Scholar 
Cleere, N. Nightjars: a guide to nightjars and related birds (A&C Black, London, 2010).
Google Scholar 
Gargallo, G. Flight feather moult in the red-necked nightjar Caprimulgus ruficollis. J. Avian Biol. 25, 119–124 (1994).
Google Scholar 
Jackson, H. D. A field survey to investigate why nightjars frequent roads at night. Ostrich 74, 97–101 (2003).
Google Scholar 
Jackson, H. D. Finding and trapping nightjars. Bokmakierie 36, 86–89 (1984).
Google Scholar 
Sénar, J. C. & Pascual, J. Keel and tarsus length may provide a good predictor of avian body size. Ardea 85, 269–274 (1997).
Google Scholar 
Svensson, L. Identification Guide To European Passerines (Lars Svensson, Cleveland, 1992).
Google Scholar 
van de Pol, M. & Wright, J. A simple method for distinguishing within-versus between-subject effects using mixed models. Anim. Behav. 77, 753–758 (2009).
Google Scholar 
Schielzeth, H. & Forstmeier, W. Conclusions beyond support: overconfident estimates in mixed models. Behav. Ecol. 20, 416–420 (2009).PubMed 

Google Scholar 
Rising, J. D. & Somers, K. M. The measurement of overall body size in birds. Auk 106, 666–674 (1989).
Google Scholar 
Magnusson, A. et al. Package “glmmTMB”. R Package Version 0.2.0. (2017).Hartig, F. DHARMa: residual diagnostics for hierarchical (multi-level/mixed) regression models. R package version 0.2, 4. (2019).Nakagawa, S. & Schielzeth, H. A general and simple method for obtaining R2 from generalized linear mixed-effects models. Methods Ecol. Evol. 4, 133–142 (2013).
Google Scholar 
Barton, K. MuMIn: Multi-Model inference. Model selection and model averaging based on information criteria (AICc and alike). R package version 1.43.17. (2020). More

Boutaiba, S., Hacene, H., Bidle, K. A. & Maupin-Furlow, J. A. Microbial diversity of the hypersaline Sidi Ameur and Himalatt Salt Lakes of the Algerian Sahara. J. Arid Environ. 75, 909–916. https://doi.org/10.1016/j.jaridenv.2011.04.010 (2011).ADS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Ventosa, A. Unusual micro-organisms from unusual habitats: hypersaline environments. Symposia Society for General Microbiology (2006).Fukuchi, S., Yoshimune, K., Wakayama, M., Moriguchi, M. & Nishikawa, K. Unique amino acid composition of proteins in halophilic bacteria. J. Mol. Biol. 327, 347–357 (2003).CAS 
Article 

Google Scholar 
Pillai, S. D., Nakatsu, C. H., Miller, R. V. & Yates, M. V. Manual of environmental microbiology. Life High-Salinity Environ. https://doi.org/10.1128/9781555818821 (2015).Article 

Google Scholar 
Poli, A. et al. Microbial diversity in extreme marine habitats and their biomolecules. Microorganisms 5, 25. https://doi.org/10.3390/microorganisms5020025 (2017).CAS 
Article 
PubMed Central 

Google Scholar 
Azpiazu-Muniozguren, M., Martinez-Ballesteros, I., Gamboa, J., Seoane, S. & Bikandi, J. Altererythrobacter muriae sp. nov., isolated from hypersaline Aana Salt Valley spring water, a continental thalassohaline-type solar saltern. Int. J. Syst. Evol. Microbiol. 71, 3 (2021).
Google Scholar 
Zhang, J. et al. Bacterial diversity in Bohai Bay Solar Saltworks, China. Curr. Microbiol. 72, 55–63 (2016).CAS 
Article 

Google Scholar 
Highfield, A., Ward, A., Pipe, R. & Schroeder, D. C. Molecular and phylogenetic analysis reveals new diversity of Dunaliella salina from hypersaline environments. J. Mar. Biol. Assoc. UK 101, 27–37. https://doi.org/10.1017/s0025315420001319 (2021).CAS 
Article 

Google Scholar 
Cycil, L. M. et al. Metagenomic insights into the diversity of halophilic microorganisms indigenous to the Karak Salt Mine, Pakistan. Front. Microbiol. https://doi.org/10.3389/fmicb.2020.01567 (2020).Article 
PubMed 
PubMed Central 

Google Scholar 
Jacob, J. H., Hussein, E. I., Shakhatreh, M. A. K. & Cornelison, C. T. Microbial community analysis of the hypersaline water of the Dead Sea using high-throughput amplicon sequencing. Microbiol. Open 6, e00500. https://doi.org/10.1002/mbo3.500 (2017).CAS 
Article 

Google Scholar 
Ben Abdallah, M. et al. Abundance and diversity of prokaryotes in ephemeral hypersaline lake Chott El Jerid using Illumina Miseq sequencing, DGGE and qPCR assay. Extremophiles 22, 811–823. https://doi.org/10.1007/s00792-018-1040-9 (2018).CAS 
Article 
PubMed 

Google Scholar 
Tazi, L., Breakwell, D. P., Harker, A. R. & Crandall, K. A. Life in extreme environments: Microbial diversity in Great Salt Lake, Utah. Extremophiles 18, 525–535. https://doi.org/10.1007/s00792-014-0637-x (2014).Article 
PubMed 

Google Scholar 
Kashi, F. J., Owlia, P., Amoozegar, M. A. & Kazemi, B. Halophilic prokaryotes in Urmia Salt Lake, a hypersaline environment in Iran. Curr. Microbiol. 78(8), 3230–3238 (2021).Article 

Google Scholar 
Sorokin, D. Y., Roman, P. & Kolganova, T. V. Halo(natrono)archaea from hypersaline lakes can utilize sulfoxides other than DMSO as electron acceptors for anaerobic respiration. Extremophiles 25, 173–180 (2021).CAS 
Article 

Google Scholar 
Hwang, K., Choe, H. & Kim, K. M. Complete genome of Nocardioides aquaticus KCTC 9944T isolated from meromictic and hypersaline Ekho Lake, Antarctica. Mar. Genom. 1, 100889 (2021).Article 

Google Scholar 
Didari, M. et al. Diversity of halophilic and halotolerant bacteria in the largest seasonal hypersaline lake (Aran-Bidgol-Iran). J. Environ. Health Sci. Eng. 18, 961–971. https://doi.org/10.1007/s40201-020-00519-3 (2020).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Oren, A. Diversity of halophilic microorganisms: Environments, phylogeny, physiology, and applications. J. Ind. Microbiol. Biotechnol. 28, 56–63 (2002).CAS 
Article 

Google Scholar 
Mutlu, M. B. et al. Prokaryotic diversity in Tuz Lake, a hypersaline environment in Inland Turkey. FEMS Microbiol. Ecol. 65, 474–483. https://doi.org/10.1111/j.1574-6941.2008.00510.x (2008).CAS 
Article 
PubMed 

Google Scholar 
Antón, J. et al. Distribution, abundance and diversity of the extremely halophilic bacterium Salinibacter ruber. Saline Syst. 4, 15. https://doi.org/10.1186/1746-1448-4-15 (2008).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Oren, A. Microbial life at high salt concentrations: phylogenetic and metabolic diversity. Saline Syst. 4, 2. https://doi.org/10.1186/1746-1448-4-2 (2008).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Abdeljabbar, H., Badiaa, E., Jean-Luc, C., Marie-Laure, F. & Najla, S. Prokaryotic biodiversity of halophilic microorganisms isolated from Sehline Sebkha Salt Lake (Tunisia). Afr. J. Microbiol. Res. 8, 355–367. https://doi.org/10.5897/ajmr12.1087 (2014).Article 

Google Scholar 
Najjari, A., Elshahed, M. S., Cherif, A., Youssef, N. H. & Löffler, F. E. Patterns and determinants of halophilic archaea (Class Halobacteria) diversity in Tunisian endorheic salt lakes and Sebkhet systems. Appl. Environ. Microbiol. 81, 4432–4441. https://doi.org/10.1128/aem.01097-15 (2015).ADS 
CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Aharon, O. The ecology of the extremely halophilic archaea. FEMS Microbiol. Rev. 1, 415–440 (1994).
Google Scholar 
Oren, A. Halophilic Archaea. FEMS Microbiol. Rev. https://doi.org/10.1016/b978-0-12-809633-8.20800-5 (2019).Article 

Google Scholar 
Feng, Y. et al. The evolutionary origins of extreme halophilic archaeal lineages. Genome Biol. Evol. 13, 8. https://doi.org/10.1093/gbe/evab166 (2021).CAS 
Article 

Google Scholar 
Ventosa, A., Nieto, J. J. & Oren, A. Biology of moderately halophilic aerobic bacteria. Microbiol. Mol. Biol. Rev. 62, 504–544 (1998).CAS 
Article 

Google Scholar 
Kushner, D. J. Halophilic bacteria. Adv. Appl. Microbiol. 10, 73–99 (1968).CAS 
Article 

Google Scholar 
Ghozlan, H., Deif, H., Kandil, R. A. & Sabry, S. Biodiversity of moderately halophilic bacteria in hypersaline habitats in Egypt. J. Gen. Appl. Microbiol. 52, 63–72 (2006).CAS 
Article 

Google Scholar 
Ali, I., Prasongsuk, S., Akbar, A., Aslam, M. & Rakshit, S. K. Hypersaline habitats and halophilic microorganisms. Maejo Int. J. Sci. Technol. 10, 330–345 (2016).CAS 

Google Scholar 
Margesin, R. & Schinner, F. Biodegradation and bioremediation of hydrocarbons in extreme environments. Appl. Microbiol. Biotechnol. 56, 650–663. https://doi.org/10.1007/s002530100701 (2001).CAS 
Article 
PubMed 

Google Scholar 
Poosarla, V. G. & Ts, C. Xylanase production by halophilic bacterium Gracilibacillus sp. TSCPVG under solid state fermentation. Res. J. Biotechnol. 16, 92–100 (2021).Article 

Google Scholar 
Foti, M. et al. Diversity, activity, and abundance of sulfate-reducing bacteria in saline and hypersaline soda lakes. Appl. Environ. Microbiol. 73, 2093–2100. https://doi.org/10.1128/aem.02622-06 (2007).ADS 
CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Boujelben, I. et al. Spatial and seasonal prokaryotic community dynamics in ponds of increasing salinity of Sfax solar saltern in Tunisia. Antonie Van Leeuwenhoek 101, 845–857. https://doi.org/10.1007/s10482-012-9701-7 (2012).CAS 
Article 
PubMed 

Google Scholar 
García-Maldonado, J. Q., Bebout, B. M., Everroad, R. C. & López-Cortés, A. Evidence of novel phylogenetic lineages of methanogenic archaea from hypersaline microbial mats. Microb. Ecol. 69, 106–117. https://doi.org/10.1007/s00248-014-0473-7 (2014).CAS 
Article 
PubMed 

Google Scholar 
Abed, R. M. M., de Beer, D. & Stief, P. Functional-structural analysis of nitrogen-cycle bacteria in a hypersaline mat from the omani desert. Geomicrobiol. J. 32, 119–129. https://doi.org/10.1080/01490451.2014.932033 (2014).CAS 
Article 

Google Scholar 
Coban, O., Rasigraf, O., Jong, A., Spott, O. & Bebout, B. M. Quantifying potential N turnover rates in hypersaline microbial mats by 15 N tracer techniques. Appl. Environ. Microbiol. 87, 8 (2021).Article 

Google Scholar 
Rodriguez-Valera, F. Introduction to Saline Environments (Springer, 1993).
Google Scholar 
Wei, H. C., Qi-Shun, F., Fu-Yuan, A., Fa-Shou, S. & Qin, Z. J. Chemical elements in core sediments of the qarhan salt lake and palaeoclimate evolution during 94–9 ka. Acta Geosci. Sin. (2016).Yu, S., Liu, X., Tan, H. & Cao, G. Sustainable Utilization of Qarhan Salt Lake Resources 27–265 (Science Press, 2009).
Google Scholar 
Zhu, D. et al. An evaluation of the core bacterial communities associated with hypersaline environments in the Qaidam Basin, China. Arch. Microbiol. 202, 2093–2103. https://doi.org/10.1007/s00203-020-01927-7 (2020).CAS 
Article 
PubMed 

Google Scholar 
Liu, W., Jiang, H., Yang, J. & Wu, G. Gammaproteobacterial diversity and carbon utilization in response to salinity in the lakes on the qinghai-tibetan plateau. Geomicrobiol. J. 35, 392–403. https://doi.org/10.1080/01490451.2017.1378951 (2018).CAS 
Article 

Google Scholar 
Zhong, Z.-P. et al. Prokaryotic community structure driven by salinity and ionic concentrations in plateau lakes of the tibetan plateau. Appl. Environ. Microbiol. 82, 1846–1858. https://doi.org/10.1128/aem.03332-15 (2016).ADS 
CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
He, C. et al. Synergistic effect of magnetite and zero-valent iron on anaerobic degradation and methanogenesis of phenol. Biores. Technol. 291, 121874. https://doi.org/10.1016/j.biortech.2019.121874 (2019).CAS 
Article 

Google Scholar 
Caporaso, J. G. et al. QIIME allows analysis of high-throughput community sequencing data. Nat. Methods 7, 335–336. https://doi.org/10.1038/nmeth.f.303 (2010).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Martin, M. Cutadapt removes adapter sequences from high-throughput sequencing reads. Embnet J. 17, 10–12 (2011).Article 

Google Scholar 
Zhang, J., Kassian, K., Tomáš, F. & Alexandros, S. PEAR: a fast and accurate Illumina Paired-End reAd mergeR. Bioinformatics 30, 614 (2014).CAS 
Article 

Google Scholar 
Schmieder, R. & Edwards, R. Quality control and preprocessing of metagenomic datasets. Bioinformatics 27, 863–864 (2011).CAS 
Article 

Google Scholar 
Edgar, R. C. UPARSE: Highly accurate OTU sequences from microbial amplicon reads. Nat. Methods 10, 996–1000 (2013).CAS 
Article 

Google Scholar 
Schloss, P. D. et al. Introducing mothur: Open-source, platform-independent, community-supported software for describing and comparing microbial communities. Appl. Environ. Microbiol. 75, 7537 (2009).ADS 
CAS 
Article 

Google Scholar 
Chen, H. & Boutros, P. C. VennDiagram: A package for the generation of highly-customizable Venn and Euler diagrams in R. BMC Bioinform. 12, 35. https://doi.org/10.1186/1471-2105-12-35 (2011).Article 

Google Scholar 
McArdle, B. H. et al. Fitting multivariate models to community data: A comment on distance-based redundancy analysis. Ecology 82, 290–290 (2001).Article 

Google Scholar 
Langille, M. G. I. et al. Predictive functional profiling of microbial communities using 16S rRNA marker gene sequences. Nat. Biotechnol. 31, 814–821. https://doi.org/10.1038/nbt.2676 (2013).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Louca, S. & Doebeli, M. Efficient comparative phylogenetics on large trees. Bioinformatics 34, 1–3 (2017).
Google Scholar 
Louca, S., Parfrey, L. W. & Doebeli, M. Decoupling function and taxonomy in the global ocean microbiome. Science 353, 1272 (2016).ADS 
CAS 
Article 

Google Scholar 
Junker, B. H. & Schreiber, F. Analysis of Biological Networks 283–304 (Analysis of biological networks, 2008).Book 

Google Scholar 
Faust, K. & Raes, J. Microbial interactions: From networks to models. Nat. Rev. Microbiol. 10, 538–550. https://doi.org/10.1038/nrmicro2832 (2012).CAS 
Article 
PubMed 

Google Scholar 
Behzad, H., Ibarra, M. A., Mineta, K. & Gojobori, T. Metagenomic studies of the Red Sea. Gene 576, 717–723. https://doi.org/10.1016/j.gene.2015.10.034 (2016).CAS 
Article 
PubMed 

Google Scholar 
Naghoni, A. et al. Microbial diversity in the hypersaline Lake Meyghan, Iran. Sci. Rep. https://doi.org/10.1038/s41598-017-11585-3 (2017).Article 
PubMed 
PubMed Central 

Google Scholar 
Kambura, A. K. et al. Bacteria and Archaea diversity within the hot springs of Lake Magadi and Little Magadi in Kenya. BMC Microbiol. https://doi.org/10.1186/s12866-016-0748-x (2016).Article 
PubMed 
PubMed Central 

Google Scholar 
Paul, D. et al. Exploration of microbial diversity and community structure of Lonar lake: The only hypersaline meteorite crater lake within basalt rock. Front. Microbiol. https://doi.org/10.3389/fmicb.2015.01553 (2016).Article 
PubMed 
PubMed Central 

Google Scholar 
Ventosa, A., de la Haba, R. R., Sánchez-Porro, C. & Papke, R. T. Microbial diversity of hypersaline environments: A metagenomic approach. Curr. Opin. Microbiol. 25, 80–87. https://doi.org/10.1016/j.mib.2015.05.002 (2015).CAS 
Article 
PubMed 

Google Scholar 
Liu, F. H. et al. Bacterial and archaeal assemblages in sediments of a large shallow freshwater lake, Lake Taihu, as revealed by denaturing gradient gel electrophoresis. J. Appl. Microbiol. 106, 1022–1032. https://doi.org/10.1111/j.1365-2672.2008.04069.x (2009).CAS 
Article 
PubMed 

Google Scholar 
Song, H., Li, Z., Du, B., Wang, G. & Ding, Y. Bacterial communities in sediments of the shallow Lake Dongping in China. J. Appl. Microbiol. 112, 79–89. https://doi.org/10.1111/j.1365-2672.2011.05187.x (2012).CAS 
Article 
PubMed 

Google Scholar 
Wu, Q. L., Zwart, G., Schauer, M., Agterveld, K. V. & Hahn, M. W. Bacterioplankton community composition along a salinity gradient of sixteen high-mountain lakes located on the Tibetan Plateau, China. Appl. Environ. Microbiol. 72, 5478–5485 (2006).ADS 
CAS 
Article 

Google Scholar 
Xing, P., Hahn, M. W. & Wu, Q. L. Low taxon richness of bacterioplankton in high-altitude lakes of the Eastern Tibetan Plateau, with a predominance of bacteroidetes and Synechococcus spp. Appl. Environ. Microbiol. 75, 7017–7025. https://doi.org/10.1128/aem.01544-09 (2009).ADS 
CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Liu, Y. et al. Bacterial diversity of freshwater Alpine Lake Puma Yumco on the Tibetan Plateau. Geomicrobiol. J. 26, 131–145. https://doi.org/10.1080/01490450802660201 (2009).CAS 
Article 

Google Scholar 
MounÃc, S., Caumette, P., Matheron, R. & Willison, J. C. Molecular sequence analysis of prokaryotic diversity in the anoxic sediments underlying cyanobacterial mats of two hypersaline ponds in Mediterranean salterns. FEMS Microbiol. Ecol. 44, 117–130. https://doi.org/10.1016/s0168-6496(03)00017-5 (2003).Article 

Google Scholar 
Valenzuela-Encinas, C. et al. Changes in the bacterial populations of the highly alkaline saline soil of the former lake Texcoco (Mexico) following flooding. Extremophiles 13, 609–621. https://doi.org/10.1007/s00792-009-0244-4 (2009).Article 
PubMed 

Google Scholar 
Kim, T. J., Lee, E. Y., Kim, Y. J., Cho, K.-S. & Ryu, H. W. Degradation of polyaromatic hydrocarbons by Burkholderia cepacia 2A–12. World J. Microbiol. Biotechnol. 19, 411–417. https://doi.org/10.1023/A:1023998719787 (2003).CAS 
Article 

Google Scholar 
Gales, G. et al. Preservation of ancestral Cretaceous microflora recovered from a hypersaline oil reservoir. Sci. Rep. https://doi.org/10.1038/srep22960 (2016).Article 
PubMed 
PubMed Central 

Google Scholar 
Kleinsteuber, S., Riis, V., Fetzer, I., Harms, H. & Müller, S. Population dynamics within a microbial consortium during growth on diesel fuel in saline environments. Appl. Environ. Microbiol. 72, 3531–3542. https://doi.org/10.1128/aem.72.5.3531-3542.2006 (2006).ADS 
CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Valenzuela-Encinas, C. et al. The archaeal diversity and population in a drained alkaline saline soil of the former lake Texcoco (Mexico). Geomicrobiol. J. 29, 18–22. https://doi.org/10.1080/01490451.2010.520075 (2012).Article 

Google Scholar 
He, S., Tan, J., Hu, W. & Mo, C. Diversity of archaea and its correlation with environmental factors in the Ebinur Lake Wetland. Curr. Microbiol. 76, 1417–1424. https://doi.org/10.1007/s00284-019-01768-8 (2019).CAS 
Article 
PubMed 

Google Scholar 
Sandaa, R. A., Enger, O. & Torsvik, V. Abundance and diversity of Archaea in heavy-metal-contaminated soils. Appl. Environ. Microbiol. 65, 3293–3297 (1999).ADS 
CAS 
Article 

Google Scholar 
Dave, B. P. & Soni, A. Diversity of halophilic archaea at salt pans around Bhavnagar Coast, Gujarat. Proc. Natl. Acad. Sci. India B 83, 225–232. https://doi.org/10.1007/s40011-012-0124-z (2012).Article 

Google Scholar 
Zafrilla, B., Martínez-Espinosa, R., Alonso, M. A. & Bonete, M. J. Biodiversity of Archaea and floral of two inland saltern ecosystems in the Alto Vinalopó Valley, Spain. Saline Syst. 6, 10 (2010).Article 

Google Scholar 
Costa, M., Santos, H. & Galinski, E. A. An overview of the role and diversity of compatible solutes in Bacteria and Archaea. Adv. Biochem. Eng. Biotechnol. 61, 117 (1998).PubMed 

Google Scholar 
Williams, R. J., Howe, A. & Hofmockel, K. S. Demonstrating microbial co-occurrence pattern analyses within and between ecosystems. Front. Microbiol. https://doi.org/10.3389/fmicb.2014.00358 (2014).Article 
PubMed 
PubMed Central 

Google Scholar 
Schmidt, T. S. B., MatiasRodrigues, J. F. & von Mering, C. A family of interaction-adjusted indices of community similarity. ISME J. 11, 791–807. https://doi.org/10.1038/ismej.2016.139 (2016).Article 
PubMed 
PubMed Central 

Google Scholar 
Oyewusi, H. A. et al. Functional profiling of bacterial communities in Lake Tuz using 16S rRNA gene sequences. Biotechnol. Biotechnol. Equip. 35, 1–10. https://doi.org/10.1080/13102818.2020.1840437 (2020).CAS 
Article 

Google Scholar  More