Philippa Kaur

Perched among the fronds of the world’s loneliest tree, Viswambharan Sarasan had an important decision to make. Sarasan had worked for years to get access to this palm — the last living member of the species Hyophorbe amaricaulis, which grows in Curepipe Botanic Gardens, Mauritius.He reached up towards a cluster of its walnut-sized, olive-green fruit. Sarasan, a botanist at the Royal Botanic Gardens at Kew, near London, had been through sensitive negotiations for permission to take the fruit, each with one crucial seed inside. He then had to wait for the tree, nicknamed the lonesome palm, to produce them. Nine metres up, 50 fruit dangling within his grasp, he had to decide how many to take: enough to give himself a chance of culturing them back at Kew, while leaving enough for local scientists to work with.“It was the only shot I could get,” he says of his visit in June 2006. “But I didn’t want to take all the seeds and then it turns out badly.”He picked ten fruit. It was not his lucky number.When the plight of trees gets publicity, deforestation is generally the reason, but it is not the only crisis they face. Nearly one-third of trees — more than 17,500 species — are threatened with extinction. This is more than twice the number of threatened mammals, birds, amphibians and reptiles combined1. Mass plantings of trees, paradoxically, often add to the problem by using single species. Now, hundreds of plant conservationists globally are fighting to save the trees speeding towards extinction.“We shouldn’t be giving up on any tree species,” says Paul Smith, head of Botanic Gardens Conservation International (BGCI), a London-based charity that co-leads the campaign to secure the future of the world’s threatened tree species.But time is short, the obstacles are formidable and both climate change and fashions in ecology are moving against them.Peter Bridgewater, a specialist in biodiversity governance at the University of Canberra, Australia, says that finding a natural home for every tree species is impossible because climate change is altering ecosystems so fundamentally. Scientists who think this goal is realistic are “living in their own cloud cuckoo land”, he says.

The ‘lonesome palm’, which lives in the Curepipe Botanic Gardens in Mauritius, is the last surviving member of the species Hyophorbe amaricaulis. Researchers have tried for years to help it reproduce, without success.Credit: Vincent Florens

Neglected trees Inextricably linked with the problem of climate change, and equally as damaging, is the disappearance of species from Earth. The rate of extinction is at historic levels and accelerating, with around one million animals and plants under threat.The plight of trees can get lost among the tales of endangered mammals or birds. To get trees more visibility, in 2016 the BGCI, working with the International Union for Conservation of Nature (IUCN), organized the largest conservation assessment in the IUCN’s history: the Global Tree Assessment. Hundreds of plant conservationists searched rainforests, mountains and strife-torn regions, sometimes with no more than a crinkly herbarium specimen or the testimony of a long-dead explorer to guide them.In a 2021 report, they announced that they had found 58,497 tree species, of which 17,510 were threatened2. Since then, almost 2,800 of those have been labelled as critically endangered. Some 142 species are thought to be extinct in the wild (see ‘Trees under threat’). This year, a separate group of modellers estimated that a further 9,000 tree species are undiscovered3.

Source: Ref. 2

It is not just the number of trees, but also their diversity that matters. A single species can be the foundation of an entire ecological network, and its disappearance could cause a cascade of extinctions that might lead to an ecosystem collapse.Strong, diverse ecosystems are also better at sequestering carbon, says Jean-Christophe Vié, director-general of the Franklinia Foundation, a private organization in Geneva, Switzerland, that funds tree conservation and supports the Global Tree Assessment. No tree species should be viewed as dispensable, says Vié, because it would set a precedent for every developer, farmer or logger to justify removing any threatened tree.But tree conservation has become lost in international biodiversity targets — partly because trees get subsumed into general plant-conservation goals, and because plants are generally less showy than birds and animals. Trees need to be assessed for ecologists to champion them, says Malin Rivers, head of conservation prioritization at the BGCI.“If you look at mammals, birds, reptiles, they have data to bring to the table when there is a policy discussion,” she says. “Taxonomy gives the species a name; conservation assessment gives it a voice.”Protect and propagateArmed with the Global Tree Assessment’s catalogue of threatened species, conservationists have begun prioritizing species and taxonomic groups. The best approach, says Smith, is to protect vulnerable trees in their natural habitats. If that’s not possible, researchers try growing them from seed in a laboratory, greenhouse or botanic garden.The Global Tree Assessment revealed that nearly two-thirds of threatened trees are found in areas that are already protected, and stressed that one important task is to strengthen or expand these havens.That might mean controlling grazing, implementing a national logging ban for a particular species or establishing plots on which the tree can be cultivated for fruit or flowers without harming the larger population. On the eastern Caribbean island of Dominica, for instance, where harvesting resin for incense was killing lansan trees (Protium attenuatum), a tweak to the tapping method has halted the damage.Sometimes, however, so few trees are left that protecting an area isn’t enough.

In its forest habitat in Tanzania, Karomia gigas is threatened by a seed-killing fungus.Credit: Kirsty Shaw/BGCI

In Tanzania, seed-biology specialist Fandey Mashimba works with a tiny population of a towering species called Karomia gigas. These trees, with their large oval leaves and distinctive, papery fruit, were thought to have gone extinct in the 1980s, but around six of them were discovered in 2011 by botanists from the University of Dar es Salaam. Protecting the habitat isn’t enough, because a fungus destroys their immature fruit. Mashimba, who oversees seed production for Tanzania’s Forest Service Agency, tries to whisk the fruit away before the fungus infects them, to sterilize and multiply the seeds for planting.Mashimba and his colleagues tried germinating hundreds of K. gigas seeds. The result: just three treasured plants, which Mashimba monitors through his office window as their giant leaves wave in the breeze. In 2018, the forestry service also dispatched 6,000 fruit to the Missouri Botanical Garden in St Louis. There, botanist Roy Gereau oversaw the extraction and cultivation of 24,000 seeds. The seeds produced only 30 plants. Last year, one sapling unfurled a small, pale purple flower, which perished within a day. When two trees flower simultaneously, botanists will attempt cross-pollination.

One of the 30 K. gigas plants at Missouri Botanical Garden flowered for a single day last year.Credit: Cassidy Moody/Missouri Botanical Garden

Mashimba is lucky in one respect: at least K. gigas produces seeds. Some trees produce none because their pollinators are gone; sometimes only one sex of a tree remains. For instance, most of the surviving specimens of the catkin yew (Amentotaxus argotaenia) in southern China are male. After a global search, a single female was discovered in the Royal Botanic Garden Edinburgh, UK; scientists there dispatched cuttings for planting near the surviving males. When they flower, reproduction can begin, says Gunter Fischer, a restoration ecologist at Missouri Botanical Garden. But this could take 30 years.Even if scientists do manage to acquire seeds from trees that are near extinction, germinating them can be tricky. Some go into dormancy, a protective state that, depending on the species, might be broken only through heating, cooling or scarring. Natural dormancy can last for years. Scientists try to circumvent it by culturing the embryo — the small section of a plant seed that will become the roots and stems — in a process known as embryo rescue.Every trick in the book The lonesome palm in the Curepipe Botanic Gardens — elderly, damaged and spindly — has seed problems, germination problems and more. It has resisted multiple rescuers since the 1980s. One obstacle is that the palm produces male and female flowers at different times, to avoid self-fertilization. Using a ladder and a brush, scientists override this process to collect, store and transfer pollen.It was the fruit of one such assisted-pollination project, each containing a single seed, that Sarasan carried back to Kew in 2006. He knew that lonesome palm seeds don’t grow if they are planted, so he used embryo rescue. With so few seeds, he felt there was no scope for experimenting with different culture media, so he made his best guess as to which blend to use.“I was so protective,” he says. “It was the responsibility, the excitement and also the fear of losing it.”The plantlets grew to 25 centimetres long. Then, one day, their fine white roots turned brown and they died, doubtless because of some nuance of the culture medium, he says.Other efforts have been derailed by mishap. In 2010, Kew horticultural scientist Carlos Magdalena negotiated to collect some freshly picked palm fruit while he was visiting Mauritius. Owing to a misunderstanding, two of the five fruit stored in a nearby fridge were eaten by a garden labourer who did not know their significance. Back at Kew, the seeds from the others failed to germinate.The failure rankles with Magdalena, who has a string of plant rescues to his name. As he roves the Kew greenhouses, steamy sanctuaries for plants that are bereft of a place in the wild, he sometimes feels he is all that stands between a species and its permanent loss.José Luis Marcelo Peña knows how he feels. In 2018, Marcelo Peña, a taxonomist at the National University of Jaén in Peru, was trekking through a steep, parched forest in Peru’s Marañón valley when he discovered a tree with light green flowers: Pradosia argentea, thought to be extinct.“It was a unique happiness that cannot be described,” says Marcelo Peña. Surveys yielded 200 more trees in the area, all of which were imminently threatened by agriculture.COVID-19 lockdowns began just as he attempted to save them. Without university facilities, but with remote help from the BGCI, he extracted 400 seeds from the purple fruit at home. More than 60 germinated: 20 survived. The following year, he tried again using fresh seeds, but a fungus got them all.As he finishes his story, he removes his glasses to wipe tears away. “It’s a big responsibility,” he says. And even with 20 little successes in the nursery, Marcelo Peña is concerned about the next step — reintroduction to the wild. Local people were unaware of P. argentea until recently, he says. They now support protecting the remaining trees — but they also need space to farm, which could put those survivors at risk.Back to the wildThriving in the wild is a distant dream for K. gigas, too. Tanzania’s forest agency and its partners are developing seed-propagation sites and nurseries for the species. But its future is uncertain, mostly because new trees could succumb to the same mysterious fungus.“We might have to content ourselves with saying, well, we have these lovely creatures in the zoo,” says Gereau.

A project at Missouri Botanical Garden produced 30 K. gigas plants.Credit: Cassidy Moody/Missouri Botanical Garden

Reintroductions can be spectacularly successful, however. The BGCI highlights a project on Malawi’s Mount Mulanje, the only natural home of the cypress Widdringtonia whytei. In 2019, just seven mature trees remained, the others victims of illegal felling. By 2022, thanks to a collaboration with Malawi’s Forestry Research Institute and local people, the slopes are alive again with 500,000 seedlings, and many locals now make a living through this endeavour.Propagation itself turned out to be fairly simple, says Smith. In Mauritius, by contrast, ecologists have a tougher task. The Mauritian Wildlife Foundation, with help from botanists elsewhere, is attempting to save multiple critically endangered species at once, but success at propagation varies widely. There have been some dramatic restorations, including of some species from which only a single tree remained. But the lonesome palm, now part of this project, continues to resist.

Scientists affix protective netting around hand-pollinated flowers on H. amaricaulis in Mauritius.Credit: Atmah Toocaram

A fourth attempt has begun. Nets hang around the tree to catch the male flowers and store their pollen for hand fertilization when the female flowers appear. In France, botanist Stéphane Buord at the National Botanical Conservatory of Brest hopes to overcome the problem that faced Sarasan — too few seeds — by tapping into the large quantities of seeds produced by Hyophorbe vaughanii, a close Mauritian relative of the lonesome palm. He and his team have spent years working out a complex technical protocol that coaxes its embryos into rooted seedlings that survive outside a test tube. Now he is waiting to try this approach on the seeds of the lonesome palm.If he succeeds, the palm might eventually be reintroduced into a national park or into the wild. Kersley Pynee, a conservation scientist at the Mauritius National Parks and Conservation Service, has reintroduced other trees and shrubs and says it is an uphill struggle. Plants can fall victim to fungi, pests and other assailants. After one recent planting of 1,000 seedlings of the flowering shrub Nesocodon mauritianus, just 5 now remain, he says.This is to be expected, says Smith. In nature, trees produce vast quantities of seeds, of which only a fraction germinate and survive because of natural dangers such as infestations, fire or competition for light or nutrients.Tree museumThe Global Trees Campaign has so far planted out hundreds of thousands of seedlings from 300 threatened tree species. But for trees that can no longer survive in the wild, the only other options are to keep a specimen in a living collection, or to store its seeds in a bank.One target of the 2011 Global Strategy for Plant Conservation, part of the Convention on Biological Diversity, was to conserve at least 75% of threatened plants in living collections or seed banks by 2020 — a goal that has not been met. What’s more, simply drying and freezing seeds doesn’t always work. Technologies such as cryopreservation — fast freezing at ultra-low temperatures — could offer an alternative, although it is expensive and impractical for many countries. And in 2018, conservationists warned4 that the seeds of one-third of tree species cannot be banked, largely because they don’t survive drying.Smith rejects this bleak diagnosis. Between seed banks, cryopreservation, nurseries, botanic gardens and arboreta, there are plenty of options to “buy us time”, he says.One trend that could help is mass tree-planting, in which governments and corporations plant trees to sequester carbon to meet emissions targets. Done badly, as many of these projects are, mass plantings can destroy biodiversity. Done well, they could rescue many species, says Smith. “This is a bandwagon we really need to jump on.”

This specimen of Encephalartos woodii, found in South Africa, was relocated in the late 1800s to the Royal Botanic Gardens at Kew, near London. It is the only one of this species to ever have been found in the wild.Credit: Andrew McRobb/RBG Kew

To help boost the usefulness of such projects to biodiversity, the BGCI and its partners have drawn up a certification programme for tree-planting projects called the Global Biodiversity Standard.Species conservation could also piggyback on the growing ecosystem-restoration movement. There are now more than 100,000 of these projects globally, helping ecosystems to capture carbon and provide essential services.Smith argues that including native species strengthens such projects. But restoration ecologists are often more concerned with overall function than with individual species, says Curt Meine, a historian of ecology at the Aldo Leopold Foundation in Baraboo, Wisconsin. And they want ecosystems to provide multiple services to humans, including sustainable livelihoods. Some acknowledge that tree conservation should have a place. “I do think it’s important work and we could learn a lot,” says Robin Chazdon, a restoration ecologist at the University of Connecticut in Storrs.But there are more threatened tree species than there are restoration projects to absorb them. “It’s not going to be the way of protecting all of those tree species,” she says.Some ecologists have deeper concerns. Bridgewater says that the efforts of conservationists and of restoration ecologists don’t factor in climate change.“They all in the end assume that nothing is going to be changing,” he says. But many trees, and whole ecosystems, just won’t survive in their current ranges, he says.“You could save every tree species but it will not be what people think — it will be in botanical gardens and larger managed conservation areas, and planting where it’s suitable for survival, not where it’s currently growing.”But the tree saviours are driven by something visceral: panic at the permanent loss of the rich, unique, irreplaceable and often-undeciphered identity of each species.“I don’t feel I am, as a humble human, here for a few decades on this planet, authorized to just cut off millions of years of evolutionary history,” says Vié. “Every species has a value.” More

Identification of a regime shift in Beaufort summer sea ice characteristicsFigure 2 shows time series of Beaufort Sea summer sea ice concentration, sea ice age, and sea ice thickness, as well as the ratio of Beaufort ice volume to that of the entire Arctic27. We define the summer to be the months of July, August, and September. A step function has been fit to the time series with a breakpoint determined by a minimization of the root-mean square fit to the data with a significance test of the difference of the means that takes into account the temporal autocorrelation of geophysical time series28; see Methods Section for further information. The first three metrics (Fig. 2a–c) indicate a transition toward less extensive, thinner and younger ice pack occurred around 2007. Furthermore, the Beaufort’s contribution to total Arctic ice volume decreased in 2007 from approximately 10% to 5% (Fig. 2d). We will refer in this study to the period from 2007-present as the “young ice regime,” while the period prior to 2007 will be referred to as the “old ice regime.” All metrics indicate that the summers of 2020 and 2021 (as well as 2013), stand out with ice characteristics above the mean for this new ice regime. This is especially true for the ice volume ratio where values for these past two summers approach those typical of conditions prior to the 2007 transition.Fig. 2: Characteristics of summer (July–September) Beaufort Sea ice.Time series of the: a sea ice concentration (%) from the NSIDC CDR dataset 1979–2021; b sea ice age (years) from the NSIDC dataset 1984–2021; c sea ice thickness (m) from PIOMAS 1979–2021 and d ratio of the volume of Beaufort sea ice to Arctic sea ice from PIOMAS 1979–2021. In all cases, the red lines represent the step function fit with the specified breakpoint that minimizes the root-mean square error in the fit. The statistical significance of the step is indicated in the legend.Full size imageSea ice conditions during 2019/2020 and 2020/2021Figure 3 shows time series of Beaufort Sea ice concentration and thickness for the 2-year period October 2019–September 2021, as well as climatological values for the first 13 years of the young ice regime (2007–2019) and anomalies with respect to these 13 years. The results show that starting in May of both years, concentration and thickness were both higher than the climatology by at least 1 standard deviation. The area-mean thickness anomaly was larger in 2020, while the sea ice concentration anomaly was larger in 2021.Fig. 3: Monthly mean Beaufort Sea ice characteristics from October 1 2019–September 30 2021.Time series (red curves) of the (a) monthly mean sea ice concentration (%) from the NSIDC CDR dataset and (b) monthly mean sea ice thickness (m) from PIOMAS with the climatological monthly mean values shown in black with one standard deviation above/below the mean indicated by the shading. The climatology is based on 2007–2019. In (c) and (d), the corresponding anomalies are shown with the shading representing +/− one standard deviation.Full size imageFigure 4 provides the Beaufort Sea ice thickness and age distributions in summer for (i) the first 13 years of the old ice regime (1979–1991) when the region was dominated by multi-year ice, (ii) the first 13 years of the young ice regime (2007–2019), (iii) the year 2020, and (iv) the year 2021. A kernel smoothing technique29 was used to fit the distributions to the data. The old ice regime was dominated by thick, old ice, with smaller contributions from thin, young ice. In contrast, the young ice regime is dominated by thin, young ice with a long “tail” of thick, old ice. The years 2020 and 2021 are representative of this young ice regime, although with thick and old ice generally ≥1 standard deviation above the mean (An exception is the amount of ice older than ~2 years in 2020, which is very close to the mean). Further analysis (Supplementary Fig. S1) indicates that many years in the young ice regime show small secondary peaks of thick or old ice (such as seen in the 2021 thickness distribution between 1.5 and 2 m). These “long-tailed” thickness and age distributions are similar to that found in summer 2020 in the Wandel Sea1. Thus, it seems that the Beaufort Sea is now dominated by thin, young ice, but a substantial component of thick, old ice remains. In the following sections, we examine the advective origins of this thick, old ice.Fig. 4: Frequency distribution of summer (July–September) sea ice characteristics in the region of interest.a PIOMAS sea ice thickness distribution and b NSIDC sea ice age distribution. Climatological distributions for 1979–1991 (1984–1991 for ice age) and 2007–2019 are shown as well as distributions for 2020 and 2021. The shading represents one standard deviation above/below the 2007–2019 mean.Full size imageImpact of sea ice transport on the observed anomalies during the summers of 2020 and 2021Recent work23,24,25 has emphasized the role that sea ice mass transport plays in determining the characteristics of pack ice in the Beaufort Sea. This transport can be decomposed into contributions from ice motion and from ice thickness; the seasonal climatology of these constituents as well as conditions during 2019/2020 and 2020/2021 are shown in Fig. 5. The climatology (Fig. 5a–d) indicates the presence of a seasonally varying anticyclonic Beaufort Gyre in the western Arctic as well as the presence of the thickest ice along the northern coast of Greenland and the Canadian Arctic Archipelago, i.e., the LIA. The spatial extent of the Beaufort Gyre is largest during the cool season, defined as fall (OND), winter (JFM) and spring (AMJ) when there is transport of ice from the LIA into the Beaufort Sea as well as transport of ice out of the Beaufort Sea into the Chukchi Sea. During summer (JAS), the Beaufort Gyre shrinks to only fill the Beaufort Sea.Fig. 5: Annual cycle in seasonal mean (OND: October–December; JFM: January–March; AMJ: April–June; JAS: July–September) sea ice thickness (shading – m) and sea ice motion (vectors- km/day).Results are shown for climatology (a–d) as well as 2019/2020 (e–h) and 2020/2021 (i–l). The polygon indicates the region along the Beaufort Coast over which statistics were computed. All fields are from PIOMAS.Full size imageThe situation during 2019/2020 (Fig. 5e–h) differs markedly from the climatology. During fall 2019 (Fig. 5e), the Beaufort Gyre was displaced southwestward with a small region of cyclonic ice motion at the boundary between the Chukchi and Beaufort Seas. Consistent with the collapse of the Beaufort High during winter 202026, ice motion during this period (Fig. 5f) is generally eastward in the Beaufort Sea and largely cyclonic over the entire Arctic Ocean. This results in ice transport from the Chukchi Sea into the Beaufort Sea and even beyond, i.e., into the LIA. In spring 2020 (Fig. 5g), transport continued from the Beaufort Sea to the LIA, although the Chukchi-to-Beaufort transport abated. By summer 2020, ice motion had reverted toward climatology (Fig. 5h).Conditions during 2020/2021 (Fig. 5i–l) were closer to climatology as compared to 2019/2020, although with some differences. Most notably during fall 2020 (Fig. 5i), the transport of thick, old ice from the LIA was restricted to a narrow region along the coast of the Canadian Arctic Archipelago, which appears to be linked to the presence of thick ice in the eastern Beaufort Sea. As discussed previously23, this strong transport continued into winter 2021 (Fig. 5j), although its width increased and thus broadly impacted the northeastern Beaufort Sea. There was also strong westward transport out of the Beaufort into the Chukchi Sea.Figure 6 shows the anomalies in sea ice motion, mass convergence, and thickness for the winters of 2020 and 2021 as well as the anomalies in sea-level pressure and 10 m wind fields for the same periods. The contrast in ice motion and sea ice thickness between the two winters is striking. During winter 2020 (Fig. 6a), anomalous cyclonic ice motion is evident as well as anomalously thick sea ice against Banks Island caused by convergence forced by eastward motion at this time (Fig. 6c, which actually started in fall 2019, Fig. 5f). Convergence also extends from the eastern Beaufort into the western LIA, where it acts to counter the long-term thinning trend; the result is enhanced negative ice thickness anomalies. This is supported by a comparison with winter 2021 (Fig. 6b, d), when thickness anomalies were much more negative and ice motion in the western LIA was closer to climatology, i.e., weakly divergent. Comparison of ice motion and thickness fields in the winters of 2020 and 2021 (Supplementary Fig. S2) demonstrates that the differences between these 2 years extend all the way from the Chukchi and Beaufort Seas into the western LIA.Fig. 6: Anomalous nature of the winter (JFM) sea ice and atmospheric circulation during 2020 and 2021.Sea ice thickness (shading – m) and sea ice motion (vectors- km/day) anomalies with respect to climatology (2007–2019) for: a 2020 and b 2021. Sea ice mass convergence (shading – m/month) and sea ice motion (vectors- km/day) anomalies with respect to climatology (2007–2019) for: c 2020 and d 2021. Sea-level pressure (contours – mb), 10 m wind (vectors- m/s) and 10 m wind speed (shading-m/s) anomalies with respect to climatology (1979–2021) for: e 2020 and f 2021. The polygon indicates the region along the Beaufort Coast over which statistics were computed. Sea ice fields are from PIOMAS. Atmospheric fields are from ERA5.Full size imageThe atmospheric circulation anomalies for these two winters highlight the role that sea-level pressure plays in forcing ice motion. During winter 2020 (Fig. 6e), the collapse of the Beaufort High26 resulted in lower sea-level pressures across the Arctic Ocean associated with a minimum 16 mb lower than climatology centered over the Barents Sea. Associated with this collapse, a cyclonic surface wind anomaly was present across the Arctic Ocean with a particularly high amplitude across the western boundary of the Beaufort Sea. In contrast, winter 2021 (Fig. 6f) was characterized by higher sea-level pressure over the Arctic Ocean with a maximum anomaly of 8 mb over the Barents Sea. As a result of this pressure perturbation, wind speeds were higher over the Arctic Ocean but did not reach the magnitudes observed during winter 2020.Quantifying the role of ice transport in anomalous Beaufort Sea ice conditions during the winters of 2019/2020 and 2020/2021Ice area and volume fluxes provide a way to quantify the transport of sea ice30. Figure 7 shows the cumulative fluxes across the boundaries of the Beaufort Sea (as defined in Fig. 1) from October 1 through the following June 1 for 2019/2020, 2020/2021, as well as a climatology for the first 13 years of the young ice period 2007–2019. Positive values indicate a flux into the region. Daily PIOMAS ice motion and ice thickness data were used to calculate these fluxes. The ice area fluxes were also computed using the NSIDC ice motion data31 with similar results obtained (Supplementary Fig. S3).Fig. 7: Variability in the PIOMAS sea ice fluxes into the region of interest.Cumulative: a ice area (105km2) flux and b ice volume flux (102km3) through the northern boundary of the region of interest. Cumulative: c ice area (105km2) flux and d ice volume flux (102km3) through the western boundary of the region of interest. The net cumulative: e ice area (105km2) flux and f ice volume flux (102km3) through the northern and western boundaries of the region of interest. Results are shown for climatology (2007–2019) as well as for 2019–2020 and 2020–2021 with the shading representing +/− one standard deviation above/below the climatological mean. Positive fluxes are into the region of interest.Full size imageWe first consider the northern boundary. The ice area and volume fluxes across this boundary are relatively small in the climatology (Fig. 7a, b), with interannual variability that includes some years in which the fluxes are negative, i.e., from the Beaufort Sea toward the LIA. During the period from October to January in 2019/2020 as well as in 2020/2021, ice area and volume fluxes are positive and growing, at rates near or above the climatological mean, indicating intensifying ice transport into the Beaufort Sea. After January, the 2 years differ. In winter 2020 the cumulative fluxes plateau, indicating near-zero values in contrast to the climatology which continues to grow. Then in spring 2020 the fluxes turn strongly negative, with values of one or more standard deviation below the mean, implying an export of ice from the Beaufort Sea into the LIA. In fact, the cool season 2019/2020 ends with an unusually large net export of ice volume from the Beaufort into the LIA. The following year, we see that the fluxes in winter 2021 continue to intensify at about 1 standard deviation above the mean. Then in spring the cumulative fluxes decline back toward the climatological mean, with end-of-cool-season values near the climatological young ice regime mean of net transport from the LIA into the Beaufort.At the western boundary, climatological ice area and ice volume fluxes are both directed out of the Beaufort Sea and into the Chukchi Sea (Fig. 7c, d). Although interannual variability is higher than that at the northern boundary, the fluxes are typically always negative. This is what makes the 2019/2020 area and volume fluxes so remarkable, in that they are nearly zero through winter 2020, and then turn strongly positive in spring, with values at or exceeding the mean by more than one standard deviation throughout the entire period. These positive fluxes reflect strong ice import from the west (Fig. 5f). In contrast, the fluxes in 2020/2021 became large and negative by early winter (greater than 1 standard deviation from the climatological mean), although this moderates later in the winter and spring. In this year, fluxes were strongly westward, from the Beaufort into the Chukchi Sea.The sum of the fluxes across the two boundaries provides a measure of the net transport into the Beaufort Sea (Fig. 7e, f). The climatology indicates that the net ice area and volume fluxes are negative, indicating a loss of ice from the Beaufort Sea. This reflects the fact that the transport out of the region through the western boundary usually exceeds the transport into the region through the northern boundary. In this context, the net fluxes during 2019/2020 again stand out as remarkable, since they are strongly positive (i.e., net transport into the Beaufort), especially for ice area flux. The net fluxes during 2020/2021 are closer to climatological values, and are well within the range of climatological variability.Impact of cool season ice fluxes on Beaufort Sea summer ice conditionsIn this section, we seek to quantify how the cool-season sea ice transport into the Beaufort Sea impacts ice conditions in the following summer using two metrics. The first metric is Beaufort Sea ice volume (i.e., the product of ice thickness and ice concentration27) on June 1. Even though melt can occur in parts of the study region prior to the beginning of June32, it is nevertheless a useful date for the start of the melt season. Our second metric is Beaufort Sea area-mean ice concentration during September, a measure of ice conditions at the end of the melt season and a closely observed indicator of climate change33,34.In Fig. 8, we correlate the net ice volume flux over the cool season, i.e., the period from October 1 to June 1 of the following year, against the Beaufort Sea June 1 ice volume anomaly, calculated by detrending the time series using a step function in 2007 that takes into account the changes between the new and old ice regimes (Fig. 2). The ice volume flux does not exhibit any trend and so no detrending was done for this time series. The correlation was done for both the old and young ice regimes. For both periods, there is a statistically significant linear relationship showing that larger net cool season ice transport into the Beaufort Sea leads to larger ice volume anomalies on June 1. However, the larger spread in the data for the old ice regime leads to a smaller percentage of the variance explained by ice transport, ~14%, as compared to ~45% for the young ice regime. The statistics are similar if May 1 is used as the end of the cool season, although using April 1 degrades the relationship to statistical insignificance, consistent with the springtime “predictability barrier”35 that arises from late-winter variability in ice-dynamics and ice growth.Fig. 8: Relationship between cumulative cool season ice volume flux into the Beaufort Sea region and June 1 Beaufort Sea ice volume 1980–2021.Scatterplot of the cool season (October 1 – June 1 following year) PIOMAS ice volume flux and June 1 PIOMAS ice volume anomaly. Linear least squares fit to the data for the two regimes are also shown as are the percentage of the variance explained. The ice volume time series has been detrended by step functions with a breakpoint in 2007.Full size imageRegarding conditions at the end of the summer, it seems intuitive that ice retreat might be slowed by the presence of thick ice. Indeed, discussions in the popular press5 have speculated that thick ice contributed to the relatively moderate September 2021 sea ice extent (12th lowest on record and the highest since 2014). On the other hand, it has also been suggested that cool atmospheric conditions during the summer of 2021 contributed to this relative maxima in sea ice extent36.To explore this question, we correlate PIOMAS-derived Beaufort Sea ice volume on June 1 with NSIDC CDR-derived September-mean sea ice concentration37. Given the nature of the underlying time series (Fig. 2), we have used step functions with a breakpoint in 2007 to detrend the data (see Materials and Methods). Although there is considerable spread, Fig. 9 indicates that there is a statistically significant linear relationship, with June 1 ice volume accounting for just under 40% of the variability in ice concentration during September for both the old and new ice regimes. Similar results are obtained if one uses the PIOMAS sea ice concentration during September (Supplementary Fig. S4).Fig. 9: Relationship between June 1 ice conditions and September mean ice concentration 1979–2021.Scatterplot of the June 1 PIOMAS ice volume anomaly and the NSIDC CDR September monthly mean ice concentration anomaly. Linear least squares fit to the data for the two regimes are also shown as are the percentage of the variance explained. Both time series have been detrended by step functions with a breakpoint in 2007.Full size imageA next logical step might be to link these two correlations together and ask, How does cool season ice transport impact end-of-summer ice concentration? Given the above results, assuming that there are no other factors related to cool season transport that impact summer ice melt and the cascade of probabilities, one would expect that the former would explain ~5% and ~16% of the variability in the latter for the old and new ice regimes. The results shown in Supplementary Fig. S5 confirm these assumptions; however, we also find that this relationship is not statistically significant in either regime. More

To date, many kinds of potential predator–prey interactions have been demonstrated between ticks and a wide range of animals, including birds, mammals, and arthropods such as beetles, spiders, and ants1,2,3. Of those, the ants were indicated to be one of the most effective tick predators4,5. More than 27 ant species of 17 genera (Anoplolepis, Aphaenogaster, Camponotus, Crematogaster, Ectatomma, Formica, Iridomyrmex, Meranoplus, Monomorium, Myrmica, Notoncus, Pheidole, Pogonomyrmex, Polyrhachis, Rhytidoponera, Solenopsis, and Tapinoma) have been reported to be effective on different tick species (Amblyomma spp., Boophilus spp., Ornithodoros spp., Ixodes spp., Argas spp., Aponomma hydrosauri, Rhipicephalus appendiculatus, Otobius megnini, and Dermacentor variabilis)2,6,7,8. It is known that ants and other predators can affect ticks in a consumptive or nonconsumptive / behavioral manner and, as a result, may reduce the abundance of ticks in the overlap ranges8,9,10. However, the effects of ants on ticks are closely related to the ant species and the species, developmental stages, and physiological status of the ticks, and as a consequence, the impact of ants on ticks can exhibit fairly high variability2,8. Furthermore, there is no sufficient data on the factors determining the tick-ant relationship11,12.Ant predation has been examined in all developmental stages of ticks, but the proportion of the studies based on the eggs is relatively low compared to the other stages2. In an egg-based study, the eggs of O. megnini, the spinose ear tick, were supplied to five different ant species, and of those, Tapinoma melanocephalum was the only species that fed on the eggs7. Conflicting results have been reported from the studies13,14 carried out to determine the predatory effects of ant species Pheidole megacephala on the eggs of Boophilus (Rhipicephalus) microplus14. Rhipicephalus sanguineus was demonstrated to secrete an acarine allomone when attacked by fire ants, Solenopsis invicta15. This allomone-based ant deterrence is known to protect ticks from being eliminated within the sympatric range. The eggs, intact and cracked, of tick species Amblyomma americanum were not attacked by S. invicta, and it was interpreted that this deterrence might be related to the possible presence of the allomone within the eggs12.This study was carried out to determine whether the ant species Lasius alienus (Förster, 1850) (Hymenoptera: Formicidae) has any predatory effect on the eggs of tick species Hyalomma marginatum, H. excavatum, and Rhipicephalus bursa, and if the tick egg wax has any protective properties against possible predation. Ticks lay eggs (each 50–100 µg in weight and 0.5–1 mm in length) with a wax coat 0.5–2.0 µm thick, which is secreted by the female-tick-specific glands and organs such as the Gené’s16,17. Different molecules have been detected in the wax, such as alkanes, fatty acids, steroids, alcohols, and some specific proteins and lipoproteins18,19,20. However, detailed data on the wax content, especially its bioactive components, are not yet available20,21. As for the function of the wax, it has been reported that it reduces water loss, waterproofs the eggs, ensures the proper gas exchange between the eggs and air and holds the eggs together16,18. The wax also provides protection against chemical and physical factors such as cold, heat, proteinase K and pronase, or microbial agents including bacteria, fungi, viruses, and protozoa19,20,21,22,23,24,25.Lasius alienus is one of the most abundant ant species in the Western Palearctic26. Its distribution area can range from natural open habitats, light forests, and forest edges to urbanized areas such as wooded residential areas and gardens. The nests can be encountered mostly in the soil, under stones, or other substances, and the nest densities may reach up to 10–50 nests/100 m2 in some endemic territories. The number of workers (2­4 mm in length) in a colony can be more than 10,000. In the active periods in hot and warm months, workers establish foraging trails on the ground, in trees, and even in dwellings for food27,28. Lasius alienus was reported to gather plant nectar, honeydew secreted by aphids and to consume both dead and small living arthropods28,29,30. However, there is no data in the literature on whether this or any other species in the Lasius genera have a predatory interaction with ticks. Furthermore, the only definitive association of L. alienus with predation on Acarina has been established recently with Dermanyssus gallinae (poultry red mite)31.Considering the fact that the workers of L. alienus can forage effectively in many different places within the wide range distribution area27,28,31, it seems quite possible that several tick species encounter this ant species in their habitat32. This ant can feed by scavenging and predating small insects, and it meets their protein needs by hunting large numbers of small invertebrates, especially during larval feeding periods using the central foraging strategy29. Whichever invertebrates are abundant in their environment, the ants undoubtedly tend to consume more of them, especially if they are easy to hunt and transport27,28,29. Engorged large female ixodid ticks (around 1–1.5 cm depending on the species) lay a single batch of a large number of eggs (hundreds or thousands depending on the species and feeding levels) for several days or weeks at the hiding points such as cracks, crevices, and spaces under stones or various objects on the ground21,33 that the ant can easily reach due to its small size. In ixodid ticks, the female ticks lay eggs at once and die. The next generation continues entirely through the eggs21. Referring to these data, it seems hypothetically possible that any level of predation of L. alienus on the eggs, which are immobile and easy to reach and carry for the ants, can have a direct effect on the tick community in the overlap ranges. At this point, of course, whether the natural distribution areas of the ant and tick species overlap and, potentially, whether there is a possible evolutionary background between them are of the expected determining factors in a possible predation34.The nests of L. alienus can be seen almost everywhere in the soil in its ranging territories, however, the density increases from dry steppe, open pasture and bushlands to cultivated areas, woodlands, and gardens29. In this study, three ixodid tick species were selected that are more or less different from each other in terms of biology, ecology, and therefore the probability of encountering L. alienus in their natural habitat. Of these, H. marginatum is a two-host tick, the immature stages (larvae and nymphs) prefer rabbits, hedgehogs and birds to feed, and the adults (male and female) use particularly cattle as host. The immature stages of mostly three-host tick H. excavatum, wild rodents are the preferable host, and a wide range of large wild animals, cattle and some other domestic animals can be the host for the adults. Both species are known as arid environment ticks, however, in accordance with their different host preferences as well, H. excavatum is more prevalent in the arid open fields, steppe, and bushlands32,35. Both immature and adult stages of two-host tick R. bursa use primarily domestic ruminants to feed. Although there is no detailed data on the natural dynamics of R. bursa, this species is suspected of having a kind of peri-farm natural dynamics32. More

Rio Conservation and Sustainability Science Centre, Department of Geography and the Environment, Pontifical Catholic University, Rio de Janeiro, BrazilBernardo B. N. Strassburg, Alvaro Iribarrem, Carlos Leandro Cordeiro, Renato Crouzeilles, Catarina C. Jakovac, André Braga Junqueira, Eduardo Lacerda & Agnieszka E. LatawiecInternational Institute for Sustainability, Rio de Janeiro, BrazilBernardo B. N. Strassburg, Alvaro Iribarrem, Carlos Leandro Cordeiro, Renato Crouzeilles, Catarina C. Jakovac, André Braga Junqueira, Eduardo Lacerda, Agnieszka E. Latawiec, Robin L. Chazdon & Carlos Alberto de M. ScaramuzzaPrograma de Pós Graduacão em Ecologia, Universidade Federal do Rio de Janeiro, Rio de Janeiro, BrazilBernardo B. N. Strassburg, Renato Crouzeilles & Fabio R. ScaranoBotanical Garden Research Institute of Rio de Janeiro, Rio de Janeiro, BrazilBernardo B. N. StrassburgSchool of Biological Sciences, University of Queensland, St Lucia, Queensland, AustraliaHawthorne L. BeyerForest Ecology and Management Group, Wageningen University, Wageningen, The NetherlandsCatarina C. JakovacInstitut de Ciència i Tecnologia Ambientals, Universitat Autònoma de Barcelona, Barcelona, SpainAndré Braga JunqueiraDepartment of Geography, Fluminense Federal University, Niterói, BrazilEduardo LacerdaDepartment of Production Engineering, Logistics and Applied Computer Science, Faculty of Production and Power Engineering, University of Agriculture in Kraków, Kraków, PolandAgnieszka E. LatawiecSchool of Environmental Sciences, University of East Anglia, Norwich, UKAgnieszka E. LatawiecDepartment of Zoology, University of Cambridge, Cambridge, UKAndrew Balmford, Stuart H. M. Butchart & Paul F. DonaldInternational Union for Conservation of Nature (IUCN), Gland, SwitzerlandThomas M. BrooksWorld Agroforestry Center (ICRAF), University of the Philippines, Los Baños, The PhilippinesThomas M. BrooksInstitute for Marine and Antarctic Studies, University of Tasmania, Hobart, Tasmania, AustraliaThomas M. BrooksBirdLife International, Cambridge, UKStuart H. M. Butchart & Paul F. DonaldDepartment of Ecology and Evolutionary Biology, University of Connecticut, Storrs, CT, USARobin L. ChazdonWorld Resources Institute, Global Restoration Initiative, Washington, DC, USARobin L. ChazdonTropical Forests and People Research Centre, University of the Sunshine Coast, Sunshine Coast, Queensland, AustraliaRobin L. ChazdonInstitute of Social Ecology, University of Natural Resources and Life Sciences Vienna, Vienna, AustriaKarl-Heinz Erb & Christoph PlutzarDepartment of Forest Sciences, ‘Luiz de Queiroz’ College of Agriculture, University of São Paulo, Piracicaba, BrazilPedro BrancalionRSPB Centre for Conservation Science, Royal Society for the Protection of Birds, Edinburgh, UKGraeme Buchanan & Paul F. DonaldSecretariat of the Convention on Biological Diversity (SCBD), Montreal, Quebec, CanadaDavid CooperInstituto Multidisciplinario de Biología Vegetal, CONICET and Universidad Nacional de Córdoba, Córdoba, ArgentinaSandra DíazUN Environment World Conservation Monitoring Centre, Cambridge, UKValerie Kapos & Lera MilesEcosystem Services Management (ESM) Program, International Institute for Applied Systems Analysis (IIASA), Laxenburg, AustriaDavid Leclère, Michael Obersteiner & Piero ViscontiEnvironmental Change Institute, Oxford University Centre for the Environment, Oxford, UKMichael ObersteinerDivision of Conservation Biology, Vegetation Ecology and Landscape Ecology, University of Vienna, Vienna, AustriaChristoph Plutzar More

Conservation biologist Kay Van Damme has spent two decades studying biodiversity on the island of Socotra in Yemen.Credit: Søren Solkær

In 1999, Kay Van Damme joined a United Nations-led multidisciplinary expedition to the Socotra archipelago, a Yemeni island group in the Arabian Sea, to explore freshwater ecosystems. Influenced by the area’s rich and unique biodiversity, Van Damme began to run annual expeditions to underground lakes on the main island, Socotra, as well as to its aquatic and terrestrial ecosystems above ground. In 2010, he earned a PhD on the evolutionary relationships of freshwater crustaceans from Ghent University in Belgium. Over the years, with the island facing the environmental challenges of climate change and a civil war that has been under way since 2014, Van Damme started applying his knowledge to conserving endangered aquatic insects and crabs, as well as the remarkable local trees. Now, alongside undertaking fieldwork on Socotra, where he works directly with local communities, Van Damme is a postdoctoral researcher at Ghent and at Mendel University in Brno, Czech Republic.What’s so special about Socotra?Socotra’s biodiversity treasures are the result of millions of years of secluded evolution. The archipelago separated from southern Arabia during the Miocene epoch (23 million to 5 million years ago). About 37% of its plant species, 90% of its reptiles and 98% of its land snails don’t exist anywhere else. It is the only Yemeni natural site on the UNESCO World Heritage List, to which it was added in 2008.Islands are often called living laboratories of evolution. Studying these habitats is important for natural history and biogeography. In addition, understanding how highly vulnerable endemic birds, plants and invertebrates have survived on these islands could help to save species in other places.How has your fieldwork changed over the years?Continuous exposure to the amazing nature and Socotra’s islanders have had a strong impact on me. The last forest of umbrella-shaped dragon’s blood trees (Dracaena cinnabari) seems out of this world, a relic of Miocene times, when this vegetation type was more widespread around the world. There’s a strange mix of ancient and recent geological features, ranging from granite mountains to coastal sand dunes and enormous limestone caves, harbouring vulnerable and isolated lifeforms.

Socotra hosts the last forest of dragon’s blood trees (Dracaena cinnabari) in the world.Credit: Kay Van Damme

Seeing the uniqueness of Socotra’s biodiversity and its fragility in the face of climate change, I feel my responsibility has grown to include more than exploration and classification. Gradually, my efforts converged on biodiversity conservation, and my fieldwork has become more strategic: doing biodiversity field surveys to assess threats to endemic species. I help Yemen’s environmental protection agency with conservation planning, including establishing and improving protected areas. We do activities in schools and with the Indigenous Soqotri people to ensure that conservation efforts are integrated into the community, while discussing their concerns about their environment and the impacts of climate change.Which species do you focus on, and why?As co-principal investigator of a conservation project funded by the Franklinia Foundation in Geneva, Switzerland, I work with colleagues from Socotra, Mendel University and the Sapienza University of Rome on protecting the dragon’s blood trees and ten endemic species of frankincense tree (Boswellia). Meanwhile, I chair a UK-based charity called the Friends of Soqotra, which runs biodiversity-conservation and environmental-awareness projects. Together with local communities in north Socotra, we have replanted mangrove trees (Avicennia marina) that had disappeared decades before. I also focus on a magnificent damselfly, the Socotra bluet (Azuragrion granti), and a colourful freshwater crab (Socotrapotamon socotrensis), which are included in the International Union for Conservation of Nature’s Red List of Threatened Species.

The Socotra bluet (Azuragrion granti), a type of damselfly, is threatened by drought, pollution and soil erosion.Credit: Kay Van Damme

What are the biggest threats to these ecosystems? Overgrazing by domestic goats, loss of traditional land-management techniques and the effects of climate change. Violent cyclones and droughts in Socotra affect both terrestrial and aquatic ecosystems. For instance, cyclones destroyed considerable proportions of unique woodlands of frankincense and dragon’s blood trees in 2015 and 2018. Likewise, threats to freshwater species include drought; landslides due to vegetation loss; pollution; and invasive species, such as a predatory fish called the Arabian toothcarp (Aphanius dispar).The most effective solution to these threats is for the local communities to get involved in leading the conservation work on the ground. For every tree felled by weather, scientists should work with locals to replant another.How have you gained the trust of the local community?I work with our local team, which continues the fieldwork, and with the Soqotri people, who own the land areas. I have great respect for their immense environmental knowledge. We have long conversations with them during visits, asking what is needed. In Socotra, there is no stronger conservation expertise than that which has been applied for centuries by the Soqotri people. For instance, by understanding the quality of a frankincense tree’s incense and the timing of flowering, they have shown us new hybrid species. Our nursery of young frankincense seedlings is maintained by an older Soqotri woman called Mona, who has traditional knowledge of how to take care of them.How does the ongoing civil war affect your fieldwork?In comparison to the mainland, Socotra has remained relatively safe for fieldwork. However, the political landscape of the island is variable and complex. This requires us to be flexible when discussing conservation issues with local decision makers, who are sometimes replaced more than once during a project.

Van Damme (centre) collaborates with community leaders — Saleh Al-Tamek, chair of a local conservation group (left), and natural-heritage specialist Haifaa Abdulhalim — to protect threatened species of mangrove tree.Credit: Marketa Jakovenko

We constantly assess risks and maintain clear communication between research team members and local communities and their leaders about where we are at which moment, and for what purpose. In such circumstances, scientists should know the local terrain and weather conditions extremely well, avoiding unnecessary risks and checking in frequently by mobile phone with their local teams.Do you have any advice for early-career conservation scientists? Local community leaders have the power to facilitate fieldwork or obstruct it. And because not all leaders prioritize nature conservation, there are some practical tips for building mutual trust with them to help form long-standing partnerships. Focus on constant communication with leaders, respect their culture and environmental knowledge, and cooperate with them in protecting the interests of local people, especially during natural disasters.Have you made any non-scientific discoveries about Socotra?My soul has been touched by Socotra’s people. Their kindness towards others is essential to their culture. They speak an endangered Semitic language that has survived through oral tradition, such as poetry and songs. Soqotri people are excellent orators and communicate using good humour. I brought my parents to Socotra to see what has stolen my heart — they understood.
This interview has been edited for length and clarity. More

Our study method includes the following stages: (1) framing the investigation problem, (2) examining the literature, (3) developing and verifying two hypotheses, (4) collecting data, (5) the multiple criteria examination of 173 countries by means of the Degree of Project Utility and Investment Value Assessments (INVAR) method, (6) calculating correlations between 33 indicators and the success of 173 countries, (7) building 12 regression models, (8) compiling eight Maps (of which seven are CSS Maps) visualizing national success and sustainability, (9) spatial perspective analysis, and (10) performing integrated linear regression, multi-variant design and multiple criteria analysis of national policy alternatives, in order to identify rational decisions.This research is a quantitative study to examine the way national success affects 12 indicators of the three dimensions of sustainability in 173 countries, and uses the data from 2020, or the latest available.As investigation methods, our CSS Maps and Models can make it easier to study interdependencies between country success and sustainability. Supplementary Section 1, 4, and 5 presents our literature analysis which is carried out to gain deeper insights into our CSS Maps and Models, and to better understand their components in the worldwide research context.The following two core hypotheses have been proposed and verified for this research:

Hypothesis 1—The increasing success of a country is generally accompanied by increasing values for the three dimensions of sustainability indicators, and declines in these indicators lead to decreases in the country’s success. Improving some sustainability indicators tends to improve other sustainability indicators.

Hypothesis 2—Changes in the number of countries and their traditional key indicators system do not make a very significant difference to the relative national sustainability and success values. Likewise, the boundaries of the seven country clusters discussed in this research do not excessively depend on specific traditional key systems of indicators used in their analysis.

Along with different sets of national 17 success (Supplementary Table S1) and 12 sustainability (Supplementary Table S2) indicators, the INVAR method46 (Supplementary Section 2 and Fig. S1) was used to measure and map the success of the 173 countries selected as the focus for this research. The traditional statistical indicator systems defining country success and the three dimensions of sustainability are based on studies from various countries analyzed and combined. The INVAR method calculates an integrated criterion characterizing the overall success of the countries. This integrated criterion is directly proportional to the relative effect the values and weights of the given criteria make on the country’s success. The multiple-criteria INVAR analysis method has been applied to various countries, including Asian nations47, ex-Soviet states48, and a group of 169 countries49.This research used data from the framework of variables taken from various databases and websites, including Transparency International, Global Data, Eurostat-OECD, the World Bank, Knoema, the World Health Organization, Global Finance, Freedom House, Heritage, the Global Footprint Network, Socioeconomic Data and Applications Center, Our World in Data, Climate Change Knowledge Portal (World Bank Group), and The Institute for Economics and Peace, as well as global and national statistics and publications. All 173 countries analyzed in this article are listed in matrices, along with their 17 detailed success (Supplementary Table S1) and 12 sustainability (Supplementary Table S2) indicators (systems of indicators, their numbering, values, and weights). The INVAR method46 was applied to perform multiple criteria analysis of the 173 countries, and the results are presented in Supplementary Table S1 and Figs. 2, 3, 4 and 5. We use equal and different weights of 17 indicators to calculate the deviation of priorities for the 173 countries, which stands at 5.34% (Supplementary Section 2 and Fig. S2).Along with different sets of 12 national sustainability and 17 success indicators, the INVAR method46 was used to measure and map the success of the 173 countries selected as the focus of this research. The traditional statistical indicator systems defining country success and the three dimensions of sustainability are based on studies from various countries analyzed and combined. The INVAR method calculates an integrated criterion characterizing the overall success of the countries. This integrated criterion is directly proportional to the relative effect the values and weights of the given criteria make on the country’s success.Supplementary Table S3 shows the correlations between all measures determined by analyzing 173 countries. Supplementary Table S4 reveals the correlation coefficient matrix of the 17 success criteria for each of the 173 countries analyzed in this survey.Along the vertical axis y we analyze seven sustainability indicators, and along the horizontal axis x we analyze the success and priority indicators (9 CSS Map dimensions). The median correlation between the survival versus self-expression values and the nine CSS Map dimensions (the x-axis and y-axis) is moderate, whereas the median correlation between the traditional versus secular–rational values and the nine CSS Map dimensions is strong (Fig. 1).Tables S5-S8 show the descriptive statistics of 12 CSS Models (Supplementary Section 3). Supplementary Table S8 shows the extent to which a 1% increase or decrease in success of country’s features can push sustainability indicators up or down, expressed as a percentage. Supplementary Table S8 also shows the degree to which the percentage changes of success or the values of country’s features explain or fail to explain the dispersion of sustainability indicators. These CSS Models (Supplementary Section 3) show that when a country’s success increases by 1%, its 12 indicators related to the three dimensions of sustainability improve by on average 0.85% (Supplementary Table S8). Furthermore, the 17 variables of country success used in the CSS Models explain 80.8% on average of the dispersion of the three dimensions of sustainability and 98.2% of the dispersion of the country success variable (Supplementary Table S8).An increase of 1% in a country’s success is accompanied by a 0.39% average increase in its social and environmental (0.84% on average) sustainability indicators (Supplementary Table S8). On average, the CSS Sustainability Models explain 76.3% of the dispersions among the environmental sustainability indicators, 83.4% of the dispersions among the social sustainability indicators, and 94.5% of the dispersion among economic (i.e. the gross national income per capita) sustainability indicators (Supplementary Table S8).The study produced the eight Maps (of which seven are CSS Maps) of the World based on an analysis of 99–150 countries (the 2020 Inglehart–Welzel Cultural Map of the World focused on 103 analogical CSS Maps countries). The two dimensions of country success on the CSS Maps are represented in a system of 17 variables (Supplementary Table S1). When a country’s success grows, its performance related to the three dimensions of sustainable development increases as well, and the eight Maps (of which seven are CSS Maps) clearly illustrate this relationship (Figs. 2, 3, 4 and 5). The CSS Maps of the World developed as part of this study are described in Supplementary Section 5.Studies from various countries and our research suggest that country success and their features (x-axis) and sustainability indicators (y-axis) are generally strongly interrelated, and move in the same direction over time. This means that successful countries also perform better on sustainability dimensions.Stage 9 involved analysis of the spatial perspective research in place for explaining and predicting globally recognised physical, spatial, and human patterns in multiple ways. We apply 12 CSS Models, alternative design and multi-criteria analysis methods for spatial perspective analysis (Supplementary Section 4).The following additional two research objectives were set: (1) to determine the impact of a country’s success factors on sustainability metrics, and (2) to offer stakeholders recommendations regarding the strategies for improving sustainability indicators. The ways to improve sustainability indicators are determined by analysing 17 dependent variables (the main paper section “Practical applications and implications”, Table S9). As previously mentioned, in stage 10, national policy options have been examined by means of integrated linear regression, multi-variant design and multiple criteria analysis to identify rational decisions. Analysis of multiple alternative options and their detailed indicators, with a consideration of the existing state of the micro, meso, and macro environment, can ensure rational country success and sustainability. Below, a brief analysis of several best global practice examples of ways to identify rational policy, activities, and strategy follows. The examples presented below suggest that multiple possible alternatives must be designed, assessed against a system of micro, meso and macro indicators, and the most effective options selected to make countries more sustainable. In Isham and Jackson’s14 opinion, materialistic lifestyles and values have been associated with adverse effects on human health as well as having detrimental effects on our planet. Therefore, activities and lifestyles should be identified that promote human well-being, yet which at the same time protect ecological security. Isham and Jackson14 identify optimal activities (arts and crafts, reading, sports, meditating) with high levels of human well-being and low environmental costs. It is important to estimate pollution impacts on health in order to come up with the right policies for better health outcomes. Yet, the task is challenging because economic activity can lead to worse pollution, but can also improve health outcomes in its own right37. Humidity, temperature, dispersal by the wind, and other environmental factors contribute to pollution levels. Certain fine particulates can stay in the atmosphere for days, and travel long distances to be inhaled in places far away from the source, even in other continents. Local conditions must be reflected in emissions-control policies, and the global flows of air pollutants must be taken into account6. The explanation for the phenomenon of demographic transition could be improved public health in developed countries which results in a move toward a slower life strategy38. Studies show that children from wealthier backgrounds undergo puberty later than those from poor socio-economic backgrounds. Early puberty can lead to a variety of health problems and a shorter life. By the early adult years, the effects of exposure to trauma, post-traumatic stress disorder, and other conditions can become apparent in the form of diseases related to aging9. Education is a very important factor in economic growth, and is also strongly related to health. In addition to health benefits, substantial increases in education, especially of women, and shrinking gender gaps have an important effect on the roles and status of women in society36.The INVAR method, statistical analysis, and the CSS Maps and Models can help generate multiple policy recommendations for various stakeholders. The possibilities are as follows:

To create alternatives for ways to develop country success and sustainability, by performing countries’ multiple criteria and statistical analysis and identifying decisions that would be rational;

to perform quantitative and qualitative analysis of the existing data and to interpret it. The results obtained this way would prompt automatic recommendations designed for different stakeholders on ways to improve country sustainability. More

Smouse, P. E. & Peakall, R. O. D. Spatial autocorrelation analysis of individual multiallele and multilocus genetic structure. Heredity 82(5), 561–573 (1999).PubMed 
Article 

Google Scholar 
Liu, J. X., Gao, T. X., Wu, S. F. & Zhang, Y. P. Pleistocene isolation in the Northwestern Pacific marginal seas and limited dispersal in a marine fish, Chelon haematocheilus (Temminck & Schlegel, 1845). Mol. Ecol. 16(2), 275–288 (2007).CAS 
PubMed 
Article 

Google Scholar 
Ding, S., Mishra, M., Wu, H., Liang, S. & Miyamoto, M. M. Characterization of hybridization within a secondary contact region of the inshore fish, Bostrychus sinensis, in the East China Sea. Heredity 120(1), 51–62 (2018).CAS 
PubMed 
Article 

Google Scholar 
Ashrafzadeh, M. R. et al. Assessing the origin, genetic structure and demographic history of the common pheasant (Phasianus colchicus) in the introduced European range. Sci. Rep. 11(1), 1–14 (2021).Article 
CAS 

Google Scholar 
Caccavo, J. A. et al. Along-shelf connectivity and circumpolar gene flow in Antarctic silverfish (Pleuragramma antarctica). Sci. Rep. 8(1), 1–16 (2018).Article 
CAS 

Google Scholar 
Otwoma, L. M., Reuter, H., Timm, J. & Meyer, A. Genetic connectivity in a herbivorous coral reef fish (Acanthurus leucosternon Bennet, 1833) in the Eastern African region. Hydrobiologia 806(1), 237–250 (2018).Article 

Google Scholar 
Li, H., Lin, H., Li, J. & Ding, S. Phylogeography of the Chinese beard eel, Cirrhimuraena chinensis Kaup, inferred from mitochondrial DNA: A range expansion after the last glacial maximum. Int. J. Mol. Sci. 15(8), 13564–13577 (2014).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Gao, B., Song, N., Li, Z., Gao, T. & Liu, L. Population genetic structure of Nuchequula mannusella (Perciformes: Leiognathidae) population in the Southern Coast of China inferred from complete sequence of mtDNA Cyt b gene. Pak. J. Zool. 51(4), 1527–1535 (2019).Article 

Google Scholar 
Qiu, F., Li, H., Lin, H., Ding, S. & Miyamoto, M. M. Phylogeography of the inshore fish, Bostrychus sinensis, along the Pacific coastline of China. Mol. Phylogenet. Evol. 96, 112–117 (2016).PubMed 
Article 

Google Scholar 
Gu, S. et al. Genetic diversity and population structure of cutlassfish (Lepturacanthus savala) along the coast of mainland China, as inferred by mitochondrial and microsatellite DNA markers. Reg. Stud. Mar. Sci. 43, 101702 (2021).
Google Scholar 
Liu, Q. et al. Genetic variation and population genetic structure of the large yellow croaker (Larimichthys crocea) based on genome-wide single nucleotide polymorphisms in farmed and wild populations. Fish. Res. 232, 105718 (2020).Article 

Google Scholar 
Song, P. et al. Genetic characteristics of yellow seabream Acanthopagrus latus (Houttuyn, 1782) (Teleostei: Sparidae) after stock enhancement in southeastern China coastal waters. Reg. Stud. Mar. Sci. 48, 102065 (2021).
Google Scholar 
Ward, R. D. Genetics in fisheries management. Hydrobiologia 420, 191–201 (2000).CAS 
Article 

Google Scholar 
Liu, X., Guo, Y., Wang, Z. & Liu, C. The complete mitochondrial genome sequence of Trichiurus nanhaiensis (Perciformes: Trichiuridae). Mitochondrial DNA 24(5), 516–517 (2013).CAS 
PubMed 
Article 

Google Scholar 
Wang, H. Y., Dong, C. A. & Lin, H. C. DNA barcoding of fisheries catch to reveal composition and distribution of cutlassfishes along the Taiwan coast. Fish. Res. 187, 103–109 (2017).Article 

Google Scholar 
Guo, Y. S., Liu, X. M., Wang, Z. D., Lu, H. S. & Liu, C. W. Isolation and characterization of microsatellite DNA loci from Naihai cutlassfish (Trichiurus nanhaiensis). J. Genet. 93(1), 109–112 (2014).Article 

Google Scholar 
Kwok, K. Y. & Ni, I. H. Reproduction of cutlassfishes Trichiurus spp. from the South China Sea. Mar. Ecol. Prog. Ser. 176, 39–47 (1999).ADS 
Article 

Google Scholar 
He, L. et al. Demographic response of cutlassfish (Trichiurus japonicus and T. nanhaiensis) to fluctuating palaeo-climate and regional oceanographic conditions in the China seas. Sci. Rep. 4(1), 1–10 (2014).
Google Scholar 
Lin, H. C., Tsai, C. J. & Wang, H. Y. Variation in global distribution, population structures, and demographic history for four Trichiurus cutlassfishes. PeerJ 9, e12639 (2021).PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Funk, W. C., McKay, J. K., Hohenlohe, P. A. & Allendorf, F. W. Harnessing genomics for delineating conservation units. Trends Ecol. Evol. 27(9), 489–496 (2012).PubMed 
PubMed Central 
Article 

Google Scholar 
Tautz, D. Hypervariability of simple sequences as a general source for polymorphic DNA markers. Nucleic Acids Res. 17, 6463–6471 (1989).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Templeton, A. R. The, “Eve” hypotheses: A genetic critique and reanalysis. Am. Anthropol. 95, 51–72 (1993).Article 

Google Scholar 
Van Oosterhout, C., Hutchinson, W. F., Wills, D. P. & Shipley, P. MICRO-CHECKER: Software for identifying and correcting genotyping errors in microsatellite data. Mol. Ecol. Notes 4(3), 535–538 (2004).Article 
CAS 

Google Scholar 
Frankham, R. Challenges and opportunities of genetic approaches to biological conservation. Biol. Conserv. 143(9), 1919–1927 (2010).Article 

Google Scholar 
Sun, P., Shi, Z., Yin, F. & Peng, S. Population genetic structure and demographic history of Pampus argenteus in the Indo-West Pacific inferred from mitochondrial cytochrome b sequences. Biochem. Syst. Ecol. 43, 54–63 (2012).CAS 
Article 

Google Scholar 
Liu, S. Y. V. et al. Genetic stock structure of Terapon jarbua in Taiwanese waters. Mar. Coast. Fish. 7(1), 464–473 (2015).Article 

Google Scholar 
Song, C. Y., Sun, Z. C., Gao, T. X. & Song, N. Structure analysis of mitochondrial DNA control region sequences and its applications for the study of population genetic diversity of Acanthogobius ommaturus. Russ. J. Mar. Biol. 46(4), 292–301 (2020).CAS 
Article 

Google Scholar 
Yan, Y. R. et al. Cryptic diversity of the spotted scat Scatophagus argus (Perciformes: Scatophagidae) in the South China Sea: Pre-or post-production isolation. Mar. Freshw. Res. 71(12), 1640–1650 (2020).Article 

Google Scholar 
Hartl, D. L., Clark, A. G., & Clark, A. G. Principles of Population Genetics, vol. 116 (Sinauer Associates, 1997).Wu, R. et al. Study on the nomenclature and taxonomie status of hairtail Trichiurus japonicus from the Chinese coastal waters. Genom. Appl. Biol. 37(9), 3782–3791 (2018).
Google Scholar 
Xu, D. et al. Genetic diversity and population differentiation in the yellow drum Nibea albiflora along the coast of the China Sea. Mar. Biol. Res. 13(4), 456–462 (2017).Article 

Google Scholar 
Wang, W. et al. Genetic diversity and population structure analysis of Lateolabrax maculatus from Chinese coastal waters using polymorphic microsatellite markers. Sci. Rep. 11(1), 1–11 (2021).Article 
CAS 

Google Scholar 
Cheng, Q., Chen, W. & Ma, L. Genetic diversity and population structure of small yellow croaker (Larimichthys polyactis) in the Yellow and East China seas based on microsatellites. Aquat. Living Resour. 32, 16 (2019).Article 

Google Scholar 
DeWoody, J. A. & Avise, J. C. Microsatellite variation in marine, freshwater and anadromous fishes compared with other animals. J. Fish Biol. 56(3), 461–473 (2000).CAS 
Article 

Google Scholar 
Song, N., Yin, L., Sun, D., Zhao, L. & Gao, T. Fine-scale population structure of Collichtys lucidus populations inferred from microsatellite markers. J. Appl. Ichthyol. 35(3), 709–718 (2019).Article 

Google Scholar 
Yin, W. et al. Species delimitation and historical biogeography in the genus Helice (Brachyura: Varunidae) in the Northwestern Pacific. Zool. Sci. 26(7), 467–475 (2009).CAS 
Article 

Google Scholar 
Hewitt, G. The genetic legacy of the Quaternary ice ages. Nature 405(6789), 907–913 (2000).ADS 
CAS 
PubMed 
Article 

Google Scholar 
Yi, M. R. et al. Genetic structure and diversity of the yellowbelly threadfin bream Nemipterus bathybius in the Northern South China Sea. Diversity 13(7), 324 (2021).CAS 
Article 

Google Scholar 
Chen, X., Wang, J. J., Ai, W. M., Chen, H. & Lin, H. D. Phylogeography and genetic population structure of the spadenose shark (Scoliodon macrorhynchos) from the Chinese coast. Mitochondrial DNA Part A 29(7), 1100–1107 (2018).Article 
CAS 

Google Scholar 
Huang, W. et al. Genetic diversity and large-scale connectivity of the scleractinian coral Porites lutea in the South China Sea. Coral Reefs 37(4), 1259–1271 (2018).ADS 
Article 

Google Scholar 
Hou, G. et al. Identification of eggs and spawning zones of hairtail fishes Trichiurus (Pisces: Trichiuridae) in Northern South China Sea, using DNA barcoding. Front. Environ. Sci. 9, 703029 (2021).Article 

Google Scholar 
Yamaguchi, K., Nakajima, M. & Taniguchi, N. Loss of genetic variation and increased population differentiation in geographically peripheral populations of Japanese char Salvelinus leucomaenis. Aquaculture 308, S20–S27 (2010).Article 

Google Scholar 
Neo, M. L., Liu, L. L., Huang, D. & Soong, K. Thriving populations with low genetic diversity in giant clam species, Tridacna maxima and Tridacna noae, at Dongsha Atoll, South China Sea. Reg. Stud. Mar. Sci. 24, 278–287 (2018).
Google Scholar 
Jeanmougin, F., Thompson, J., Gouy, M., Higgins, D. & Gibson, T. Multiple sequence alignment with Clustal X. Trends Biochem. Sci. 23, 403–405 (1998).CAS 
PubMed 
Article 

Google Scholar 
Librado, P. & Rozas, J. DnaSP v5: A software for comprehensive analysis of DNA polymorphism data. Bioinformatics 25(11), 1451–1452 (2009).CAS 
PubMed 
Article 

Google Scholar 
Kumar, S., Stecher, G., Li, M., Knyaz, C. & Tamura, K. MEGA X: Molecular evolutionary genetics analysis across computing platforms. Mol. Biol. Evol. 35, 1547–1549 (2018).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Zhang, D. et al. PhyloSuite: An integrated and scalable desktop platform for streamlined molecular sequence data management and evolutionary phylogenetics studies. Mol. Ecol. Resour. 20(1), 348–355 (2020).PubMed 
Article 

Google Scholar 
Excoffier, L. & Lischer, H. E. Arlequin suite ver 3.5: A new series of programs to perform population genetics analyses under Linux and Windows. Mol. Ecol. Resour. 10(3), 564–567 (2010).PubMed 
Article 

Google Scholar 
Tajima, F. Statistical method for testing the neutral mutation hypothesis by DNA polymorphism. Genetics 123, 585–595 (1989).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Fu, Y. X. Statistical tests of neutrality of mutations against population growth, hitchhiking and background selection. Genetics 147(2), 915–925 (1997).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Drummond, A. J., Suchard, M. A., Xie, D. & Rambaut, A. Bayesian phylogenetics with BEAUti and the BEAST 1.7. Mol. Biol. Evol. 29(8), 1969–1973 (2012).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Rambaut, A., Suchard, M. A., Xie, D., & Drummond, A. J. Tracer v1.6. (2014). http://tree.bio.ed.ac.uk/software/tracer/ (accessed 23 Jan 2018).Bohonak, A. J. IBD (isolation by distance): A program for analyses of isolation by distance. J. Hered. 93(2), 153–154 (2002).CAS 
PubMed 
Article 

Google Scholar 
Weir, B. S. & Cockerham, C. C. Estimating F-statistics for the analysis of population structure. Evolution 38, 1358–1370 (1984).CAS 
PubMed 

Google Scholar 
Goudet, J. FSTAT (version 2.9.3): A program to estimate and test gene diversities and fixation indices. www.unil.ch/izea/softwares/fstat.html (2001).Pritchard, J. K., Stephens, M. & Donnelly, P. Inference of population structure using multilocus genotype data. Genetics 155(2), 945–959 (2000).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Earl, D. A. & VonHoldt, B. M. STRUCTURE HARVESTER: A website and program for visualizing STRUCTURE output and implementing the Evanno method. Conserv. Genet. Resour. 4(2), 359–361 (2012).Article 

Google Scholar 
Peakall, R. O. D. & Smouse, P. E. GENALEX 6: Genetic analysis in Excel. Population genetic software for teaching and research. Mol. Ecol. Notes 6(1), 288–295 (2006).Article 

Google Scholar 
Jombart, T., Devillard, S. & Balloux, F. Discriminant analysis of principal components: A new method for the analysis of genetically structured populations. BMC Genet. 11(1), 1–15 (2010).Article 

Google Scholar 
R Core Team. R: A Language and Environment for Statistical Computing (R Foundation for Statistical Computing, 2015).Cornuet, J. M. et al. DIYABC v2.0: A software to make approximate Bayesian computation inferences about population history using single nucleotide polymorphism, DNA sequence and microsatellite data. Bioinformatics 30(8), 1187–1189 (2014).CAS 
PubMed 
Article 

Google Scholar 
Cornuet, J. M., Ravigne, V. & Estoup, A. Inference on population history and model checking using DNA sequence and microsatellite data with the software DIYABC (v1.0). BMC Bioinform. 11, 401 (2010).Article 
CAS 

Google Scholar  More

Highland lakes in southwestern China supply water to more than 1.4 billion people. Increasingly subject to eutrophication, biodiversity loss, drought and pollution, the lakes urgently need integrated management by government, community stakeholders and scientists to guide development of watershed policy and address these challenges.
Competing Interests
The authors declare no competing interests. More