Ecology

Single plasmid, single-population modelsTo understand the dynamics of parasitic plasmids in complex ecologies, we first need to understand their behavior in simple scenarios. In this section, we analyze the dynamics of plasmids spreading by different HGT mechanisms in single populations. We begin by modeling competition between plasmid-free cells and cells containing a conjugative plasmid. A nutrient, with concentration (C), is supplied to the system at rate (S). Cells grow at a rate proportional to (C) with proportionality constant (alpha) for plasmid-free cells or ((1 ,-, {Delta})alpha) for plasmid-containing cells. Since we are interested in parasitic plasmids, we assume that ({Delta} in (0,1)). Cells of both types die at a rate (delta). When a plasmid-containing cell divides there is a loss probability, (p_ell), for one of the daughter cells to contain no plasmids. As long as a daughter cell contains at least one plasmid, the original plasmid copy number (the number of copies of the plasmid maintained per cell) is regenerated (as depicted in Fig. 1A). Plasmids can spread horizontally by conjugation, as illustrated in Fig. 1B, wherein a plasmid-free cell and a plasmid-containing cell interact to produce two plasmid-containing cells. We model the rate of conjugation by a mass-action term with rate (gamma _{mathrm{c}}). The equations governing the dynamics of conjugation are therefore:$$ frac{{drho }}{{dt}} ,=, alpha Crho ,-, gamma _{mathrm{c}}rho rho _{mathrm{p}} ,+, p_ell (1 ,-, {Delta})alpha Crho _{mathrm{p}} ,-, delta rho ,\ frac{{drho _{mathrm{p}}}}{{dt}} ,=, (1 ,-, {Delta})alpha Crho _{mathrm{p}} ,+, gamma _{mathrm{c}}rho rho _{mathrm{p}} ,-, p_ell (1 ,-, {Delta})alpha Crho _{mathrm{p}} ,-, delta rho _{mathrm{p}},\ frac{{dC}}{{dt}} ,=, S ,-, alpha Crho ,-, (1 ,-, {Delta})alpha Crho _{mathrm{p}}.$$
(1)
Fig. 1: Different modeled mechanisms of plasmid transfer lead to distinct ecological phase diagrams, but all such mechanisms leave individual populations susceptible to runaway plasmid invasion.A At each division, plasmids are randomly segregated between daughter cells. Original plasmid copy number is regenerated if at least one plasmid remains in a daughter cell. B Schematic of plasmid transfer mechanisms. Left: spread of plasmids by plasmid-containing cells conjugating with plasmid-free cells. Right: spread of plasmids by extracellular plasmids infecting plasmid-free cells via transformation. C Phase diagram for conjugative plasmids as a function of plasmid cost, ({Delta}), and (gamma _{mathrm{c}}); (delta ,=, 0.1), (S ,=, 1), (p_ell ,=, 0), and (alpha ,=, 1) (see Eq. 4). D Phase diagram for transformative plasmids as a function of ({Delta}) and (gamma _{mathrm{t}}). Parameters as in C with (delta _{mathrm{p}} ,=, 0.3) and (n_{{mathrm{eff}}} ,=, 0.6) (see Eq. 9). See “Methods” for details. E In model multiplasmid cells, plasmid types segregate independently. If at least one plasmid of a given type remains in a daughter cell, the full copy number of that plasmid type is regenerated. F Fitness cost as a function of number of unique plasmid types in a cell for multiplicative case ({Delta}_{{mathrm{tot}}} ,=, 1 ,-, (1 ,-, {Delta})^m) with ({Delta} ,=, 0.05). G Steady-state distribution of number of plasmid types per cell at different conjugation rates, measured relative to (gamma _{mathrm{c}}^ ast) (the critical conjugation rate necessary for invasion of a single plasmid into a plasmid-free population, see Eq. 4). Results for eight unique plasmid types with (delta ,=, 1), ({Delta} ,=, 0.1), (alpha ,=, 1), (S ,=, 1), and (p_ell ,=, 0.05).Full size imageIn this model, what are the conditions for a parasitic conjugative plasmid to be able to invade a plasmid-free population? Invasibility implies that the equilibrium containing only plasmid-free cells is locally unstable, which occurs when$$qquadqquadqquadgamma _{mathrm{c}}rho ^ ast , > , delta {Delta} ,+, delta p_ell (1 ,-, {Delta}),$$
(2)
where (rho ^ ast ,=, S/delta) is the steady-state abundance of the plasmid-free cells at the plasmid-free equilibrium. This invasibility condition has an intuitive physical interpretation: to invade, the rate of conjugation must overcome losses due to reduced host growth rate as well as plasmid loss during division. This condition is similar to those found in previous studies [15].Given the condition for plasmid invasion in Eq. 4, what is the optimal behavior for a parasitic conjugative plasmid? The left-hand-side of the expression is linear in the plasmid-free population, meaning that it is more difficult for a plasmid to invade smaller populations. To favor invasion, the plasmid can minimize the right-hand-side of the equation. For a plasmid that relies on random segregation upon cell division, both the plasmid cost ({Delta}) and the loss probability (p_ell) are functions of plasmid copy number, (n_{mathrm{p}}), a property controlled by the plasmid itself. If the primary cost of a plasmid is its replication and its gene products, plasmid cost will scale with copy number such that ({Delta} ,=, {Delta}_{mathrm{p}}n_{mathrm{p}}), where ({Delta}_{mathrm{p}}) is the cost of an individual plasmid copy. The loss probability will be (p_ell ,=, 2^{1 ,-, n_{mathrm{p}}}), i.e., the probability that a daughter cell receives zero plasmids from random segregation. The right-hand-side of the invasion condition Eq. 4 is therefore (delta ({Delta}_{mathrm{p}}n_{mathrm{p}} ,+, 2^{1 ,-, n_{mathrm{p}}}(1 ,-, {Delta}_{mathrm{p}}n_{mathrm{p}}))), which has a minimum at finite (n_{mathrm{p}}). The minimum in the invasion boundary at finite (n_{mathrm{p}}) indicates that in our framework optimal conjugative plasmids have a moderate copy number.What kinds of ecological dynamics does our model for a conjugative parasitic plasmid exhibit? To answer this question, we characterize the stability of the system’s equilibria (see SI Appendix 1 for details). For conjugative plasmids with the optimal copy number, the dominant form of loss will be from reduced host fitness (see SI Fig. S1), and thus we characterize the case of negligible loss rate (p_ell ,=, 0) (we consider the case of finite loss rates in SI Fig. S2 and find similar results). In Fig. 1C we show the phase diagram of possible ecological outcomes as a function of plasmid cost ({Delta}) and conjugation rate (gamma _{mathrm{c}}). For high values of plasmid cost and low values of conjugation rate, the plasmid is unable to invade and the plasmid-free equilibrium is the only stable state. As plasmid cost decreases or conjugation rate increases, plasmids are able to invade and there is a state of stable coexistence between plasmid-free and plasmid-containing cells. The range of conjugation rates permitting coexistence is larger for costlier plasmids. Once the plasmid cost is sufficiently low or the conjugation rate is sufficiently high, the unique stable state consists only of plasmid-containing cells (note that for finite values of loss rate (p_ell), this plasmid-only state will contain a small fraction of plasmid-free cells due to plasmid loss).Conjugation is the best studied mechanism of plasmid transmission, but plasmids can instead be transmitted by transformation, whereby plasmid-free cells are infected by free-floating plasmids, as illustrated in Fig. 1B. We therefore consider a model for plasmid-spread via transformation in which cell death results in release of free-floating plasmids which can then infect cells by mass action at rate (gamma _{mathrm{t}}). For every cell death, (n_{{mathrm{eff}}}) free-floating plasmids are released and these plasmids decay at a rate (delta _{mathrm{p}}). The dynamics of transformative plasmids are therefore:$$ frac{{drho }}{{dt}} ,=, alpha Crho – gamma _{mathrm{t}}rho P ,+, p_ell (1 ,-, {Delta})alpha Crho _{mathrm{p}} ,-, delta rho ,\ frac{{drho _{mathrm{p}}}}{{dt}} ,=, (1 ,-, {Delta})alpha Crho _{mathrm{p}} ,+, gamma _{mathrm{t}}rho P ,-, p_ell (1 ,-, {Delta})alpha Crho _{mathrm{p}} ,-, delta rho _{mathrm{p}},\ frac{{dC}}{{dt}} ,=, S ,-, alpha Crho ,-, (1 ,-, {Delta})alpha Crho _{mathrm{p}},\ frac{{dP}}{{dt}} ,=, n_{{mathrm{eff}}}delta rho _{mathrm{p}} ,-, gamma _{mathrm{t}}rho P ,-, delta _{mathrm{p}}P.$$
(3)
What is the condition for transformative plasmid invasion? The plasmid-free equilibrium is unstable if$$qquadqquadquadgamma _{mathrm{t}}rho ^ ast , > , delta _{mathrm{p}}left( {frac{{{Delta} ,+, p_ell (1 ,-, {Delta})}}{{n_{{mathrm{eff}}} ,-, {Delta} ,-, p_ell (1 ,-, {Delta})}}} right).$$
(4)
The left-hand-side of Eq. 9 is similar to the conjugative plasmid invasion condition, with the conjugation rate (gamma _{mathrm{c}}) replaced by the transformation rate (gamma _{mathrm{t}}). The numerator of the right-hand-side is also similar, with the cell death rate (delta) replaced with the plasmid decay rate (delta _{mathrm{p}}). The primary difference is in the denominator, which is the difference between the number of plasmids released on cell death, (n_{{mathrm{eff}}}), and the total replication deficit of plasmid-containing cells. If this denominator is negative, the inequality reverses and the plasmid-free equilibrium is always stable.The invasion condition in Eq. 9 determines the optimal (n_{mathrm{p}}) of transformative plasmids: if each plasmid within a cell has a fixed probability of remaining viable after cell death, (p_{mathrm{v}}), then (n_{{mathrm{eff}}}) will scale linearly with (n_{mathrm{p}}) such that (n_{{mathrm{eff}}} ,=, p_{mathrm{v}}n_{mathrm{p}}). If the denominator of Eq. 9 is positive, the optimal copy number will be (n_{mathrm{p}} ,=, 1/{Delta}_{mathrm{p}}), the point at which the host’s growth rate is driven to zero and the plasmid relies entirely on horizontal transfer to survive. These results are substantially different than in the case of conjugation: instead of restricting itself to a limited portion of the host’s metabolic budget, a transformative parasite maximizes its spread by using as much of the host’s resources as possible. This is reminiscent of the behavior of phages—suggesting a possible evolutionary link between parasitic plasmids and phages.As in the conjugation case, we now explore the ecological outcomes possible with transformative plasmids. We again consider the case of negligible loss rate (p_ell ,=, 0) and characterize the stability of the equilibria (see SI Appendix 1 for details). For (n_{{mathrm{eff}}} , > , 1), the system has similar ecological outcomes to the conjugative case, with the system transitioning through no-plasmid, coexistence, and plasmid-only equilibria as ({Delta}) decreases and (gamma _{mathrm{t}}) increases. Interestingly, when (n_{{mathrm{eff}}} , , 0.} end{array}$$
(5)
Fig. 2: Competition between populations may prevent runaway plasmid invasion.A Illustration of multiple populations, each occupying an isolated “deme”. During each epoch, populations compete for demes, with plasmid invasion occurring randomly (see Eq. 11 for details). In the example shown, in the first epoch, the population with two plasmids is replaced by the population with zero plasmids. In the second epoch, the population with magenta plasmids is invaded by the green plasmid. B Multiplasmid fitness costs for different types of epistasis. With no epistasis, fitness burden is multiplicative as in Fig. 1F. With positive epistasis, fitness burden increases sub-multiplicatively (pictured: ({Delta}_{{mathrm{tot}}} ,=, {Delta}) for (m , > , 0)). For negative epistasis, fitness burden increases super-multiplicatively (pictured: ({Delta}_{{mathrm{tot}}} ,=, 1 ,-, (1 ,-, {Delta})^{m^{3/2}})). C Steady-state distributions of number of plasmid types per cell in the Wright–Fisher model (see SI Appendix 3). Parameters ({Delta} ,=, 0.01) and plasmid invasion probability for each time period (q ,=, 0.005).Full size imageA population’s fitness is dependent on the number of unique plasmid types it contains. Thus far, we have considered a simple multiplicative model. However, it has been demonstrated that plasmid–plasmid interactions can modulate plasmid properties. For example, one study found that the presence of a plasmid can reduce the fitness cost of an invading plasmid [12]. To account for this epistasis between plasmids, we also consider fitness costs that increase sub-multiplicatively (positive epistasis) or super-multiplicatively (negative epistasis). We show examples of positive epistasis, negative epistasis, and no epistasis in Fig. 2B.What is the distribution of unique plasmid types across populations in our model with HGT barriers? We derive the stationary distribution of this model for the three different epistasis functions in Fig. 2B and plot them in Fig. 2C (see SI Appendix 3 for details). For the case of no epistasis, the stationary distribution is Poisson-like. Positive epistasis favors carriage of multiple plasmids and results in an exponential-like distribution with a long tail. Negative epistasis has the opposite effect: it penalizes carriage of multiple plasmids and results in a sub-Poissonian distribution with a reduced tail. Importantly, in all cases the runaway invasion of plasmids is stopped. While there is nothing stopping individual populations from being overrun by invading plasmids, these populations are more likely to be out-competed by populations with fewer plasmids. Thus, the single-population “tragedy of the commons” is counteracted at a higher level of selection.Analysis of natural genomesHow does our predicted distribution of unique plasmid types per cell compare to that in natural genomes? To make this comparison, we downloaded all complete bacterial genomes from NCBI (a total of 17,725 genomes) and analyzed their plasmid content. In Fig. 3A, we show the overall distribution of unique plasmid types per genome and corresponding model fits for both positive and no epistasis cases (see “Methods” for fitting details). The natural distribution is exponential-like and is well-fit by a model with positive epistasis. The model fit with no epistasis has too short a tail to be able to fit the data, and this problem becomes even more severe for negative epistasis. Thus, interestingly, we find that the distribution of unique plasmid types in real-world genomes is consistent with parasitic plasmids that ameliorate each other’s fitness costs. The degree of positive epistasis suggested by the data is quite strong—the distribution is nearly a pure exponential. In our model, this corresponds to the case in which the cost of all plasmids beyond the first is zero, such that for (m , > , 1) the parameters controlling both population replication and plasmid invasion are independent of plasmid number. This means that the ratio between consecutive elements of the distribution is constant, yielding an exponential tail. In order to determine whether our conclusions are influenced by oversampling of clinically relevant species, we excluded 91 genera known to be clinically relevant or human-associated and repeated our analysis. The remaining dataset contains nearly 5000 genomes and still shows clear exponential behavior (see SI Fig. S4). We also analyzed whether the presence of engineered strains within the NCBI database influences our results. We found that there are only a small number of these engineered strains and that removing them had negligible impact on our results (see SI Fig. S5).Fig. 3: Comparison of distributions of number of unique plasmid types per cell in natural genomes to Wright–Fisher model.A Distribution of number of plasmid types per cell in 17,725 complete NCBI genomes. Positive epistasis distribution fit with the fitness function ({Delta}_{{mathrm{tot}}} ,=, {Delta}) for (m , > , 0) (best-fit parameters: ({Delta} ,=, 9.8 ,times, 10^{ – 3}), (q ,=, 5.4 ,times, 10^{ – 3})), no epistasis distribution fit with ({Delta}_{{mathrm{tot}}} ,=, 1 ,-, (1 ,-, {Delta})^m) (best-fit parameters: ({Delta} ,=, 3.9 ,times, 10^{ – 3}), (q ,=, 1.4 ,times, 10^{ – 2})). B Distribution of number of plasmid types per cell in 1153 complete Escherichia genomes, with a positive epistasis fit using the fitness function ({Delta}_{{mathrm{tot}}} ,=, 1 ,-, (1 ,-, {Delta})^{m^a}) (best-fit parameters: ({Delta} ,=, 8.3 ,times, 10^{ – 3}), (q ,=, 8 ,times, 10^{ – 3}), (a ,=, 0.33)). C Distribution of number of plasmid types per cell in 576 complete Klebsiella genomes, with a positive epistasis fit using the fitness function as in (B) (best-fit parameters: ({Delta} ,=, 7 ,times, 10^{ – 3}), (q ,=, 9.7 ,times, 10^{ – 3}), (a ,=, 0.43)). Note that in certain limits of our models, only the ratio of (q) and ({Delta}) can be properly estimated, effectively reducing them to single parameter (see SI Appendix 3). D Distribution of number of plasmid types per cell in genomes containing and not containing cas genes. Genomes are considered cas containing if at least one chromosome or plasmid within the genome contains a cas gene. See “Methods” for details.Full size imageCan our model capture variation within smaller, related groups of genomes? In Fig. 3B we show the distribution of unique plasmid types per cell within the genus Escherichia. As can be seen, the data is very well fit by a model of parasitic plasmids with positive epistasis. However, our model was not able to capture some of the within-genus distributions we encountered. A notable exception is the distribution of unique plasmid types per cell in the genus Klebsiella, shown in Fig. 3C. In this genus, there is a substantial discontinuity between the zero-plasmid class and the rest of the distribution. While our simple Wright–Fisher model with some positive epistasis can capture the tail of the distribution, it then fails to capture the first few classes. Despite such exceptions, we find that the positive epistasis model is generally able to capture the overall trends in plasmid distributions over the bulk of natural genomes (see SI Fig. S6).It should be noted that our current model of constant plasmid invasion probability and strong positive epistasis is not the only Wright–Fisher model that can produce an exponential distribution matching the data. We analyzed a more general form of the Wright–Fisher model in which the invasion probability and total fitness cost are arbitrary functions of unique plasmid number (see SI Appendix). We find that the general condition to yield an exponential is that the plasmid invasion probability and total fitness cost must be comparable regardless of the number of plasmids in the cell. These results indicate that even if there is no epistasis in fitness cost, an exponential can still result if there is positive epistasis in the invasion probability (i.e., if existing plasmids make it more likely for a new plasmid to successfully invade).HGT barriers are not the only mechanism that can plausibly limit runaway plasmid invasion. Cells also have specialized systems to defend against foreign DNA, notably the CRISPR-Cas system [32]. To explore whether CRISPR-Cas is responsible for limiting plasmid invasion in natural genomes, we searched for cas genes within the NCBI complete bacterial genomes using HMMER (see “Methods” for details). We expect that if CRISPR-Cas plays a major role in limiting the spread of plasmids, the distribution of unique plasmid types per cell would be shifted towards lower plasmid numbers in genomes containing cas genes versus those lacking cas genes. In Fig. 3D, we show the distribution of unique plasmid types per genome in genomes containing at least one cas gene and those not containing any cas genes. The distributions are very similar, with no large differences between them. These results suggest that CRISPR-Cas is not a major mechanism limiting the spread of plasmids in bacteria. There are additional defense systems that may also influence plasmid carriage. However, a prior bioinformatics study found results similar to ours for restriction-modification (RM) systems, another defense system that protects against foreign DNA; the study examined the distribution of RM systems in bacterial genomes and found almost no relationship between the number of RM systems a genome encodes and the presence of plasmids (in one subset of data the authors actually found a positive relation) [33]. More

Here we present five current shortcomings identified in China’s biodiversity protection framework.Varying threat-assessment quality and uniform treatment of speciesIn this section, we highlight how the threat classifications of the Catalogue of Wildlife under Special State Conservation can lead to sentences that are not commensurate with the species’ threat level. In recent amendments to the catalogue, insect species occur in the highest protection classes (3 species out of 234 in Class I and 72 species out of 746 in Class II; Fig. 2) with similar sentencing standards as for large mammals and birds. For instance, killing more than six individuals of Class I protected insects is treated equally to killing one giant panda, with a punishment of at least ten years’ imprisonment according to the Judicial Interpretation of Several Questions Concerning the Application of Law in the Trial of Criminal Cases of Destruction of Wildlife Resources.Fig. 2: Example species with the highest protection status but considerably different life histories.a,b, Mammals such as the giant panda (a) and insects such as the butterfly T. aureus (b) both occur in the highest protection category in the Catalogue of Wildlife under Special State Conservation. Credit: Juping Zeng (b).Full size imageIn June 2002, 10 poachers captured 263 adults of the butterfly Teinopalpus aureus, meant to be sold on the black market. As T. aureus is listed in Class I of the Catalogue of Wildlife under Special State Conservation, based on the assumption of being rare, the punishment was 5 to 13 years’ imprisonment20. However, recent observations indicate both a wider distribution range21,22 and larger population sizes than initially assumed23. Further, the reproduction rate of insects is generally much higher than that of mammals, which usually makes insects more resistant to the removal of specimens. This case raised some controversy about the scientific basis for classification and the financial profit that can be made with insects compared with mammals24. On the black market, T. aureus can be sold for 700 Chinese yuan per male (~US$100; US$1 = 6.9932 yuan, 21 July 2020; gross domestic product (GDP) per capita: 30,808 yuan in 2010, 54,139 yuan in 2016) and 3,500 yuan per female (~US$500; personal communication with collectors in 2011), while a pair of giant pandas is usually rented to abroad zoos for about 7 million yuan (~US$1 million) per year25.In 2015, a college student and a farmer took 16 fledglings of the Eurasian hobby (Falco subbuteo), a Class II protected species, and were sentenced to 10.5 and 10 years’ imprisonment and fines of 10,000 and 5,000 yuan, respectively26. However, ecological studies indicate that the distribution range, population density and reproduction rate of F. subbuteo in China seem sufficient for sustaining viable populations27, highlighting the potential of overly harsh punishment when classification lacks scientific basis.In contrast to valuation according to (black) market prices, wild species also provide higher-level socioeconomic benefits28. For instance, the value of insect pollination services in China was estimated to be 886.5 billion yuan (US$131 billion) in 201529. In comparison, the ecosystem services related to the giant panda were estimated at between 18 billion and 48 billion yuan per year (US$2.6–6.9 billion) in 2010, but they seem more indirect via regulating, provisioning and cultural services provided by the panda reserves30. However, pollination services are provided by multiple species within a highly flexible network31,32 and the impact of removing a particular amount of specimens is hard to assess, whereas large mammals, such as the giant panda, are irreplaceable in ecosystems and their roles as umbrella species. Thus, differences between insects and mammals are striking not only in terms of direct financial profit but also in terms of ecological and socioeconomic damage, and therefore it is questionable that they are both listed in the highest protection class with the same stringent punishment.Lack of quantitative sentencing standards for herbaceous plants, fungi and algaeHere, we discuss how limited scientific knowledge for particular species groups can lead to legal uncertainties and consequently to limited protection or overly harsh punishment. The Regulations of the People’s Republic of China on the Protection of Wild Plants identify the legal responsibilities for the protection of wild plants (excluding trees), but have not yet reached the status of a law and thus are without judicial interpretation of the Supreme People’s Court and respective sentencing standards. Instead, stipulations of ‘seriousness’ are used with regard to the sentences used for trees, defined in the Judicial Interpretation of Several Questions Concerning the Specific Application of Law in the Trial of Criminal Cases of Destruction of Forest Resources (Box 1), and respective sentencing standards, defined in the Criminal Law of the People’s Republic of China, are applied (up to seven years’ imprisonment). With this analogy, an offender was sentenced to three years in prison in 2016 (suspended sentence) and a fine of 1,000 yuan for digging out three stems of Cymbidium faberi33, an orchid listed in Appendix II of the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES34; Fig. 3d) but with high market value. Some uncertainty in the legal position regarding herbaceous plants is expressed by another case in the same year, in which an offender was sentenced to one year of imprisonment (fine of 5,000 yuan) for digging out 55 stems of C. faberi35, and the later revocation of the sentences given that C. faberi is not listed in the Catalogue of Wild Plants under Special State Conservation36.Fig. 3: Example species with changing threat status.a–d, Wildlife protection laws need to be adaptive to reflect the recovery of formerly threatened species, such as the snow leopard (Panthera uncia; a) or the kiang (E. kiang; b), or the increasing endangerment of initially non-threatened species, such as the butterfly Bhutanitis lidderdalii (c) or the orchid C. faberi (d). Credit: Zhi Lu (a, b); Lixin Zhu (c); Yu Ren (d).Full size imageSimilar to the non-discrimination of large mammals and insects, we find such an approach also questionable for precious trees and other plants. Such analogies might become almost impossible when applied to algae such as Nostoc flagelliforme, an important water and soil conservation and high-priced food algae but under Class I protection37. The main reason for the lack of quantitative sentencing standards for these organisms is limited evidence. Therefore, we think it is necessary to raise the Regulations of the People’s Republic of China on the Protection of Wild Plants to become law with respective judicial interpretations and to establish comprehensive scientific assessments targeting herbaceous plants, fungi and algae to provide a solid basis for the development of sentencing standards.Lack of legislative flexibility to reflect dynamic changes in status and taxonomyWe identified a lack of regular updates of the Catalogues of Wildlife and Wild Plants under Special State Conservation needed to address the dynamic changes in taxonomy and threat status. Since its promulgation, the Wildlife Protection Law of the People’s Republic of China has been revised four times and the Regulations of the People’s Republic of China on the Protection of Wild Plants was amended once in 200138, but the Catalogues of Wildlife and Wild Plants under Special State Conservation have basically remained unchanged for the past 32 and 20 years, respectively, with the exception of a recent amendment of the Catalogue of Wildlife in February 2021 and a pending amendment of the Catalogue of Wild Plants (Box 1). Taxonomies change dynamically, which can lead to considerable incongruences among scientifically accepted species names and those in the respective protection lists39. Until this recent amendment, there has been a mismatch in the names of 25 threatened species as listed under CITES compared with the Catalogue of Wildlife under Special State Conservation, putting them at particular risk because their protection status might be questioned, for example, when species such as the Himalayan goral (Naemorhedus goral), or even genera such as the leaf monkeys (Presbytis spp.), have been split into different units with different names that are not listed in the respective catalogues40. Although the Catalogue of Wildlife under Special State Conservation has been updated very recently, it is still recommended that such updates are done regularly and in a coordinated manner, not only in China but across all CITES signatory nations40.Additional legislative flexibility is also needed when formerly endangered species have recovered11, while others have become endangered16,41 (Fig. 3). Recently, several mammals such as the giant panda, snow leopards or the kiang (Equus kiang)11,42 have considerably recovered and their threat status has been reduced by the International Union for Conservation of Nature (IUCN)11. Although the Chinese government does not follow such a downgrade because of precautionary reasons, we think that the sentencing threshold for such species should be adapted in the Judicial Interpretation of Several Questions Concerning the Application of Law in the Trial of Criminal Cases of Destruction of Wildlife Resources. On the other hand, species whose endangerment has increased since the promulgation of the Catalogues of Wildlife and Wild Plants under Special State Conservation, such as the narrow-ridged finless porpoise43, many birds44, snakes45, turtles46, frogs40, butterflies47 or herbaceous (medicinal) plants2, have long been with low or no protection until the recent amendment. Cultivation can also increase endangerment of wild species by hybridization between the cultivars and the wild populations (for example, rice, wheat, soybean and cotton)48.Outdated punishment standards based on economic profitsSimilar to the lack of flexibility covering species’ taxonomic and threat status, here we highlight that punishment standards are outdated and regular updates are required to reflect economic developments and guarantee balanced sentencing. For instance, according to the Judicial Interpretation of Several Questions Concerning the Application of Law in the Trial of Criminal Cases of Destruction of Wildlife Resources, the illegal purchase, transport and sale of precious and endangered wildlife products will be considered as a ‘serious crime’ if the financial profit is more than 100,000 yuan and as an ‘extremely serious crime’ if the profit is 200,000 yuan or more. The sentencing standard was developed in the year 2000, but with the rapid development of China’s economy, nationwide per capita income has increased more than fourfold from 6,279 yuan in 2000 to 28,228 yuan in 201849. To reflect economic developments, the penalty standards need to be adjusted to comply with the principle of balanced sentencing. In comparison, the Chinese standards for corruption and bribery have been increased from 4,886 yuan in 1997 to currently 30,715 yuan for crimes involving a ‘relatively large amount’, which might serve as a guideline for adapting the sentencing standards for wildlife protection50.Potential for excessive punishment because of non-discrimination between organized and individual wildlife crimeIn this section, we highlight that ignoring the motivational, educational and economic backgrounds of offenders is against the principle of proportionality and may lead to inappropriate deterrence strategies. China’s laws are very strict with quite harsh penalty sentencing; for example, 10.5 years’ imprisonment and a fine of 10,000 yuan for a student taking birds26, 12 years and a fine of 10,000 yuan for a farmer killing a giant panda51 or 13 years and a fine of 2,000 yuan for a farmer taking butterflies20, all cases representing ‘extremely serious crimes’ with a minimum sentencing standard of 10 years’ imprisonment (no maximum defined). Even in comparison with other criminal fields in China and internationally, these standards seem very stringent. For instance, sentences of more than 10 years’ imprisonment apply to larceny only if the value of the stolen goods is larger than 500,000 yuan, or to the theft of first-class cultural relics (all valued in the millions; Criminal Law of the People’s Republic of China, Article 264). Also in comparison, the United Nations Convention Against Transnational Organized Crime52 defines much lower sentencing standards, with at least four years’ imprisonment for a ‘serious crime’. In contrast to China, the wildlife protection laws of Western and many other developing countries prioritize monetary fines over imprisonment. Under European wildlife law53, for example, hunting or destroying Class I protected species is generally punishable by a fine and will be sentenced with fixed-term imprisonment only if the case is ‘extremely serious’. In the United States, the maximum imprisonment is a year, with fines of up to US$50,000 (340,000 yuan)54; in the UK, 6 months and fines of up to £20,000 (177,000 yuan)55,56; in India, 3–7 years and a minimum fine of 25,000 rupees (2,300 yuan)57; or in Brazil, 3 months to a year plus fines58.The wildlife protection laws of such countries may provide useful examples for China, but to adhere to the principle of proportionality, motivational, educational and economic backgrounds, in particular a differentiation between organized wildlife crimes and individual violations needs to be considered. Individual and organized crimes are currently not differentiated in the Criminal Law of the People’s Republic of China. Historically, wildlife crime was considered a local activity performed by single individuals. However, at present criminal networks are highly involved59 and resulting economic damage from environmental crime has been estimated to range between US$91 billion and US$259 billion globally60, with the profits of illegal wildlife trade ranging between US$7 billion and US$23 billion61, which is of similar orders to human trafficking, and arms and drug dealing62. In China, the consumption of illegal wildlife products has increased with growing economic wealth63, while China has also been identified as one of the major exporters of such products64. Key players in both cases are organized crime groups65,66, causing severe ecological damage while making enormous financial profits67. In such cases, high fines might be simply factored in as part of the ‘business model’. Thus, the current focus on severe jail sentences seems appropriate, and the level is comparable to other Southeast Asian countries (Indonesia: 10 years; Singapore: 2 years; Thailand: 7 years; Vietnam: 15 years)68,69.In contrast to organized wildlife crime, we also noticed that many cases of harvesting or poaching protected wildlife happened in remote and less-developed regions, conducted by individuals seeking to earn some extra income but without good knowledge of the protection laws20,51. The resulting ecological damage and profits gained are much lower compared with cases of organized wildlife crime, and thus applying the same harsh punishments, as shown in our earlier examples, is clearly against the principle of proportionality. Moreover, it has been shown that the mentality of different types of offender and how they perceive different punishments (imprisonment, fines or both) need to be considered for designing appropriate deterrence strategies for different offence categories, suggesting that imprisonment as the main policy instrument is inappropriate70. Imprisonment is not necessarily a deterrent for every offender, especially when the price of time in prison falls relative to the price of time outside71. Consequently, a penalty that eliminates any financial gain should eliminate the incentive to engage in such conduct72. A shift in focus from imprisonment to fines, at best coupled with local or regional GDP per capita and in combination with raising public awareness, might not only increase proportionality and effectiveness of environmental laws but also comply with other international standards, where, for example, the Council of Europe’s Recommendation (92)17, concerning consistency in sentencing, paragraph B5(2), states that “custodial sentences should be regarded as a sanction of last resort, and should therefore be imposed only in cases where, taking due account of other relevant circumstances, the seriousness of the offence would make any other sentence clearly inadequate”. More

1.Tapia, J., Audry, S., Townley, B. & Duprey, J. L. Geochemical background, baseline and origin of contaminants from sediments in the mining-impacted Altiplano and Eastern Cordillera of Oruro, Bolivia. Geochemistry 12, 3–20. https://doi.org/10.1144/1467-7873/10-RA-049 (2012).CAS 
Article 

Google Scholar 
2.Sarricolea, E. P. & Romero, H. Variabilidad y cambios climáticos observados y esperados en el Altiplano del norte de Chile. Revista de Geografía Norte Grande. 62, 169–183 (2015).Article 

Google Scholar 
3.Garreaud, R., Vuille, M. & Clement, C. A. The climate of the Altiplano: observed current conditions and mechanisms of past changes. Palaeogeogr. Palaeoclimatol. Palaeoecol. 194(1–3), 5–22. https://doi.org/10.1016/S0031-0182(03)00269-4 (2003).Article 

Google Scholar 
4.Vuille, M. & Keiming, F. Interannual variability of summertime convective cloudiness and precipitation in the central andes derived from ISCCP-B3 data. J. Clim. 17(17), 3334–3348. https://doi.org/10.1175/15200442(2004)017%3c3334:IVOSCC%3e2.0.CO;2 (2004).ADS 
Article 

Google Scholar 
5.Cerveny, R. Present climates of South America. In Climates of the Southern Continents: Present, Past and Future (ed. Hobbs, J. E.) 107–135 (Wiley, Chichester, 1998).
Google Scholar 
6.Garreaud, R. & Aceituno, P. Interannual rainfall variability over the South American Altiplano. J. Clim. 14(12), 2779–2789. https://doi.org/10.1175/1520-0442(2001)014%3c2779:IRVOTS%3e2.0.CO;2 (2001).ADS 
Article 

Google Scholar 
7.Coronel, J., Declerck, S. & Brendonck, L. High-altitude peatland temporary pools in Bolivia house a high cladoceran diversity. Wetlands 27(4), 1166–1174. https://doi.org/10.1672/0277-5212(2007)27[1166:HPTPIB]2.0.CO;2 (2007).Article 

Google Scholar 
8.Dorador, C., Vila, I., Witzel, K. P. & Imhoff, J. F. Bacterial and archaeal diversity in high altitude wetlands of the Chilean Altiplano. Fundam. Appl. Limnol. 182(2), 135–159. https://doi.org/10.1127/1863-9135/2013/0393 (2013).CAS 
Article 

Google Scholar 
9.Garcia, E. & Otto, M. Caracterización ecohidrológica de humedales alto andinos usando imágenes de satélite multitemporales en la cabecera de cuenca del río Santa, Ancash, Perú. Ecología Aplicada 14(2), 115–125 (2013).
Google Scholar 
10.Buytaert, W., Camacho, F. C. & Tobón, C. Potential impacts of climate change on the environmental services of humid tropical alpine regions. Glob. Ecol. Biogeogr. 20(1), 19–33. https://doi.org/10.1111/j.1466-8238.2010.00585.x (2011).Article 

Google Scholar 
11.Hribljan, J. A. et al. Carbon storage and long-term rate of accumulation in high-altitude Andean peatlands of Bolivia. Mires Peat 15(12), 1–14 (2015).
Google Scholar 
12.Yager, K. et al. Dimensiones socioecológicas del cambio del paisaje pastoral andino: puente entre el conocimiento ecológico tradicional y el análisis de imágenes satelitales en Sajama Parque Nacional, Bolivia. Cambio ambiental regional 17, 27–37 (2019).
Google Scholar 
13.Urrutia, R. & Vuille, M. Climate change projections for the tropical Andes using a regional climate model: temperature and precipitation simulations for the end of the 21st century. J. Geophys. Res. 114, D02108. https://doi.org/10.1029/2008JD011021 (2009).ADS 
Article 

Google Scholar 
14.Buytaert, W. et al. Uncertainties in climate change projections and regional downscaling in the tropical Andes: implications for water resources management. Hydrol. Earth. Syst. Sci. 14, 1247–1258. https://doi.org/10.5194/hess-14-1247-2010 (2010).ADS 
Article 

Google Scholar 
15.Rabatel, A. et al. Current state of glaciers in the tropical Andes: a multi-century perspective on glacier evolution and climate change. Cryosphere 7, 81–102. https://doi.org/10.5194/tc-7-81-2013 (2013).ADS 
Article 

Google Scholar 
16.Prieto, G. et al. A mass sacrifice of children and camelids at the Huanchaquito-Las Llamas site, Moche Valley, Peru . PLoS ONE 14(3), e0211691. https://doi.org/10.1371/journal.pone.0211691 (2019).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
17.Babidge, S., Kalazich, F., Prieto, M. & Yager, K. That’s the problem with that lake; it changes sides’: mapping extraction and ecological exhaustion in the Atacama. J. Political Ecol. 26(1), 738–760. https://doi.org/10.2458/v26i1.23169 (2019).Article 

Google Scholar 
18.Prieto, M., Fragkou, M. & Calderón, M. Water policy and management in Chile. In The Wiley Encyclopedia of Water: Science, Technology, and Society (ed. Strickland, C.) 2589–2600 (Wiley, New York, 2020).
Google Scholar 
19.Fritz, S. C., Baker, P. A., Tapia, P., Spanbauer, T. & Westover, K. Evolution of the Lake Titicaca basin and its diatom flora over the last 370,000 years. Palaeogeogr. Palaeoclim. Palaeoecol. 317–318, 93–103 (2012).ADS 
Article 

Google Scholar 
20.Cohen, S. C. Scientific drilling and biological evolution in ancient lakes: lessons learned and recommendations for the future. Hydrobiologia 682(1), 3–25. https://doi.org/10.1007/s10750-010-0546-7 (2012).CAS 
Article 

Google Scholar 
21.Tapia, P. M., Fritz, S. C., Baker, P. A., Seltzer, G. A. & Dunbar, R. B. A Late Quaternary diatom record of tropical climatic history from Lake Titicaca (Peru and Bolivia). Palaeogeogr. Palaeoclimatol. Palaeoecol. 194, 1–3. https://doi.org/10.1016/S0031-0182(03)00275-X (2003).Article 

Google Scholar 
22.Vining, B. R., Steinman, B. A., Abbott, M. B. & Woods, A. Paleoclimatic and archaeological evidence from Lake Suches for highland Andean refugia during the arid middle-Holocene. The Holocene 29(2), 328–344. https://doi.org/10.1177/0959683618810405 (2019).ADS 
Article 

Google Scholar 
23.Fritz, S. C., Baker, P. A., Tapia, P. & Garland, J. Spatial and temporal variation in cores from Lake Titicaca, Bolivia/Peru during the last 13,000 years. Quat. Int. 158(1), 23–29. https://doi.org/10.1016/j.quaint.2006.05.014 (2006).Article 

Google Scholar 
24.Hernández, A. et al. Biogeochemical processes controlling oxygen and carbon isotopes of diatom silica in Late Glacial to Holocene lacustrine rhythmites. Palaeogeogr. Palaeoclimatol. Palaeoecol. 299(3–4), 413–425. https://doi.org/10.1016/j.palaeo.2010.11.020 (2012).Article 

Google Scholar 
25.Placzek, C. et al. Climate in the dry central Andes over Geologic, millennial, and interannual timescales. Ann. Mo. Bot. Gard. 96(3), 386–397. https://doi.org/10.3417/2008019 (2009).Article 

Google Scholar 
26.Cerda, M. et al. A new 20th century lake sedimentary record from the Atacama Desert/Chile reveals persistent PDO (Pacific Decadal Oscillation) impact. J. S. Am. Earth Sci. 95, 102302. https://doi.org/10.1016/j.jsames.2019.102302 (2019).Article 

Google Scholar 
27.Aránguiz-Acuña, A. et al. Aquatic community structure as sentinel of recent environmental changes unraveled from lake sedimentary records from the Atacama Desert, Chile . PLoS ONE 15(2), e0229453. https://doi.org/10.1371/journal.pone.0229453 (2020).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
28.Flores-Varas, A. et al. Ascotán and Carcote salt flats as sensors of humidity fluctuations and anthropic impacts in the transition zone of the Andean Altiplano. J. S. Am. Earth Sci. 105, 102934. https://doi.org/10.1016/j.jsames.2020.102934 (2021).CAS 
Article 

Google Scholar 
29.Maher, B. A. & Taylor, R. M. Formation of ultrafine-grained magnetite in soils. Nature 336, 368–370 (1988).ADS 
CAS 
Article 

Google Scholar 
30.Dearing, J. A. et al. Frequency-dependent susceptibility measurements of environmental materials. Geophys. J. Int. 124, 228–240. https://doi.org/10.1111/j.1365-246X.1996.tb06366.x (1996).ADS 
Article 

Google Scholar 
31.Evans, M. & Heller, F. Environmental magnetism: principles and applications of enviromagnetics. Int. Geophys. 86, 202 (2003).
Google Scholar 
32.Pizarro, H. et al. The origin of the magnetic record in Eocene-Miocene coarse-grained sediments deposited in hyper-arid/arid conditions: examples from the Desert. Palaeogeogr. Palaeoclimatol. Palaeoecol. 516, 322–335. https://doi.org/10.1016/j.palaeo.2018.12.009 (2019).Article 

Google Scholar 
33.Risacher, F., Alonso, H. & Salazar, C. Geoquímica de Aguas en Cuencas cerradas: I, II y III Regiones-Chile. Volumen III Estudio de Cuencas de la II Región. (Convenio de Cooperación DGA-UCN-IRD. S.I.T. Nº51, 1999).34.Tapia, R. et al. Glacial differences of Southern Ocean Intermediate Waters in the Central South Pacific. Quat. Sci. Rev. 208, 105–117. https://doi.org/10.1016/j.quascirev.2019.01.016 (2019).ADS 
Article 

Google Scholar 
35.Horne, D. J. Life-cycles of podocopid Ostracoda – a review (with particular reference to marine and brackish-water species). In Applications of Ostracoda. Proceedings of the Eighth International Symposium on Ostracoda (ed. Maddocks, R.) 581–590 (University of Houston, Texas, 1983).
Google Scholar 
36.Cohen, A. C. & Morin, J. G. Patterns of reproduction in ostracodes; a review. J. Crust. Biol. 10(2), 184–211. https://doi.org/10.2307/1548480 (1990).Article 

Google Scholar 
37.Mesquita-Joanes, F., Smith, A. J. & Viehberg, F. A. The ecology of ostracoda across levels of biological organization from individual to ecosystem. J. Quat. Sci. 17, 15–35. https://doi.org/10.1016/B978-0-444-53636-5.00002-0 (2012).Article 

Google Scholar 
38.McLay, C. L. The population biology of Cyprinotus carolinensis and Herpetocypris reptans (Crustacea, Ostracoda). Can. J. Zool. 56(5), 1170–1179. https://doi.org/10.1139/z78-161 (1978).ADS 
Article 

Google Scholar 
39.Hamouda, S. A., Sames, B., Mohammed, A. & Bensalah, M. First record of non-marine ostracods from the Paleogene “hamadian deposits” of Méridja area, west of Bechar (southwestern Algeria). Annales de Paléontologie 104(1), 27–44. https://doi.org/10.1016/j.annpal.2017.12.001 (2018).Article 

Google Scholar 
40.Bergue, C. T., Maranhao, M. D. S. A. S. & Fauht, G. Paleolimnological inferences based on Oligocene ostracods (Crustacea: Ostracoda) from Tremembé Formation. Southeast Brazil. An. Acad. Bras. Cienc. 87(3), 1531–1544. https://doi.org/10.1590/0001-3765201520140366 (2015).Article 
PubMed 

Google Scholar 
41.Sylvestre, F., Servant-Vildary, S. & Roux, M. Diatom-based ionic concentration and salinity models from the south Bolivian Altiplano (15–23°S). J. Paleolimnol. 25, 279–295 (2001).ADS 
Article 

Google Scholar 
42.Nunnery, J. A., Fritz, S. C., Baker, P. A. & Selenbien, W. Lake-level variability in Salar de Coipasa, Bolivia during the past ∼40,000 yr. Quat. Res. https://doi.org/10.1017/qua.2018.108 (2018).Article 

Google Scholar 
43.Herrera, C. et al. Investigaciones hidrogeológicas en la laguna Tuyajto perteneciente a la Reserva Nacional de los Flamencos (Atacama, Chile). Bol. Geol. Min. 130(4), 789–806. https://doi.org/10.21701/bolgeomin.130.4.011 (2019).Article 

Google Scholar 
44.Houston, J. Variability of precipitation in the Atacama Desert: its causes and hydrological impact. Q. J. R. Meteorol. Soc. 26(15), 2181–2198. https://doi.org/10.1002/joc.1359 (2006).Article 

Google Scholar 
45.Herrera, C. et al. Groundwater flow in a closed basin with a saline shallow lake in a volcanic area: Laguna Tuyajto, northern Chilean Altiplano of the Andes. Sci. Total Environ. 541, 303–318. https://doi.org/10.1016/j.scitotenv.2015.09.060 (2016).ADS 
CAS 
Article 
PubMed 

Google Scholar 
46.Munk, L. A., Boutt, D. F., Hynek, S. A. & Moran, B. J. Hydrogeochemical fluxes and processes contributing to the formation of lithium-enriched brines in a hyper-arid continental basin. Chem. Geol. 493, 37–57. https://doi.org/10.1016/j.chemgeo.2018.05.013 (2018).ADS 
CAS 
Article 

Google Scholar 
47.Godfrey, L. & Álvarez-Amado, F. Volcanic and Saline Lithium Inputs to the Salar de Atacama. Minerals 10(2), 201. https://doi.org/10.3390/min10020201 (2020).CAS 
Article 

Google Scholar 
48.Marazuela, M. A., Ayora, C., Vázquez-Suñé, E., Olivella, S. & García-Gil, A. Hydrogeological constraints for the genesis of the extreme lithium enrichment in the Salar de Atacama (NE Chile): A thermohaline flow modelling approach. Sci. Total Environ. 739, 139959. https://doi.org/10.1016/j.scitotenv.2020.139959 (2020).ADS 
CAS 
Article 
PubMed 

Google Scholar 
49.Bobst, A. L. et al. A 106 ka paleoclimate record from drill core of the Salar de Atacama, northern Chile. Palaeogeogr. Palaeoclimatol. Palaeoecol. 173(1–2), 21–42. https://doi.org/10.1016/S0031-0182(01)00308-X (2001).Article 

Google Scholar 
50.Baspineiro, C. F., Franco, J. & Flexer, V. Potential water recovery during lithium mining from high salinity brines. Sci. Total Environ. 720, 137523. https://doi.org/10.1016/j.scitotenv.2020.137523 (2020).ADS 
CAS 
Article 
PubMed 

Google Scholar 
51.Marazuela, M. A., Vázquez-Suñé, E., Ayora, C. & García-Gil, A. Towards more sustainable brine extraction in salt flats: Learning from the Salar de Atacama. Sci. Total Environ. 703, 135605. https://doi.org/10.1016/j.scitotenv.2019.135605 (2020).ADS 
CAS 
Article 
PubMed 

Google Scholar 
52.Babidge, S. Sustaining ignorance: the uncertainties of groundwater and its extraction in the Salar de Atacama, northern Chile. J. R. Anthropol. Inst. 25(1), 83–102. https://doi.org/10.1111/1467-9655.12965 (2018).Article 

Google Scholar 
53.Sonter, L. J., Dade, M. C., Watson, J. E. M. & Valenta, R. K. Renewable energy production will exacerbate mining threats to biodiversity. Nat. Commun. 11, 4174. https://doi.org/10.1038/s41467-020-17928-5 (2020).ADS 
CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
54.Stahl, A. T., Fremier, A. K. & Cosens, B. A. Mapping legal authority for terrestrial conservation corridors along streams. Conserv. Biol. 34(4), 943–955. https://doi.org/10.1111/cobi.13484 (2020).Article 
PubMed 
PubMed Central 

Google Scholar 
55.García, M., Prieto, M. & Kalazich, F. The protection of the mountain ecosystems of the Southern Central Andes: tensions between Aymara herding practices and conservation policies. Eco. Mont. 13(1), 22–30 (2021).
Google Scholar 
56.Vila, T. Geología de los depósitos salinos andinos, provincia de Antofagasta, Chile. Revista de Geología de Chile 2, 41–55 (1975).
Google Scholar 
57.CIREN. Catastro Agrícola, estudio de metodología para la realización y actualización de catastro agrícola regional en base a la utilización de tecnología geoespacial 1–35 (CCIRA, Atacama, 2013).
Google Scholar 
58.Villagrán, C., Kalin-Arroyo, M. T. & Marticorena, C. Efectos de la destización en la distribución de la flora andina de Chile. Rev. Chil. Hist. Nat. 56, 137–157 (1983).
Google Scholar 
59.CONAF. Actualización Plan de Manejo Participativo Reserva Nacional Los Flamencos, Region de Antofagasta (2008).60.Núñez, L., Grosjean, M. & Cartajena, I. Ocupaciones humanas y paleoambientes en la Puna de Atacama (Universidad Católica del Norte-Taraxacum, Antofagasta, 2005).
Google Scholar 
61.Los Ostracodos, M. P. VI. 4f. In El lago Titicaca, síntesis del conocimiento limnológico actual (eds Dejoux, C. & Iltis, A.) 345–352 (Orstom, New Caledonia, 1991).
Google Scholar 
62.Karanovic, I. Recent Freshwater Ostracods of the World, Crustacea, Ostracoda, Podocopida (Springer , Berlin, 2012).
Google Scholar 
63.Palacios-Fest, M. R., Cusminsky, G. C. & McGlue, M. M. Late Quaternary lacustrine ostracods (Ostracoda, Crustacea) and charophytes (Charophyta, Charales) from the Puna Plateau, Argentina. Micropaleontology 35, 66–78 (2016).
Google Scholar 
64.Brandão, S. N., Angel, M. V., Karanovic, I., Perrier, V. & Meidla, T. World Ostracoda Database. http://www.marinespecies.org/ostracoda/aphia.php?p=taxdetails&id=1091 on 2020–01–15 (2018).65.Fatela, F. & Taborda, R. Confidence limits of species proportions in microfossil assemblages. Mar. Micropaleontol. 45(2), 169–174. https://doi.org/10.1016/S0377-8398(02)00021-X (2002).ADS 
Article 

Google Scholar 
66.Díaz, C. P. & Maidana, N. I. Diatomeas de los Salares Atacama y Punta Negra, II Región-Chile (Centro de Ecología Aplicada , La Reina, 2005).
Google Scholar 
67.Diatoms of North America. The source for diatom identification and ecology. https://diatoms.org (2019).68.Hammer, Ø. & Harper, D. Paleontological Data Analysis (Blackwell Publishing , Hoboken, 2006).
Google Scholar 
69.Oksanen, J. et al. vegan: Community Ecology Package. Ordination methods, diversity analysis and other functions for community and vegetation ecologists. Version 2.5–1. URL https://CRAN.R-project.org/package=vegan (2019).70.McArdle, B. H. & Anderson, M. J. Fitting multivariate models to community data: a comment on distance-based redundancy analysis. Ecology 82(1), 290–297. https://doi.org/10.1890/0012-9658(2001)082[0290:FMMTCD]2.0.CO;2 (2001).Article 

Google Scholar  More

Study areaThe study was conducted along the Agi-gawa River, a tributary of the Kiso-gawa River system in central Japan (35°23 42″–35°26 49″N, 137°25 12″–137°28 01″E; Fig. 1), with the Agi-gawa Dam (110 km from the river mouth, 418 m a.s.l.). The Agi-gawa River is a 3rd to 4th-order river with a naturally sand-rich bed derived from weathered granite that characterizes the local geology36. The Agi-gawa Dam (35°25 32″N, 137°25 55″E) had begun operations in 1990; it is a 102 m high rockfill dam with a catchment area of 82 km2, a storage capacity of 4.8 × 107 m3, a mean depth of ~ 45 to 50 m at the dam site, and a hydraulic residence time of 71 days. Although three small sub-dams at the upstream end of the impoundment trap particulates, the sediment speed in the reservoir has been 1,000,000 m3 for 24 years. The dam serves multiple purposes, including flood control, industrial and urban water supply, and the maintenance of baseflow. Further information on the Agi-gawa Dam is available in Katano et al.37.Figure 1The study area shows six study reaches in three stream segments along the Agi-gawa River and Iinuma-gawa Stream, Gifu Prefecture, Japan. Gray circles denote reaches, which are numbered from upstream to downstream within each segment: UD1 and UD2 are upstream of the dam, DD1 and DD2 are downstream of the dam, whilst TR1 and TR2 are in the tributary. The two black circles denote the sediment replenished reaches (S1 and S2). The three small rectangles at the upstream ends of the impoundment are sub-dams, constructed to reduce the inputs of particulates to the impoundment. This map is based on the Digital Topographic Map 25,000 published by Geospatial Information Authority of Japan.Full size imageSediment replenishment and sampling sitesSediment replenishment was undertaken 0.8 and 1.8 km downstream of the Agi-gawa Dam (S1 and S2, Fig. 1) on February 16 and 27, 2005. A total of 1,200 m3 of sediment (D50 ≈ 0.6 mm; mainly sand) was mined from the upstream sub-dams and transported to S1 and S2. We estimate that this constituted 0.086% of the annual sedimentation in the Agi-gawa Dam (e.g., in 2007, replenished sediment per year × 100/sedimentation in the reservoir). The sediment (800 and 400 m3) was replenished at high-flow banks in both sites. The replenished sediment was gradually washed during the high flows at the end of June (visual observation by dam administrators) (Fig. 2). We confirmed that this replenished sediment remained on both banks in March, and no sediments remained on both banks in early July.Figure 2(a) Precipitation (mm·d) (b) mean inflow to the impoundment per day (m3·s-1); and (c) mean outflow from the Agi-gawa Dam per day (m3·s-1). The vertical broken line indicates the study period. Note that the y-axes for (b) and (c) have a logarithmic scale.Full size imageField sampling was conducted twice between March 15 and 18, 2005, prior to sediment flushing and between August 22 and 24, 2005, following sediment flushing [7 weeks after the end of the sediment drift out (Fig. 2)]. The later sampling date was scheduled to investigate the continuous effects (i.e., not immediate effects) of replenished sediment on the riverbed environment and macroinvertebrate assemblages before the replenished sediment had completely been transported further downstream from S1 and S2.Three study segments (length: 1–2 km each) were selected: (1) upstream of the dam and impounded area (UD); (2) downstream of the dam (DD); and (3) in the tributary (TR). These sites were along a 6.0 km stretch of the Agi-gawa River and a 1.0 km stretch of the Iinuma-gawa Stream (catchment area = 24 km2); the latter is a tributary that flows into the Agi-gawa River 2.7 km, downstream of the dam (Fig. 1, Table 1). Each segment contained two study reaches (six reaches in total), and each study reach was 160 m long with several pool–riffle sequences; all reaches were > 300 m apart. DD1 and DD2 were located immediately downstream of the sediment-displaced banks (S1 and S2; 100 m upstream of DD1 and DD2, respectively). Measurements at the two reaches within the same segment were completed on the same day, and the reaches were surveyed in an upstream direction. The dominant land use along the study area was paddy fields, with sparse riparian forest.Table 1 General characteristics of the three study segments and two seasons.Full size tableAlthough the most suitable reference site for DD is the DD prior to dam construction, we were unable to investigate the site prior to dam construction. Therefore, we treated the reference sites as sites that were less affected by the dam than DD on the present day. Katano et al.37 indicated that the difference between the TR and UD sites was smaller than that between DD and UD/TR sites in terms of biota and geology. However, UD was characterized by a wider channel and higher discharge than TR, due to differences in their catchment areas (Table 1). As we did not have a definitive reference, we treated both UD and TR as reference sites (see “Statistical analysis” section). Therefore, how DD in March and DD in August is different from UD and TR can be interpreted as the effect of sediment reduction.Physical environment and water qualitySix riffles were selected at each study reach, and a sampling location (50 × 50 cm quadrat) was established in the mid-channel area of each riffle. Prior to invertebrate sampling, physical environmental variables were measured.Substrate coarseness was measured by gently floating a Plexiglas observation box (50 × 50 × 10 cm deep) divided into four grid squares (25 × 25 cm) on the surface water such that the grid had projected onto the streambed. The size of the substrate material was coded based on the intermediate-axis length: 1 = sand (particles  16 mm) and sieved through a 0.25 mm mesh sieve. Sieved samples and substrate material smaller than pebbles were mixed in a container and preserved in 5% formalin in the field.The material in each container was later divided into two size fractions using 1-and 0.25 mm mesh sieves. To simplify the sorting process, all material retained in the 0.25 mm sieve was mixed and divided into 2n subsamples (maximum n = 32) using a splitter (Idea Co., Tokyo, Japan), following the method described by Vinson and Hawkins43. All macroinvertebrates in subsamples in the 1 mm sieve were counted and identified to the lowest taxonomic level possible, usually to genus or species level using the taxonomic keys of Kawamura and Ueno44, Merritt and Cummins34, Kathman and Brinkhurst45, Kawai and Tanida35, and Torii46.Macroinvertebrate taxa were also classified into five functional feeding groups (FFGs) according to Kawamura and Ueno44, Merritt and Cummins34, Kathman and Brinkhurst45, Kawai and Tanida (2005)35, and Torii46. FFGs were defined as collector-filterers, collector-gatherers, predators, scrapers, and shredders. If a species belonged to ≥ 2 FFGs, the number of individuals was apportioned across the FFGs. We also counted the number of burrowers (#burrowers), inorganic case-bearing caddisflies (#ICB), and net-spinners (#net spinners) of macroinvertebrate assemblages according Kawamura and Ueno44, Merritt and Cummins34, Kathman and Brinkhurst45, Kawai and Tanida35, and Torii46 (see Supplementary Table S1). This classification was carried out as such life-habit traits are important for surviving in a regulated river containing reduced quantities of sand and gravel on the riverbed37. The Chironomidae family was excluded in the life-habit analysis as they consist of various life forms. Once all invertebrates were removed, dry mass (mg m−2) and ash-free dry mass (AFDM, mg m−2) of benthic coarse particulate organic matter (BCPOM,  > 1 mm), and benthic fine particulate organic matter (BFPOM,  0.25 mm) were obtained by drying in an oven at 60 °C for 1 day and combusting in a muffle furnace at 550 °C for 4 h. BCPOM and BFPOM were calculated based on the difference between the dry mass and the AFDM.The total number of invertebrate individuals and the AFDM of BFPOM in each sample were estimated by multiplying by the corresponding 2n value. The number of taxa and density of invertebrates in each sample were calculated as the sum of the values in both size fractions. Additionally, we determined Shannon’s diversity index (H), Simpson’s evenness index, and the percentage of Ephemeroptera, Plecoptera, and Trichoptera (%EPT)47. A sample from UD2 in March had been lost and therefore could not be included in the analyses.Periphyton was sampled from cobbles adjacent to each sampling location. Periphyton was removed from a 5 × 5 cm area on the upper surface of each cobble with a toothbrush. Each sample was placed in a separate container with 250 mL of water. Within 24 h of sample collection, a subsample of the well-mixed content in each container was filtered using a glass-fiber filter (GF/C; Whatman Co., Maidstone, UK). Each filter was placed in a separate vial with 20 mL of 99.5% ethanol and stored in a dark refrigerator at 4 °C for 24 h. The extracted pigments were measured using a spectrophotometer (U-1800; Shimadzu Co., Kyoto, Japan), following the method of Lorenzen48.Analysis of case materials of an inorganic case-bearing caddisflyWe compared the particle size structure of replenished sediment, riverbed sediment, and case materials for case-bearing caddisfly. The replenished sediment was directly sampled in a 1 L polyethylene jar at the upstream replenished bank (S1) on March 16, 2005 (Fig. 1). Riverbed sediment was sampled at two stations; 100 m upstream of S1, and 100 m upstream of DD1 between August 22 and 24, 2005. At each station of the river, a metallic narrow cup (200 mL) with a lid was pushed into a vacancy between the cobbles, which had been randomly selected, and fine sediments (up to small gravel) in the vacancies were sampled by closing the lid underwater. Sampling was carried out three times (i.e., three different vacancies in the cobbles), and subsamples were pooled for measurement. The replenished and riverbed sediment was combusted at 550 °C for 2 h in a muffle furnace to remove organic contamination. Combusted samples were separated with eight sieves with a mesh size range of 0.075–9.5 mm (JIS A 1204). Each fraction was weighed, and the grain size accumulation curve of each type of sediment and its D50 were obtained.In a macroinvertebrate sample at DD1 between August 22 and 24, 2005, ten individuals from two case-bearing caddisfly larvae, Glossosoma sp. and Gumaga orientalis, which were prevalent at DD1 during this period (see Results), were randomly selected from the formalin-fixed sample. The case was carefully removed from the larvae and combusted as described above for the replenished and riverbed sediment. The number of case material grains was measured using a dissection microscope.Statistical analysesWe described results based on two main assumptions: (1) the DD in March is the dam-affected reach (cf. unregulated reaches UD and TR), and (2) the changes in DD from March to August were mainly a result of sediment replenishment. In the statistical analyses, the p criterion (⍺) was set at 0.05.To consider the effects of the segment, replicate reach, and season on variables, nested multivariate analysis of variance (MANOVA) was used to test whether any measured variables at the riffle scale differed between segments (UD, DD, and TR). Three segments and two replicate reaches were nested within each season (March and August) and segment (i.e., Season/Segment/Reach), whereby measurements within each reach were treated as subsamples. In the MANOVA, we also consider the interactions of the variances to interpret the interactions among the sampling segments and seasons to consider the independent effects on the factors.To perform MANOVA, we assumed that temporal variability was greater than spatial variability within each reach for variables measured over 24 h (e.g., water quality), and the opposite would hold true for variables measured only once (e.g., macroinvertebrates). Therefore, subsamples within each reach were either spatially or temporally replicated, depending on the variable type. Temporal replicates (four samples collected every 6 h) were treated as a repeated factor (time factor). A nested MANOVA was used for variables quantified once at each location (e.g., macroinvertebrates), and nested repeated-measures MANOVA (rm-MANOVA) were used for variables quantified over a 24 h period at each reach (e.g., water quality). When a significant difference was detected by MANOVA with non-significant interactions, each variable was tested separately with a nested ANOVA for variable groups once at each location or the nested rm-ANOVA for repeated-measured variables, as appropriate for the particular variable. The risk of inflating Type 1 errors for the ANOVA was reduced using Bonferroni adjustments.These MANOVA and ANOVA tests were conducted with R version 3.6.049. The residuals of each variable in each MANOVA and ANOVA model were verified using the Shapiro–Wilk normality test prior to analyses, and normality was improved using arcsine(x) or log (x + 1) transformation when appropriate.Tukey’s multiple comparison test in a one-way ANOVA model (Season/Segment/Reach) was used for comparisons between segments. Any significant changes in values for variables from UD to DD were interpreted as the effects of the dam based on the assumption that conditions in UD and DD were similar prior to dam construction; this was because replenished sediment had not been supplied in March (see before). However, UD may be unsuitable as a reference site compared with TR as the former may be at least partly affected by the dam. This may particularly be the case for benthic invertebrates, such as the interruption of the upstream flight of adult females50. Therefore, UD and TR were treated as reference sites for reservoir and tributary effects, respectively. This was because both were unaffected by the dam, and sediment replenishment as tributaries may function as sites for resource recovery for the dam-affected mainstem of the river37,51,52, despite differing watershed areas. Therefore, the similarity of variables between the TR and UD sites was statistically confirmed such that they could be treated as reference sites. As such, the recovery from March to August could reliably demonstrate the effect of sediment replenishment. For example, although the value at DD differed from that at TR and/or UD in March, it was similar to that at UD and/or TR in August.Multivariate analyses were conducted using the R “vegan” package version 2.5.6 to compare invertebrate assemblage structures between segments. Bray–Curtis coefficients based on species abundance were used to calculate a dissimilarity matrix, and dissimilarities between UD and DD, and between TR and DD in each season were tested using two-way ANOVA and Tukey post-hoc tests.Macroinvertebrate assemblage organization in relation to environmental gradients was analyzed using redundancy analysis (RDA) with the “rda” function of “vegan” package. This was because the preliminary analysis using detrended correspondence analysis (DCA) showed that the gradient lengths of DCA were More

Experimental study and protein domain analysisInsectsHosts and parasitoids were maintained as previously described25. Five Leptopilina heterotoma (Hymenoptera: Figitidae) populations were used for experiments: a population from Japan (Sapporo), two populations from the United Kingdom (1: Whittlesford; 2: Great Shelford) and two populations from Belgium (1: Wilsele; 2: Eupen). Information on collection sites, including GPS coordinates, can be found in25.Determination of host fat contentD. simulans and D. melanogaster hosts were allowed to lay eggs during 24 h in glass flasks containing ~ 50 mL standard medium25. After two days, developing larvae were sieved and ~ 200 were larvae placed in a Drosophila tube containing ~ 10 mL medium. Seven days after egg laying, newly formed pupae were frozen at – 18 °C, after which fat content was determined as described in25, where dry weight before and after neutral fat extraction was used to calculate absolute fat amount (in μg) for each host. The host pupal stage was chosen for estimating fat content, because at this point the host ceases to feed, while the parasitoid starts consuming the entire host36. All data were analysed using R Project version 3.4.360. Fat content of hosts was compared using a one-way ANOVA with host species as fixed factor.Manipulation of host fat contentTo generate leaner D. melanogaster hosts, we adapted our standard food medium25 to contain 100 times less (0.5 g) sugar per litre water. Manipulating sugar content did not alter the structure of the food medium, thus maintaining similar rearing conditions, with the exception of sugar content. Fat content of leaner and fatter D. melanogaster hosts was determined and analysed as described above.Fat synthesis quantification of wasp populationsMated female L. heterotoma were allowed to lay eggs on host fly larvae collected as described above with ad libitum access to honey as a food source until death. Honey consists of sugars and other carbohydrates that readily induce fat synthesis. After three weeks, adult offspring emergence was monitored daily and females were haphazardly placed in experimental treatments. Females were either killed at emergence (to measure teneral lipid reserves) or after feeding for 7 days on honey. Wasps were frozen at − 18 °C after completion of experiments. Fat content was determined as described above for hosts. The ability for fat synthesis was then determined by comparing mean fat levels of recently emerged compared to fed individuals, similar to procedures described in15,25,28. An increase in fat levels after feeding is indicative of active fat synthesis; equal or lower fat levels suggest fat synthesis did not take place. Each population tested on D. melanogaster or D. simulans represented an independent dataset that was analysed separately, as in Visser et al. 201825, because we are interested in the response of each population on each host species. We used T-tests when data was normally distributed and variances equal, log-transformed data for non-normal data, and a Welch’s t-test when variances were unequal. We corrected for multiple testing using Benjamini and Hochberg’s False Discovery Rate61.Fat synthesis quantification using a familial design and GC–MS analysesTo tease apart the effect of wasp genotype and host environment, we used a split-brood design where the offspring of each mother developed on lean D. simulans or fat D. melanogaster hosts in two replicated experiments (experiment 1 and 2). In both experiments, mothers were allowed to lay eggs in ~ 200 2nd to 3rd instar host larvae of one species for four days, after which ~ 200 host larvae of the other species were offered during four days. The order in which host larvae were presented was randomized across families. Following offspring emergence, daughters were allocated into two treatment groups: a control where females were fed a mixture of honey and water (1:2 w/w) or a treatment group fed a mixture of honey and deuterated water (Sigma Aldrich) (1:2 w/w; stable isotope treatment) for 7 days. Samples were prepared for GC–MS as described in 28. Incorporation of up to three deuterium atoms can be detected, but percent incorporation is highest when only 1 deuterium atom is incorporated. As incorporation of a single atom unequivocally demonstrates active fat synthesis, we only analysed percent incorporation (in relation to the parent ion) for the abundance of the m + 1 ion. Percent incorporation was determined for five fatty acids, C16:1 (palmitoleic acid), C16:0 (palmitate), C18:2 (linoleic acid), C18:1 (oleic acid), and C18:0 (stearic acid), and the internal standard C17:0 (margaric acid). Average percent incorporation for C17:0 was 19.4 (i.e. baseline incorporation of naturally occurring deuterium) and all values of the internal standard remained within 3 standard deviations of the mean (i.e. 1.6). Percent incorporation of control samples was subtracted from treatment sample values to correct for background levels of deuterium (i.e. only when more deuterium is incorporated in treatment compared to controls fatty acids are actively being synthesized). For statistical analyses, percent incorporation was first summed for C16:1, C16:0, C18:2, C18:1 and C18:0 to obtain overall incorporation levels, as saturated C16 and C18 fatty acids are direct products of the fatty acid synthesis pathway (that can subsequently be desaturated).Data (presented in Fig. 1) was analysed by means of a linear mixed effects model (GLMM, lme4 package) with host (lean D. simulans and fat D. melanogaster) and experiment (conducted twice) as fixed effect, family nested within population (Japan, United Kingdom 1 and 2, Belgium 1 and 2) as random factor, and percentage of incorporation of stable isotopes as dependent variable (log transformed; n = 138). Non-significant terms (i.e., experiment) were sequentially removed from the model to obtain the minimal adequate model as reported in Table 2. When referring to “families,” we are referring to the comparison of daughters of singly inseminated females, which (in these haplodiploid insects) share 75% of their genome.Identification of functional acc and fas genes in distinct parasitoid speciesTo obtain acc and fas nucleotide sequences for L. clavipes, G. legneri, P. maculata and A. bilineata, we used D. melanogaster mRNA ACC transcript variant A (NM_136498.3 in Genbank) and FASN1-RA (FBtr0077659 in FlyBase) and blasted both sequences against transcripts of each parasitoid (using the blast function available at http://www.parasitoids.labs.vu.nl62,63). Each nucleotide sequence was then entered in the NCBI Conserved Domain database64 to determine the presence of all functional protein domains. All sequences were then translated using the Expasy translate tool (https://web.expasy.org/translate/), where the largest open reading frame was selected for further use and confirming no stop codons were present. Protein sequences were then aligned using MAFFT v. 7 to compare functional amino acid sequences between all species (Supplementary files 1 and 2)65.Simulation studyWe consider the general situation where phenotypic plasticity is only sporadically adaptive and ask the question whether and under what circumstances plasticity can remain functional over long evolutionary time periods when the regulatory processes underlying plasticity are gradually broken down by mutations. We consider a regulatory mechanism that switches on or off a pathway (like fat synthesis) in response to environmental conditions (e.g., host fat content).Fitness considerationsWe assume that the local environment of an individual is characterized by two factors: fat content F and nutrient content N, where nutrients represent sugars and other carbohydrates that can be used to synthesize fat. Nutrients are measured in units corresponding to the amount of fat that can be synthesized from them. We assume that fitness (viability and/or fecundity) is directly proportional to the amount of fat stored by the individual. When fat synthesis is switched off, this amount is equal to F, the amount of fat in the environment. When fat synthesis is switched on, the amount of fat stored is assumed to be (N – c + (1 – k)F). This expression reflects the following assumptions: (i) fat is synthesized from the available nutrients, but this comes at a fitness cost c; (ii) fat can still be absorbed from the environment, but at a reduced rate ((1 – k)). It is adaptive to switch on fat synthesis if (N – c + (1 – k)F) is larger than F, or equivalently if (F < tfrac{1}{k}(N - c)).The right-hand side of this inequality is a straight line, which is illustrated by the blue line in Fig. 4. The three boxes in Fig. 4 illustrate three types of environmental conditions. Red box low-fat environments. Here, (F < tfrac{1}{k}(N - c)) is always satisfied, implying that fat synthesis should be switched on constitutively. Yellow box high-fat environments. Here, (F > tfrac{1}{k}(N – c)), implying that fat synthesis should be switched off constitutively.

Orange box intermediate-fat environments. Here, fat synthesis should be plastic and switched on if for the given environment (N, F) the fat content is below the blue line and switched off otherwise.

Figure 4Environmental conditions encountered by the model organisms. For a given combination of environmental nutrient content N and environmental fat content F, it is adaptive to switch on fat synthesis if (N, F) is below the blue line (corresponding to (F < tfrac{1}{k}(N - c))) and to switch it off otherwise. The three boxes illustrate three types of environment: a low-fat environment (red) where fat synthesis should be switched on constitutively; a high-fat environment (yellow) where fat synthesis should be switched off constitutively; and an intermediate-fat environment (orange) where a plastic switch is selectively favoured.Full size imageThe simulations reported here were all run for the parameters (k = tfrac{1}{2}{text{ and }}c = tfrac{1}{4}). We also investigated many other combinations of these parameters; in all cases, the results were very similar to those reported in Fig. 3.Gene regulatory networks (GRN)In our model, the switching device was implemented by an evolving gene regulatory network (as in van Gestel and Weissing66). The simulations shown in Fig. 3 of the main text are based on the simplest possible network that consists of two receptor nodes (sensing the fat and the nutrient content in the local environment, respectively) and an effector node that switches on fat synthesis if the combined weighted input of the two receptor nodes exceeds a threshold value T and switches it off otherwise. Hence, fat synthesis is switched on if (w_{F} F + w_{N} N > T) (and off otherwise). The GRN is characterized by the weighing factors (w_{F} {text{ and }}w_{N}) and the threshold T. These parameters are transmitted from parents to offspring, and they evolve subject to mutation and selection. We also considered alternative network structures (all with two receptor nodes and one effector node, but with a larger number of evolvable weighing factors67, and obtained very similar results, see below).For the simple GRN described above, the switching device is 100% adaptive when the switch is on (i.e., (w_{F} F + w_{N} N > T)) if (F < tfrac{1}{k}(N - c)) and off otherwise. A simple calculation yields that this is the case if: (w_{N} > 0{, }w_{F} = – k{kern 1pt} w_{N} {text{ and }}T = c{kern 1pt} w_{N}).Evolution of the GRNFor simplicity, we consider an asexual haploid population with discrete, non-overlapping generations and fixed population size (N = 10,000). Each individual has several gene loci, each locus encoding one parameter of the GRN. In case of the simple network described above, there are three gene loci, each with infinitely many alleles. Each individual harbours three alleles, which correspond to the GRN parameters (w_{F} {, }w_{N} {text{ and }}T), and hence determine the functioning of the genetic switch. In the simulations, each individual encounters a randomly chosen environment ((N{, }F)). Based on its (genetically encoded) GRN, the individual decides on whether to switch on or off fat synthesis. If synthesis is switched on, the individual’s fitness is given by (N – c + (1 – k)F); otherwise its fitness is given by F. Subsequently, the individuals produce offspring, where the number of offspring produced is proportional to the amount of fat stored by an individual. Each offspring inherits the genetic parameters of its parent, subject to mutation. With probability μ (per locus) a mutation occurs. In such a case the parental value (in case of a simple network: the parent’s allelic value (w_{F} {, }w_{N} {text{ or }}T)) is changed to a mutated value ((w_{F} { + }delta {, }w_{N} { + }delta {text{ or }}T + delta)), where the mutational step size δ is drawn from a normal distribution with mean zero and standard deviation σ. In the reported simulations, we chose (mu = 0.001) and (sigma = 0.1). The speed of evolution is proportional to (mu cdot sigma^{2}), implying that the rate of change in Fig. 3 (both the decay of plasticity and the rate of regaining adaptive plasticity) are positively related to μ and σ.Preadaptation of the GRNsStarting with a population with randomly initialized alleles for the GRN parameters, we first let the population evolve for 10,000 generations in the intermediate-fat environment (the orange box in Fig. 4). In all replicate simulations, a “perfectly adapted switch” (corresponding to (w_{N} > 0{, }w_{F} = – k{kern 1pt} w_{N} {text{ and }}T = c{kern 1pt} w_{N})) evolved, typically within 1,000 generations. Still, the evolved GRNs differed across replicates, as they evolved different values of (w_{N} > 0). These evolved networks were used to seed the populations in the subsequent “decay” simulations.Evolutionary decay of the GRNsFor the decay experiments reported in Fig. 3 of the main text, we initiated a large number of monomorphic replicate populations with one of the perfectly adapted GRNs from the preadaptation phase. These populations were exposed for an extended period of time (1,000,000 generations) to a high-fat environment (the yellow box in Fig. 4), where all preadapted GRNs switched off fat synthesis. However, in some scenarios, the environmental conditions changed back sporadically (with probability q) to the intermediate-fat environment (the orange box in Fig. 4), where it is adaptive to switch on fat metabolism in 50% of the environmental conditions (when (N, F) is below the blue line in Fig. 4). In Fig. 3, we report on the changing rates (q = 0.0) (no changing back; red), (q = 0.001) (changing back once every 1,000 generations; purple), and (q = 0.01) (changing back once every 100 generations; pink). When such a change occurred, the population was exposed to the intermediate-fat environment for t generations (Fig. 3 is based on t = 3).Throughout the simulation, the performance of the network was monitored every 100 generations as follows: 100 GRNs were chosen at random from the population, and each of these GRNs was exposed to 100 randomly chosen environmental conditions from the intermediate-fat environment (orange box in Fig. 4). From this, we could determine the average percentage of “correct” decisions (where the network should be switched on if and only if (F < tfrac{1}{k}(N - c)). 1.0 means that the GRN is still making 100% adaptive decisions; 0.5 means that the GRN only makes 50% adaptive decision, as would be expected by a random GRN or a GRN that switches the pathway constitutively on or off. This measure for performance in the “old” intermediate-fat environment was determined for 100 replicate simulations per scenario and plotted in Fig. 3 (mean ± standard deviation).Evolving robustness of the GRNsThe simulations in Fig. 3 are representative for all networks and parameters considered. Whenever (q = 0.0), the performance of the regulatory switch eroded in evolutionary time, but typically at a much lower rate in case of the more complex GRNs. Whenever (q = 0.01), the performance of the switch went back to levels above 90% and even above 95% for the more complex GRNs. Even for (q = 0.001), a sustained performance level above 75% was obtained in all cases.Intriguingly, in the last two scenarios the performance level first drops rapidly (from 1.0 to a much lower level, although this drop is less pronounced in the more complex GRNs) and subsequently recovers to reach high levels again. Apparently, the GRNs have evolved a higher level of robustness, a property that seems to be typical for evolving networks8. For the simple GRN studied in Fig. 3, this outcome can be explained as follows. The initial network was characterized by the genetic parameters (w_{N} > 0{, }w_{F} = – k{kern 1pt} w_{N} {text{ and }}T = c{kern 1pt} w_{N}) (see above), where (w_{N}) was typically a small positive number. In the course of evolutionary time, the relation between the three evolving parameters remained approximately the same, but (w_{N}) (and with it the other parameters) evolved to much larger values. This automatically resulted in an increasingly robust network, since mutations with a given step size distribution affect the performance of a network much less when the corresponding parameter is large in absolute value.Costs of plasticityPhenotypically plastic organisms can incur different types of costs68. In our simple model, we only consider the cost of phenotype-environment mismatching, that is, the costs of expressing the ‘wrong’ phenotype in a given environment. When placed in a high-fat environment, the preadapted GRNs in our simulations take the ‘right’ decision to switch off fat metabolism. Accordingly, they do not face any costs of mismatching. Yet, the genetic switch rapidly decays (as indicated in Fig. 3 by the rapid drop in performance when tested in an intermediate-fat environment), due to the accumulation of mutations.It is not unlikely that there are additional fitness costs of plasticity, such as the costs for the production and maintenance of the machinery underlying plasticity68. In the presence of such constitutive costs, plasticity will be selected against when organisms are living in an environment where only one phenotype is optimal (as in the high- and low-fat environments in Fig. 4). This would obviously affect the evolutionary dynamics in Fig. 3, but the size of the effect is difficult to judge, as the constitutive costs of plasticity are notoriously difficult to quantify. In case of the simple switching device considered in our model, we consider the constitutive costs of plasticity as marginal, but these costs might be substantial in other scenarios. More

A reef manta ray visits a cleaning station at Lady Elliot Island, Australia. Credit: A. O. Armstrong et al./Ecol. Evol. (CC BY 4.0)

Ecology
08 April 2021
What manta rays remember: the best spots to get spruced up

Giant fish preserve a mental map of where cleaning fish provide the highest-quality pest removal.

Share on Twitter
Share on Twitter

Share on Facebook
Share on Facebook

Share via E-Mail
Share via E-Mail

Even sea creatures need pampering. Manta rays make regular visits to ‘cleaning stations’, where small fish rid the rays of skin parasites at the coral-reef equivalent of a day spa. Now it seems that rays can identify and remember spots where they have received quality cleaning.Cleaning stations are often centred on corals inhabited by cleaner shrimp or fish. To understand how these stations influence rays’ movements, Asia Armstrong at the University of Queensland in St Lucia, Australia, and her colleagues tracked 34 reef manta rays (Mobula alfredi) off the coast of eastern Australia for about 1.5 years.The highest density of rays was found at places where cleaning fish called blue-streak cleaner wrasses (Labroides dimidiatus) were most abundant. Rays typically visited cleaning stations during the day, when cleaner wrasses are most active, and favoured stations close to foraging regions.Rays are thought to prefer stations that provide superior cleaning — where the cleaners don’t bite them, for example. The rays’ behaviour suggests that they have a mental map of spots that offer both high-quality cleaning and proximity to foraging grounds.

Ecol. Evol. (2021)

Ecology More

1.Fushimi, H. Production of juvenile marine finfish for stock enhancement in Japan. Aquaculture 200, 33–53. https://doi.org/10.1016/S0044-8486(01)00693-7 (2001).Article 

Google Scholar 
2.Masuda, R. & Tsukamoto, K. Stock enhancement in Japan: review and perspective. Bull. Mar. Sci. 62, 337–358. https://www.ingentaconnect.com/content/umrsmas/bullmar/1998/00000062/00000002/art00005 (1998).3.Sekino, M., Saitoh, K., Yamada, T., Hara, M. & Yamashita, Y. Genetic tagging of released Japanese flounder (Paralichthys olivaceus) based on polymorphic DNA markers. Aquaculture 244, 49–61. https://doi.org/10.1016/j.aquaculture.2004.11.006 (2005).CAS 
Article 

Google Scholar 
4.Arnold, W. S. Bivalve enhancement and restoration strategies in Florida, USA. Hydrobiologia 465, 7–19. https://doi.org/10.1023/A:1014596909319 (2001).Article 

Google Scholar 
5.Castell, L. L., Naviti, W. & Nguyen, F. Detectability of cryptic juvenile Trochus niloticus Linnaeus in stock enhancement experiments. Aquaculture 144, 91–101. https://doi.org/10.1016/S0044-8486(96)01320-8 (1996).Article 

Google Scholar 
6.McCormick, T. B., Herbinson, K., Mill, T. S. & Altick, J. A review of abalone seeding, possible significance and a new seeding device. Bull. Mar. Sci. 55, 680–693. https://www.ingentaconnect.com/content/umrsmas/bullmar/1994/00000055/f0020002/art00035 (1994).7.Zohar, Y. et al. The Chesapeake Bay blue crab (Callinectes sapidus): A multidisciplinary approach to responsible stock replenishment. Rev. Fish. Sci. 16, 24–34. https://doi.org/10.1080/10641260701681623 (2008).Article 

Google Scholar 
8.Funge-Smith, S., Briggs, M. & Miao, W. Regional Overview of Fisheries and Aquaculture in Asia and the Pacific 2012 (RAP Publication (FAO), 2012).
Google Scholar 
9.Mao, Y. et al. Chapter 4. Bivalve production in China. In Goods and Services of Marine Bivalves (eds Smaal, A. C. et al.) 51–72 (Springer, 2019).
Google Scholar 
10.Ran, Z. et al. Fatty acid and sterol changes in razor clam Sinonovacula constricta (Lamarck 1818) reared at different salinities. Aquaculture 473, 493–500. https://doi.org/10.1016/j.aquaculture.2017.03.017 (2017).CAS 
Article 

Google Scholar 
11.Suzuki, T., Inoue, K. & Ozawa, T. Environmental degradation in Ise and Mikawa Bays after 1960’s as viewed from intertidal molluscan community. Boll. Nagoya Univ. Museum 22, 31–64. https://doi.org/10.18999/bulnum.022.04 (2006).Article 

Google Scholar 
12.Nakamura, T. et al. Marine reservoir effect deduced from 14C dates on marine shells and terrestrial remains at archeological sites in Japan. Nucl. Instrum. Methods Phys. Res. B 259, 453–459. https://doi.org/10.1016/j.nimb.2007.01.186 (2007).ADS 
CAS 
Article 

Google Scholar 
13.Ministry of the Environment, Japan. Ariake Sea and Yatsushiro Sea Comprehensive Survey Evaluation Committee Report. https://www.env.go.jp/council/20ari-yatsu/rep061221/all.pdf (2006).14.Ito, S., Eguchi, T. & Kawahara, I. Rearing experiment on planktonic larvae of the JackKnife clam, Sinonovacula constricta. Boll. Saga Prefect. Ariake Fish. Res. Dev. Cent. 20, 49–53. https://agriknowledge.affrc.go.jp/RN/2010813185.pdf (2001).15.Ohkuma, H., Ymagachi, T., Kawahara, I., Eguchi, T. & Ito, S. A study on the development of techniques for mass production of seeds of jackknife clam, Sinonovacula constricta. Boll. Saga Prefect. Ariake Fish. Res. Dev. Cent. 22, 47–54. https://agriknowledge.affrc.go.jp/RN/2030813210.pdf (2004).16.Tsukuda, M. et al. Variation in the Distribution of the JackKnife Clam, Sinonovacula constricta, on the Muddy Tidal flat of Ariake Sound off Saga Prefecture. Boll. Saga Prefect. Ariake Fish. Res. Dev. Cent. 28, 47–49. https://agriknowledge.affrc.go.jp/RN/2010925685.pdf (2017).17.Holman, L. E., de la Garcia, S. D., Onoufriou, A., Hillestad, B. & Johnston, I. A. A workflow used to design low density SNP panels for parentage assignment and traceability in aquaculture species and its validation in Atlantic salmon. Aquaculture 476, 59–64. https://doi.org/10.1016/j.aquaculture.2017.04.001 (2017).CAS 
Article 

Google Scholar 
18.Li, Y. H. & Wang, H. P. Advances of genotyping-by-sequencing in fisheries and aquaculture. Rev. Fish Biol. Fish. 27, 535–559. https://doi.org/10.1007/s11160-017-9473-2 (2017).Article 

Google Scholar 
19.Robledo, D., Palaiokostas, C., Bargelloni, L., Martínez, P. & Houston, R. Applications of genotyping by sequencing in aquaculture breeding and genetics. Rev. Aquac 10, 670–682. https://doi.org/10.1111/raq.12193 (2018).Article 
PubMed 

Google Scholar 
20.You, X., Shan, X. & Shi, Q. Research advances in the genomics and applications for molecular breeding of aquaculture animals. Aquaculture 526, 735357. https://doi.org/10.1016/j.aquaculture.2020.735357 (2020).CAS 
Article 

Google Scholar 
21.Arthington, A. H. Ecological and genetic impacts of introduced and translocated freshwater fishes in Australia. Can. J. Fish. Aquat. Sci. 48, 33–43. https://doi.org/10.1139/f91-302 (1991).Article 

Google Scholar 
22.Habtemariam, B. T., Arias, A., García-Vázquez, E. & Borrell, Y. J. Impacts of supplementation aquaculture on the genetic diversity of wild Ruditapes decussatus from northern Spain. Aquacult. Environ. Interact. 6, 241–254. https://doi.org/10.3354/aei00128 (2015).Article 

Google Scholar 
23.Williams, S. L. & Orth, R. J. Genetic diversity and structure of natural and transplanted eelgrass populations in the Chesapeake and Chincoteague Bays. Estuaries 21, 118–128. https://doi.org/10.2307/1352551 (1998).Article 

Google Scholar 
24.Yamakawa, A. Y. & Imai, H. PCR-RFLP typing reveals a new invasion of Taiwanese Meretrix (Bivalvia: Veneridae) to Japan. Aquat. Invasions 8, 407–415. https://doi.org/10.3391/ai.2013.8.4.04 (2013).Article 

Google Scholar 
25.Niu, D. H., Feng, B. B., Liu, D. B., Zhong, Y. M., Shen, H. D. & Li, J. L. Significant genetic differentiation among ten populations of the razor clam Sinonovacula constricta along the coast of china revealed by a microsatellite analysis. Zool. Stud. 51, 406–414. http://zoolstud.sinica.edu.tw/Journals/51.3/406.pdf. (2012).26.DeFaveri, J., Shikano, T., Ghani, N. I. A. & Merilä, J. Contrasting population structures in two sympatric fishes in the Baltic Sea basin. Mar. Biol. 159, 1659–1672. https://doi.org/10.1007/s00227-012-1951-4 (2012).Article 

Google Scholar 
27.Jeffery, N. W. et al. RAD sequencing reveals genomewide divergence between independent invasions of the European green crab (Carcinus maenas) in the Northwest Atlantic. Ecol. Evol. 7, 2513–2524. https://doi.org/10.1002/ece3.2872 (2017).Article 
PubMed 
PubMed Central 

Google Scholar 
28.Kato, D. et al. Evaluation of the population structure and phylogeography of the Japanese Genji firefly, Luciola cruciata, at the nuclear DNA level using RAD-Seq analysis. Sci. Rep. 10, 1533. https://doi.org/10.1038/s41598-020-58324-9 (2020).ADS 
CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
29.Yokota, M., Harada, Y. & Iizuka, M. Genetic drift in a hatchery and the maintenance of genetic diversity in hatchery-wild systems. Fish Sci. 69, 101–109. https://doi.org/10.1046/j.1444-2906.2003.00593.x (2003).CAS 
Article 

Google Scholar 
30.Harada, Y., Yokota, M. & Iizuka, M. Genetic risk of domestication in artificial fish stocking and its possible reduction. Popul. Ecol. 40, 311–324. https://doi.org/10.1007/BF02763463 (1998).Article 

Google Scholar 
31.Etter, P. D., Bassham, S., Hohenlohe, P. A., Johnson, E. A. & Cresko, W. A. SNP discovery and genotyping for evolutionary genetics using RAD sequencing, in: Molecular methods for evolutionary genetics. Methods Mol. Biol. Humana Press 772, 157–178. https://doi.org/10.1007/978-1-61779-228-1_9 (2011).CAS 
Article 

Google Scholar 
32.Rochette, N. C. & Catchen, J. M. Deriving genotypes from RAD-seq short-read data using stacks. Nat. Protoc. 12, 2640–2659. https://doi.org/10.1038/nprot.2017.123 (2017).CAS 
Article 
PubMed 

Google Scholar 
33.Ran, Z. et al. Chromosome-level genome assembly of the razor clam Sinonovacula constricta (Lamarck, 1818). Mol. Ecol. Resour. 19, 1647–1658. https://doi.org/10.1111/1755-0998.13086 (2019).CAS 
Article 
PubMed 

Google Scholar 
34.Li, H. Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. arXiv preprint arXiv: 1303.3997. https://arxiv.org/abs/1303.3997 (2013).35.Zheng, X. et al. A high-performance computing toolset for relatedness and principal component analysis of SNP data. Bioinformatics 28, 3326–3328. https://doi.org/10.1093/bioinformatics/bts606 (2012).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
36.Danecek, P. et al. The variant call format and VCFtools. Bioinformatics 27, 2156–2158. https://doi.org/10.1093/bioinformatics/btr330 (2011).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
37.Alexander, D. H., Novembre, J. & Lange, K. Fast model-based estimation of ancestry in unrelated individuals. Genome Res. 19, 1655–1664. https://pubmed.ncbi.nlm.nih.gov/19648217/ (2009).38.Purcell, S. et al. PLINK: A tool set for whole-genome association and population-based linkage analyses. Am. J. Hum. Genet. 81, 559–575. https://doi.org/10.1086/519795 (2007).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
39.Eklund, A. R package beeswarm: the bee swarm plot, an alternative to stripchart. https://cran.r-project.org/web/packages/beeswarm/beeswarm.pdf (2016).40.Stamatakis, A. RAxML version 8: A tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics 30, 1312–1313. https://doi.org/10.1093/bioinformatics/btu033 (2014).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
41.Chifman, J. & Kubatko, L. Quartet inference from SNP data under the coalescent model. Bioinformatics 30, 3317–3324. https://doi.org/10.1093/bioinformatics/btu530 (2014).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
42.Swofford, D. L. PAUP*. Phylogenetic analysis using parsimony (*and other methods). Version 4 (Sinauer Associates, 2002).
Google Scholar  More

Species threat abatement and restoration (STAR) metricWe developed and analysed a STAR metric that evaluates the potential benefit for threatened species of actions to reduce threats and restore habitat. Like the Red List Index7,8, STAR is derived from existing data in the IUCN Red List and is intended to help address an urgent need. STAR is spatially explicit, enabling identification of specific threat abatement and habitat restoration opportunities in particular places, which, if implemented, could reduce species extinction risk to levels that would exist without ongoing human impact. Abatement of threats to species encompasses reduction in threat intensity and/or action to mitigate the impacts of threats. Positive population and/or distribution changes, along with the resulting reduction of species extinction risk, have been documented in response to threat abatement13. STAR assumes that, for the great majority of species (see Supplementary Discussion), complete alleviation of threats would reduce extinction risk through halting the decline and/or permitting sufficient recovery in population and distribution, such that the species could be downlisted to the IUCN Red List category of Least Concern. We recognize that complete threat reduction is difficult, incremental conservation gains will need to be tracked at the species level14 and species recovery will vary across a species’ range14.For each species, a global STAR threat abatement (START) score is defined. This varies from zero for species of Least Concern to 100 for Near Threatened, 200 for Vulnerable, 300 for Endangered and 400 for Critically Endangered species (using established weighting ratios7,8). The sum of START values across all species represents the global threat abatement effort needed for all species to become Least Concern. START scores can be disaggregated spatially, based on the area of habitat (AOH) currently available for each species in a particular location (as a proxy for population proportion). This shows the potential contribution of conservation actions in that location to reducing the extinction risk for all species globally. The local START score can be further disaggregated by threat, based on the known contribution of each threat to the species’ risk of extinction (see Methods). This quantifies how actions that abate a specific threat at a particular location contribute to the global abatement of extinction risk for all species.The STAR metric also includes a complementary habitat restoration component to reflect the potential benefits to species of restoring lost habitat. During the United Nations Decade on Ecosystem Restoration (2021–2030), restoration efforts are likely to expand. The STAR restoration component applies a similar logic to the STAR threat abatement component, but for habitat that has been lost and is potentially restorable (that is, restorable AOH). The STAR restoration component does not make assumptions about the extent of habitat restoration required for individual species, but instead quantifies the potential contribution that habitat restoration activities could make to reducing species’ extinction risk. For a particular species at a particular location, the STAR restoration (STARR) score reflects the proportion that restorable habitat at the location represents of the global area of remaining habitat for that species. Importantly, a multiplier is applied to STARR scores to reflect the slower and lower success rate in delivering benefits to species from restored habitat compared with conserved existing habitat15. Again, STARR scores can be disaggregated by threat and summed across species within the location.STAR is intended to provide a metric to underpin the establishment of science-based targets as explicit contributions from individual actors towards the post-2020 biodiversity framework, by allowing assessment of actions and locations according to their potential ability to deliver towards international conservation targets. Individual spatially based START and STARR scores, for all species present in a particular location or country, represent a proportion of the global opportunity to reduce species’ extinction risk through threat abatement and restoration, respectively. For each species, the total START score could be achieved by the complete abatement of all threats in remaining habitat, or an equivalent value of the STAR metric can be achieved by a combination of threat abatement in the remaining habitat and restoration of lost habitat (with concomitant threat abatement therein). The metric can support establishment of science-based targets by a range of actors across spatial scales. By enabling governments and non-state actors to quantify their potential contributions, STAR, along with other tools, could facilitate achievement of global policy goals, notably the species component of the Sustainable Development Goals and the expected post-2020 Global Biodiversity Framework.STAR uses existing publicly available datasets: species’ extinction risk categories and threats available from the IUCN Red List6 (or, for country endemics not yet assessed globally, from national red lists); and species’ AOH estimated using species’ ranges, habitat associations, and elevation limits, along with digital elevation models and current and historical land cover maps (here, we used backcast land cover maps of the distribution of habitat pre-human impact, as in ref. 16). To demonstrate the utility of STAR, we calculated global STAR scores for the groups of terrestrial vertebrate species that are comprehensively assessed on the IUCN Red List (that is, threatened and Near Threatened species of amphibians, birds and mammals globally; n = 5,359).Potential to reduce species extinction riskGlobally, the greatest contribution that could be made to reduce the extinction risk of these groups is tackling threats from annual and perennial non-timber crop production, which account for 24.5% of the global START score (Fig. 1). A further 16.4% is contributed by logging and wood harvesting. There are likely to be specific targets for reducing agriculture and forestry threats in the post-2020 framework3, and applying STAR quantifies the large potential contribution that mitigating these threats could make to the goal for species conservation. Appropriate activities to deliver on such targets range along a continuum from land sharing through to land sparing17.Fig. 1: Contribution to the global START score of different threats and the potential contribution of habitat restoration.The total global START score represents the global threat abatement effort needed for all Near Threatened and threatened (Vulnerable, Endangered and Critically Endangered, according to the IUCN Red List) amphibian, bird and mammal species to be reclassified as Least Concern. This score can be disaggregated by threat type, based on the known contribution of each threat to species’ risk of extinction. The STARR score quantifies the potential contribution that habitat restoration activities could make to reducing overall species’ extinction risk. The total START score could thus be achieved by the complete abatement of all threats in existing natural habitat, or through a combination of threat abatement in existing habitat and restoration of lost habitat (with concomitant threat abatement therein).Full size imageSTAR can be used in combination with existing policy and planning tools to quantify the potential contribution of action targets towards species conservation outcomes. The proposed post-2020 framework includes an action target for the protection of sites of particular importance to biodiversity3. Key Biodiversity Areas11, which include Important Bird and Biodiversity Areas18 and Alliance for Zero Extinction sites19, correspond to such sites. Key Biodiversity Areas so far cover 8.8% of the terrestrial surface (www.keybiodiversityareas.org; identification is ongoing), but already capture 47% of the global START score for the vertebrate groups analysed. They represent large proportions of some national START scores: >70% in Mexico and Venezuela and >50% in Madagascar, Ecuador, the Philippines and Tanzania.START scores can also support target setting at national and sub-national scales to help meet international policy goals. The control and eradication of invasive species forms one of the CBD’s proposed post-2020 action targets3. New Zealand has already set a Predator Free 2050 goal that aims to eradicate three invasive mammal species by 2050. New Zealand contributes 0.8% to the global START score for the three vertebrate groups included in this study. Achieving the Predator Free 2050 goal would contribute 30% of the total START score for New Zealand, amounting to 0.2% of the global START score.All countries contribute towards the global START score, but scores are highly skewed, with a few countries having high START scores and most having low scores for the vertebrate groups analysed (Fig. 2a and Extended Data Fig. 1). The highest-scoring countries are located in biodiverse regions with many threatened endemic species20: Indonesia contributes 7.1% of the global START score, Colombia 7.0%, Mexico 6.1%, Madagascar 6.0% and Brazil 5.2%. These top five countries contribute 31.3% of the global START score. In contrast, the lowest-scoring 88 countries together contribute only 1% of the global START score. This does not imply that these low-scoring countries have negligible species conservation responsibilities; the global decline in even common species indicates that all countries must act to reverse the degradation of nature and restore the diversity and abundance of species and integrity of ecosystems21, as well as preventing extinctions at a national scale. Moreover, most countries have a Red List Index22, or an equivalent, quantifying their progress or failure in reducing the global extinction risk of assessed species relative to their national responsibility for global species conservation. STAR provides a means to guide the reduction of extinction risk and so assist all countries in meeting national species conservation targets.Fig. 2: Global distribution of START and STARR scores.a,b, Global STAR scores for amphibians, birds and mammals at a 50-km grid cell resolution for START scores (a) and STARR scores (b). Each species has a global START score, weighted relative to their extinction risk. This global START score can be disaggregated spatially, based on the AOH currently available for each species in a particular location. The total START score per grid cell (a) is thus the sum of the individual species’ START scores per grid cell across all Near Threatened and threatened species of amphibians, birds and mammals included in this study. The global STARR score per species reflects the potential contribution that habitat restoration activities could make to reducing species’ extinction risk, and is spatially disaggregated based on the availability of restorable habitat. Thus, the total STARR score per grid cell (b) is the sum of the individual species’ STARR scores per grid cell across all species included in this study. For the legends in a and b, each range excludes the lower bound and includes the upper bound.Full size imageAt the global level, we estimated that an equivalent to 55.9% of the global START score for vertebrates could, theoretically, be achieved by restoring lost habitat within the current range (Fig. 1). Ecosystem restoration objectives have been identified in many national biodiversity strategies for the CBD, as well as in many countries’ commitments under the Bonn Challenge, and as part of Nationally Determined Contributions under the United Nations Framework Convention on Climate Change. The STAR metric has the potential to support restoration initiatives alongside species conservation targets by quantifying the potential benefit to particular species of restoring habitat in specific places23 (Fig. 2b). Restoration may be particularly important for some species, including those assessed under Red List sub-criteria D/D1 (with a very small population) or Bac (with a small range with severe fragmentation, plus extreme fluctuations). For species uniquely assessed under these criteria (2.8% of those included in this study), threat abatement alone is unlikely to eliminate extinction risk, so this might need to be complemented by restoration in order to achieve Least Concern status (see Supplementary Discussion). Moreover, depending on habitat loss and threat type, restoration of habitat may be beneficial for a larger proportion of threatened species.Application of STAR at the landscape scaleWe tested the landscape-scale application of the STAR metric in the southern part of Bukit Tigapuluh landscape, in central Sumatra, Indonesia (Fig. 3a). The Bukit Tigapuluh Sustainable Landscape and Livelihoods Project is a sustainable commercial rubber initiative. The study area (approximately 88,000 ha) includes a 5-km buffer (which is set aside to support local livelihoods, wildlife conservation areas and forest protection and restoration) and two ecosystem restoration areas (which form a conservation management zone that protects the Bukit Tigapuluh National Park from encroachment).Fig. 3: STAR results for the Bukit Tigapuluh Sustainable Landscape and Livelihoods Project.The Bukit Tigapuluh Sustainable Landscape and Livelihoods Project is a sustainable commercial rubber initiative. The study area (approximately 88,000 ha) includes a 5-km buffer, which is set aside to support local livelihoods, wildlife conservation areas and forest protection and restoration, and two ecosystem restoration areas, which form a conservation management zone that protects the Bukit Tigapuluh National Park from encroachment. a, Mapped START scores in areas with remaining forest (green) and STARR scores in areas where forest has been lost (purple) at the 30-m grid cell resolution. b, START scores per threat for the top five highest-scoring threats across the study area (the concession, 5-km buffer and ecosystem restoration areas combined).Full size imageThe total START score for the study area represents 0.2% of the START score for Sumatra, 0.04% of the START score for Indonesia and 0.003% of the global START. The major threats are from annual and perennial non-timber crops, logging and wood harvesting, and the collection of terrestrial animals (Fig. 3b). The proximate causes of these pressures in the project area are rubber cultivation, oil palm cultivation, industrial logging, subsistence wood cutting and hunting. STAR analysis shows that areas with the greatest potential to contribute to species conservation through threat mitigation are in remaining natural habitat close to the national park, with a small area of high potential also to the west, where the relatively small distribution of the orbiculus leaf-nosed bat (Hipposideros orbiculus) overlaps the site (Fig. 3a). Additionally, due to recent forest loss, 47% of the START score for the study area could be achieved through habitat restoration (that is, STARR). Investment in these management actions has the potential to deliver these quantified contributions to national and global biodiversity targets.Operationalization and future developmentThe STAR metric makes use of the best available data, producing results that are relevant to policy and practice. However, there is scope for future refinement as the underlying data improve. Here, the STAR metric covers amphibians, birds and mammals globally, constituting a well-studied but small proportion of taxonomic diversity (see Extended Data Figs. 2 and 3 for variation among taxa). STAR can be expanded to other taxonomic groups, including freshwater and marine species, as data become available (reptiles, cacti, cycads, conifers, freshwater fish and reef-building corals are among the groups imminently available for incorporation). Global application of STAR will require comprehensive assessment of taxonomic groups, testing of the transferability of the STAR metric assumptions among taxa as Red List coverage expands, and further development of methods to calculate AOH. AOH calculation does not currently capture spatial variation in species’ population density, which will be important for many species14; such data have not been gathered on a global scale yet and could be incorporated as available.The completeness of threat data in the IUCN Red List is uneven but is continually improving. The STAR metric does not currently reflect spatial variation in threat magnitude within species’ ranges; more broadly, there is a lack of information on the spatial distribution of threats24. Most species included in this study have relatively small ranges; the total current AOH is More