Ecology

1.
Pringle, R. M. A mountain of ecological interactions. Nature 568, 38–39 (2019).
CAS  Article  ADS  Google Scholar 
2.
Sayre, R. et al. A new high-resolution map of world mountains and an online tool for visualizing and comparing characterizations of global mountain distributions. Mt Res Dev 38, 240–249. https://doi.org/10.1659/mrd-journal-d-17-00107.1 (2018).
Article  Google Scholar 

3.
Peters, M. K. et al. Climate-land-use interactions shape tropical mountain biodiversity and ecosystem functions. Nature 568, 88–92. https://doi.org/10.1038/s41586-019-1048-z (2019).
CAS  Article  PubMed  ADS  Google Scholar 

4.
Bian, J., Li, A., Lei, G., Zhang, Z. & Nan, X. Global high-resolution mountain green cover index mapping based on Landsat images and Google Earth Engine. ISPRS J Photogramm Remote Sens 162, 63–76. https://doi.org/10.1016/j.isprsjprs.2020.02.011 (2020).
Article  ADS  Google Scholar 

5.
Antonelli, A. et al. Geological and climatic influences on mountain biodiversity. Nat Geo 11, 718–725. https://doi.org/10.1038/s41561-018-0236-z (2018).
CAS  Article  ADS  Google Scholar 

6.
Price, M. F., Arnesen, T., Gløersen, E. & Metzger, M. J. Mapping mountain areas: learning from Global, European and Norwegian perspectives. J Mt Sci 16, 1–15. https://doi.org/10.1007/s11629-018-4916-3 (2019).
Article  Google Scholar 

7.
Körner, C., Paulsen, J. & Spehn, E. M. A definition of mountains and their bioclimatic belts for global comparisons of biodiversity data. Alp Bot 121, 73–78. https://doi.org/10.1007/s00035-011-0094-4 (2011).
Article  Google Scholar 

8.
Körner, C. et al. A global inventory of mountains for bio-geographical applications. Alp Bot 127, 1–15. https://doi.org/10.1007/s00035-016-0182-6 (2017).
Article  Google Scholar 

9.
Karagulle, D. et al. Modeling global Hammond landform regions from 250-m elevation data. Trans GIS 21, 1040–1060. https://doi.org/10.1111/tgis.12265 (2017).
Article  Google Scholar 

10.
Payne, D., Spehn, E. M., Snethlage, M. & Fischer, M. Opportunities for research on mountain biodiversity under global change. Curr Opin Environ Sustain 29, 40–47. https://doi.org/10.1016/j.cosust.2017.11.001 (2017).
Article  Google Scholar 

11.
Silveira, F. A. O. et al. Tropical mountains as natural laboratories to study global changes: A long-term ecological research project in a megadiverse biodiversity hotspot. Perspect Plant Ecol Evol Syst 38, 64–73. https://doi.org/10.1016/j.ppees.2019.04.001 (2019).
Article  Google Scholar 

12.
Elsen, P. R., Monahan, W. B. & Merenlender, A. M. Topography and human pressure in mountain ranges alter expected species responses to climate change. Nat Commun 11, 1974. https://doi.org/10.1038/s41467-020-15881-x (2020).
CAS  Article  PubMed  PubMed Central  ADS  Google Scholar 

13.
Cheng, G. et al. Rebirth after death: forest succession dynamics in response to climate change on Gongga Mountain, Southwest China. J Mt Sci 15, 1671–1681. https://doi.org/10.1007/s11629-017-4435-7 (2018).
Article  Google Scholar 

14.
Trant, A., Higgs, E. & Starzomski, B. M. A century of high elevation ecosystem change in the Canadian Rocky Mountains. Sci Rep 10, 9698. https://doi.org/10.1038/s41598-020-66277-2 (2020).
CAS  Article  PubMed  PubMed Central  ADS  Google Scholar 

15.
Strinella, E., Scridel, D., Brambilla, M., Schano, C. & Korner-Nievergelt, F. Potential sex-dependent effects of weather on apparent survival of a high-elevation specialist. Sci Rep 10, 8386. https://doi.org/10.1038/s41598-020-65017-w (2020).
CAS  Article  PubMed  PubMed Central  ADS  Google Scholar 

16.
Durán-Romero, C., Medina-Sánchez, J. M. & Carrillo, P. Uncoupled phytoplankton-bacterioplankton relationship by multiple drivers interacting at different temporal scales in a high-mountain Mediterranean lake. Sci Rep 10, 350. https://doi.org/10.1038/s41598-019-57269-y (2020).
CAS  Article  PubMed  PubMed Central  ADS  Google Scholar 

17.
Mair, D. et al. Fast long-term denudation rate of steep alpine headwalls inferred from cosmogenic 36Cl depth profiles. Sci Rep 9, 11023. https://doi.org/10.1038/s41598-019-46969-0 (2019).
CAS  Article  PubMed  PubMed Central  ADS  Google Scholar 

18.
Liu, H., Chen, Y., Ye, Z., Li, Y. & Zhang, Q. Recent lake area changes in Central Asia. Sci Rep 9, 16277. https://doi.org/10.1038/s41598-019-52396-y (2019).
CAS  Article  PubMed  PubMed Central  ADS  Google Scholar 

19.
Spandre, P. et al. Climate controls on snow reliability in French Alps ski resorts. Sci Rep 9, 8043. https://doi.org/10.1038/s41598-019-44068-8 (2019).
CAS  Article  PubMed  PubMed Central  ADS  Google Scholar 

20.
Tichavský, R., Ballesteros-Cánovas, J. A., Šilhán, K., Tolasz, R. & Stoffel, M. Dry spells and extreme precipitation are the main trigger of landslides in Central Europe. Sci Rep 9, 14560. https://doi.org/10.1038/s41598-019-51148-2 (2019).
CAS  Article  PubMed  PubMed Central  ADS  Google Scholar 

21.
Stoffel, M., Ballesteros Cánovas, J. A., Luckman, B. H., Casteller, A. & Villalba, R. Tree-ring correlations suggest links between moderate earthquakes and distant rockfalls in the Patagonian Cordillera. Sci Rep 9, 12112. https://doi.org/10.1038/s41598-019-48530-5 (2019).
CAS  Article  PubMed  PubMed Central  ADS  Google Scholar 

22.
de Haas, T., Nijland, W., de Jong, S. M. & McArdell, B. W. How memory effects, check dams, and channel geometry control erosion and deposition by debris flows. Sci Rep 10, 14024. https://doi.org/10.1038/s41598-020-71016-8 (2020).
CAS  Article  PubMed  PubMed Central  ADS  Google Scholar 

23.
Yousefi, S. et al. A machine learning framework for multi-hazards modeling and mapping in a mountainous area. Sci Rep 10, 12144. https://doi.org/10.1038/s41598-020-69233-2 (2020).
CAS  Article  PubMed  PubMed Central  ADS  Google Scholar  More

Identification of the risk variables and their correlations with the COVID-19 damages
We have investigated a series of factors contributing to the risk of an epidemic diffusion and its impact on the population. Among many possible, we selected the following variables: mobility index, housing concentration, healthcare density, air pollution, average winter temperature and age of population. In paragraph 1 of Methods section we motivate our choice on such variables (mainly based on epidemics literature and features of the COVID-19 outbreak), show the related data (see Table 1) and explain the adopted normalization.
The first step is, of course, to estimate to what extent the chosen normalized variables individually correlate with the main impact indicators of the COVID-19 epidemic, i.e., total cases and total deaths detected in each Italian region, cumulated up to July 14, 20204, when the first epidemic wave seemed to have finished, and the intensive care occupancy recorded on April 2, 20204, when the epidemic peak was reached. In the first two rows of Fig. 2, from panel (a) to panel (f), the spatial distributions of the six risk indicators, multiplied by the population of each region, are reported as chromatic maps and thus can be visually compared with the analogous maps of the three impact indicators, panels (g), (h) and (i) in the third row. As detailed in Table 2, in paragraph 2 of Methods section, pairwise correlations between risk indicators are, with a few exceptions, quite weak; furthermore, in Table 3, results of the linear least squares fit of each individual risk indicator to damages are reported. We found correlation coefficients ranging from 0.71 to 0.96, always higher than those observed as a function of the population, which can be considered the null model; however, the relative quadratic errors stay quite high (from 0.26 to 0.62). This suggests that some opportune combination of risk indicators could better capture the risk associated to each region. In the next paragraph, we propose a risk assessment framework aimed to this.
Figure 2

The geographical distribution of the six risk factors (a–f) can be compared with the COVID-19 total cases (g), the total deaths (h) and the intensive care occupancy (i). Cases and deaths have been cumulated up to July 14, 2020, i.e. at the end of the first epidemic wave; the intensive care data have been recorded on April 2, 2020, i.e. just before the epidemic peak. The risk indicators have been multiplied for the population of each region and normalized between 0 and 1 (the color scale for temperature has been reversed, i.e. dark colors mean low temperatures, see Methods). A concentration of dark colors in the northern regions is roughly visible for almost all the indicators and the correlations between the single factors and the damages range from 0.70 to 0.95. Maps were realized with QGIS 3.10 (https://qgis.org/en/site/). (l) Crichton’s Risk Triangle. (m) Risk Index assessment framework: risk indicators (factors) are reported in red, risk components in black.

Full size image

Definition of a risk assessment framework and calibration with COVID-19 data
Conventional risk assessment theory relies on “Crichton’s Risk Triangle”24,25, shown in panel (l) of Fig. 2. In this framework, risk is evaluated as a function of three components: Hazard, Exposure and Vulnerability. Hazard is the potential for an event to cause harm (e.g., earthquake, flooding, epidemics); Exposure measures the amount of assets exposed to harm (e.g., buildings, infrastructures, population); Vulnerability is the harm proneness of assets if exposed to hazard events (e.g., building characteristics, drainage systems, age of population). The risk is present only when all of the three components co-exist in the same place. Used for the first time in the insurance industry24, this approach has been extended to assess spatially distributed risks in many fields of disaster management, such as those related to climate change impact27,28,29,30,31 and earthquakes32.
In the present paper, we consider Hazard as the degree of diffusion of the virus over the population of an Italian region (influenced by a set of factors, related to spatial and socio-economic characteristics of the region itself); Exposure is the amount of people who might potentially be infected by the virus as a consequence of the Hazard (it should coincide with the size of the population of the region); Vulnerability is the propensity of an infected person to become sick or die (in general, it is strongly related to the age and pre-existing health conditions prior to infection). The combination of Vulnerability and Exposure provides a measure of the absolute damage (i.e., the number of ill people due to pathologies related to the virus in the region), which we called Consequences.
In paragraph 3 of Methods section we propose two models that differ in the way the risk indicators are aggregated into the three components of the Crichton’s risk triangle. In particular, we consider the E_HV model, where the effect of Hazard and Vulnerability are combined in a single affine function of the six indicators, and the E_H_V model, where Hazard and Vulnerability are considered as affine functions of, respectively, mobility index, housing concentration and healthcare density, on one hand, and air pollution, average winter temperature and age of population on the other hand (see Fig. 2 (m) for a summary). In both models the Exposure is represented by the population of each region. Furthermore, two versions of each model have been considered: an optimized one, where the weights of the risk indicators are obtained through a least-square fitting versus real COVID-19 data, and an a-priori one, where all the weights are assumed to be equal.
As shown in Tables 4 and 5 of Methods section, models based on data fitting perform better, both in terms of relative mean quadratic error and correlation coefficient, as expected. In particular, the E_H_V model fits the best. Furthermore, in agreement with the strong correlation of the variables with the targets, most coefficients are positive. Indeed, all coefficients obtained by fitting the number of cases and the intensive care occupancy are positive, and only one negative coefficient appears in each model, when fitting the number of deceased. However, the numerical value of the coefficients strongly depends on both models and targets, making these models not very robust. On the other hand, the a-priori models are independent of the targets, depending only on the choice of the variables we decided to include in the risk evaluation.
Among the two considered a-priori models, where all coefficients assume the same value, we observe that the E_H_V model produces a smaller error with respect to real COVID-19 data and better correlation coefficients than the E_HV model, thus justifying the multiplicative approach which define the risk intensity in terms of the product between Hazard and Vulnerability (we used data at April 2, 2020 for this preliminary analysis but similar results would be obtained using data at July 14, 2020). Moreover, the aggregation of risk indicators in the three components of the E_H_V model follows better our motivations to choose those indicators (as explained in Methods, paragraph 1).
Validation of the a-priori E_H_V model on COVID-19 data
Once we established the robustness of the a-priori E_H_V model, let us now build the corresponding regional risk ranking and validate the model with the regional COVID-19 data as a case study. In particular, following the scheme of Fig. 2 (m), by multiplying Exposure and Vulnerability for the k-th region, we first calculate the Consequences ((C_{k} = E_{k} cdot V_{k}), k = 1,…,20). Then, by multiplying Hazard and Consequences, we obtain the global risk index (R_{k}) for each region ((R_{k} = H_{k} cdot C_{k}), k = 1,…, 20). In this respect, the risk index can be interpreted as the product of what is related to the occurrence of causes of the virus diffusion in a given region ((H_{k})) and what is related to the severity of effects on people ((C_{k})).
In Fig. 3a we can appreciate the predictive capability of our model by looking at the a-priori risk ranking of the Italian regions, compared with the COVID-19 data4, in terms of total cases (cumulated), deaths (cumulated) and intensive care occupancy (daily, not cumulated), updated both at April 2, 2020 and July 14, 2020. The values of (R_{k}) have been normalized to their maximum value, so that Lombardia results to have (R_{k}) = 1. The average of (R_{k}) over all the regions is (R_{av} = 0.15) and can be considered approximately a reference level for the Italian country (even if, of course, it has only a relative value).
Figure 3

(a) A-priori normalized risk ranking of Italian regions, emerging from our analysis of risk indicators, compared with the corresponding total cases, deaths and intensive care occupancy updated, respectively, at April 2, 2020 (just before the epidemic peak) and at July 14, 2020 (at the end of the first wave). Regions are organized in four risk groups, corresponding to different colors: very high, high, medium and low risk. The agreement with the observed effects Data referring to overestimations or underestimations of risk are also colored in green and red, respectively. (b–d) Comparison between the spatial distribution of COVID-19 total cases at July 14, 2020 (b), the most struck regions (in terms of severe cases and deaths) from 2019–2020 seasonal flu (d) according to the ISS data19 and our a-priori risk map (c). The geographical correlation with the risk map is evident for both kind of epidemic flus. Maps were realized with QGIS 3.10 (https://qgis.org/en/site/).

Full size image

As already explained, due to the intrinsic limitations of the official COVID-19 data, it is convenient to make the comparison at the aggregate level of groups of regions, without expecting to predict the exact rank within each group. Let us therefore arrange the 20 regions in four risk groups, each one characterized by a different color and ordered according to decreasing values of the risk index: very high risk ((0.4 < R_{k} le 1), in red), high risk ((0.2 < R_{k} le 0.4), in brown), medium risk ((0.03 < R_{k} le 0.2), in beige) and low risk ((R_{k} le 0.03), in pink). With this choice, our model is clearly able to correctly identify the four northern regions where the epidemic effects have been far more evident, in terms of cases, deaths and intensive care occupancy: the first in the ranking, i.e. Lombardia (whose risk score is about three times the second classified) and the group of the three regions immediately after it, Veneto, Piemonte and Emilia Romagna (even if not in the exact order of damage). A quite good agreement can be observed also for the other two groups: only for Sardegna the effects on both total cases and deaths seem to have been slightly overestimated (its insularity might play a role), while for other two regions, Umbria and Valle d’Aosta, some impact indicators have been slightly underestimated. Notice that the proposed risk classification seems quite robust, since it holds both near to the peak of April and at the end of the first wave, in July, when the intensive care occupancy of the majority of the regions was zero. In Table 6 reported in Methods, a further analysis of the robustness of this classification has been performed by eliminating, one by one, single indicators from the risk index definition: results show that the position of some regions slightly changes inside each group, but the composition of the four risk groups remains for the mostly unchanged with just few exceptions worsening the agreement with the impact indicators shown in Fig. 3a. This confirms the advantage of including all indicators in the risk index. The clear separation between northern regions from central and southern ones is also confirmed in the bottom part of Fig. 3, where the a-priori risk color map, in panel (c), is compared with the map of COVID-19 total cases in July, panel (b), and the map of the serious cases and deaths of the seasonal flu 2019/20 in Italy, panel (d) (ISS data19). The agreement is clearly visible. In Fig. 4 we show the correlations between the a-priori risk index and the three main impact indicators related to the outbreak, i.e. the total number of cases (a) and the total number of deaths (b), cumulated up to July 14, 2020, and the intensive care occupancy (c), registered at April 2, 2020. For each plot, a linear regression has been performed, with Pearson correlation coefficients always taking values greater or equal to 0.97, indicating a strong positive correlation. On the right of each plot we report the corresponding percentages of damage observed in the three Italian macro-regions—North, Center and South, see the geographic map (d). Also in this case the correlation is evident, if compared with the percentage of cumulated a-priori risk associated to the same macro-regions (e). Figure 4 The three main impact indicators for COVID-19—the total number of cases (a) and the total number of deaths (b) cumulated up to July 14, 20204, and the intensive care occupancy (c) at April 2, 20204—are reported as function of the a-priori risk index for all the Italian regions. The size of the points is proportional to the risk index score. A linear regression has been performed for each plot. The Pearson correlation coefficients are very good, always greater or equal than 0.97. The corresponding percentages of damages, aggregated for the three Italian macro-regions (North, Center and South (d)) are also reported to the right and can be compared with the percentages of cumulated a-priori risk (e). It is clear that our a-priori risk index is able to explain the anomalous damage discrepancies between these different parts of Italy. Maps were realized with QGIS 3.10 (https://qgis.org/en/site/). Full size image Another interesting way to visualize these correlations is to represent the a-priori risk index through its two main aggregated components, Hazard and Consequences, and plotting each region as a point of coordinates ((H_{i} ,C_{i} )) in the plane (left{ {H times C} right}). This Risk Diagram is reported in Fig. 5a, where the points have been also characterized by the same color of the corresponding risk group of Fig. 3. It is evident that the iso-risk line described by the equation C = Rav/H (being Rav = 0.15 the average regional risk value) is correctly able to separate the four more damaged and highly risky, northern regions (plus Lazio) from all the others. The value of the risk index is reported in parentheses next to each region name. As shown in Fig. 5b, where the ranking of the Italian regions has been disaggregated for both Hazard and Consequences, it is interesting to notice that some regions (such as Friuli, Trentino or Valle d’Aosta) exhibit high values of Hazard and quite low values of Consequences, while for other regions (such as Campania or Piemonte) the opposite is true. See also the colored geographic maps in Fig. 5c,d for a visual comparison. This confirms that it is necessary to aggregate such two main components in a single global index to have a more reliable indication of the regional a-priori risk. Figure 5 (a) Risk Diagram. Each region is represented as a point in the plane (left{ {H times C} right}) while the color is proportional to the corresponding risk group updated at July 14, 2020 (see Fig. 3a). The most damaged regions lie with a good approximation above the C = Rav/H hyperbole (i.e. the iso-risk line related to the average regional risk index), while the less damaged ones lie below this line. The a-priori risk index score is also reported for each region. (b) The rankings of Italian regions according to either Hazard (on the left) or Consequences (on the right). The corresponding colored geographic maps are also shown in panels (c) and (d) for comparison. Maps were realized with QGIS 3.10 (https://qgis.org/en/site/). Full size image Let us close this paragraph by showing, in Fig. 6, three sequences of the geographic distribution of the total cases (a), total number of deaths (b) and current intensive care occupancy (c) as a function of time, from March 9 to July 14, 2020. These sequences are compared with the geographic map of the a-priori risk level (the bordered image on the right in each sequence), the latter being independent of time. In all the plots, damages seem to spread over the regions with a variable intensity (expressed by the color scale) quite correctly predicted by our a-priori risk analysis. The intensive care occupancy map compared with the risk map is dated April 2, since the occupancy on July 14 is zero almost everywhere (with the exception of Lombardia and a few other regions). Figure 6 The geographic distributions of damage in the various Italian regions—cumulated total cases (a), cumulated total deaths (b) and daily intensive care occupancy (c)—are reported as function of time, from March 9, 2020 to July 14, 2020 and compared with the geographic distribution of the a-priori risk. Obviously, the intensive care occupancy to compare with the risk map is that of April, since in July, at the end the epidemic wave, this variable is zero everywhere except for a few regions (among which only Lombardia has a score slightly higher than 25). Maps were realized with QGIS 3.10 (https://qgis.org/en/site/). Full size image In the next paragraph, the methodology proposed in this paper, and in particular this representation in terms of risk diagram, will be used to build a policy model aimed at mitigating damages in case of an epidemic outbreak similar to the COVID-19 one. A proposal for a policy protocol to reduce the epidemic risk We have seen how the risk can be thought as composed in two components, one related to the causes of the infection diffusion and the other to the consequences. In this paragraph we will interpret the consequences in terms of protection and required support to people with the goal of improving the social result and/or reducing the economic cost. It is evident that enhancing the capability of the healthcare system appears to be the most important action: basically, the insufficient carrying capacity creates the emergency. Beyond specific factors explained above, the epidemic crisis in Lombardia essentially showed a breakdown of its healthcare system, caused by high demand rate for hospital admissions, long permanence times in intensive care, insufficient health assistance (diagnosis equipment, staff, spaces, etc.). Previously illustrated data provide a positive analysis of an epidemic disease (i.e., how things are, in a given state of the world). The normative approach here described presents a viable framework to assess possible policy protocols. Several variables affecting the diffusion of an infection can be looked at as suitable policy instruments to manage both the spreading process and the stress level to the healthcare system of a given district (such as a country, a region, an urban area, etc.). Following the evidence suggested by data, we propose a theoretical model (whose details are presented in the Methods section, paragraph 4) based on two independent variables influencing the level of risk, namely the infection ratio, i.e., the proportion of infected individuals over the total population, and the number of per capita hospital beds, as a measure of the impact of consequences caused by the spreading of the disease. We adopt an approach based on a standard model of economic policy, in which a series of instruments explicitly affecting the infection ratio and the per capita hospital beds endowment can be used to approach the target, i.e., the minimization of the risk level. A similar rationale, covering other topics, can be found in Samuelson and Solow33 (1960) and builds upon a widely consolidated literature which dates back in time34,35,36,37,38,39 (among many others). Despite the analysis concerns a collective problem, the model here proposed describes elements of a possible decision process followed by an individual policy-maker, thus remaining microeconomic in nature. Panel (a) in Fig. 7 shows the risk function, while the right panel provides an illustration of the family of its convex contours, for a finite set of risk levels (limited for graphic convenience): Figure 7 (a,b) The Risk function and its convex contours: an example for (R = x^{0.5} b^{0.5}). (c,d) The carrying capacity function and effects of policy interventions on the supply-side. (e,f) Comparative statics of equilibrium and disequilibrium. (g,h) Two examples of model implementation, see the main text. Full size image Panel (b) in Fig. 7 replicates the meaning of Fig. 5a by translating the consequences indicated by data as the required per capita hospital beds, while explaining that the position of each iso-risk curve corresponds to the different actual composition of the scenario at hand. We assume a unique care strategy based on the structural carrying capacity of the healthcare system, defined as the available number of per capita hospital beds. Such a carrying capacity derives from the health expenditure (G_{H}), which is set to a level considered sufficient. Such a choice is based on political decisions and is reasonably inferred from past experience, structural elements of population, such as age and territorial density, etc. A part of the deliberated budget is dedicated to set up intensive care beds, as an advanced assistance service provision. During an emergency, possibly deriving from an epidemic spreading, the number of beds can suddenly reveal insufficient. In other words, it is possible that the amount of hospital beds required at a certain point is greater than the current availability. In the model, we assume the number of hospital beds, H, and the proportion of intensive care beds, (alpha), as exogenously determined by the policy-maker who fixes (G_{H}). The actual carrying capacity is shown as a function of the infection ratio, x, computed as the infected population over the total, as shown in panel (c) of Fig. 7, and detailed in paragraph 4 of Methods. Changes in the proportion of per capita intensive care hospital beds over the total, cause instead, a variation in the slope of the line (which becomes steeper for reduction in the proportion of intensive care beds). Finally, changes in the overall expenditure shift the line with the same slope (above for increments of the expenditure). In particular, it is worth to notice that the political choice of the ratio (alpha = HH/H) may imply that the overall capacity to assist the entire population is not guaranteed (i.e. the intercept on the (x) axis might be less than (1)). A direct comparison of elements contained in panels (a-b) and (c-d) of Fig. 7 provides a quick inspection of the policy problem, focused to control the epidemic spreading. The constraint should be considered as a dynamic law, but since the speed of adjustment is reasonably low, we will proceed by means of a comparative statics perspective, in which a comparison of different strategies can be presented, by starting from different, static, scenarios. Further, by definition, an emergency challenges the usual policy settings, since the speed of damages is greater than that of policy tools. In panel (e) of Fig. 7 a hypothetic country has a given carrying capacity to sustain the risk level represented by the iso-risk curve. Without an immediate availability of funds to increase the carrying capacity, the main policy target could easily be described as the transposition of the iso-risk curve to the bottom-left: the closer the curve to the origin, the higher the satisfaction for the community. Secondly, the meaning of the relationship between the curve and the line is that until the curve touches the line, the policy maker has a sort of measure of how much the problem is out of control, given by the distance between the curve and the constraint. Third, policies may try to transpose the curve to lower levels or, equivalently, the constraint upwards (with or without modification of the slope). A minimal result is reached if both are at least tangent, as depicted in panel (f) of Fig. 7. Whenever such a tangency condition has been reached, the highest infection rate that the given health care system can sustain has been found. Further policy actions are possible to approach a lower iso-risk curve or to save resources and/or re-allocate them differently. A policy can be considered satisfactory when any of points belonging to the arc TT’ is reached, e.g. the point L. Alternative policies are neither equivalent, nor requiring the same actions, and the policy-maker has to choose actions with reference to the actual data collected by its own Country. Points F and G, although carrying the same risk level as E, still represent out-of-control positions. Different regions of the plot have a different signaling power: at point F, the infection rate is low and, thus, very difficult to be further reduced. In such a case, for example, it would be advisable to suggest health protocols which improve people safety. On the contrary, at point G, the infection rate is so high that a limit on social interaction easily appears to be much more urgent than medical protocols. The right mix between a demand-side and a supply-side policy to adopt is a decision of political nature. A distinction can be made by saying that demand-side policies are devoted to reduce the number of newly infected people (by means of restrictions to movements, quarantine regulations, rules of conduct, etc.) and their effects are able to lower the iso-risk curves; supply-side policies are, instead, aimed at incrementing the carrying capacity of the system (by means of expenditure for the healthcare system, increments of dedicated personnel and intensive care beds, in-house medical protocols) and their effects can shift the constraint representing the carrying capacity of the system. Politics has, then, to decide when the risk is low enough or the constraint is sufficiently high. Specific calibration of the model will allow, in a forthcoming research, a detailed analysis of policy implications, by considering actual conditions and risk factors of specific districts, thus providing the policy-maker with a toolbox for normative directions. For instance, the model can be read to analyze differences in proposed actions in Lombardia and Veneto, and in other regions or countries. More

Photographs of crabs and backgrounds
We sampled crabs and backgrounds to obtain images for the game. The population used was located in Falmouth (50.141888, −5.063811) on the south coast of the UK, comprising a stretch of shoreline encompassing neighbouring Castle and Gyllyngvase beaches. The crab habitats at the site comprise rock pools with rocky crevices with stony or gravel substrates in the pools and, lower down on the shore, increasing abundance of seaweed21. Together these create visually variable textures and heterogeneity in crab habitat types.
Photographs of natural backgrounds (rock pools) were taken by Samsung NX1000 digital camera converted to full spectrum and attached with a Nikon EL 80 mm lens. Background sampling was conducted along three ~100 m long transects placed parallel to the shoreline across different tide-zones (i.e. low, middle, high) spaced evenly down the beach (following21). Each of the backgrounds photographed were at least 5 m apart from each other (i.e. transect was subdivided approximately into 5-m-intervals) ensuring the variability in background types across transect. These sampling quadrats were photographed during low-tide to avoid specular light reflecting back from the water. To obtain images that capture naturalistic colour variation, the images were taken in RAW format with manual white balance and a fixed aperture setting. For human visible photos as used here, we placed a UV and infra-red (IR) blocking filter in front of the lens, which transmits wavelengths only between 400–680 nm (Baader UV/IR Cut Filter). We have previously characterised the spectral sensitivity of our cameras39. For calibration purposes, each photograph included a grey reflectance standard, which reflects light equally at 7 and 93% between 300 and 750 nm.
Quadrats were searched for shore crabs for a period of ~5 min. We searched for crabs by raking gravel by hand, moving small boulders aside, turning seaweed over and checking crevices to ensure any crabs were unlikely to be missed. After crabs were found we transported them to laboratory facilities at the University of Exeter Penryn campus for standardised photography. During the transportation all crabs were kept on standard average grey buckets. Photographs of crabs were taken with the same camera set up as above. In the laboratory a bulb simulating D65 illuminant (Iwasaki eyeColor bulb) was used while crabs were photographed against grey standard background. We included grey standards and scale bars in the photographs. Images were then calibrated and converted to normalised reflectance images (relative to the grey standard)39,40.
Crab images were scaled into the same pixel/mm aspect ratio to show crabs against the background images in natural size with respect to the background scale. Following past work25, crab outlines were cut out from the image by custom software was designed (called ‘autocrab’) to automate the process of background subtraction. This software allowed us to step through hundreds of images, automatically loading, thresholding and flood filling background areas, saving them with an appropriate transparency channel in the correct format and resolution needed for the game. This created usable crab images for 80% of the photographs easily, with some additional cleaning up required for the rest using GIMP2 image manipulation software (https://zenodo.org/record/1101057; DOI for the source code: https://doi.org/10.5281/zenodo.1099634). The crab images were PNGs (portable network graphic) with a variable alpha level to ensure there were no jagged edges visible.
Selection of crabs
We aimed to ensure that we had an ecologically relevant range of crab phenotypes used in the game. We also sought to test how different types or ‘morphs’ of crab would affect search image formation and detection. Therefore, we used a procedure to categorise crabs into one of six categories prior the experiment. Note that, statistically crab variation may be more continuous rather than falling into true morphs, but there are a number of common crab patterns and features that frequently arise in the wild20, potentially reflecting ‘modules’ of development and pattern expression. We emphasise that our aim here was not to test specifically whether shore crabs occur in discrete morphs, but rather to capture some of the variation and common features that exist in this species in order to explore the effects of different pattern types on search image formation and whether effects differ among common categories of appearance.
Game design
The design of the experiment generally followed the approach of previous citizen science camouflage games24. Ethical approval was granted by Exeter University (ID: 2015/736). Subjects were recruited via social media and word of mouth. On loading the webpage, subjects were taken to a start screen and informed that the game was an experiment and that by playing they consented to their data being used. They were free to leave the game at any time and no personal or identifying data were collected. Subjects also asked if they had played the game before.
The game was programmed in HTML5 (including JavaScript, CSS and PHP), and was available to play on all standard internet browsers. Upon loading the game each participant was shown a series of photographs of 24 natural rock pool backgrounds (randomly sampled from 105 natural background images) with a single crab (randomly sampled from 155 natural crab images) in each image (Fig. 1). Participants were asked to detect the crab (by clicking on it) as quickly as possible, which would progress them to the next slide. If the crab was not found within 15 s the crab was highlighted with a circle for 1 s, and then the participant progressed to the next slide. During the experiment, the probability of being shown the same individual crab phenotype in the next slide was always 80% (although the crab’s position and rotation, and the background image were all randomised), meaning that subjects were likely to have runs of the same individual crab in succession, often up to 10 encounters (the median run length for each crab being ~5 encounters). This approach mimicked a situation where there is no intraspecific variation in pattern, and allowed us to test which aspects of crab/morph appearance affected search image formation and switching.
Analysis of crab appearance and camouflage
Following our previous work testing how different types of camouflage metric predict detection26, we analysed a large number of metrics linked to camouflage efficacy, these include edge disruption, colour, luminance (lightness), and pattern metrics. The metrics included crab-only appearance measures (such as the crab’s intrinsic colour, brightness, and dominant marking size), and also comparative metrics where each crab is compared to its local surroundings (within a radius of one body-length, where body length is described as the diameter of a circle which best fits the crab’s outline), and also the crab compared to the entire background image. In total there were 45 metrics, all described in Supplementary Data 1. All image analysis was performed using ImageJ v1.5041, code available on request.
Images were converted from sRGB to CIELAB colour space before measuring them given that humans were the participants used in this study. Each crab was measured by recreating its exact position and rotation on each background for image analysis.
Luminance distribution difference was measured from the CIE L channel in 100 bins following the methods described in Troscianko et al.26, effectively the sum of absolute differences between the crab’s luminance histogram and the background or surrounding’s luminance histogram. The highly variable nature of the crab’s colour and background colours mean that calculating a mean colour for the background or crab may not be appropriate because it creates intermediate colours which do not represent the scene as a whole. Therefore, a colour equivalent of the luminance distribution difference method was also developed, where pixel CIE A and B values were plotted in a two-dimensional histogram to create a proportional frequency “map”. Each axis had 200 bins ranging from −100 to 100, meaning the bins are smaller than the human colour discrimination threshold in CIE LAB space. The absolute differences in the crab’s colour map and its background or surround colour maps were used as a non-parametric method for describing background colour matching. Edge disruption was also measured following the GabRat approach described in Troscianko et al. (2017), however in addition to measuring the CIE L image, the chromatic opponent channel images (CIE A and B images) were also measured (i.e. as a measure of chromatic edge disruption). Pattern energy difference was measured by creating a series of bandpass images, filtering each crab and surround into different spatial scales, then measuring the degree of “energy” standard deviation in pixel values) at each spatial scale to create an energy spectrum. Pattern energy difference calculates the absolute sum of energy differences at each spatial scale between the crab and its background following Troscianko et al.26.
Statistics and reproducibility
Survival models were used to determine how crab capture times were affected by experimental treatments and camouflage variables. Survival models offer the ability to count crabs reaching “timeout” (where participants still could not find the crab after 15 s) as surviving up to this point (termed censored in survival models). Mixed effects survival models (coxme version 2.2–1027) were used to reflect the fact that within-session data are not independent. All statistical analyses were performed in R (version 3.4.4), with the raw data and R script available as supplementary material (“Supplementary Data 2”, and “Supplementary Data 3” respectively). We used four different models to test each of our key predictions: (i) models ranking each of the camouflage metrics in order to find the best predictor of human performance, within each camouflage strategy the best predictor was selected and used in the subsequent tests; (ii) models testing the rate of improvement in capture time for each phenotype; (iii) models comparing the capture time and appearance of each crab relative to those of the previously encountered crab; (iv) models comparing the capture time of each crab given its morph, and the morph of the previous crab (i.e. interaction between individual phenotype and overall morph). We describe each in turn here:
First, based on our metrics of camouflage, we worked out the best predictor of human performance within each of these metrics. An example of the survival model is:
coxme(Surv(cTime, hit) ~ screenScale + playedBefore + poly(crab_circular_fit_centre_x,2) + poly(crab_circular_fit_centre_y,2) + L_GabRat_sig2.0 + crab_area + (1|sessionID), data).
This model takes into account the screen resolution, whether subjects have played before, the slide number (learning within session), the screen coordinates of the crabs (crabs in the corners of the screen take longer to find), the camouflage metric (GabRat luminance edge disruption in this example), the size of the crab (bigger crabs are easier to find), and session ID as a random factor. From these models we could calculate the metrics that were most effective in predicting detection times26, and narrowed the metrics down to the best predictors of luminance, colour, pattern and edge disruption.
Second, we tested how the number of previous encounters with the current crab phenotype affected capture times. This is testing for speed-of-improvement within each phenotype, and how different types of camouflage (determined above) affect this. An example survival model is:
coxme(Surv(cTime, hit) ~ screenScale + playedBefore + slide + poly(crab_circular_fit_centre_x,2) + poly(crab_circular_fit_centre_y,2) + L_GabRat_sig2.0 * encounters + crab_area + (1|sessionID), data). Where ‘encounters’ codes for the number of previous encounters with the current phenotype.
Third, we tested capture time differences when switching between crabs, comparing the camouflage of the previous crab with the current one (note the previously encountered crab was sometimes the same phenotype, and sometimes would switch to a new one). The dependent variable (timeDiff) was log(current crab capture time) – log(previous crab capture time). The camouflage variables are calculated in the same manner, e.g. the current level of disruption minus the previous level of disruption. Here, an interaction with the number of prior encounters with the current crab phenotype shows how switching is affected by prior experience of this camouflage type. An example model is:
lmer(timeDiff ~ crab_area + pArea + playedBefore + slide + poly(crab_circular_fit_centre_x,2) + poly(crab_circular_fit_centre_y,2) + poly(pX,2) + poly(pY,2) + drpLDiff*novelCrab + (1|sessionID), diffData). The values pArea, pX and pY denote the size and screen location of the previous crab.
Finally, we analysed capture time differences when switching between each of the six crab morphs (rather than comparing camouflage metric differences), using the timeDiff value as above. An example model is:
lmer(timeDiff ~ crab_area + pArea + slide + poly(crab_circular_fit_centre_x,2) + poly(crab_circular_fit_centre_y,2) + poly(pX,2) + poly(pY,2) + slide + morphSwitch*novelCrab + (1|sessionID), morphData). Here ‘morphSwitch’ has two levels which describe whether a switch event was to the same, or a different morph. The random factor ‘sessionID’ explained almost zero variance in this dataset, and where this occurred the models were cross-validated with GLMs (see Supplementary Data 3).
Selection of crab phenotypes
We asked 10 naïve participants (who had no prior experience of crab phenotype discrimination) to subjectively sort images of crabs into distinct categories. People were not instructed on how many groups they should form – they were simply asked to group crabs based on their colour and patterning (i.e. phenotypic variation). This resulted in six categories (the actual numbers of the crab images representing that phenotype are given in brackets as follows): Black (22), Disruptive (15), Green (50), Mottled (28), Pale (20) and Spotted (20). Although this is subjective, we subsequently analysed the appearance of crabs from these categories and showed that ‘crab morph’ is a significant predictor of a range of appearance metrics, including colour, luminance, mean pattern energy, and dominant marking size (P  More

1.
d’Errico, F. & Banks, W. E. Identifying mechanisms behind Middle Paleolithic and Middle Stone Age cultural trajectories. Curr. Anthropol. 54, S371–S387 (2013).
Article  Google Scholar 
2.
Richerson, P. J., Bettinger, R. L. & Boyd, R. Evolution on a restless planet: Were environmental variability and environmental change major drivers of human evolution? In Handbook of Evolution: The Evolution of Living Systems (Including Hominids) Vol. 2 (eds Wuketits, F. M. & Ayala, F. J.) 223–242 (Wiley-VCH, New York, 2005).
Google Scholar 

3.
Pedersen, J., Maier, A. & Riede, F. A punctuated model for the colonisation of the Late Glacial margins of northern Europe by Hamburgian hunter-gatherers. Quartär 65, 85–104 (2018).
Google Scholar 

4.
Riede, F. & Pedersen, J. B. Late glacial human dispersals in Northern Europe and disequilibrium dynamics. Hum. Ecol. 46, 621–632 (2018).
Article  Google Scholar 

5.
Langley, M. C., Clarkson, C. & Ulm, S. Behavioural complexity in Eurasian Neanderthal Populations: A chronological examination of the archaeological evidence. Camb. Archaeol. J. 18, 289–307 (2008).
Article  Google Scholar 

6.
Roebroeks, W. & Soressi, M. Neandertals revised. Proc. Natl. Acad. Sci. USA 113, 6372–6379 (2016).
CAS  PubMed  Article  Google Scholar 

7.
Zilhão, J. et al. Last Interglacial Iberian Neandertals as fisher-hunter-gatherers. Science 367, 6485 (2020).
Article  CAS  Google Scholar 

8.
Benito, B. M. et al. The ecological niche and distribution of Neanderthals during the Last Interglacial. J. Biogeogr. 44, 51–61 (2017).
Article  Google Scholar 

9.
Nielsen, T. K. et al. Investigating Neanderthal dispersal above 55°N in Europe during the Last Interglacial Complex. Quat. Int. 431, 88–103 (2017).
Article  Google Scholar 

10.
Bocquet-Appel, J.-P. & Tuffreau, A. Technological responses of neanderthals to macroclimatic variations (240,000–40,000 BP). Hum. Biol. 81, 287–307 (2009).
PubMed  Article  Google Scholar 

11.
Daujeard, C. et al. Neanderthal subsistence strategies in Southeastern France between the plains of the Rhone Valley and the mid-mountains of the Massif Central (MIS 7 to MIS 3). Quat. Int. 252, 32–47 (2012).
Article  Google Scholar 

12.
Discamps, E., Jaubert, J. & Bachellerie, F. Human choices and environmental constraints: deciphering the variability of large game procurement from Mousterian to Aurignacian times (MIS 5–3) in southwestern France. Quat. Sci. Rev. 30, 2755–2775 (2011).
Article  ADS  Google Scholar 

13.
Hublin, J. J. The origin of Neandertals. Proc. Natl. Acad. Sci. USA 106, 16022–16027 (2009).
CAS  PubMed  Article  ADS  Google Scholar 

14.
Rogers, A. R., Bohlender, R. J. & Huff, C. D. Early history of Neanderthals and Denisovans. Proc. Natl. Acad. Sci. USA 114, 9859–9863 (2017).
CAS  PubMed  Article  Google Scholar 

15.
Moncel, M.-H. et al. Early Levallois core technology between Marine Isotope Stage 12 and 9 in Western Europe. J. Hum. Evol. 139, 102735 (2020).
PubMed  Article  Google Scholar 

16.
Castellano, S. et al. Patterns of coding variation in the complete exomes of three Neandertals. Proc. Natl. Acad. Sci. USA 111, 6666–6671 (2014).
CAS  PubMed  Article  ADS  Google Scholar 

17.
Mafessoni, F. & Prüfer, K. Better support for a small effective population size of Neandertals and a long shared history of Neandertals and Denisovans. Proc. Natl. Acad. Sci. USA 114, E10256–E10257 (2017).
CAS  PubMed  Article  Google Scholar 

18.
Prüfer, K. et al. A high-coverage Neandertal genome from Vindija Cave in Croatia. Science 358, 655–658 (2017).
PubMed  PubMed Central  Article  ADS  CAS  Google Scholar 

19.
Moncel, M.-H., Fernandes, P., Willmes, M., James, H. & Grün, R. Rocks, teeth, and tools: New insights into early Neanderthal mobility strategies in South-Eastern France from lithic reconstructions and strontium isotope analysis. PLoS ONE 14, e0214925 (2019).
CAS  PubMed  PubMed Central  Article  Google Scholar 

20.
Peterson, A. T. et al. Ecological Niches and Geographic Distributions (Princeton University Press, Princeton, 2011).
Google Scholar 

21.
Rogers, A. R., Bohlender, R. J. & Huff, C. D. Reply to Mafessoni and Prüfer: Inferences with and without singleton site patterns. Proc. Natl. Acad. Sci. USA 114, E10258–E10260 (2017).
CAS  PubMed  Article  Google Scholar 

22.
Vaissié, E. et al. Techno-économie et signification culturelle de l’occupation moustérienne supérieure de Baume-Vallée (Haute-Loire). C.R. Palevol 16, 804–819 (2017).
Article  Google Scholar 

23.
Bocquet-Appel, J.-P., Demars, P.-Y., Noiret, L. & Dobrowsky, D. Estimates of Upper Palaeolithic meta-population size in Europe from archaeological data. J. Archaeol. Sci. 32, 1656–1668 (2005).
Article  Google Scholar 

24.
Delagnes, A., Jaubert, J. & Meignen, L. Les technocomplexes du Paléolithique moyen en Europe occidentale dans leur cadre diachronique et géographique. In Les Néandertaliens Biologie et cultures (eds Vandermeersch, B. & Maureille, B.) 213–229 (Editions du Comité des Travaux Historiques et Scientifiques, Aubervilliers, 2007).
Google Scholar 

25.
Faivre, J.-P., Gravina, B., Bourguignon, L., Discamps, E. & Turq, A. Late Middle Palaeolithic lithic technocomplexes (MIS 5–3) in the northeastern Aquitaine Basin: Advances and challenges. Quat. Int. 433, 116–131 (2017).
Article  Google Scholar 

26.
Jaubert, J., Bordes, J.-G., Discamps, E. & Gravina, B. A new look at the end of the Middle Palaeolithic Sequence in Southwestern France. In Characteristic Features of the Middle to Upper Paleolithic transition in Eurasia (eds Derevianko, A. P. & Shunkov, M. V.) 102–115 (Asian Palaeolithic Association, Tokyo, 2011).
Google Scholar 

27.
Boëda, E. Levallois: A volumetric construction, methods, A technique. In The Definition and Intrepretation of Levallois Technology (eds Dibble, H. L. & Bar-Yosef, O.) 41–68 (Prehistory Press, Madison, 1995).
Google Scholar 

28.
Boëda, E. L. débitage discoïde et le débitage Levallois récurrent centripède. Bull. Soc. Préhist. Fr. 90, 392–404 (1993).
Article  Google Scholar 

29.
Bourguignon, L. Le Moustérien de type Quina: Nouvelles définitions d’une entité technique (University of Paris 10, Paris, 1997).
Google Scholar 

30.
Turq, A. L. Moustérien de type Quina. Paléo Rev. Archéol. Préhist. 2, 310–343 (2000).
Google Scholar 

31.
Turq, A. Approche technologique et économique du faciès Moustérien de type Quina: Étude préliminaire. Bull. Soc. Préhist. Fr. 86, 244–256 (1989).
Article  Google Scholar 

32.
Collard, M., Vaesen, K., Cosgrove, R. & Roebroeks, W. The empirical case against the ‘demographic turn’ in Palaeolithic archaeology. Philos. Trans. R. Soc. B 371, 20150242 (2016).
Article  Google Scholar 

33.
Soberón, J. & Nakamura, M. Niches and distributional areas: Concepts, methods, and assumptions. Proc. Natl. Acad. Sci. USA 106, 19644–19650 (2009).
PubMed  Article  ADS  Google Scholar 

34.
Cobos, M. E., Peterson, A. T., Barve, N. & Osorio-Olvera, L. kuenm: An R package for detailed development of ecological niche models using Maxent. PeerJ 7, e6281 (2019).
PubMed  PubMed Central  Article  Google Scholar 

35.
Cobos, M. E., Osorio-Olvera, L., Soberón, J. & Peterson, A. T. ellipsenm: An R package for ecological niche’s characterization using ellipsoids. (2020).

36.
Waelbroeck, C. et al. Sea-level and deep water temperature changes derived from benthic foraminifera isotopic records. Quat. Sci. Rev. 21, 295–305 (2002).
Article  ADS  Google Scholar 

37.
Peterson, A. T., Papeş, M. & Soberón, J. Rethinking receiver operating characteristic analysis applications in ecological niche modeling. Ecol. Model. 213, 63–72 (2008).
Article  Google Scholar 

38.
Antoine, P. et al. Paléoenvironnements pléistocènes et peuplements paléolithiques dans le bassin de la Somme (nord de la France). Bull. Soc. Préhist. Fr. 100, 5–28 (2003).
Article  Google Scholar 

39.
Locht, J.-L. et al. Timescales, space and culture during the Middle Palaeolithic in northwestern France. Quat. Int. 411, 129–148 (2016).
Article  Google Scholar 

40.
Raynal, J.-P. et al. Land-use strategies, related tool-kits and social organization of lower and middle Palaeolithic groups in the South-East of the Massif Central, France. Quartär 60, 29–59 (2013).
Google Scholar 

41.
Turq, A., Faivre, J.-P., Gravina, B. & Bourguignon, L. Building models of Neanderthal territories from raw material transports in the Aquitaine Basin (southwestern France). Quat. Int. 433, 88–101 (2017).
Article  Google Scholar 

42.
Mathias, C., Bourguignon, L., Brenet, M., Grégoire, S. & Moncel, M.-H. Between new and inherited technical behaviours: A case study from the Early Middle Palaeolithic of Southern France. Archaeol. Anthropol. Sci. 12, 1–39 (2020).
Article  Google Scholar 

43.
Lebegue, F. & Meignen, L. Quina ou pas ? Révision techno-économique d’un site moustérien charentien en Languedoc oriental: La grotte de la Roquette à Conqueyrac (Gard, France). Bull. Soc. Préhist. Fr. 111, 603–630 (2014).
Article  Google Scholar 

44.
Moncel, M.-H. et al. La grotte du Figuier (Saint-Martin-d’Ardèche): Bilan des travaux récents sur un site du Paléolithique moyen et supérieur de la moyenne vallée du Rhône (Sud-Est de la France). Bull. Soc. Préhist. Fr. 109, 35–67 (2012).
Article  Google Scholar 

45.
Slimak, L. Moustériens Quina Rhodaniens et Quina classiques dans le sud-est de la France. In Territoires, Déplacements, Mobilité, Echanges durant la Préhistoire (eds Jaubert, J. & Barbaza, M.) 95–113 (Comité des travaux historiques et scientifiques, Aubervilliers, 2005).
Google Scholar 

46.
Sánchez Goñi, M. F., Bard, E., Landais, A., Rossignol, L. & d’Errico, F. Air–sea temperature decoupling in western Europe during the last interglacial–glacial transition. Nat. Geosci. 6, 837–841 (2013).
Article  ADS  CAS  Google Scholar 

47.
Antoine, P., Munaut, A.-V. & Sommé, J. Réponse des environnements aux climats du début glaciaire weichsélien: Données de la France du Nord-Ouest [Responses of the environments to Early Weichselian climates. Records in north­western France]. Quaternaire 5, 151–156 (1994).
Article  Google Scholar 

48.
Fletcher, W. J. et al. Millennial-scale variability during the last glacial in vegetation records from Europe. Quat. Sci. Rev. 29, 2839–2864 (2010).
Article  ADS  Google Scholar 

49.
Baena, J., Moncel, M.-H., Cuartero, F., Chacón Navarro, M. G. & Rubio, D. Late Middle Pleistocene genesis of Neanderthal technology in Western Europe: The case of Payre site (south-east France). Quat. Int. 436, 212–238 (2017).
Article  Google Scholar 

50.
Geneste, J.-M., Jaubert, J., Lenoir, M., Meignen, L. & Turq, A. Approche technologique des Moustériens Charentiens du Sud-Ouest de la France et du Languedoc oriental. Paléo Rev. Archéol. Préhist. 9, 101–142 (1997).
Google Scholar 

51.
Geneste, J.-M. & Plisson, H. Production et utilisation de l’outillage lithique dans le Moustérien du sud-ouest de la France: les Tares à Sourzac, Vallé de l’Isle, Dordogne. Quat. Nova 6, 343–367 (1996).
Google Scholar 

52.
Mathias, C. & Bourguignon, L. Cores-on-flakes and ramification during the middle palaeolithic in Southern France: A gradual process from the early to late middle palaeolithic?. J. Archaeol. Sci. Rep. 31, 102336 (2020).
Google Scholar 

53.
Halstead, P. & O’Shea, J. Introduction: Cultural responses to risk and uncertainty. In Bad Year Economics: Cultural Responses to Risk and Uncertainty (eds Halstead, P. & O’Shea, J.) 1–7 (Cambridge University Press, Cambridge, 1989).
Google Scholar 

54.
d’Errico, F. et al. Identifying early modern human ecological niche expansions and associated cultural dynamics in the South African Middle Stone Age. Proc. Natl. Acad. Sci. USA 114, 7869–7876 (2017).
PubMed  Article  CAS  Google Scholar 

55.
Delagnes, A. & Meignen, L. Diversity of lithic production systems during the Middle Paleolithic in France. In Transitions Before the Transition: Evolution and Stability in the Middle Paleolithic and Middle Stone Age (eds Hovers, E. & Kuhn, S. L.) 85–107 (Springer Verlag, New York, 2006).
Google Scholar 

56.
Hiscock, P., Turq, A., Faivre, J.-P. & Bourguignon, L. Quina procurement and tool production. In Lithic Materials and Paleolithic Societies (eds Adams, B. & Blades, B. S.) 232–246 (Wiley-Blackwell, New York, 2009).
Google Scholar 

57.
Binford, L. R. Willow smoke and dogs’ tails: Hunter-gatherer settlement systems and archaeological site formation. Am. Antiq. 45, 4–20 (1980).
Article  Google Scholar 

58.
Dibble, H. L. et al. Context, curation, and bias: An evaluation of the Middle Paleolithic collections of Combe-Grenal (France). J. Archaeol. Sci. 36, 2540–2550 (2009).
Article  Google Scholar 

59.
R Core Team. R: A Language and Environment for STATISTICAL Computing (R Foundation for Statistical Computing, Vienna, 2019).
Google Scholar 

60.
Lê, S., Josse, J. & Husson, F. FactoMineR: An R package for multivariate analysis. J. Stat. Softw. 25, 1–18 (2008).
Article  Google Scholar 

61.
Sarkar, D. Lattice: Multivariate Data Visualization with R (Springer, Berlin, 2008).
Google Scholar 

62.
Fernandes, P., Raynal, J.-P. & Moncel, M.-H. Middle Palaeolithic raw material gathering territories and human mobility in the southern Massif Central, France: first results from a petro-archaeological study on flint. J. Archaeol. Sci. 35, 2357–2370 (2008).
Article  Google Scholar 

63.
Dufresne, J.-L. et al. Climate change projections using the IPSL-CM5 Earth System Model: from CMIP3 to CMIP5. Clim. Dyn. 40, 2123–2165 (2013).
Article  Google Scholar 

64.
Argus, D. F. & Peltier, W. R. Constraining models of postglacial rebound using space geodesy: A detailed assessment of model ICE-5G (VM2) and its relatives. Geophys. J. Int. 181, 697–723 (2010).
ADS  Google Scholar 

65.
Petit, J. R. et al. Climate and atmospheric history of the past 420,000 years from the Vostok ice core, Antarctica. Nature 399, 429–436 (1999).
CAS  Article  ADS  Google Scholar 

66.
Laskar, J. et al. A long-term numerical solution for the insolation quantities of the Earth. Astron. Astrophys. 428, 261–285 (2004).
Article  ADS  Google Scholar 

67.
Vrac, M., Marbaix, P., Paillard, D. & Naveau, P. Non-linear statistical downscaling of present and LGM precipitation and temperatures over Europe. Clim. Past 3, 669–682 (2007).
Article  Google Scholar 

68.
Pfeiffer, M., Spessa, A. & Kaplan, J. O. A model for global biomass burning in preindustrial time: LPJ-LMfire (v1.0). Geosci. Model Dev. 6, 643–685 (2013).
Article  ADS  CAS  Google Scholar 

69.
Phillips, S. J., Anderson, R. P., Dudík, M., Schapire, R. E. & Blair, M. E. Opening the black box: An open-source release of Maxent. Ecography 40, 887–893 (2017).
Article  Google Scholar 

70.
Cobos, M. E., Peterson, A. T., Osorio-Olvera, L. & Jiménez-García, D. An exhaustive analysis of heuristic methods for variable selection in ecological niche modeling and species distribution modeling. Ecol. Inform. 53, 100983 (2019).
Article  Google Scholar 

71.
Anderson, R. P., Lew, D. & Peterson, A. T. Evaluating predictive models of species’ distributions: Criteria for selecting optimal models. Ecol. Model. 162, 211–232 (2003).
Article  Google Scholar 

72.
Warren, D. L. & Seifert, S. N. Ecological niche modeling in Maxent: the importance of model complexity and the performance of model selection criteria. Ecol. Appl. 21, 335–342 (2011).
PubMed  Article  Google Scholar 

73.
Owens, H. L. et al. Constraints on interpretation of ecological niche models by limited environmental ranges on calibration areas. Ecol. Model. 263, 10–18 (2013).
Article  Google Scholar 

74.
Nuñez-Penichet, C., Cobos, M. E. & Soberon, J. Non-overlapping climatic niches and biogeographic barriers explain disjunct distributions of continental Urania moths. Front. Biogeogr. 13(2), e52142 (2021).
Google Scholar 

75.
Qiao, H. et al. NicheA: Creating virtual species and ecological niches in multivariate environmental scenarios. Ecography 39, 805–813 (2016).
Article  Google Scholar 

76.
Mammola, S. Assessing similarity of n-dimensional hypervolumes: Which metric to use?. J. Biogeogr. 46, 2012–2023 (2019).
Article  Google Scholar 

77.
Van Aelst, S. & Rousseeuw, P. Minimum volume ellipsoid. WIREs. Comput. Stat. 1, 71–82 (2009).
Article  Google Scholar 

78.
Murdoch, D. J. & Chow, E. D. A graphical display of large correlation matrices. Am. Stat. 50, 178–180 (1996).
Google Scholar  More

Sampling stations
Baywatchers sampling stations were generally concentrated in the upper half of estuaries and major sub-estuaries to better characterize water quality changes over time. The program has grown over time and expanded to include additional stations within sub-estuaries as well as adding additional sub-estuaries. In 2012, stations were added in Vineyard Sound, which adjoins Buzzards Bay, and in 2017, sampling began at additional stations in the coastal ponds connected to Vineyard Sound.
We assign each station a unique identification code (station ID) and data are integrated into geographic information systems. Station maps are given to monitors (historically using ArcView GIS overlaid on scanned U.S.G.S. quadrangle maps, but more recently, using Google Earth to produce the maps overlaid on aerial images). In a small number of cases, the location of a monitoring site has varied slightly over time—for example, nutrient samples were collected in the Agawam River in a rowboat 250 feet from shore from 1998 to 2007 (station AG2A), but have been collected from a nearby dock since 2008 (station AG2). When a monitoring site’s location has moved, it is given a unique station ID in the database (i.e., AG2 vs AG2A in the example above).
Field sampling
Sampling occurs from late May to September to document conditions when biological activity is highest. Field water sampling is separated into “basic” sampling and “laboratory” sampling days. On all sampling dates, water temperature, salinity, Secchi depth, and total depth are measured in the field. Monitors record these results on hard copy datasheets along with the station ID, sampling date, collection time, name of person sampling, and name of the sub-estuary.
On basic sampling days, D.O. is measured in the early morning (between 6:00 and 9:00 am) to capture typical daily minimum oxygen concentrations before peak daytime photosynthetic oxygen production. Basic sampling occurs on a schedule roughly every five days between late May and mid-September. On laboratory sampling days, oxygen measurements are only made if the monitor has a water quality sonde, as the focus is the collection of samples for laboratory analysis of NH4+, NO3− + NO2−, PO43−, TDN, PON, POC, Chl a, and Pheo. At designated fresh water and low salinity stations, TP and DOC are also measured. Laboratory sampling occurs on four scheduled days each summer (2 in July, 2 in August) during the last three hours of an outgoing tide when concentrations of solutes in estuarine water are expected to most strongly reflect the influence of watershed inputs. While the vast majority of observations were made between late May and mid-September (Fig. 3), some additional basic and laboratory sampling occurred at other times of the year when short-term projects provided opportunities for expanded sampling.
Fig. 3

Frequency of Baywatchers samples collected by month between 1992 and 2018. Note difference in y-axis scale for panel c.

Full size image

The Buzzards Bay Coalition pairs some nearby basic sampling stations with laboratory sampling stations for analysis of sub-estuary water quality. These station pairs have station IDs that end in either the suffix X or N to indicate that they are sampled on basic or laboratory sampling days, respectively.
Basic sampling procedures
Water samples are collected for water temperature, salinity, and D.O. from near the bottom of the water column (0.3 m above the bottom). Where the water column is deeper than 1.2 m, a sample near the surface (0.15 m depth) is also collected to provide information on potential water column stratification. The depth of 0.15 m below the surface prevents entrainment of floating particles and overlying air into the sample bottles. Sampling 0.3 m above the bottom prevents resuspension and capture of bottom sediments by the sampling apparatus.
Water samples for temperature, salinity, and D.O. are collected either with a steel sampling pole or measured in situ with water quality sondes (YSI models 600XL, 600XLM, 6600, EXO2, ProDSS). Sampling poles are 1.5 or 3 m long and marked in 5 to 10 cm depth increments. Sampling poles have 1 L and 0.5 L plastic (HDPE) bottles with rubber stopper closures connected to strings. Poles are lowered to the appropriate depth and then bottles are opened by pulling the strings, first the 0.5 L bottle is opened, followed by the 1 L bottle. D.O. is measured from the 0.5 L bottle, so it is opened first to prevent entrainment of air bubbles into the D.O. sample. Temperature and salinity are measured from the 1 L bottle. Water temperature is measured directly in the 1 L bottle using a thermometer that is calibrated annually. Monitors have primarily used analog thermometers; however, some digital thermometers have been used since 2016. Salinity is measured by then transferring 0.5 L of sample from the 1 L bottle to a 0.5 L graduated cylinder. Specific gravity is measured using a hydrometer that is calibrated annually. Temperature is measured in the graduated cylinder and salinity is determined from a table of specific gravity and temperature.
The 0.5 L bottle has been modified with plastic fittings and tubing at the bottom so that water can be extracted from the bottom of the bottle. Water is siphoned through the tubing to the bottom of a glass-stoppered bottle, overflowing the glass bottle until the 0.5 L bottle is only one-quarter full. Monitors measure D.O. in the glass bottle directly in the field using a modified Winkler titration (Hach Test Kit, Model OX-2P). Briefly, pre-weighed aliquots of manganese sulphate and a lithium hydroxide monohydrate/potassium iodide mixture are added to the sample bottle, which is stoppered and vigorously shaken. The resultant floc is allowed to settle, then reshaken and settled again, before the addition of a pre-weighed aliquot of a sodium phosphate dibasic/sodium sulphate/citric acid mixture. The sample is shaken until this dissolves and the sample is clear, then an aliquot of sample is measured into a separate vial. The sample aliquot is titrated drop-wise using a sodium thiosulphate standard until the sample becomes colorless. The D.O. concentration (mg L−1) is calculated from the number of titration drops.
Secchi depth is recorded by lowering a Secchi disk into water slowly from the shady side of a boat, dock or pier until it just disappears from view. It is then raised and lowered slightly to ensure the proper average depth of disappearance. If the Secchi disk hits the bottom before it disappears, no Secchi depth value is recorded. Total depth is determined when slack is felt in the measuring tape of the Secchi disk.
On basic sampling days, monitors also record the tidal direction (ebb or flood) and the time of the nearest low tide according to the Eldridge Tide and Pilot Book, wave conditions according to the Beaufort scale, and weather status (based on eight potential choices: cloudless, partly cloudy, overcast, fog/haze, drizzle, intermittent rain, rain, snow). Precipitation in the previous 24 hours is noted as either none, light, or heavy. Wind direction is also recorded. All data are recorded on a paper data sheet.
Approximately 20,000 of the temperature, salinity, and D.O. and roughly 5,000 of the pH and Chl a measurements were made in situ using water quality sondes. The majority of these measurements were made since 2000 (YSI models 600XL, 600XLM, 6600, EXO2, ProDSS), though a few were made in the early years of the program (YSI model 51B). The sonde measurements are made following the manufacturers’ standard operating procedures. Instruments are calibrated for temperature, salinity, D.O., pH and Chl a at the beginning of each sampling season. The instrument D.O. calibration is checked prior to each sampling day and re-calibrated in the field if necessary.
Laboratory sampling procedures
On laboratory sampling days, water samples for analysis of dissolved and particulate constituents are primarily collected from near the surface (0.15 m), though a small portion have been collected from near the bottom (0.3 m above bottom). Bottom water samples were generally collected near where there is significant freshwater input that could cause water column stratification. Station ID, water temperature, salinity, Secchi depth, total depth, sample depth, and collection time are recorded in the field on hard copy data sheets. Monitors with sondes also record D.O.
Samples are collected directly into 1 L acid-washed plastic HDPE bottles either by hand or using the sampling pole. All bottles used in water collection were acid washed by the analytical laboratory. Samples bottles are rinsed once with sample prior to filling with the sample. When using the sampling pole, sample bottles are attached and removed using hose clamps.
Monitors filter 60 mL of sample from the 1 L bottle using a 0.2 μm cellulose acetate membrane filter. Filters are first rinsed with 30 mL of sample, which is discarded. A subsequent 30 mL of filtered sample is used to rinse the 60 mL sampling bottle before an additional 60 mL are filtered directly into the 60 mL bottle that was previously acid washed by the analytical laboratory. The remaining unfiltered sample and the 60 mL filtered sample are stored in coolers with ice packs and delivered on the day they are collected to the analytical laboratory. A new membrane filter is used for each sample and the filter holder and syringes are rinsed with tap water after a sample is filtered.
Laboratory analyses
Laboratory analyses were conducted under the supervision of B. Howes at the Woods Hole Oceanographic Institution (1992–1997) and at the University of Massachusetts, Dartmouth (1998–2007), and under H. Ducklow (2008–2012) and C. Neill (2013–2018) at the Marine Biological Laboratory. The methods and instruments described below are those currently used. In some cases, the instruments used have changed, but there has been a significant effort to maintain consistency over the lifetime of the program and to intercalibrate methods/instruments when a change has been made.
Laboratory analyses are designed to accommodate the samples that range from fresh water to nearly full strength seawater and analyte concentrations that range from at or below the detection limit for a method to 1,000 times the detection limit in some cases. Laboratory staff used aliquots of the 60 mL field-filtered sample to perform the dissolved analyses (NH4+, NO3− + NO2−, PO43−, TDN). The filtration for the particulate analyses (Chl a, Pheo, PON, POC) was performed by laboratory staff using water from the 1 L dark sample bottle.
NH4+ is measured colorimetrically by the indophenol-hypochlorite method18. Analyses are conducted in pre-reacted test tubes to reduce blank corrections and absorbance is read on a Cary spectrophotometer with an automatic sipper attachment. NH4+ is analysed on the day samples are collected.
NO3− + NO2− is measured colorimetrically after cadmium reduction19 on a Lachat flow injection analyser (Hach, Loveland, CO). The method was modified for flow injection analysis by Lachat Instruments20. Prior to analysis, samples are refrigerated at 4 °C. Samples are analysed within one week of collection if collected July-August, or if collected outside of July-August, frozen and analysed within 90 days.
PO43− is measured colorimetrically by the molybdenum blue method21 on a Lachat flow injection analyser (Hach, Loveland, CO). The method was modified for flow injection analysis by Lachat Instruments22. Samples are analysed within one week of collection if collected July-August, or if collected outside of July-August, frozen and analysed within 90 days.
TDN is analysed by persulphate digestion23 that oxidizes dissolved nitrogen to NO3− and subsequent analysis of NO3− by colorimetry on a Lachat flow injection analyser. To reduce the magnitude of the reagent blank, persulphate is recrystallized prior to analysis. Samples are stored at 4 °C prior to analysis and are analysed within two weeks of collection if collected July-August, or collected outside of July-August, frozen and analysed within 90 days.
Chl a and Pheo are measured with method of Arar et al.24 by filtering a known volume of water through a 25 mm glass fiber filter (GFF). On the day of sample collection, samples are filtered under low vacuum pressure ( More

1.
Serre, D. & Paabo, S. Evidence for gradients of human genetic diversity within and among continents. Genome Res. 14, 1679–1685 (2004).
CAS  PubMed  PubMed Central  Article  Google Scholar 
2.
Ohta, T. Population size and rate of evolution. J. Mol. Evol. 1, 305–314 (1972).
PubMed  Article  PubMed Central  Google Scholar 

3.
Slatkin, M. Gene flow and selection in a cline. Genetics 75, 733–756 (1973).

4.
Potts, R. Variability selection in hominid evolution. Evol. Anthropol. 7, 81–96 (1998).
Article  Google Scholar 

5.
Potts, R. Hominin evolution in settings of strong environmental variability. Quat. Sci. Rev. 73, 1–13 (2013).
Article  Google Scholar 

6.
Prüfer, K. et al. The bonobo genome compared with the chimpanzee and human genomes. Nature 486, 527–531 (2012).
PubMed  PubMed Central  Article  CAS  Google Scholar 

7.
Hill, W. C. O. in The Nomenclature, Taxonomy and Distribution of Chimpanzees, Vol. 1 (ed. Bourne, G. H.) 22–49 (Karger, 1969).

8.
Pruetz, J. D. & Bertolani, P. Chimpanzee (Pan troglodytes verus) behavioral responses to stresses associated with living in a savannah-mosaic environment: implications for hominin adaptations to open habitats. Paleoanthropology 2009, 252–262 (2009).
Article  Google Scholar 

9.
Fünfstück, T. et al. The sampling scheme matters: Pan troglodytes troglodytes and P. t. schweinfurthii are characterized by clinal genetic variation rather than a strong subspecies break. Am. J. Phys. Anthropol. 156, 181–191 (2014).
PubMed  PubMed Central  Article  Google Scholar 

10.
Fischer, A., Pollack, J., Thalmann, O., Nickel, B. & Paabo, S. Demographic history and genetic differentiation in apes. Curr. Biol. 16, 1133–1138 (2006).
CAS  PubMed  Article  Google Scholar 

11.
Gonder, M. K. et al. A new west African chimpanzee subspecies? Nature 388, 337 (1997).
CAS  PubMed  Article  Google Scholar 

12.
Becquet, C., Patterson, N., Stone, A. C., Przeworski, M. & Reich, D. E. Genetic structure of chimpanzee populations. PLoS Genet. 3, e66 (2007).
PubMed  PubMed Central  Article  CAS  Google Scholar 

13.
Bowden, R. et al. Genomic tools for evolution and conservation in the chimpanzee: Pan troglodytes ellioti is a genetically distinct population. PLoS Genet. 8, e1002504 (2012).
CAS  PubMed  PubMed Central  Article  Google Scholar 

14.
Prado-Martinez, J. et al. Great ape genetic diversity and population history. Nature 499, 471–475 (2013).
CAS  PubMed  PubMed Central  Article  Google Scholar 

15.
de Manuel, M. et al. Chimpanzee genomic diversity reveals ancient admixture with bonobos. Science 354, 477–481 (2016).
PubMed  PubMed Central  Article  CAS  Google Scholar 

16.
Langergraber, K. E. et al. Genetic and ‘cultural’ similarity in wild chimpanzees. Proc. R. Soc. B. 278, 408–416 (2010).
PubMed  Article  PubMed Central  Google Scholar 

17.
Garcin, Y. et al. Early anthropogenic impact on Western Central African rainforests 2,600 y ago. Proc. Natl Acad. Sci. USA 115, 3261–3266 (2018).
CAS  PubMed  Article  PubMed Central  Google Scholar 

18.
Vicente, M. & Schlebusch, C. M. African population history: an ancient DNA perspective. Curr. Opin. Genet. Dev. 62, 8–15 (2020).
CAS  PubMed  Article  PubMed Central  Google Scholar 

19.
IUCN Red List. IUCN red list of threatened species. http://www.iucnredlist.org (2020).

20.
Wentworth, C. K. Natural bridges and glaciation. Am. J. Sci. 26, 577–584 (1933).
Article  Google Scholar 

21.
Maley, J. The African rainforest: main characteristics of changes in vegetation and climate from the Upper Cretaceous to the Quaternary. Proc. R. Soc. Edinb. 104B, 31–73 (1996).
Google Scholar 

22.
Langergraber, K. E. et al. Generation times in wild chimpanzees and gorillas suggest earlier divergence times in great ape and human evolution. Proc. Natl Acad. Sci. USA 109, 15716–15721 (2012).
CAS  PubMed  Article  Google Scholar 

23.
Shea, B. T. & Coolidge, H. J. Craniometric differentiation and systematics in the genus Pan. J. Hum. Evol. 13, 671–685 (1988).
Article  Google Scholar 

24.
Albrecht, G. H. & Miller, J. M. A. in Geographic Variation in Primates (eds Kimbel, W. H. & Martin, L. B.) 123–161 (Springer Science & Business Media, 2013).

25.
Shea, B. T., Leigh, S. R. & Groves, C. P. in Multivariate Craniometric Variation in Chimpanzees (eds Kimbel, W. H. & Martin, L. B.) 265–296 (Springer Science & Business Media, 2013).

26.
Kühl, H. S. et al. Chimpanzee accumulative stonethrowing. Sci. Rep. 6, 22219 (2016).

27.
Wright, S. Isolation by distance. Genetics 28, 114–138 (1943).
CAS  PubMed  PubMed Central  Google Scholar 

28.
Pritchard, J. K., Stephens, M. & Donnelly, P. Inference of population structure using multilocus genotype data. Genetics 155, 945–959 (2000).
CAS  PubMed  PubMed Central  Google Scholar 

29.
Pritchard, J. K., Wen, X. & Falush, D. Documentation for structure version 2.3 software: version 2.3. 1–39. http://pritchardlab.stanford.edu/structure_software/release_versions/v2.3.4/structure_doc.pdf (2009).

30.
Meirmans, P. G. The trouble with isolation by distance. Mol. Ecol. 21, 2839–2846 (2012).
PubMed  Article  Google Scholar 

31.
Perez, M. F. et al. Assessing population structure in the face of isolation by distance: are we neglecting the problem? Divers. Distrib. 24, 1883–1889 (2018).
Article  Google Scholar 

32.
Thalib, L., Kitching, R. L. & Bhatti, M. I. Principal component analysis for grouped data – a case study. Environmetrics 10, 565–574 (1999).
Article  Google Scholar 

33.
Fischer, A. et al. Bonobos fall within the genomic variation of chimpanzees. PLoS ONE 6, e21605 (2011).
CAS  PubMed  PubMed Central  Article  Google Scholar 

34.
Frantz, A. C., Cellina, S., Krier, A., Schley, L. & Burke, T. Using spatial Bayesian methods to determine the genetic structure of a continuously distributed population: clusters or isolation by distance? J. Appl. Ecol. 46, 493–505 (2009).
Article  Google Scholar 

35.
Kalinowski, S. T. The computer program STRUCTURE does not reliably identify the main genetic clusters within species: simulations and implications for human population structure. Heredity 106, 625–632 (2010).
PubMed  PubMed Central  Article  Google Scholar 

36.
Schwartz, M. K. & McKelvey, K. S. Why sampling scheme matters: the effect of sampling scheme on landscape genetic results. Conserv. Genet. 10, 441–452 (2008).
Article  Google Scholar 

37.
Meirmans, P. G. Seven common mistakes in population genetics and how to avoid them. Mol. Ecol. 24, 3223–3231 (2015).
PubMed  Article  PubMed Central  Google Scholar 

38.
Arandjelovic M, et al. Pan African Programme—The cultured chimpanzee. Guidelines for research and data collection. http://panafrican.eva.mpg.de/english/approaches_and_methods.php (2014).

39.
Arandjelovic, M. & Vigilant, L. Non-invasive genetic censusing and monitoring of primate populations. Am. J. Primatol. 80, e22743 (2018).
PubMed  Article  Google Scholar 

40.
Arthofer, W., Heussler, C., Krapf, P., Schlick-Steiner, B. C. & Steiner, F. M. Identifying the minimum number of microsatellite loci needed to assess population genetic structure: a case study in fly culturing. Fly 12, 13–22 (2018).

41.
Wenburg, J. K., Bentzen, P. & Foote, C. J. Microsatellite analysis of genetic population structure in an endangered salmonid: the coastal cutthroat trout (Oncorhyncus clarki clarki). Mol. Ecol. 7, 733–749 (1998).
CAS  Article  Google Scholar 

42.
Guo, X.-Z. et al. Phylogeography and populationgenetics of Schizothorax o’connori: strong subdivision in the Yarlung Tsangpo River inferred from mtDNA and microsatellite markers. Sci. Rep. 6, 29821 (2016).

43.
Kleinhans, C. & Willows-Munro, S. Low genetic diversity andshallow population structure inthe endangered vulture, Gyps copotheres. Sci. Rep. 9, 5536 (2019).
PubMed  PubMed Central  Article  CAS  Google Scholar 

44.
Bonato, L. et al. Diversity among peripheral populations: genetic and evolutionary differentiation of Salamandra atraat the southern edge of the Alps. J. Zool. Syst. Evol. Res. 56, 533–548 (2018).
Article  Google Scholar 

45.
Balkenhol, N. et al. A multi-method approach for analyzing hierarchical genetic structures: a case study with cougars Puma concolor. Ecography 37, 552–563 (2014).
Article  Google Scholar 

46.
Kobayashi, T. & Sota, T. Contrasting effects of habitat discontinuity on three closely related fungivorous beetle species with diverging host‐use patterns and dispersal ability. Ecol. Evol. 9, 2475–2486 (2019).
PubMed  PubMed Central  Article  Google Scholar 

47.
Rousset, F. Genetic differentiation and estimation of gene flow from F-statistics under isolation by distance. Genetics 145, 1219–1228 (1997).

48.
Valdes, A. M., Slatkin, M. & Freimer, N. B. Allele frequencies at microsatellite loci: the stepwise mutation model revisited. Genetics 133, 737–749 (1993).
CAS  PubMed  PubMed Central  Article  Google Scholar 

49.
Goldstein, D. B., Ruiz Linares, A., Cavalli-Sforza, L. L. & Feldman, M. W. An evaluation of genetic distances for use with microsatellite loci. Genetics 139, 463–471 (1995).
CAS  PubMed  PubMed Central  Article  Google Scholar 

50.
Calabrese, P. P., Durrett, R. T. & Aquadro, C. F. Dynamics of microsatellite divergence under stepwise mutation and proportional slippage/point mutation models. Genetics 159, 839–852 (2001).
CAS  PubMed  PubMed Central  Google Scholar 

51.
Petkova, D., Novembre, J. & Stephens, M. Visualizing spatial population structure with estimated effective migration surfaces. Nat. Genet. 48, 94–100 (2015).
PubMed  PubMed Central  Article  CAS  Google Scholar 

52.
Rich, A. M., Wasserman, M. D., Hunt, K. D. & Kaestle, F. A. Chimpanzee (Pan troglodytes schweinfurthii) population spans multiple protected areas in the Albertine Rift. Folia. Primatol. 91, 595–609 (2020).

53.
Baldwin, P. J., McGrew, W. C. & Tutin, C. E. G. Wide-ranging chimpanzees at Mt. Assirik, Senegal. Int. J. Primatol. 3, 367–385 (1982).
Article  Google Scholar 

54.
Lemoine, S. et al. Group dominance increases territory size and reduces neighbour pressure in wild chimpanzees. R. Soc. Open Sci. 7, 200577 (2020).
PubMed  PubMed Central  Article  Google Scholar 

55.
Newton-Fisher, N. E. The home range of the Sonso community of chimpanzees from the Budongo Forest, Uganda. Afr. J. Ecol. 41, 150–156 (2003).
Article  Google Scholar 

56.
Allendorf, F. W. Genetic drift and the loss of alleles versus heterozygosity. Zoo. Biol. 5, 181–190 (1986).
Article  Google Scholar 

57.
Barratt, C. D., et al. Late Quaternary habitat suitability models for chimpanzees (Pan troglodytes) since the Last Interglacial (120,000 BP). Preprint at BioRxiv https://www.biorxiv.org/content/10.1101/2020.05.15.066662v1 (2019).

58.
Bertola, L. D. et al. Phylogeographic patterns in Africa and high resolution delineation of genetic clades in the lion (Panthera leo). Sci. Rep. 6, 30807 (2016).

59.
Marchesi, P., Marchesi, N., Fruth, B. & Boesch, C. Census and distribution of chimpanzees in Côte D’Ivoire. Primates 36, 591–607 (1995).
Article  Google Scholar 

60.
Sommer, V., Adanu, J., Faucher, I. & Fowler, A. Nigerian chimpanzees (Pan troglodytes vellerosus) at Gashaka: two years of habituation efforts. Folia Primatol. 75, 295–316 (2004).
Article  Google Scholar 

61.
Chancellor, R. L., Langergraber, K. E., Ramirez, S., Rundus, A. S. & Vigilant, L. Genetic sampling of unhabituated chimpanzees (Pan troglodytes schweinfurthii) in Gishwati Forest Reserve, an isolated forest fragment in western Rwanda. Int. J. Primatol. 33, 479–488 (2012).
Article  Google Scholar 

62.
Piel, A. K. et al. Population status of chimpanzees in the Masito-Ugalla Ecosystem, Tanzania. Am. J. Primatol. 77, 1027–1035 (2015).
PubMed  Article  Google Scholar 

63.
Wessling, E. G. et al. Seasonal variation in physiology challenges the notion of chimpanzees as a forest-adapted species. Front. Ecol. Evol. 6, 1–21 (2018).
Article  Google Scholar 

64.
Whiten, A. et al. Cultures in chimpanzees. Nature 399, 682–685 (1999).
CAS  PubMed  Article  Google Scholar 

65.
Kühl, H. S. et al. Human impact erodes chimpanzee behavioral diversity. Science 363, 1453–1455 (2019).
PubMed  Article  CAS  Google Scholar 

66.
Luncz, L. V., Mundry, R. & Boesch, C. Evidence for cultural differences between neighboring chimpanzee communities. Curr. Biol. 22, 922–926 (2012).
CAS  PubMed  Article  Google Scholar 

67.
Boesch, C., Marchesi, P., Marchesi, N., Fruth, B. & Joulian, F. Is nut cracking in wild chimpanzees a cultural behaviour? J. Hum. Evol. 26, 325–338 (1994).
Article  Google Scholar 

68.
Kalan, A. K. et al. Environmental variability supports chimpanzee behavioural diversity. Nat. Commun. 11, 4451 (2020).

69.
Hockings, K. J. in Chimpanzees of Bossou and Nimba, 221–219 (Springer Science & Business Media, 2011).

70.
McCarthy, M. S., Lester, J. D. & Stanford, C. B. Chimpanzees (Pan troglodytes) flexibly use introduced species for nesting and bark feeding in a human-dominated habitat. Int. J. Primatol. 38, 321–337 (2016).
PubMed  PubMed Central  Article  Google Scholar 

71.
McLennan, M. R. et al. Surviving at the extreme: chimpanzee ranging is not restricted in a deforested human‐dominated landscape in Uganda. Afr. J. Ecol. 8, e57872 (2020).
Google Scholar 

72.
Junker, J. et al. Recent decline in suitable environmental conditions for African great apes. Divers. Distrib. 18, 1077–1091 (2012).
Article  Google Scholar 

73.
Kühl, H. S. et al. The Critically Endangered western chimpanzee declines by 80%. Am. J. Primatol. 79, e22681 (2017).
Article  Google Scholar 

74.
Walsh, P. D., Breuer, T., Sanz, C., Morgan, D. B. & Doran-Sheehy, D. M. Potential for Ebola transmission between gorilla and chimpanzee social groups. Am. Nat. 169, 684–689 (2007).
PubMed  Article  Google Scholar 

75.
Baden, A. L. et al. Anthropogenic pressures drive population genetic structuring across a critically endangered lemur species range. Sci. Rep. 9, 16276 (2019).
PubMed  PubMed Central  Article  CAS  Google Scholar 

76.
Nsubuga, A. M. et al. Factors affecting the amount of genomic DNA extracted from ape faeces and the identification of an improved sample storage method. Mol. Ecol. 13, 2089–2094 (2004).
CAS  PubMed  Article  Google Scholar 

77.
Arandjelovic, M. et al. Two-step multiplex polymerase chain reaction improves the speed and accuracy of genotyping using DNA from noninvasive and museum samples. Mol. Ecol. Resour. 9, 28–36 (2009).
CAS  PubMed  Article  Google Scholar 

78.
Kalinowski, S. T., Taper, M. L. & Marshall, T. C. Revising how the computer program CERVUS accommodates genotyping error increases success in paternity assignment. Mol. Ecol. 16, 1099–1106 (2007).
PubMed  Article  Google Scholar 

79.
Waits, L. P., Luikart, G. & Taberlet, P. Estimating the probability of identity among genotypes in natural populations: cautions and guidelines. Mol. Ecol. 10, 249–256 (2001).
CAS  PubMed  Article  Google Scholar 

80.
Meirmans, P. G. & Hedrick, P. W. Assessing population structure: FST and related measures. Mol. Ecol. Resour. 11, 5–18 (2010).
PubMed  Article  PubMed Central  Google Scholar 

81.
Mondol, S. et al. New evidence for hybrid zones of forest and savanna elephants in Central and West Africa. Mol. Ecol. 24, 6134–6147 (2015).
CAS  PubMed  Article  PubMed Central  Google Scholar  More

1.
Bindoff, N. L. et al. Changing Ocean, Marine Ecosystems, and Dependent Communities. in IPCC Special Report on the Ocean and Cryosphere in a Changing Climate (eds. Pörtner, H.-O. et al.) 447–588 (IPCC, 2019).
2.
Collins, M. et al. Extremes, Abrupt Changes and Managing Risks. in IPCC Special Report on the Ocean and Cryosphere in a Changing Climate (eds. Pörtner, H.-O. et al.) 589–656 (IPCC, 2019).

3.
Hoegh-Guldberg et al. The Ocean. in Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part B: Regional Aspects. Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change (eds. Barros, V. R. et al.) 1655–1731 (Cambridge University Press, 2014).

4.
Laufkotter, C., Zscheischler, J. & Frolicher, T. L. High-impact marine heatwaves attributable to human-induced global warming. Science 369, 1621–1625 (2020).
CAS  PubMed  PubMed Central  Google Scholar 

5.
Holbrook, N. J. et al. A global assessment of marine heatwaves and their drivers. Nat. Commun. 10, 1–13 (2019).
CAS  Article  Google Scholar 

6.
Oliver, E. C. J. et al. Longer and more frequent marine heatwaves over the past century. Nat. Commun. 9, 1–12 (2018).
CAS  Article  Google Scholar 

7.
Carr, M. & Kearns, E. J. Production regimes in four Eastern Boundary Current systems. Deep. Res. Part II 50, 3199–3221 (2003).
CAS  Article  Google Scholar 

8.
Castro, C. G. et al. Introduction to ‘ The 1997 – 8 El Nino Atlas of oceanographic conditions along the west coast of North America (23°N – 50°N)’. Prog. Oceanogr. 54, 503–511 (2002).
Article  Google Scholar 

9.
Kendrick, G. A. et al. A systematic review of how multiple stressors from an extreme event drove ecosystem-wide loss of resilience in an iconic seagrass community. Front. Mar. Sci. 6, 1–15 (2019).
Article  Google Scholar 

10.
Nohaïc, M. L. et al. Marine heatwave causes unprecedented regional mass bleaching of thermally resistant corals in northwestern Australia. Sci. Rep. 7, 1–11 (2017).
Article  CAS  Google Scholar 

11.
Smale, D. A. Impacts of ocean warming on kelp forest ecosystems. N. Phytol. 225, 1447–1454 (2020).
Article  Google Scholar 

12.
Smale, D. A. et al. Marine heatwaves threaten global biodiversity and the provision of ecosystem services. Nat. Clim. Chang. 9, 306–312 (2019).
Article  Google Scholar 

13.
Schiel, D. R. & Foster, M. S. The Biology and Ecology of Giant Kelp Forests (University of California Press, 2015).

14.
Wernberg, T., Krumhansl, K., Filbee-dexter, K. & Pedersen, M. F. in World Seas: An Environmental Evaluation (ed. Sheppard, C.) 57–78 (Elsevier Ltd., 2019).

15.
Wernberg, T. et al. An extreme climatic event alters marine ecosystem structure in a global biodiversity hotspot. Nat. Clim. Chang. 3, 78–82 (2012).
Article  Google Scholar 

16.
Wernberg, T. et al. Climate-driven regime shift of a temperate marine ecosystem. Science 353, 169–172 (2016).
CAS  PubMed  PubMed Central  Article  Google Scholar 

17.
Oliver, E. C. J. et al. The unprecedented 2015/16 Tasman Sea marine heatwave. Nat. Commun. 8, 1–12 (2017).
Article  Google Scholar 

18.
Thomsen, M. S. et al. Local extinction of bull kelp (Durvillaea spp.) due to a marine heatwave. Front. Mar. Sci. 6, 1–10 (2019).
Article  Google Scholar 

19.
Cavanaugh, K. C. et al. Spatial variability in the resistance and resilience of giant kelp in Southern and Baja California to a multiyear heatwave. Front. Mar. Sci. 6, 1–14 (2019).
Article  Google Scholar 

20.
Arafeh-dalmau, N., Montaño-moctezuma, G., Martínez, J. A. & Smale, D. A. Extreme marine heatwaves alter kelp forest community near its equatorward distribution limit. Front. Mar. Sci. 6, 1–18 (2019).
Article  Google Scholar 

21.
Filbee-Dexter, K., Feehan, C. J. & Scheibling, R. E. Large-scale degradation of a kelp ecosystem in an ocean warming hotspot. Mar. Ecol. Prog. Ser. 543, 141–152 (2016).

22.
Rogers-Bennett, L. & Catton, C. A. Marine heat wave and multiple stressors tip bull kelp forest to sea urchin barrens. Sci. Rep. 9, 1–9 (2019).
CAS  Article  Google Scholar 

23.
Sanford, E., Sones, J. L., García-reyes, M., Goddard, J. H. R. & Largier, J. L. Widespread shifts in the coastal biota of northern California during the 2014- 2016 marine heatwaves. Sci. Reportscientific Rep. 9, 1–14 (2019).
Article  CAS  Google Scholar 

24.
Beas-Luna, R. et al. Geographic variation in responses of kelp forest communities of the California current to recent climatic changes. Glob. Chang. Biol. https://doi.org/10.1111/gcb.15273 (2020).

25.
Filbee-Dexter, K. & Wernberg, T. Rise of turfs: a new battlefront for globally declining kelp forests. Bioscience 68, 64–76 (2018).
Article  Google Scholar 

26.
Filbee-Dexter, K. & Scheibling, R. E. Sea urchin barrens as alternative stable states of collapsed kelp ecosystems. Mar. Ecol. Prog. Ser. 495, 1–25 (2014).
Article  Google Scholar 

27.
Ling, S. D., Johnson, C. R., Frusher, S. D. & Ridgway, K. R. Overfishing reduces resilience of kelp beds to climate-driven catastrophic phase shift. Proc. Natl. Acad. Sci. USA 106, 22341–22345 (2009).

28.
Estes, J. A., Tinker, M. T., Williams, T. M. & Doak, D. F. Killer whale predation on sea otters linking oceanic and nearshore ecosystems. Science 282, 473–477 (1998).
CAS  PubMed  Article  PubMed Central  Google Scholar 

29.
Hamilton, S. L. & Caselle, J. E. Exploitation and recovery of a sea urchin predator has implications for the resilience of southern California kelp forests. Proc. R. Soc. B 282, 1–10 (2014).

30.
Jacox, M. G. et al. Forcing of multiyear extreme ocean temperatures that impacted California current living marine resources in 2016. Bull. Am. Meteorol. Soc. 99, 27–33 (2018).
Article  Google Scholar 

31.
Montecino-Latorre, D. et al. Devastating transboundary impacts of sea star wasting disease on subtidal asteroids. PLoS ONE 11, 1–21 (2016).
Article  CAS  Google Scholar 

32.
Eisaguirre, J. M., Davis, K., Carlson, P. M., Gaines, S. D. & Caselle, J. E. Trophic redundancy and predator size class structure drive differences in kelp forest ecosystem dynamics. Ecology 101, 1–11 (2020).
Article  Google Scholar 

33.
Harrold, C. & Reed, D. C. Food availability, sea urchin grazing, and kelp forest community structure. Ecology 66, 1160–1169 (1985).
Article  Google Scholar 

34.
Cowen, R. K. The effect of sheephead (Semicossyphus pulcher) predation on red sea urchin (Strongylocentrotus franciscanus) populations: an experimental analysis. Oecologia 58, 249–255 (1983).
PubMed  Article  Google Scholar 

35.
Burt, J. M. et al. Sudden collapse of a mesopredator reveals its complementary role in mediating rocky reef regime shifts. Proc. R. Soc. B 285, 1–9 (2018).

36.
Springer, Y. P., Hays, C., Carr, M. H. & Mackey, M. R. Toward ecosystem-based management of marine macroalgae-the bull kelp, Nereocystis luetkeana. Oceanogr. Mar. Biol. Annu. Rev. 48, 1–42 (2010).
Google Scholar 

37.
Springer, Y., Hays, C., Carr, M., Mackey, M. & Bloeser, J. Ecology and management of the bull kelp, Nereocystis luetkeana: a synthesis with recommendations for future research. Vol. 48, 1–45 (Lenfest Ocean Program at The Pew Charitable Trusts, 2006).

38.
Bell, T. W., Allen, J. G., Cavanaugh, K. C. & Siegel, D. A. Three decades of variability in California’s giant kelp forests from the Landsat satellites. Remote Sens. Environ. 238, 110811 (2020).

39.
Pfister, C. A., Berry, H. D. & Mumford, T. The dynamics of Kelp Forests in the Northeast Pacific Ocean and the relationship with environmental drivers. J. Ecol. https://doi.org/10.1111/1365-2745.12908 (2017).

40.
Griggs, G. B. & Hein, J. R. Sources, dispersal, and clay mineral composition of fine-grained sediment off Central and Northern California. J. Geol. 88, 541–566 (1979).
Article  Google Scholar 

41.
Harvell, C. D., Caldwell, J. M., Burt, J. M., Bosley, K. & Keller, A. Disease epidemic and a marine heat wave are associated with the continental-scale collapse of a pivotal predator (Pycnopodia helianthoides). Sci. Adv. 5, 1–9 (2019).
Article  Google Scholar 

42.
Okamoto, D. K., Schroeter, S. C. & Reed, D. C. Effects of ocean climate on spatiotemporal variation in sea urchin settlement and recruitment. Limnol. Oceanogr. https://doi.org/10.1002/lno.11440 (2020).

43.
Estes, J. A., Burdin, A. & Doak, D. F. Sea otters, kelp forests, and the extinction of Steller’s sea cow. Proc. Natl Acad. Sci. USA 113, 880–885 (2016).
CAS  PubMed  Article  PubMed Central  Google Scholar 

44.
Scheffer, M., Carpenter, S., Foley, J. A., Folke, C. & Walker, B. Catastrophic shifts in ecosystems. Nature 413, 591–596 (2001).
CAS  PubMed  Article  PubMed Central  Google Scholar 

45.
Ebert, T. A., Schroeter, S. C., Dixon, J. D. & Kalvass, P. Settlement patterns of red and purple sea urchins (Strongylocentrotus franciscanus and S. purpuratus) in California, USA. Mar. Ecol. Prog. Ser. 111, 41–52 (1994).
Article  Google Scholar 

46.
Frölicher, T. L., Fischer, E. M. & Gruber, N. Marine heatwaves under global warming. Nature 560, 360–366 (2018).
PubMed  Article  CAS  PubMed Central  Google Scholar 

47.
Steneck, R. S. et al. Kelp forest ecosystems: biodiversity, stability, resilience and future. Environ. Conserv. 29, 436–459 (2002).
Article  Google Scholar 

48.
Smith, J. G. et al. Behavioral responses across a mosaic of ecosystem states restructure a sea otter-urchin trophic cascade. Proc. Natl. Acad. Sci. USA 118, 202012493 (2020). https://doi.org/10.1073/pnas.2012493118.

49.
Eurich, J. G., Selden, R. L. & Warner, R. R. California spiny lobster preference for urchins from kelp forests: implications for urchin barren persistence. Mar. Ecol. Prog. Ser. 498, 217–225 (2014).
Article  Google Scholar 

50.
Graham, M. H. Effects of local deforestation on the diversity and structure of Southern California giant kelp forest food webs. Ecosystems 7, 341–357 (2004).
Article  Google Scholar 

51.
Babcock, R. C. et al. Decadal trends in marine reserves reveal differential rates of change in direct and indirect effects. Proc. Natl Acad. Sci. USA 107, 18256–18261 (2010).
CAS  PubMed  Article  PubMed Central  Google Scholar 

52.
Hewson, I., Bistolas, K. S. I., Cardé, E. M. Q. & Button, J. B. Investigating the complex association between viral ecology, environment, and Northeast Pacific Sea Star Wasting. Front. Mar. Sci. 5, 1–14 (2018).

53.
Pearse, J. S. & Hines, A. H. Expansion of a central California kelp forest following the mass mortality of sea urchins. Mar. Biol. 51, 83–91 (1979).

54.
Ebeling, A. W., Laur, D. R., Rowley, R. J. & Barbara, S. Severe storm disturbances and reversal of community structure in a southern California kelp forest. Mar. Biol. 294, 287–294 (1985).
Article  Google Scholar 

55.
Martínez, B. et al. Distribution models predict large contractions of forming seaweeds in response to ocean warming. Divers. Distrib. 24, 1350–1366 (2018).
Article  Google Scholar 

56.
DeYoung, B. et al. Regime shifts in marine ecosystems: detection, prediction and management. Trends Ecol. Evol. 23, 402–409 (2008).
PubMed  Article  Google Scholar 

57.
Crépin, A., Biggs, R., Polasky, S., Troell, M. & Zeeuw, A. De. Regime shifts and management. Ecol. Econ. 84, 15–22 (2012).
Article  Google Scholar 

58.
Roberts, D. A. et al. Mapping chaparral in the Santa Monica Mountains using multiple endmember spectral mixture models. Remote Sens. Environ. 65, 267–279 (1998).
Article  Google Scholar 

59.
Cavanaugh, K. C., Siegel, D. A., Reed, D. C. & Dennison, P. E. Environmental controls of giant-kelp biomass in the Santa Barbara Channel, California. Mar. Ecol. Prog. Ser. 429, 1–17 (2011).
Article  Google Scholar 

60.
Hobday, A. J. et al. A hierarchical approach to defining marine heatwaves. Prog. Oceanogr. 141, 227–238 (2016).
Article  Google Scholar 

61.
García-Reyes, M., Largier, J. L. & Sydeman, W. J. Progress in Oceanography Synoptic-scale upwelling indices and predictions of phyto- and zooplankton populations. Prog. Oceanogr. 120, 177–188 (2014).
Article  Google Scholar 

62.
McHugh, T., Abbott, D. & Freiwald, J. Phase shift from kelp forest to urchin barren along California’s North Coast. (Western Society of Naturalists, 2018).

63.
Rogers-Bennett, L., Kashiwada, J. V., Taniguchi, I. K., Kawana, S. K. & Catton, C. A. Using density-based fishery management strategies to respond to mass mortality events. J. Shellfish Res. 38, 1–11 (2019).
Article  Google Scholar 

64.
Carrascal, L. M. & Galva, I. Partial least squares regression as an alternative to current regression methods used in ecology. Oikos 118, 681–690 (2009).
Article  Google Scholar 

65.
Vadas, R. L. Ecological implications of culture studies on nereocystis luetkeana. J. Phycol. 8, 196–203 (1972).
Google Scholar 

66.
Finger, D. J. I., McPherson, M. L., Houskeeper, H. F. & Kudela, R. M. Mapping bull kelp canopy in northern California using Landsat to enable long-term monitoring. Remote Sens. Environ. 254, 112243 (2021).
Article  Google Scholar  More

1.
Tsukamoto, K. Discovery of the spawning area for Japanese eel. Nature 356, 789–791 (1992).
ADS  Article  Google Scholar 
2.
Tsukamoto, K. Spawning of eels near a seamount. Nature 439, 929 (2006).
ADS  CAS  PubMed  Article  PubMed Central  Google Scholar 

3.
Chow, S. et al. Discovery of mature freshwater eels in the open ocean. Fish. Sci. 75, 257–259 (2009).
CAS  Article  Google Scholar 

4.
Kurogi, H. et al. First capture of post-spawning female of the Japanese eel Anguilla japonica at the southern West Mariana Ridge. Fish. Sci. 77, 199–205 (2011).
CAS  Article  Google Scholar 

5.
Tsukamoto, K. et al. Positive buoyancy in eel leptocephali: an adaptation for life in the ocean surface layer. Mar. Biol. 156, 835–846 (2009).
Article  Google Scholar 

6.
Cheng, P. W. & Tzeng, W. N. Timing of metamorphosis and estuarine arrival across the dispersal range of the Japanese eel Anguilla japonica. Mar. Ecol. Prog. Ser. 131, 87–96 (1996).
ADS  Article  Google Scholar 

7.
Chen, J. Z., Huang, S. L. & Han, Y. S. Impact of long-term habitat loss on the Japanese eel Anguilla japonica. Estuar. Coast. Shelf Sci. 151, 361–369 (2014).
ADS  CAS  Article  Google Scholar 

8.
Tanaka, E. Stock assessment of Japanese eels using Japanese abundance indices. Fish. Sci. 80, 1129–1144 (2014).
CAS  Article  Google Scholar 

9.
Jacoby, D. & Gollock, M. Anguilla anguilla. The IUCN red list of threatened species, version 2014.2. IUCN 2014 e.T60344A45833138. https://doi.org/10.1108/ICS-04-2017-0025 (2014).

10.
Onda, H. et al. Vertical distribution and assemblage structure of leptocephali in the North Equatorial Current region of the western Pacific. Mar. Ecol. Prog. Ser. 575, 119–136 (2017).
ADS  Article  Google Scholar 

11.
Saijo, Y., Iizuka, S. & Asaoka, O. Chlorophyll maxima in Kuroshio and adjacent area. Mar. Biol. 4, 190–196 (1969).
CAS  Article  Google Scholar 

12.
Furuya, K. Subsurface chlorophyll maximum in the tropical and subtropical western Pacific Ocean: Vertical profiles of phytoplankton biomass and its relationship with chlorophylla and particulate organic carbon. Mar. Biol. 107, 529–539 (1990).
CAS  Article  Google Scholar 

13.
Otake, T., Nogami, K. & Maruyama, K. Dissolved and particulate organic matter as possible food sources for eel leptocephali. Mar. Ecol. Prog. Ser. 92, 27–34 (1993).
ADS  Article  Google Scholar 

14.
Mochioka, N. & Iwamizu, M. Diet of anguilloid larvae: Leptocephali feed selectively on larvacean houses and fecal pellets. Mar. Biol. 125, 447–452 (1996).
Google Scholar 

15.
Miller, M. J., Otake, T. & Aoyama, J. Observations of gut contents of leptocephali in the North Equatorial current and Tomini Bay Indonesia. Coast. Mar. Sci. 35, 277–288 (2012).
Google Scholar 

16.
Tomoda, T. et al. Observations of gut contents of anguilliform leptocephali collected in the western North Pacific. Nippon Suisan Gakkaishi 84, 32–44 (2018).
Article  Google Scholar 

17.
Deibel, D., Parrish, C. C., Grønkjær, P., Munk, P. & GisselNielsen, T. Lipid class and fatty acid content of the leptocephalus larva of tropical eels. Lipids 47, 623–634 (2012).
CAS  PubMed  Article  PubMed Central  Google Scholar 

18.
Liénart, C. et al. Geographic variation in stable isotopic and fatty acid composition of anguilliform leptocephali and particulate organic matter in the South Pacific. Mar. Ecol. Prog. Ser. 544, 225–241 (2016).
ADS  Article  CAS  Google Scholar 

19.
Miller, M. J. et al. A low trophic position of Japanese eel larvae indicates feeding on marine snow. Biol. Lett. 9, 20120826 (2013).
PubMed  PubMed Central  Article  Google Scholar 

20.
Miyazaki, S. et al. Stable isotope analysis of two species of anguilliform leptocephali (Anguilla japonica and Ariosoma major) relative to their feeding depth in the North Equatorial Current region. Mar. Biol. 158, 2555–2564 (2011).
CAS  Article  Google Scholar 

21.
Chow, S. et al. Japanese eel Anguilla japonica do not assimilate nutrition during the oceanic spawning migration: evidence from stable isotope analysis. Mar. Ecol. Prog. Ser. 402, 233–238 (2010).
ADS  CAS  Article  Google Scholar 

22.
Chow, S. et al. Onboard rearing attempts for the Japanese eel leptocephali using POM-enriched water collected in the Western North Pacific. Aquat. Living Resour. 30, 1–7 (2017).
Article  CAS  Google Scholar 

23.
Miller, M. J., Hanel, R., Feunteun, E. & Tsukamoto, K. The food source of Sargasso Sea leptocephali. Mar. Biol. 167, 57 (2020).
CAS  Article  Google Scholar 

24.
Pompanon, F. et al. Who is eating what: Diet assessment using next generation sequencing. Mol. Ecol. 21, 1931–1950 (2012).
CAS  PubMed  Article  PubMed Central  Google Scholar 

25.
Wang, M. & Jeffs, A. G. Nutritional composition of potential zooplankton prey of spiny lobster larvae: a review. Rev. Aquac. 6, 270–299 (2014).
Article  Google Scholar 

26.
Ho, T. W., Hwang, J. S., Cheung, M. K., Kwan, H. S. & Wong, C. K. Dietary analysis on the shallow-water hydrothermal vent crab Xenograpsus testudinatus using Illumina sequencing. Mar. Biol. 162, 1787–1798 (2015).
CAS  Article  Google Scholar 

27.
Chow, S. et al. Molecular diet analysis of Anguilliformes leptocephalus larvae collected in the western North Pacific. PLoS ONE 14, e0225610 (2019).
CAS  PubMed  PubMed Central  Article  Google Scholar 

28.
Riemann, L. et al. Qualitative assessment of the diet of European eel larvae in the Sargasso Sea resolved by DNA barcoding. Biol. Lett. 6, 819–822 (2010).
CAS  PubMed  PubMed Central  Article  Google Scholar 

29.
Ayala, D. J. et al. Gelatinous plankton is important in the diet of European eel (Anguilla anguilla) larvae in the Sargasso Sea. Sci. Rep. 8, 6156 (2018).
ADS  PubMed  PubMed Central  Article  CAS  Google Scholar 

30.
Estrada, M. et al. Phytoplankton across tropical and subtropical regions of the Atlantic Indian and Pacific Oceans. PLoS ONE 11, e0151699 (2016).
PubMed  PubMed Central  Article  CAS  Google Scholar 

31.
Lundgreen, R. B. C. et al. Eukaryotic and cyanobacterial communities associated with marine snow particles in the oligotrophic Sargasso Sea. Sci. Rep. 9, 1–12 (2019).
CAS  Article  Google Scholar 

32.
Ayala, D., Riemann, L. & Munk, P. Species composition and diversity of fish larvae in the Subtropical Convergence Zone of the Sargasso Sea from morphology and DNA barcoding. Fish. Oceanogr. 25, 85–104 (2016).
Article  Google Scholar 

33.
Arai, M. N. Active and passive factors affecting aggregations of hydromedusae: a review. Sci. Mar. 56, 99–108 (1992).
Google Scholar 

34.
Boero, F. et al. Gelatinous plankton: Irregularities rule the world (sometimes). Mar. Ecol. Prog. Ser. 356, 299–310 (2008).
ADS  Article  Google Scholar 

35.
Purcell, J. E. Feeding and growth of the siphonophore Muggiaea atlantica (Cunningham 1893). J. Exp. Mar. Bio. Ecol. 62, 39–54 (1982).
Article  Google Scholar 

36.
Alldredge, A. Particle aggregation dynamics. In Encyclopedia of Ocean Sciences, 2nd edn, 330–337 (Elsevier Inc., 2008). https://doi.org/10.1016/B978-012374473-9.00468-9

37.
Hosia, A. & Bamstedt, U. Seasonal abundance and vertical distribution of siphonophores in western Norwegian fjords. J. Plankton Res. 30, 951–962 (2008).
Article  Google Scholar 

38.
Lo, W. T., Yu, S. F. & Hsieh, H. Y. Effects of summer mesoscale hydrographic features on epipelagic siphonophore assemblages in the surrounding waters of Taiwan, western North Pacific Ocean. J. Oceanogr. 69, 495–509 (2013).
Article  Google Scholar 

39.
Lo, W.-T., Yu, S.-F. & Hsieh, H.-Y. Hydrographic processes driven by seasonal monsoon system affect siphonophore assemblages in tropical-subtropical waters (Western North Pacific Ocean). PLoS ONE 9, e100085 (2014).
ADS  PubMed  PubMed Central  Article  CAS  Google Scholar 

40.
Li, K. Z., Yin, J. Q., Huang, L. M. & Song, X. Y. Comparison of siphonophore distributions during the southwest and northeast monsoons on the northwest continental shelf of the South China Sea. J. Plankton Res. 34, 636–641 (2012).
Article  Google Scholar 

41.
López-López, L., Molinero, J. C., Tseng, L.-C., Chen, Q.-C. & Hwang, J.-S. Seasonal variability of the gelatinous carnivore zooplankton community in Northern Taiwan. J. Plankton Res. 35, 677–683 (2013).
Article  Google Scholar 

42.
Price, J. F. Upper ocean response to a hurricane. J. Phys. Ocean. 11, 153–175 (1981).
ADS  Article  Google Scholar 

43.
Toratani, M. Primary production enhancement by typhoon Ketsana in 2003 in western North Pacific. In Remote Sensing of Inland, Coastal, and Oceanic Waters (eds. Frouin, R. J. et al.) 7150, 715013 (SPIE, 2008).

44.
Lin, I. I. Typhoon-induced phytoplankton blooms and primary productivity increase in the western North Pacific subtropical ocean. J. Geophys. Res. Ocean. 117, C03039 (2012).
ADS  Article  Google Scholar 

45.
Ishida, H., Furusawa, K., Makino, T., Ishizaka, J. & Watanabe, Y. The effect of typhoons on phytoplankton communities and settling particle flux in the western North Pacific subtropical region. Oceanogr. Jpn. 25, 17–41 (2016).
Article  Google Scholar 

46.
Siswanto, E., Ishizaka, J., Yokouchi, K., Tanaka, K. & Tan, C. K. Estimation of interannual and interdecadal variations of typhoon-induced primary production: a case study for the outer shelf of the East China Sea. Geophys. Res. Lett. 34, L03604 (2007).
ADS  Article  Google Scholar 

47.
Chen, Y. L. L., Houng-Yung, C., Jan, S. & Tuo, S. H. Phytoplankton productivity enhancement and assemblage change in the upstream Kuroshio after typhoons. Mar. Ecol. Prog. Ser. 385, 111–126 (2009).
ADS  CAS  Article  Google Scholar 

48.
Tsuchiya, K. et al. Typhoon-induced response of phytoplankton and bacteria in temperate coastal waters. Estuar. Coast. Shelf Sci. 167, 458–465 (2015).
ADS  CAS  Article  Google Scholar 

49.
Typhoon information. Japan Meteorological Agency. https://www.data.jma.go.jp/fcd/yoho/typhoon/index.html. Accessed 10 Dec 2020.

50.
Miller, M. J. et al. Morphology and gut contents of anguillid and marine eel larvae in the Sargasso Sea. Zool. Anz. 279, 138–151 (2019).
Article  Google Scholar 

51.
Singh, P., Liu, Y., Li, L. & Wang, G. Ecological dynamics and biotechnological implications of thraustochytrids from marine habitats. Appl. Microbiol. Biotechnol. 98, 5789–5805 (2014).
CAS  PubMed  Article  Google Scholar 

52.
Tanaka, H., Kagawa, H., Ohta, H., Unuma, T. & Nomura, K. The first production of glass eel in captivity: fish reproductive physiology facilitates great progress in aquaculture. Fish Physiol. Biochem. 28, 493–497 (2003).
Article  Google Scholar 

53.
Stenly, W. et al. Ingestion by Japanese eel Anguilla japonica larvae on various minute zooplanktons. Aquac. Sci. 61, 341–347 (2013).
Google Scholar 

54.
Butts, I. A. E., Sørensen, S. R., Politis, S. N. & Tomkiewicz, J. First-feeding by European eel larvae: a step towards closing the life cycle in captivity. Aquaculture 464, 451–458 (2016).
Article  Google Scholar 

55.
Tsukamoto, K. & Miller, M. J. The mysterious feeding ecology of leptocephali: a unique strategy of consuming marine snow materials. Fish. Sci. 87, 11–29 (2020).
Article  CAS  Google Scholar 

56.
Bouilliart, M., Tomkiewicz, J., Lauesen, P., De Kegel, B. & Adriaens, D. Musculoskeletal anatomy and feeding performance of pre-feeding engyodontic larvae of the European eel (Anguilla anguilla). J. Anat. 227, 325–340 (2015).
PubMed  PubMed Central  Article  Google Scholar 

57.
Westeberg, H. A proposal regarding the source of nutrition of leptocephalus larvae. Int. Rev. Hydrobiol. Hydrogr. 75, 863–864 (1990).
Article  Google Scholar 

58.
Miller, M. Ecology of anguilliform leptocephali: remarkable transparent fish larvae of the ocean surface layer. Aqua-BioScience Monogr. https://doi.org/10.1093/gbe/evy021 (2009).
ADS  Article  Google Scholar 

59.
Strom, S., Bright, K., Fredrickson, K. & Brahamsha, B. The Synechococcus cell surface protein SwmA increases vulnerability to predation by flagellates and ciliates. Limnol. Oceanogr. 62, 784–794 (2017).
ADS  Article  Google Scholar 

60.
Benner, R. & Kaiser, K. Abundance of amino sugars and peptidoglycan in marine particulate and dissolved organic matter. Limnol. Oceanogr. 48, 118–128 (2003).
ADS  CAS  Article  Google Scholar 

61.
Seymour, J., Ahmed, T., Durham, W. & Stocker, R. Chemotactic response of marine bacteria to the extracellular products of Synechococcus and Prochlorococcus. Aquat. Microb. Ecol. 59, 161–168 (2010).
Article  Google Scholar 

62.
Biller, S. J. et al. Bacterial vesicles in marine ecosystems. Science (80-) 343, 183–186 (2014).
ADS  CAS  Article  Google Scholar 

63.
Scanlan, D. Bacterial vesicles in the ocean. Science 343, 143–144 (2014).
ADS  CAS  PubMed  Article  PubMed Central  Google Scholar 

64.
Cisternas-Novoa, C., Lee, C. & Engel, A. Transparent exopolymer particles (TEP) and Coomassie stainable particles (CSP): Differences between their origin and vertical distributions in the ocean. Mar. Chem. 175, 56–71 (2015).
CAS  Article  Google Scholar 

65.
Long, R. A. & Azam, F. Abundant protein-containing particles in the sea. Aquat. Microb. Ecol. 10, 213–221 (1996).
Article  Google Scholar 

66.
Tanoue, E., Ishii, M. & Midorikawa, T. Discrete dissolved and particulate proteins in oceanic waters. Limnol. Oceanogr. 41, 1334–1343 (1996).
ADS  CAS  Article  Google Scholar 

67.
Simon, M., Alldredge, A. L. & Azam, F. Bacterial carbon dynamics on marine snow. Mar. Ecol. Prog. Ser. 65, 205–211 (1990).
ADS  CAS  Article  Google Scholar 

68.
Godhe, A. et al. Quantification of diatom and dinoflagellate biomasses in coastal marine seawater samples by real-time PCR. Appl. Environ. Microbiol. 74, 7174–7182 (2008).
CAS  PubMed  PubMed Central  Article  Google Scholar 

69.
Zhu, F., Massana, R., Not, F., Marie, D. & Vaulot, D. Mapping of picoeucaryotes in marine ecosystems with quantitative PCR of the 18S rRNA gene. FEMS Microbiol. Ecol. 52, 79–92 (2005).
CAS  PubMed  Article  PubMed Central  Google Scholar 

70.
Gong, W. & Marchetti, A. Estimation of 18S gene copy number in marine eukaryotic plankton using a next-generation sequencing approach. Front. Mar. Sci. 6, 219 (2019).
Article  Google Scholar 

71.
Furuya, K. & Marumo, R. The structure of the phytoplankton community in the subsurface chlorophyll maxima in the western North Pacific Ocean. J. Plankton Res. 5, 393–406 (1983).
Article  Google Scholar 

72.
Kuroki, M., Okamura, A., Yamada, Y., Hayasaka, S. & Tsukamoto, K. Evaluation of optimum temperature for the early larval growth of Japanese eel in captivity. Fish. Sci. 85, 801–809 (2019).
CAS  Article  Google Scholar 

73.
Okamura, A. et al. Effects of water temperature on early development of Japanese eel Anguilla japonica. Fish. Sci. 73, 1241–1248 (2007).
CAS  Google Scholar 

74.
Kurokawa, T. et al. Influence of water temperature on morphological deformities in cultured larvae of Japanese eel, Anguilla japonica, at completion of yolk resorption. J. World Aquac. Soc. 39, 726–735 (2008).
Article  Google Scholar 

75.
Tsukamoto, K. et al. Oceanic spawning ecology of freshwater eels in the western North Pacific. Nat. Commun. 2, 179 (2011).
ADS  PubMed  PubMed Central  Article  CAS  Google Scholar 

76.
Shirai, K. et al. Temperature and depth distribution of Japanese eel eggs estimated using otolith oxygen stable isotopes. Geochim. Cosmochim. Acta 236, 373–383 (2018).
ADS  CAS  Article  Google Scholar 

77.
Ichikawa, T. Particulate organic carbon and nitrogen in the adjacent seas of the Pacific Ocean. Mar. Biol. 68, 49–60 (1982).
CAS  Article  Google Scholar 

78.
Hebel, D. V. & Karl, D. M. Seasonal, interannual and decadal variations in particulate matter concentrations and composition in the subtropical North Pacific Ocean. Deep Sea Res Part II Top. Stud. Oceanogr. 48, 1669–1695 (2001).
ADS  CAS  Article  Google Scholar 

79.
MacIntyre, S., Alldredge, A. L. & Gotschalk, C. C. Accumulation of marines now at density discontinuities in the water column. Limnol. Oceanogr. 40, 449–468 (1995).
ADS  Article  Google Scholar 

80.
Tomas, C. R. & Hasle, G. R. Identifying Marine Phytoplankton (Academic Press, New York, 1997).
Google Scholar 

81.
Suzuki, K. et al. Responses of phytoplankton and heterotrophic bacteria in the northwest subarctic Pacific to in situ iron fertilization as estimated by HPLC pigment analysis and flow cytometry. Prog. Oceanogr. 64, 167–187 (2005).
ADS  Article  Google Scholar 

82.
Nagai, S. et al. Influences of diurnal sampling bias on fixed-point monitoring of plankton biodiversity determined using a massively parallel sequencing-based technique. Gene 576, 667–675 (2016).
CAS  PubMed  Article  PubMed Central  Google Scholar 

83.
Tanabe, A. S. et al. Comparative study of the validity of three regions of the 18S-rRNA gene for massively parallel sequencing-based monitoring of the planktonic eukaryote community. Mol. Ecol. Resour. 16, 402–414 (2016).
CAS  PubMed  Article  PubMed Central  Google Scholar 

84.
Dzhembekova, N., Moncheva, S., Ivanova, P., Slabakova, N. & Nagai, S. Biodiversity of phytoplankton cyst assemblages in surface sediments of the Black Sea based on metabarcoding. Biotechnol. Biotechnol. Equip. 32, 1507–1513 (2018).
CAS  Article  Google Scholar 

85.
Dzhembekova, N., Urusizaki, S., Moncheva, S., Ivanova, P. & Nagai, S. Applicability of massively parallel sequencing on monitoring harmful algae at Varna Bay in the Black Sea. Harmful Algae 68, 40–51 (2017).
PubMed  Article  PubMed Central  Google Scholar 

86.
Bolger, A. M., Lohse, M. & Usadel, B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30, 2114–2120 (2014).
CAS  PubMed  PubMed Central  Article  Google Scholar 

87.
Schloss, P. D., Gevers, D. & Westcott, S. L. Reducing the effects of PCR amplification and sequencing artifacts on 16S rRNA-based studies. PLoS ONE 6, e27310 (2011).
ADS  CAS  PubMed  PubMed Central  Article  Google Scholar 

88.
Edgar, R. C., Haas, B. J., Clemente, J. C., Quince, C. & Knight, R. UCHIME improves sensitivity and speed of chimera detection. Bioinformatics 27, 2194–2200 (2011).
CAS  PubMed  PubMed Central  Article  Google Scholar 

89.
Cheung, K. L. Y., Huen, J., Houry, W. A. & Ortega, J. Comparison of the multiple oligomeric structures observed for the Rvb1 and Rvb2 proteins. Biochem. Cell Biol. 88, 77–88 (2010).
CAS  PubMed  PubMed Central  Article  Google Scholar 

90.
Camacho, C. et al. BLAST+: architecture and applications. BMC Bioinform. 10, 421 (2009).
Article  CAS  Google Scholar 

91.
Horton, T. et al. World register of marine species (WoRMS) (2018).

92.
R Core team. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing , Vienna, Austria. ISBN 3-900051-07-0, URL http://www.R-project.org/ (2017). https://doi.org/10.2788/95827.

93.
Bray, J. R. & Curtis, J. T. An ordination of the upland forest communities of southern Wisconsin. Ecol. Monogr. 27, 325–349 (1957).
Article  Google Scholar 

94.
Anderson, M. J. A new method for non-parametric multivariate analysis of variance. Aust. Ecol. 26, 32–46 (2001).
Google Scholar 

95.
Oksanen, J. et al. vegan: Community ecology package. R package version 2.5-2. CRAN R (2018). ISBN 0-387-95457-0. More