Ecology

Here we provide a unified analysis of predator-prey biomass scaling in complex food webs. Doing so reveals a consistent sub-linear scaling pattern across levels of organization – from populations within webs to whole ecosystems – for freshwater, marine and terrestrial systems. This regularity in sub-linear predator-prey scaling among complex food webs from diverse ecosystem types has important implications for understanding energy flows in natural systems across large spatial gradients.Within food webs, predator-prey biomass scaling was characterised by a near three-quarter power scaling relationship ((bar{k}) = 0.71 across ecosystem types), revealing an approximately three-fold increase in predator biomass for every five-fold increase in prey biomass. When summing all predator and prey biomasses within a food web (Fig. 4), predator-prey scaling across webs followed a similar sub-linear scaling regime, with k ranging from 0.65 to 0.67 between ecosystem types. That is, biomass pyramids became systematically more bottom-heavy as pyramid size increased along a biomass gradient (Fig. 1a). These ecosystem-level patterns are quantitatively consistent with previous analysis of predator-prey biomass scaling among distinct trophic groups, which also found sub-linear scaling with k values between 0.66 to about 0.768,17,18. The approach we introduce here permits expanding these analyses to more complex omnivorous feeding relations both among populations within webs and across webs in diverse ecosystems. The similarity in the scaling exponents (and overlap in confidence intervals) of within- and across-web scaling suggest the existence of a general sub-linear scaling pattern, possibly signifying that fundamental constraints apply across levels of biological organization.These results beg the question: where do these sub-linear scaling patterns originate? We are not aware of any ecological theory that is sufficiently general to encompass the diversity of community types in which sub-linear biomass scaling is observed (Appendix S2). Size spectrum theory, which aims to explain the observation that, for whole ecosystems, biomass is approximately evenly distributed across logarithmic body size classes19,20 would appear to be particularity relevant. However, static size spectrum models typically assume that the predator-prey body mass ratio (PPmR) and trophic transfer efficiency (ratio of predator to prey production), whilst inherently variable21,22, do not vary systematically with prey biomass19,23. These measures indicate from which size class energy is obtained relative to predator body mass, and how efficiently that energy is utilized by any given predator to maintain its biomass. While these variables are thought to drive size spectra scaling3, they do not appear to be consistent with predator-prey biomass scaling observed in natural communities. Assuming both an even distribution of biomass across size classes, and a constant PPmR or transfer efficiency across a prey biomass gradient suggests an unchanging trophic biomass pyramid (all else being equal; Appendix S2), Therefore it is not clear how current size-spectrum models might account for sub-linear predator-prey biomass scaling.Predator-prey theory, on the other hand, which models the dynamics of feeding interactions, has traditionally focused on two distinct trophic levels, rather than on networks of highly omnivorous food webs24. Equilibrium predictions from a range of simple predator-prey models are also not consistent with sub-linear predator-prey scaling without additional and likely questionable assumptions (Appendix S2). Although predator-prey theory can be made to accord with our observed patterns, it requires tuning the scaling of prey growth or other terms of the model. Furthermore, questions remain about how best to simulate a biomass gradient as well as how models should be generalized to multi-trophic food webs (Appendix S2).Despite the lack of any general mechanism, it is reasonable to assume that predator biomass, at steady state, is maintained in proportion to prey production8,10. This would suggest that as prey biomass increases, their total production should scale near ~¾ to match the predator biomass they support. Density-dependent processes, such as competition for resources and other negative interactions among prey species, could thus cause per capita growth to decline sub-exponentially. We observed that changes in prey biomass were primarily driven by changes in prey density, rather than average prey body size, consistent with density dependent effects driving the sub-linear nature of predator-prey biomass relations, rather than allometric body mass effects. Clearly, however, ecological theory has further work yet to knit together the various patterns and processes to explain the consistency and generality of predator-prey scaling patterns.Addressing predator-prey biomass scaling from a food web perspective allowed us to assess which node properties were associated with greater predator-prey biomass ratios. Our results go beyond prior theoretical studies6,7 to provide empirical evidence that populations of highly omnivorous predators, as well as predator populations that feed down the food web on smaller, more productive, prey (i.e. a high predator-to-prey body mass ratio), tend to attain higher biomass stocks than predicted by their prey biomass alone. Interestingly, the role of these variables in driving predator biomass deviations appear to vary between ecosystem types: predator biomass increases more strongly with PPmR in rock pool webs, whereas predator omnivory only proved to correlate with predator biomass residuals in soil webs (Fig. 3). Further research would be instructive to understand if these are general patterns across different types of terrestrial and aquatic ecosystems. For instance, whilst rock pool webs display very similar network topology and PPmR scaling as open marine webs25,26, we might expect different scaling patterns in pelagic marine webs where trophic interactions take place in three dimensions21, where ontogenetic diet shifts are common27, and where food chains are long13. Adapting our food-web approach to quantify biomass scaling among size classes would likely be informative for tackling these additional complexities. Whilst predator biomass was associated with PPmR and omnivory (in soil webs), the consistent sub-linear predator-prey scaling regime within ecosystem types and across levels of organization, suggests that density dependent population growth might be the overriding driver of predator-prey biomass scaling.The regularity in predator-prey scaling we observed could provide insight into baselines for the biomass structure of natural communities, which could be informative for assessing the effects of environmental impacts within ecological communities and ecological status. For instance within webs, deviations away from these baselines in the form of smaller power-law exponents (shallower slopes) could reflect local perturbations (e.g. acidification, warming, over-exploitation) which have a disproportionate impact among larger organisms at higher trophic levels28. Predator-prey biomass scaling could therefore offer a complementary approach to body size distributions and size spectra for evaluating ecosystem health29. A similar approach could be applied for scaling relations within species, where the same species occur in multiple webs. Doing so could reveal how the biomass of a given predator species responds to variation in prey availability. For instance, among the stream food webs studied here, two common fish species displayed the characteristic near ¾-power scaling pattern, whilst the biomass of salmonids, and particularly brown trout (Salmo trutta), was invariant with prey biomass across webs (Fig. S4). These results are consistent with previous work in these sites which has highlighted the importance of terrestrial prey for subsidizing the biomass production of these apex predators30,31. Deviations from the expected general scaling pattern could therefore be valuable for identifying the importance of environmental factors that permit some species an ‘escape’ from the predator-prey power law (see also32), and offers promising avenues for future research.Our study, which takes a first step towards investigating predator-prey biomass scaling in complex food webs, has some notable limitations. First, information on the weighting of feeding links was not available for the food webs studied here, and a more comprehensive investigation should require specific interactions strengths and vulnerabilities of each species, data that is, as yet, unavailable. Although our results are robust to alternative assumptions in how these factors are treated (Table S5), any systematic variation in feeding interactions could play an important role. Second, information on the biomass of all basal resources was also not generally available, so our analysis focused on higher trophic predators feeding on (animal) prey. While our approach may equally apply more generally to consumers and resources (e.g. aquatic snails feeding on biofilm), further work is required to test the generality of the empirical patterns we observed using more detailed datasets where this information, and data on interaction strengths, is widely available.Overall, our study reveals a consistent sub-linear predator-prey scaling regime in complex food webs and makes a strong case for the existence of a systematic form of density-dependent population growth that governs the biomass structure of freshwater, marine and terrestrial ecosystems. The highly conserved predator-prey scaling we observed within and across food webs implies a relatively simple scaling-up of predator and prey population biomasses across levels of biological organization. These general patterns in energy flow between predator and prey could facilitate improvements in modelling trophic structure and community dynamics, as well as the corresponding ecosystem functions4,5. Our findings suggest sub-linear predator-prey biomass scaling holds within complex omnivorous food webs, urging ecologists to understand the origin of this large scale, cross-system pattern. More

Different proxies for changes in structure and/or function of microbiomes have been developed, allowing assessing microbiome dynamics at multiple levels. However, the lack and differences in understanding the microbiome dynamics are due to the differences in the choice of proxies in different studies and the limitations of proxies themselves. Here, using both amplicon and metatranscriptomic sequencings, we compared four different proxies (16/18S rRNA genes, 16/18S rRNA transcripts, mRNA taxonomy and mRNA function) to reveal the impact of a severe summer drought in 2018 on prokaryotic and eukaryotic microbiome structures and functions in two rewetted fen peatlands in northern Germany. We found that both prokaryotic and eukaryotic microbiome compositions were significantly different between dry and wet months. Interestingly, mRNA proxies showed stronger and more significant impacts of drought for prokaryotes, while 18S rRNA transcript and mRNA taxonomy showed stronger drought impacts for eukaryotes. Accordingly, by comparing the accuracy of microbiome changes in predicting dry and wet months under different proxies, we found that mRNA proxies performed better for prokaryotes, while 18S rRNA transcript and mRNA taxonomy performed better for eukaryotes. In both cases, rRNA gene proxies showed much lower to the lowest accuracy, suggesting the drawback of DNA based approaches. To our knowledge, this is the first study comparing all these proxies to reveal the dynamics of both prokaryotic and eukaryotic microbiomes in soils. This study shows that microbiomes are sensitive to (extreme) weather changes in rewetted fens, and the associated microbial changes might contribute to ecological consequences. More

Having introduced our novel explore-and-fuse method and the Physarum Steiner Algorithm we shall dedicate this section to discussing how the algorithm’s parameters influence the model, and how the method can be used towards diverse applications.In what follows we shall consider how different parameters such as the different shapes of cells, as well as their number, influence the results obtained by the Physarum Steiner Algorithm. We shall then conclude the section by studying different applications that our methods have.Cell shapeAlthough13 and6 considered diamond shaped CELLs, we shall consider here CELLs with other shapes. The primary benefit of square cells is that their shape allows for more cytoplasm to be placed on the grid. As a result, the foraging phase is very fast so using square cells tends to result in shorter run times than using diamond-shaped cells. In addition, large square cells are able to more completely cover the standard square grid than diamond-shaped cells. On the other hand, diamond-shaped cells result in less cytoplasm and more time spent in the foraging stage. This gives the cytoplasm time to move towards a centralized location which results in better solutions.Example A In order to illustrate the above point, in Fig. 3a.i., we begin with squares that are tightly packed. Since the squares are so tightly packed (1 apart), if any piece of cytoplasm in a square is moved, it will lead to a connection with a neighboring cell. As a result, all the points are found very quickly. In fact, many of the squares are connected and part of the network even if they are not close to any of the points, as shown in Fig. 3 (a.ii.). Shrinking these extra squares takes a long time and can also result in long paths which are far out of the way as seen in Fig. 3a.iii.Example B In contrast to Example A, in Fig. 3b, we consider diamond-shaped cells. The cells start off diamond-shaped and with less overall cytoplasm than the square cells. The cells then spend quite a few iterations in the foraging phase. Although this does take time, it allows the cytoplasm to move towards a centralized location around the active zones as seen in Fig. 3 (b.ii.). When the cell finally proceeds to the shrinking phase, there is less cytoplasm to remove and no out of the way paths, resulting in shorter solutions. The downside to this is the increased time which in some cases can be very long (over 100 million iterations) and in some cases the algorithm may not even complete.The effect of multiple cellsIn what follows we shall examine the effects of the number of cells used. We run 10 trials on 10 grids for a total of 100 trials on each cell size and number of cells. For each trial, we measure the total amount or area of cytoplasm that is initially spawned. This is used to normalize the search area which is the number of squares in the grid (for example a (100 times 100) grid has search area 10,000).Success rate: The algorithm may sometimes be unsuccessful at connecting all the points. For example, the cells may miss a point early on and move far away from that point, making it almost impossible to ever find that point. There may also simply not be enough cytoplasm for two far away cells to fuse into one. For each number of cells (1, 9, 25, 100), we try various sizes/amounts of cytoplasm and compute the proportion of trials (out of 100) that successfully terminate within 10 million iterations.Figure 4(a) Proportion of trials that are successful versus the search area as a percentage of cytoplasm for trials with 1, 9, 25, and 100 cells. (b) Length of solutions versus the search area as a percentage of cytoplasm. (c) Number of iterations versus the search area as a percentage of cytoplasm. Failed trails excluded from graphs.Full size imageIn Fig. 4a, we see that the black line (100 cells) extends much further to the right than the cyan line (one cell). Thus, the more cells there are, the larger of a search area we can explore. This is mainly because with more cells, we can spread out our cytoplasm instead of having it be concentrated in certain areas.Solution length Another important metric to consider is the solution length. We measure how good the solution is by counting the amount of cytoplasm when the algorithm terminates. We ignore any cytoplasm that is part of a disjoint cell that does not contain an active zone, or in other words is separate from the cell that actually forms the tree. In Fig. 4b, we see that as the search area as a percentage of cytoplasm increases, the quality of the solution improves. This is because there is comparatively less cytoplasm to begin with. In addition, we see that as the number of cells increases, it is possible to find a better solution. This correlates with the earlier result shown in Fig. 4a that using more cells allows solutions to be found with less cytoplasm. Trials with 100 cells found the shortest solutions (rightmost data point).Run time The last metric we consider is the run time. We consider the true number of iterations the algorithm runs for. By true iterations, we account for the fact that in a parallel algorithm or set of real-world Physarum organisms, multiple cells will be introducing and moving bubbles at the same time. As a result, the iteration count is scaled by the number of disjoint cells. In Fig. 4c, we see that the more cells there are, the lower the number of iterations. This may be because with more cells, the cytoplasm is more spread out and therefore there are less out of the way points which may take a very long time to find. From the above analysis, we see that using more cells allows us to explore bigger search areas, find shorter solutions, and solve problems faster.ApplicationsThe behavior of Physarum and the models it has inspired have found many different uses among which are drug repositioning, developing bio-computing chips, approximating highways layouts, and designing subway systems2,8,9,10. In order to illustrate the operation of the Physarum Steiner Algorithm and demonstrate its applicability to real world problems, we consider the following:

sep0em

Network design We use the algorithm to develop a road network in the United States.

Obstacle-avoidance We use the algorithm to solve the obstacle-avoiding Euclidean Steiner tree problem.

VLSI routing We use the algorithm to route connections between pads in chip design.

Topological surfaces We discuss the algorithm’s adaptability to varying surfaces and boundaries by considering topological surfaces such as the sphere, torus, Klein bottle, and (mathbb{RP}mathbb{}^2).

Road networks The Physarum Steiner Algorithm can be used to build a road network between the largest one hundred cities in the lower 48 United States (excluding Alaska and Hawaii). We use data32 containing the longitude and latitude of the 100 cities with the highest population to generate a rectangular grid of active zones.We spawn diamond-shaped cells of size 7 with a spacing of 1 as shown in Fig. 3. After many iterations, the final road network is shown in Fig. 5a. The algorithm is particularly suited to the problem of designing transportation systems because it first connects all the points before optimizing the network into a tree. The algorithm can thus be terminated early depending on how much redundant connectivity is desired in the transportation network.For example, in Fig. 5b, we have a network that still contains loops in high-traffic routes between the Bay Area, Los Angeles, and Las Vegas. If we allow the algorithm to continue running, we will get networks with fewer loops and eventually a tree.Figure 5Road network generated by the algorithm. (a) shows the final solution with no loops while (b) displays a solution that has some redundancy resulting from terminating the algorithm early.Full size imageWe believe that this algorithm can be applied to many similar problems such as designing fiber optic or electric cable networks. Moreover, as discussed in the last section, it will be very interesting to compare this study to that of33, where in vitro slime mold is used to investigate the construction of transportation networks over a USA map.Obstacle avoidance Due to the cellular automaton nature of this algorithm, it is straightforward to define boundaries or other obstacles that need to be avoided. This is very useful in cases where certain areas need to be avoided such as a lake or the boundary of a county. And, unlike the current standard obstacle-avoiding Euclidean Steiner algorithm27 which takes multiple hours for graphs with only 150 points, the run time of the Physarum Steiner Algorithm is not affected by the need to avoid obstacles.As an example, consider the boundary given in Fig. 6a. Here, the grey area represents the search area and the 100 white squares outlined in dark grey are the points. There are many possible real world situations similar to this. For example, the grey area could be a county and all the points represent homes that subscribe to a certain Internet service provider (ISP). The big white area in the center could be a lake and the smaller white area could be a dog park. The ISP company could utilize the Physarum Steiner Algorithm to find networks to lay fiber optic cables.Figure 6(a) Sample boundary map. Grey area is search area and small white squares are points. (b) Initial deployment of Physarum. (c) Solution at the end of the foraging stage. (d) The final network.Full size imageWe begin by deploying square Physarum cells of size 7 in Fig. 6b. In Fig. 6c, the cells begin to fuse, share intelligence, and find all the points. We choose a solution that still has some loops to increase reliability and ease of future modification to the network. Our final solution is shown in Fig. 6d. This solution is generated in 300,000 iterations and less than 30 seconds.VLSI Routing for VLSI (very large-scale integration) chip design19 is one of the largest real-world manifestations of the Steiner tree problem, especially as modern chips may contain upwards of 10 billion transistors. Solving the VLSI problem would require additional modification to the Physarum Steiner Algorithm since VLSI design is typically presented as a group Steiner tree problem and has very large problem sizes, the Physarum Steiner Algorithm. Due to the usage of a square grid in the Physarum Steiner Algorithm, the algorithm is easily applied to find rectilinear networks such as those required for routing chips. In addition, our empirical results suggest that it should scale well to the large problem sizes common in chip design. Using data from34, we consider a set of pads that need to be connected. In Fig. 7, we represent the pads as active zones and generate a tree between them.Figure 7(a) Graphical representation of 131-point VLSI data set34. (b) Routing solution obtained by the Physarum Steiner Algorithm.Full size imageTopological surfaces Finally, the Physarum Steiner Algorithm is easily applicable to finding Steiner trees on other topological surfaces. Given the nature of the algorithm, we are able to map coordinates on one edge to another. In Fig. 8, we use square identification spaces to find Steiner trees on the torus, sphere, Klein bottle, and (mathbb{RP}mathbb{}^2). These solutions on identification spaces can be seen on a torus and a sphere in Fig. 8a,b.Figure 8Steiner trees on topological surfaces we defined by identification space and obtained through our code. (a) Torus. (b) Sphere. (c) Klein Bottle. (d) (mathbb{RP}mathbb{}^2). Images generated using manim35.Full size imageConcluding remarksWe have presented here a novel explore-and-fuse approach to solve problems that cannot be solved by traditional divide-and-conquer.Our approach is inspired by Physarum, a unicellular slime mold capable of solving the traveling salesman and Steiner tree problems. Besides exhibiting individual intelligence, Physarum can also share information with other Physarum organisms through fusion. These characteristics of Physarum inspire us to spawn many Physarum organisms to independently explore the problem space and collect information in parallel before sharing the information with other organisms through fusion. Eventually, all the organisms fuse into one large Physarum that can then globally optimize using the knowledge collected earlier. Explore-and-fuse can be seen as a less rigid form of divide-and-conquer that can better handle problems that cannot be decomposed into independent subproblems.We demonstrate the explore-and-fuse approach on the Steiner tree problem by creating the Physarum Steiner Algorithm. This algorithm has the ability to incrementally find Steiner trees. The first solution tends to contain many loops that are removed with additional iterations of the algorithm. This incremental improvement is particularly useful for applications such as road and cable networks where some degree of redundancy in the connectivity is desired. In particular, it will be very interesting to compare our work to the the one done in33 where a protoplasmic network created by in vivo Physarum is considered to study and asses show the slime mold imitates the United States Interstate System. We foresee several applications of our algorithm in this direction, leading to similar findings to those appearing in the studies done in33.The algorithm operates on a rectilinear grid and is particularly applicable to rectilinear Steiner tree problems such as those that often arise in VLSI design. In addition, the algorithm performs well on the obstacle-avoidance Euclidean Steiner tree problem.In comparison to the existing Physarum-inspired Steiner tree algorithms described in Section “The Steiner tree problem”, the Physarum Steiner Algorithm uses a completely different mechanism. While the existing algorithms use a system of equations modeling the thickening of tubes as protoplasm flows through them, the Physarum Steiner Algorithm is based on modeling Physarum spatially moving around a grid and finding a tree between squares of the grid. In addition, it should be noted that the approach taking in existing algorithms would not work on the Euclidean Steiner tree problem as in the Euclidean Steiner tree problem, there are an infinite number of possible points that could be part of the Steiner tree (essentially any point in the plane). It would not be possible to write a system of equations representing the infinite possible points and edges. In the future, we believe further work could be done to improve the Physarum Steiner Algorithm. Since the Physarum Steiner Algorithm is an approximate algorithm, future improvements could be made so its approximations are closer to the actual optimal solution. In addition, it would be interesting to see this approach applied to other problems Physarum has been able to solve such as the traveling salesmen problem. More

Graham, N. A. J. & Nash, K. L. The importance of structural complexity in coral reef ecosystems. Coral Reefs https://doi.org/10.1007/s00338-012-0984-y (2013).Drake, J. L. et al. How corals made rocks through the ages. Glob. Change Biol. 26, 31–53 (2020).Zawada, K. J. A., Madin, J. S., Baird, A. H., Bridge, T. C. L. & Dornelas, M. Morphological traits can track coral reef responses to the Anthropocene. Funct. Ecol. 33, 962–975 (2019).Article 

Google Scholar 
Wehrberger, F. & Herler, J. Microhabitat characteristics influence shape and size of coral-associated fishes. Mar. Ecol. Prog. Ser. 500, 203–214 (2014).Article 

Google Scholar 
Munday, P. & Jones, G. The ecological implications of small body size among coral-reef fishes. Oceanogr. Mar. Biol. 36, 373–411 (1998).
Google Scholar 
Pereira, P. H. C. & Munday, P. L. Coral colony size and structure as determinants of habitat use and fitness of coral-dwelling fishes. Mar. Ecol. Prog. Ser. 553, 163–172 (2016).Article 

Google Scholar 
Doszpot, N., McWilliam, M., Pratchett, M., Hoey, A. & Figueira, W. Plasticity in three-dimensional geometry of branching corals along a cross-shelf gradient. Diversity 11, 44 (2019).Article 

Google Scholar 
Ow, Y. X. & Todd, P. A. Light-induced morphological plasticity in the scleractinian coral Goniastrea pectinata and its functional significance. Coral Reefs 29, 797–808 (2010).Article 

Google Scholar 
Soto, D., De Palmas, S., Ho, M. J., Denis, V. & Chen, C. A. Spatial variation in the morphological traits of Pocillopora verrucosa along a depth gradient in Taiwan. PLoS ONE 13, 1–20 (2018).Article 
CAS 

Google Scholar 
Bruno, J. F. & Edmunds, P. J. Clonal variation for phenotypic plasticity in the coral Madracis Mirabilis. Ecology 78, 2177–2190 (1997).Article 

Google Scholar 
Todd, P. A. Morphological plasticity in scleractinian corals. Biol. Rev. 83, 315–337 (2008).Willis, B. L. Phenotypic plasticity versus phenotypic stability in the reef corals Turbinaria mesenterina and Pavona cactus. Proc. Fifth Int. Coral Reef. Symp. 4, 107–112 (1985).
Google Scholar 
Grottoli, A. G. et al. The cumulative impact of annual coral bleaching can turn some coral species winners into losers. Glob. Change Biol. 20, 3823–3833 (2014).Article 

Google Scholar 
Smith, L. W., Barshis, D. & Birkeland, C. Phenotypic plasticity for skeletal growth, density and calcification of Porites lobata in response to habitat type. Coral Reefs 26, 559–567 (2007).Article 

Google Scholar 
Barnes, D. Growth in colonial scleractinians. Bull. Marine Sci. 280–298 (1973).Anthony, K. R. N. & Hoegh-Guldberg, O. Variation in coral photosynthesis, respiration and growth characteristics in contrasting light microhabitats: an analogue to plants in forest gaps and understoreys? Funct. Ecol. 17, 246–259 (2003).Article 

Google Scholar 
Kramer, N., Tamir, R., Eyal, G. & Loya, Y. Coral morphology portrays the spatial distribution and population size-structure along a 5–100 m depth gradient. Front. Mar. Sci. 7, 615 (2020).Article 

Google Scholar 
Dubé, C. E., Mercière, A., Vermeij, M. J. A. & Planes, S. Population structure of the hydrocoral Millepora platyphylla in habitats experiencing different flow regimes in Moorea, French polynesia. PLoS ONE 12, 1–20 (2017).
Google Scholar 
Chappell, J. Coral morphology, diversity and reef growth. Nature 286, 249–252 (1980).Article 

Google Scholar 
Paz-García, D. A. et al. Morphological variation and different branch modularity across contrasting flow conditions in dominant Pocillopora reef-building corals. Oecologia 178, 207–218 (2015).PubMed 
Article 

Google Scholar 
Laverick, J. H., Tamir, R., Eyal, G. & Loya, Y. A generalized light-driven model of community transitions along coral reef depth gradients. Glob. Ecol. Biogeogr. 29, 1554–1564 (2020).Article 

Google Scholar 
Bongaerts, P., Ridgway, T., Sampayo, E. M. & Hoegh-Guldberg, O. Assessing the “deep reef refugia” hypothesis: focus on Caribbean reefs. Coral Reefs 29, 1–19 (2010).Article 

Google Scholar 
Sherman, C. E., Locker, S. D., Webster, J. M. & Weinstein, D. K. In Mesophotic Coral Ecosystems (eds. Loya, Y., Puglise, K. A. & Bridge, T. C. L.) 849–878 (Springer International Publishing, 2019).Kahng, S. E., Copus, J. M. & Wagner, D. Recent advances in the ecology of mesophotic coral ecosystems (MCEs). Curr. Opin. Environ. Sustainability 7, 72–81 (2014).Article 

Google Scholar 
Tamir, R., Eyal, G., Kramer, N., Laverick, J. H. & Loya, Y. Light environment drives the shallow to mesophotic coral community transition. Ecosphere 10, e02839 (2019).Article 

Google Scholar 
Lichtenthaler, H. K., Ač, A., Marek, M. V., Kalina, J. & Urban, O. Differences in pigment composition, photosynthetic rates and chlorophyll fluorescence images of sun and shade leaves of four tree species. Plant Physiol. Biochem. 45, 577–588 (2007).CAS 
PubMed 
Article 

Google Scholar 
Bragg, J. G. & Westoby, M. Leaf size and foraging for light in a sclerophyll woodland. Funct. Ecol. 16, 633–639 (2002).Article 

Google Scholar 
Sæbø, A., Krekling, T. & Appelgren, M. Light quality affects photosynthesis and leaf anatomy of birch plantlets in vitro. Plant Cell, Tissue Organ Cult. 41, 177–185 (1995).Article 

Google Scholar 
Einbinder, S. et al. Changes in morphology and diet of the coral Stylophora pistillata along a depth gradient. Mar. Ecol. Prog. Ser. 381, 167–174 (2009).Article 

Google Scholar 
Mass, T. et al. Photoacclimation of Stylophora pistillata to light extremes: metabolism and calcification. Mar. Ecol. Prog. Ser. 334, 93–102 (2007).CAS 
Article 

Google Scholar 
Kramer, N. et al. Efficient light-harvesting of mesophotic corals is facilitated by coral optical traits. Funct. Ecol. 36, 406–418 (2022).CAS 
Article 

Google Scholar 
Einbinder, S. et al. Novel adaptive photosynthetic characteristics of mesophotic symbiotic microalgae within the reef-building coral, Stylophora pistillata. Front. Mar. Sci. 3, 195 (2016).Article 

Google Scholar 
Martinez, S. et al. Energy sources of the depth-generalist mixotrophic coral Stylophora pistillata. Front. Mar. Sci. 7, 1–16 (2020).CAS 
Article 

Google Scholar 
Anthony, K. R. N., Hoogenboom, M. O. & Connolly, S. R. Adaptive variation in coral geometry and the optimization of internal colony light climates. Funct. Ecol. 19, 17–26 (2005).Article 

Google Scholar 
Kahng, S. E., Watanabe, T. K., Hu, H.-M., Watanabe, T. & Shen, C.-C. Moderate zooxanthellate coral growth rates in the lower photic zone. Coral Reefs https://doi.org/10.1007/s00338-020-01960-4 (2020).Wangpraseurt, D. et al. The in situ light microenvironment of corals. Limnol. Oceanogr. 59, 917–926 (2014).Article 

Google Scholar 
Dustan, P. Depth-dependent photoadaption by zooxanthellae of the reef coral Montastrea annularis. Mar. Biol. 68, 253–264 (1982).CAS 
Article 

Google Scholar 
Dubinsky, Z., Falkowski, P. G., Porter, J. W. & Muscatine, L. Absorption and utilization of radiant energy by light- and shade-adapted colonies of the hermatypic coral Stylophora pistillata. Proc. R. Soc. B: Biol. Sci. 222, 203–214 (1984).CAS 

Google Scholar 
Falkowski, P. G. & Dubinsky, Z. Light-shade adaptation of Stylophora pistillata, a hermatypic coral from the Gulf of Eilat. Nature 289, 172–174 (1981).Article 

Google Scholar 
Kahng, S. E. et al. In Mesophotic Coral Ecosystems (eds. Loya, Y., Puglise, K. A. & Bridge, T. C. L.) 801–828 (Springer International Publishing, 2019).Hoogenboom, M. O., Connolly, S. R. & Anthony, K. R. N. Interactions between morphological and physiological plasticity optimize energy acquisition in corals. Ecology 89, 1144–1154 (2008).PubMed 
Article 

Google Scholar 
House, J. E. et al. Moving to 3D: relationships between coral planar area, surface area and volume. PeerJ 6, e4280 (2018).PubMed 
PubMed Central 
Article 

Google Scholar 
Zawada, K. J. A., Dornelas, M. & Madin, J. S. Quantifying coral morphology. Coral Reefs 38, 1281–1292 (2019).Article 

Google Scholar 
Malik, A. et al. Molecular and skeletal fingerprints of scleractinian coral biomineralization: From the sea surface to mesophotic depths. Acta Biomaterialia 1–14 https://doi.org/10.1016/j.actbio.2020.01.010 (2020).Todd, P. A., Ladle, R. J., Lewin-Koh, N. J. I. & Chou, L. M. Genotype x environment interactions in transplanted clones of the massive corals Favia speciosa and Diploastrea heliopora. Mari. Ecol. Prog. Ser. https://doi.org/10.3354/meps271167 (2004).Crabbe, M. J. C. & Smith, D. J. Modelling variations in corallite morphology of Galaxea fascicularis coral colonies with depth and light on coastal fringing reefs in the Wakatobi Marine National Park (S.E. Sulawesi, Indonesia). Computational Biol. Chem. 30, 155–159 (2006).CAS 
Article 

Google Scholar 
Studivan, M. S., Milstein, G. & Voss, J. D. Montastraea cavernosa corallite structure demonstrates distinct morphotypes across shallow and mesophotic depth zones in the Gulf of Mexico. PLoS ONE 14, e0203732 (2019).Wangpraseurt, D., Larkum, A. W. D., Ralph, P. J. & Kühl, M. Light gradients and optical microniches in coral tissues. Front. Microbiol. 3, 1–9 (2012).Article 

Google Scholar 
Enríquez, S., Méndez, E. R. & Iglesias-Prieto, R. Multiple scattering on coral skeletons enhances light absorption by symbiotic algae. Limnol. Oceanogr. 50, 1025–1032 (2005).Article 

Google Scholar 
Wangpraseurt, D., Jacques, S. L., Petrie, T. & Kühl, M. Monte Carlo modeling of photon propagation reveals highly scattering coral tissue. Front. Plant Sci. 7, 1–10 (2016).Article 

Google Scholar 
Muko, S., Kawasaki, K., Sakai, K., Takasu, F. & Shigesada, N. Morphological plasticity in the coral Porites sillimaniani and its adaptive significance. Bull. Mar. Sci. 66, 225–239 (2000).
Google Scholar 
Klaus, J. S., Budd, A. F. & Fouke, B. Environmental controls on corallite morphology in the reef coral Montastraea annularis Hot springs microbiology View project Positive Accretion in the Deep: Carbonate budget analysis of Caribbean mesophotic coral reef habitats View project. Bull. Mar. Sci. 80, 233–260 (2007).
Google Scholar 
Enríquez, S., Méndez, E. R., Hoegh-Guldberg, O. & Iglesias-Prieto, R. Key functional role of the optical properties of coral skeletons in coral ecology and evolution. Proc. R. Soc. B: Biol. Sci. 284, (2017).Wangpraseurt, D. et al. Microscale light management and inherent optical properties of intact corals studied with optical coherence tomography. J. R. Soc. Interface 16, 20180567 (2019).Groves, S. H. et al. Growth rates of Porites astreoides and Orbicella franksi in mesophotic habitats surrounding St. Thomas, US Virgin Islands. Coral Reefs 37, 345–354 (2018).Article 

Google Scholar 
Shlesinger, T., Grinblat, M., Rapuano, H., Amit, T. & Loya, Y. Can mesophotic reefs replenish shallow reefs? Reduced coral reproductive performance casts a doubt. Ecology https://doi.org/10.1002/ecy.2098 (2018).Kaniewska, P. et al. Importance of macro- versus microstructure in modulating light levels inside coral colonies. J. Phycol. 47, 846–860 (2011).PubMed 
Article 

Google Scholar 
Swain, T. D. et al. Relating coral skeletal structures at different length scales to growth, light availability to symbiodinium, and thermal bleaching. Front. Mar. Sci. 5, (2018).Bongaerts, P. & Smith, T. B. In Mesophotic Coral Ecosystems (eds. Loya, Y., Puglise, K. A. & Bridge, T. C. L.) 881–895 (Springer International Publishing, 2019).Bongaerts, P., Riginos, C., Brunner, R., Englebert, N. & Smith, S. R. Deep reefs are not universal refuges: reseeding potential varies among coral species. Sci. Adv. 3, 1–40 (2017).Article 

Google Scholar 
Serrano, X. M. et al. Long distance dispersal and vertical gene flow in the Caribbean brooding coral Porites astreoides. Sci. Rep. 6, 21619 (2016).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Fantazzini, P. et al. Gains and losses of coral skeletal porosity changes with ocean acidification acclimation. Nat. Commun. 6, 7785 (2015).Mollica, N. R. et al. Ocean acidification affects coral growth by reducing skeletal density. Proc. Natl Acad. Sci. USA 115, 1754–1759 (2018).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Fordyce, A. J., Ainsworth, T. D., Leggat, W. & Katherine, M. Light capture, skeletal morphology, and the biomass of corals’ boring endoliths. mSphere 6, e00060–21 (2021).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Muscatine, L. The role of symbiotic algae in carbon and energy flux in coral reefs. Ecosyst. World 25, 75–87 (1990).
Google Scholar 
Houlbrèque, F. & Ferrier-Pagès, C. Heterotrophy in tropical scleractinian corals. Biol. Rev. 84, 1–17 (2009).PubMed 
Article 

Google Scholar 
Tremblay, P., Gori, A., Maguer, J. F., Hoogenboom, M. & Ferrier-Pagès, C. Heterotrophy promotes the re-establishment of photosynthate translocation in a symbiotic coral after heat stress. Sci. Rep. 6, 38112 (2016).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Lesser, M. P. et al. Photoacclimatization by the coral Montastraea cavernosa in the mesophotic zone: light, food, and genetics. Ecology https://doi.org/10.1890/09-0313.1 (2010).Pinheiro, H. T., Eyal, G., Shepherd, B. & Rocha, L. A. Ecological insights from environmental disturbances in mesophotic coral ecosystems. Ecosphere 10, e02666 (2019).Article 

Google Scholar 
Wangpraseurt, D. et al. In vivo microscale measurements of light and photosynthesis during coral bleaching: evidence for the optical feedback loop? Front. Microbiol. 8, 1–12 (2017).Article 

Google Scholar 
Veron, C., Stafford-Smith, M., Turak, E. & DeVantier, L. Corals of the World (Australian Institute of Marine Science, 2000).Loya, Y. The Red Sea coral Stylophora pistillata is an r strategist. Nature 259, 478–480 (1976).Article 

Google Scholar 
Drake, J. L. et al. Physiological and transcriptomic variability indicative of differences in key functions within a single coral colony. Front. Mar. Sci. 8, 768 (2021).Article 

Google Scholar 
Jacques, S., Li, T. & Prahl, S. mcxyz. c, a 3D Monte Carlo simulation of heterogeneous tissues. omlc.org/software/mc/mcxyz (2013).Tuchin, V. V. Tissue Optics: Light Scattering Methods and Instruments for Medical Diagnostics (SPIE PRESS, 2015).Wang, L., Jacques, S. L. & Zheng, L. MCML—Monte Carlo modeling of light transport in multi-layered tissues. Computer Methods Prog. Biomed. 47, 131–146 (1995).CAS 
Article 

Google Scholar 
Jacques, S. L., Wangpraseurt, D. & Kühl, M. Optical properties of living corals determined with diffuse reflectance. Spectrosc. Front Mar. Sci. 6, 1–9 (2019).Article 

Google Scholar 
Hill, R. et al. Spatial heterogeneity of photosynthesis and the effect of temperature-induced bleaching conditions in three species of corals. Mar. Biol. 144, 633–640 (2004).Article 

Google Scholar 
Ritchie, R. J. & Larkum, A. W. D. Modelling photosynthesis in shallow algal production ponds. Photosynthetica 50, 481–500 (2012).CAS 
Article 

Google Scholar 
Team, R. C. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/ (2022).Bates, D., Mächler, M., Bolker, B. & Walker, S. Fitting linear mixed-effects models using {lme4}. J. Stat. Softw. 67, 1–48 (2015).Article 

Google Scholar 
Luo, D., Ganesh, S. & Koolaard, J. predictmeans: calculate predicted means for linear models. Preprint at https://CRAN.R-project.org/package=predictmeans (2021).Clarke, K. R. Non‐parametric multivariate analyses of changes in community structure. Aust. J. Ecol. 18, 117–143 (1993).Article 

Google Scholar 
Kramer, N., Guan, J., Chen, S., Wangpraseurt, D. & Loya, Y. Morpho-functional traits of the coral Stylophora pistillata enhance light capture for photosynthesis at mesophotic depths. Dryad, Dataset https://doi.org/10.5061/dryad.7d7wm37w7 (2022). More

The generalist accumulates extracellular nitriteWe first tested whether the generalist accumulates substantial extracellular nitrite under our experimental conditions, and thus creates a niche for the specialist. To accomplish this, we grew the generalist alone in bioreactors with anoxic ACS medium amended with 12 mM nitrate as the growth-limiting substrate and measured the extracellular concentrations of nitrate and nitrite over time. We performed these experiments at pH 6.5 (strong nitrite toxicity) and 7.5 (weak nitrite toxicity).We observed a substantial accumulation of extracellular nitrite regardless of the pH (Fig. 3A, B). When grown at pH 6.5 (strong nitrite toxicity), extracellular nitrite accumulated to a concentration comparable to the initial nitrate concentration (measured maximum extracellular nitrite concentration, 11.8 mM; measured initial nitrate concentration, 12.0 mM) and was subsequently consumed to below the detection limit (Fig. 3A). When grown at pH 7.5 (weak nitrite toxicity), extracellular nitrite again accumulated to a concentration comparable to the initial nitrate concentration (measured maximum extracellular nitrite concentration, 11.7 mM; measured initial nitrate concentration, 12.9 mM) and was subsequently consumed to below the detection limit (Fig. 3B). During growth at pH 6.5, substantial nitrite consumption did not begin until a prolonged period of time after nitrate consumption was complete, resulting in a relatively long duration of nitrite availability (Fig. 3A). During growth at pH 7.5, in contrast, substantial nitrite consumption began immediately after nitrate consumption was complete, resulting in a relatively short duration of nitrite availability (Fig. 3B). The longer duration of nitrite availability at pH 6.5 indicates that the duration of the niche created by the generalist for the specialist depends on pH.Fig. 3: Growth and nitrogen oxide dynamics of the generalist in batch culture.We grew the generalist alone in a bioreactor at A pH 7.5 (weak nitrite toxicity) or B pH 6.5 (strong nitrite toxicity) under anoxic conditions with nitrate as the growth-limiting substrate. Blue squares are measured extracellular nitrate concentrations, yellow triangles are measured extracellular nitrite concentrations, and black circles are measured cell densities. We measured extracellular nitrate and nitrite concentrations with IC and cell densities with FC. C Measured durations of nitrite availability for the generalist growing in batch culture. We grew the generalist alone in 96-well microtiter plates under anoxic conditions with nitrate as the growth-limiting substrate. Open symbols are durations of nitrite availability at pH 6.5 and closed symbols are durations of nitrite availability at pH 7.5. Each symbol is an independent biological replicate.Full size imageTo routinely quantify the duration of nitrite availability, we grew the generalist alone with varying amounts of nitrate as the growth-limiting substrate. We then quantified the length of time from when the growth rate with nitrate was maximum to when the growth rate with nitrite was maximum. This cell density-based proxy measure is valid because the growth of the generalist is directly linked to the consumption of nitrate and nitrite (Fig. 3A, B). The cell density of the generalist was initially linearly correlated with nitrate consumption at both pH 6.5 (strong nitrite toxicity) (two-sided Pearson correlation test; r = −0.96, p = 1.5 × 10–8, n = 15) (Fig. 3A) and 7.5 (weak nitrite toxicity) (two-sided Pearson correlation test; r = −1.00, p = 2.2 × 10–16, n = 30) (Fig. 3B). After nitrate was depleted, the cell density of the generalist became linearly correlated with nitrite consumption at both pH 6.5 (strong nitrite toxicity) (two-sided Pearson correlation test; r = −0.97, p = 3 × 10–4, n = 7) (Fig. 3A) and 7.5 (weak nitrite toxicity) (two-sided Pearson correlation test; r = −0.97, p = 6.8 × 10–10, n = 16) (Fig. 3B). We further validated our cell density-based approach by testing for concordance with our IC-based direct measures of the duration of nitrite availability. We observed a significant positive and linear relationship between the cell density- and IC-based measures (two-sided Pearson correlation test; r = 0.999, p = 0.023, n = 3) (linear regression model; slope = 1.19, intercept = −2.31, r2 = 0.99) (Supplementary Fig. S2), which further validates our cell density-based approach to routinely estimate the duration of nitrite availability.Using our cell density-based approach, we found that the duration of nitrite availability was significantly longer at pH 6.5 (strong nitrite toxicity) than at 7.5 (weak nitrite toxicity) regardless of the initial nitrate concentration (two-sample two-sided t-tests; Holm-adjusted p  0.92, Holm-adjusted p  0.6), and thus followed model predictions (Fig. 4A). However, when the specialist was initially rare (measured initial log rS/Gs of –3.19, –2.65, and –0.88), the relative abundances of the specialist continuously decreased between the third and twelfth transfers (Mann–Kendall trend tests; tau = –0.61 to –0.89, p  0 were dominated by phenotype C (dominant ancestral phenotype with a long time delay between nitrate and nitrite consumption), while generalist isolates from co-cultures with initial rS/Gs  More

Hydrological changes in high-latitude peatlandsWe observed three hydrological pathways, i.e., drying, wetting, and fluctuating, for both peatland clusters, non-permafrost and permafrost peatlands (Fig. 2). Approximately 54% of the studied peatlands have shifted towards drier surface conditions since 1800 CE and more intensively since 1900 CE (Fig. 2a, d), which is in line with the post-LIA warming. The overall change point to drier conditions was dated to ca. 1950 CE for non-permafrost sites and ca. 1890 CE for the permafrost compiled group. Approximately 32% of the studied peatlands have shifted towards wetter conditions (Fig. 2b, e). The overall shifting point to wetter conditions was dated to ca. 1995 CE for non-permafrost peatlands and to ca. 1990 CE for permafrost peatlands. Wetting has been especially intensive since 1900 CE for non-permafrost peatlands and since 1950 CE for permafrost peatlands. Interestingly, the data showed that in permafrost peatlands a significant dry shift always preceded a wet shift (Fig. 2e). Approximately 14% of the studied peatlands indicated no clear trend, with fluctuating hydrological conditions (Fig. 2c, f).Non-permafrost peatlands generally showed spatially extensive drying across the northern high latitudes, apart from northeastern Canada, where a wetting trend was more frequently observed. Permafrost peatlands, however, were more variable, with some drying, some wetting, and no overall coherent regional pattern was visible (Fig.1a, b). It should be noted that peatlands synthesized here have undergone little or no direct human impact, i.e., their surface hydrology was not significantly affected by human disturbances such as drainage, when compared to, for example, central European peatlands discussed in Swindles et al. (2019)5. This implies that climate and/or local topographical forcing are the predominant hydrological drivers in this study. The dataset is to some extent biased as there are more non-permafrost records from northeastern Canada but more permafrost records from northern Sweden and this might result in regional overestimation to either wetting or drying trends. Nevertheless, the pattern of diverse timing of the hydrological shifts between the individual coring points (Fig. 2) indicates the variability in sensitivity of different regions/peatlands to climate changes.Potential links to climate change and permafrost dynamicsThe comparison of the reconstructed water table and climate data suggests that climate, especially summer temperature, has played an important role in shaping the peatland water table (Fig. 1c–f). The pattern detected here for non-permafrost peatlands, an extensive drying, is comparable to that observed for central European peatlands5. In addition to direct climate forcing, a recent acceleration of peat accumulation might partly explain the drying trend by disconnecting the peatland surface from the water table17. However, for northeastern Canada a wetting trend has been observed more often, possibly regulated by the regional climate that shows clearly less warming in the focused period compared to other regions (Fig. 1c, d).Permafrost initiation in the past caused a peat surface uplift and is probably detected as a dry phase (Fig. 2d, e). Post-LIA warming-induced increase in evapotranspiration may have strengthened the surface drying which originally resulted from surface-uplift and probably mitigated the gradual wetting related to permafrost thawing11. The level of warming has varied among the regions. In some areas such as northeastern Canada temperature has increased less and, when combined with higher precipitation or higher effective moisture level, may have caused surface wetting in permafrost peatlands. This is a direct climate forcing rather than permafrost thawing, which is a consequence of climate warming, i.e., more indirect climate forcing. To date, it is yet challenging to estimate any one tipping point of warming that might trigger permafrost thawing, as the local conditions vary from bottom ground soil conditions to hydrology and vegetation. The consequent wetting or drying depends on evapotranspiration and ice richness etc., which further challenges the prediction of hydrological conditions of permafrost peatlands.The divergent three moisture patterns may occur in the same region and even in the same peatland, especially if the permafrost is present. This complex response pattern is well supported by the records from the Abisko region, Sweden, where replicated sampling was carried out, and captured different successional stages of local permafrost peatlands7,18. In contrast to permafrost peatlands, non-permafrost peatlands are more likely to experience a more consistent ecosystem response pattern19 as supported by the replicated records suggesting the same pattern happening simultaneously in several regions (Fig. 1a). The fluctuating pattern of many records reported here suggests that the past and recent climate has not yet caused a state change in hydrological conditions.Insights into carbon dynamics and future perspectiveGenerally, our results suggest that the recent climate warming has caused hydrological shifts in most high-latitude peatlands, highlighting its pronounced effect on shaping peatland moisture balance, and further on driving peatland C balance. It has been reported that a 1-cm water-table drawdown would increase 3.3–5.0 mg CO2 m−2 h−1 and decrease 2.2–3.6 mg CO2-eq m−2 h−1 (CH4) to the atmosphere, and the average sensitivity of CO2 and CH4 combined was 0.8–2.3 mg CO2-eq m−2 h−1 cm−1 according to a global scale analysis, including sites from high-latitudes3. However, it should be noted that the sensitivity of greenhouse gas fluxes to the magnitude of hydrological changes might vary among different regions and peatland types. It appears that most pan-Arctic peatlands are undergoing a drying trend, that may lead to a decreased C sink capacity3,19, if not compensated by increased C uptake from the atmosphere20.It is very likely that over the 21st century warming in high latitudes will continue to be more pronounced than the global average21. Precipitation is projected to increase, albeit with large regional variability. Also, extreme events with heavy rainfall and drought are becoming more frequent and intense22. It is estimated that about 20% of permafrost zone is experiencing accelerated and abrupt permafrost thaw that is likely causing wetting conditions4, while gradual permafrost thaw has been observed across the circumpolar regions23. Both an increase in precipitation and permafrost thaw might mitigate the drying pressure caused by warming and increased evapotranspiration. However, abrupt permafrost thaw in peatlands can result in a rapid (over years to decades) loss of C from the formerly frozen permafrost peat, causing these peatlands to be a net source of C to the atmosphere before post-thaw accumulation returns them to a net sink (centuries to millennia)12,13. The future C sink and source function of peatlands is a key element in contributing to climate change, but the observed divergent pathways of peatland hydrological successions further challenge the projections of high-latitude peatland C sink and source dynamics. Conversely, it clearly highlights the importance of climate forcing in peatland succession scenarios. Our study reveals that the response of high-latitude peatlands to changing climate conditions is complex. We detect variable ecohydrological trajectories, and in the future, these will determine the C sink capacity of northern peatlands. The observed patterns inevitably create challenges for the climate change modelling community. How to capture the highly heterogenic successional pathways of northern peatlands needs to be a key research focus. More

To assess coral holobiont health, we used the commercially available Urinalysis strips (Accutest URS-14 100 strips Urinalysis Reagent Test Paper; ca. $11–35 USD/ bottle) that are designed to detect disease markers in humans, potentially indicating diabetes, metabolic abnormalities, liver diseases, kidney function, and urinary tract infections. There are ten tests on each strip that provide an initial assessment of health using the following standardized markers: leukocytes, nitrite, urobilinogen, protein, pH, blood, specific gravity, ketone, bilirubin, and glucose. To apply this test to coral extracts, we used ambient and stressed nubbins in a comparative approach to identify trends in the data. We presumed that some of the tests are likely not applicable to corals, whereas others such as ketones and leukocytes, which target conserved animal pathways (see below) could prove useful. Samples from three different coral experiments (two bleaching experiments and one environmental survey) were used to assess the response and utility of the different tests on each strip. The first bleaching experiments was done in 2019 on two Hawaiian species: Montipora capitata (stress resistant) and Pocillopora acuta (stress sensitive); this research has been previously described in detail11,12. Briefly, three M. capitata colonies (different genotypes) were selected, with samples collected from each in triplicate at three timepoints (referred to as T1, T3, and T5) that span a period of prolonged thermal stress (16 days). The second bleaching experiment was done in 2021 and included nubbins from M. capitata, P. acuta, and P. compressa. These corals were maintained for 9 days under heat treatment or ambient conditions, with samples collected at the beginning and end of the experimental period from both conditions (see Methods). Finally, the third experiment was done in June 2021 on wild colonies of M. capitata, P. acuta, and P. compressa (Fig. 1A) from Kāneʻohe Bay, Oʻahu, Hawaiʻi (Fig. 1B) and analyzed to assess the extent of natural variability in the test strip results. It should be noted that the test strip results from each experiment are independent (i.e., analyzed separately to measure relative differences within the test population) due to differences in tissue freezing time, storage length, and handling, that affect the results.Figure 1Analysis of Hawaiian corals. (A) The three targeted species in Kāneʻohe Bay. Images created by D. Bhattacharya. (B) Sites of wild coral collection (marked by the yellow circles) in Kāneʻohe Bay in June 2021. Nubbins from two colonies (n = 3) of M. capitata were collected from six reefs: Reef 8, 9, Reef 13, Reef 27, Reef 41, and Reef 43 and from three sites near the Hawaiʻi Institute of Marine Biology (HIMB) on Coconut Island (Moku o Loʻe). This image was adapted from the Pacific Islands ocean observing system (https://www.pacioos.hawaii.edu/projects/coral/)33.Full size imagePortable device for test strip analysisTo facilitate field testing and allow accurate measurement of test strip results in the R, G, and B channels, a combination of 3D printing technology and computer vision was applied to the problem. A custom opaque phone holder was produced using 3D printing (Ultimaker Inc. [https://ultimaker.com]) to control light source quality and quantity during the procedure and for calculating test strip scores (Fig. 2A). This portable phone holder was designed to allow field personnel to conduct experiments and collect results in a convenient and user-friendly manner. Inside the black opaque phone holder, two LED diodes served as a controllable and consistent light source to eliminate noise introduced by any change in light levels (between or during) the test strip reactions, allowing for more accurate and consistent data collection19. Frames of the videos (taken using the smart phone positioned in the holder) were captured for different reagents according to the recommended reading time given by the user manual. For test strip RGB value measurement (Fig. 2B), an automated machine learning method, shown in Fig. 2C, was used as a replacement for traditional, manual methods. Our computer vision workflow, named TestStripDX, uses YOLOv418, which is one of the most mature, accurate, and popular computer vision models available, to isolate a target feature in an image and provide the metadata associated with the object (in this case, each test along the strip), such as position within the coordinates of a rectangular box. Sample pictures of strips were manually annotated and used for training the TestStripDX pipeline (Supplementary Data 1): i.e., these images were used to train a custom detector to identify the different reagent tests along a strip and thereafter, to provide the relevant RGB values for each test (Fig. 2C). Figure 2D shows the comparison between data collected using the manual (ImageJ) versus automated (TestStripDX) methods for a partial list of samples and reagent strips (data shown in Table S1). The very close association of the regression line (R2 = 0.996) with the artificial diagonal line validates the utility and accuracy of the computer vision method.Figure 2Portable instrument for test strip analysis. (A) The 3D-printed phone holder used to analyze the test strips. (B) Example test strip. (C) Flow chart for the training and application of the machine learning method (TestStripDX) used to analyze test strips. (D) Comparison of test strip colorimetric measurement values produced by TestStripDX and ImageJ, for a set of representative images (Table S1). The enforced diagonal matches very closely to the regression line.Full size imageCoral color score measurementsAs described above, color scores provide a proxy for coral health, and are generated by measuring the RGB channels of the bleached (unhealthy) and brown (healthy: i.e., pigmentation provided by the algal symbionts) areas of coral nubbins. For the CoralDX workflow, we trained a custom detector that can recognize coral nubbins, as well as red, green, and blue colored blocks (standards) which are used to normalize the R, G, and B channels in each image (see Supplementary Data 2). The images taken in this case were from a lab environment (Fig. 3A), but this approach could potentially be used in any location provided a background with a uniform color is used in the image. To achieve this goal, a background panel containing the red, blue, and green color blocks would be placed behind (smaller) coral nubbins to make the measurements. It is clear that for large colonies, this approach may prove challenging to apply but we expect that additional testing and modification to the method will allow us to design a better-suited tool for field use. To achieve our goal of automatically obtaining colorimetric measurements of coral nubbins, we needed to devise a method that provides YOLOv4 with areas that are not limited by the standard rectangular boxes used as input for this model. To accommodate irregular nubbin shapes, an additional step (training and testing steps are shown in Fig. 3B) was added to the automated method (Fig. 2C). Our approach uses computer vision-based edge detection to eliminate most of the background surrounding the edges of the coral nubbins, allowing for accurate quantification of the RGB values of the targeted piece of coral. After edge detection, we obtain a picture with a black background highlighting the coral shape as a “mask” (Fig. 3C; 2nd image from left). We then place the mask onto the original coral image, measuring non-zero R, G, and B values (Fig. 3C; 3rd and 4th images from left). This method is superior to selecting coral areas for manual analysis using tools such as the handsfree selection function in ImageJ, which are generally hard to manipulate, have low fault tolerance, and are time-consuming. The RGB values extracted from the coral nubbins and color blocks were investigated using principal component analysis (PCA) to generate Euclidean distances (color scores) among coral nubbins according to treatment, time point, and colony11,16. In Fig. 3D, the correlation between bleaching scores generated using CoralDX and ImageJ for M. capitata and P. acuta are presented for a representative set of nubbins cultured under ambient or thermal stress conditions in the 2019 bleaching experiment (Tables S2, S3). As is apparent, the regression lines show strong correlations (R2 = 0.968 for M. capitata and R2 = 0.991 for P. acuta), supporting the utility of the automated method. The data in both cases are very closely associated with the artificial diagonal line shown in the images, substantiating the strong positive correlation between the scores from the two approaches. The small differences in score values between the methods is primarily explained by difficulties in cropping the edges of coral nubbins that are heavily bleached; the contrast between the coral nubbin and the white background is lessened, creating discrepancies between the nubbin edges identified by CoralDX and ImageJ. This issue can be addressed by testing different color backgrounds to find the optimal set-up. Another potential contributing factor is the differences in the area selected for the color blocks between manual (ImageJ) versus automated methods, although this might only have a minor effect on the results. The CoralDX workflow is not computationally intensive and easily portable to other platforms, such as personal computers and smart phones, allowing for easy deployment in the field.Figure 3Analysis of coral color scores. (A) Example image of M. capitata coral nubbins used for automated color score analysis, with target areas marked with the black (coral nubbins) and yellow (color standards) boxes. The nubbin on the far right is used to demonstrate the masking procedure. (B) Flow chart for the training and application of the machine learning method (CoralDX) used to analyze coral color scores. (C) Example of image processing for one M. capitata coral nubbin showing (from left to right) the original, masked, segmented, and masked RGB nubbin, with the final (right) image used for color score measurements. (D) Comparison of color score values produced by CoralDX and ImageJ, for a set of representative coral nubbins from the 2019 bleaching experiment (Tables S2, S3). The enforced diagonal matches very closely to the regression line in both analyses.Full size imageTest strip resultsArmed with these new tools, we generated the test strip data for different coral samples, where larger “relative enzymatic activity” (REA, measured using the RGB values) values indicate increased reactivity (i.e., increased levels of products targeted by the test). The Accutest URS-14 100 strips test for ketones (using the sodium nitroprusside reaction) measures acetoacetate and assumes the presence of β-hydroxybutyrate and acetone. The former (β-hydroxybutyrate) acts as a signal to regulate metabolism and maintain energy homeostasis during nutrient deprivation. In this process, β-hydroxybutyrate is converted to acetoacetate. Ketone bodies are transported into tissues and converted into acetyl-CoA by thiolases, which then enters the TCA cycle and is oxidized in the mitochondria for energy. Bleaching in corals which are incapable of obtaining adequate energy stores through heterotrophy results in diminished growth rates, degraded reproductive capacity, amplified susceptibility to disease, and elevated mortality rates for the entire colony20. Although ketosis has not been explored in cnidarians, transcriptomic data generated from the M. capitata samples measured in this study12 demonstrate expression of the KEGG pathway for degradation of ketone bodies (Fig. S1). The combination of time-point and treatment (field, T1-AT, T1-HT, T3-HT, T5-HT) was the most significant factor impacting the ketone REA scores (p-value = 0.010) (see details of the PERMANOVA analysis in the “Methods” section). Given this framework, we find that the M. capitata samples remain steady throughout the bleaching period, except for a decrease in enzymatic activity at T3-HT (Fig. 4A). This result is supported by the transcriptomic data, which shows this pathway to be uniformly expressed at all timepoints, except T3-HT, when acetyl-CoA C-acetyltransferase is up-regulated in comparison to T1-HT (fold change [FC] = 1.52) and down-regulated at T5-HT (FC = − 1.84) (Table S4). A possible explanation for this result is that at T3-HT, when the first evidence of bleaching was present, the photosynthetic rate of the symbiotic algae was elevated due to the thermal stress, resulting in greater energy production and a decrease in the abundance of ketone bodies within the coral, but without sufficient stress to cause significant bleaching. During this time, acetyl-CoA C-acetyltransferase enzymatic activity favored the production of acetoacetyl-CoA. However, as dysbiosis continued and the corals no longer had access to the photosynthetic products provided by the symbionts, they produced more acetoacetate and ketosis was detected by the test strips. This response occurs despite the observation that M. capitata can persist heterotrophically and meet much of its energy needs in the absence of algal symbionts21. Interestingly, the three M. capitata field samples show similar amounts of ketone bodies to lab stressed corals, suggesting the presence of stressors in the natural environment.Figure 4Test strip results from the 2019 bleaching experiment. (A) Ketone test strip results for M. capitata, showing genotype-specific (see legend) differences in response. (B) Leukocytes test strip results for M. capitata, showing genotype-specific (see legend) differences in response. These are standard box plots, with the boxes representing the first (Q1) to third (Q3) quartiles. The lines in the boxes are the median (Q2) values and lines (“whiskers”) extending beyond the boxes are the minimum and maximum values, excluding outliers. (C) PCA of the ketone and leukocytes test strip data for the field, ambient, and T1-HT, T3-HT, and T5-HT timepoints.Full size imageThe leukocyte test measures the activity of leukocyte esterase (presence of white blood cells) and other signs of infection in human subjects. This test putatively assesses the coral innate immunity response, which includes the same phases in response to infection and loss of tissue integrity as other invertebrates: recognition, signaling, and effector response22. Corals contain multiple types of immune cells, such as amoebocytes and fibroblasts23. Amoebocytes are amoeboid cells residing in the mesoglea that remove necrotic tissue, encapsulate foreign particles, and generally display phagocytic activity to aid in organism defense against pathogens, which is the cnidarian principal mechanism of immunity24. Amoebocytes can be melanin-containing, agranular, or granular based cells, depending on the signaling pathway22. M. capitata shows an overall increase in enzymatic activity at T1-HT, signaling a heightened immune response (Fig. 4B). This relatively higher level of enzymatic activity decreases at T3 and T5. The field samples show levels that are comparable to the T3-HT and T5-HT thermal stress acclimated colonies and the T1 ambient (Amb) colonies. Again, a combination of time-point and treatment was the most significant factor impacting the leukocyte REA scores (p-value = 0.001). Additional research needs to be done to understand the cause of this response. Nonetheless, the results are consistent with the widely accepted hypothesis that M. capitata adapts well to bleaching conditions12. PCA of the M. capitata test strip results shows a clear separation of the T1-HT coral data from the T3-HT and T5-HT values, with the ambient and field samples intermixed among the latter two timepoints (Fig. 4C). This result again highlights the initial robust response to stress by M. capitata followed by acclimation to the heat treatment that is reflected in the field samples.A noteworthy aspect of the test strip results is the divergence in response to thermal stress among different colonies. This result has also been found for coral metabolomic data11. As described above, each holobiont integrates a complex set of biotic interactions between the host animal and microbiome, explaining the high variation in ketone and leukocytes test results, often between replicate nubbins from one colony and more frequently, between different colonies (Fig. 4A). Existing data using omics methods demonstrate that the stress response of the coral holobiont varies from colony to colony15. The metabolome is controlled by the coral animal genotype, microbial consortium, and environmental conditions, among other factors, and can fluctuate greatly based on individual metabolite turnover rates and the timing of sampling25,26. Therefore, accounting for natural variation in the stress response phenotype and its importance for effective testing methods is a crucial aspect of our work. Our results demonstrate that broad population level sampling (dozens to 100 s of colonies/genotypes) is likely needed to account for the inherent genetic and metabolic variation present in wild coral populations.We also did a more limited analysis of coral stress responses using the ketone and leukocyte test strips with three species (M. capitata, P. acuta, P. compressa) in a 2021 bleaching experiment in which we sampled multiple coral genotypes at time 0 and after 9 days of heat treatment (3ºC increase from ambient; see Methods). These results are based on analysis of 8–9 different coral genotypes (summarized in Fig. 5). Interpreted in the same way as described above, we see that there is substantial genotype-based variation in the stress response. Nonetheless, consistent with the 2019 data, M. capitata shows evidence of a thermal stress response in the ketone and leukocytes tests (Fig. 5A, B). P. acuta shows a more limited response, whereas P. compressa appears to have fully acclimated to the stress regime with lowered reactivity at the end of the experiment. Species identity was the most significant factor impacting the leukocyte and ketone REA scores (p-value = 0.002 and 0.033, respectively), but a combination of species and treatment (AT vs. HT) was found to be significant for leukocytes (p-value = 0.026). PCA of the M. capitata test results shows separation between the ambient samples and some of the high temperature treated samples along PC1; both the ketone and leukocytes tests contribute to the spread of samples along PC1. This reinforces the conclusion that there is evidence for a thermal stress response in the ketone and leukocytes tests of M. capitata however, the significant genotypic variability (particularly apparent when compared with Fig. 4) obscures the differences between the conditions for some of the samples. PCA of the P. acuta and P. compressa test results mirrors our conclusion that the ketone and leukocytes tests in these species show a limited response to stress, with no separation between the samples from the different conditions.Figure 5Test strip results from the 2021 bleaching experiment for three Hawaiian coral species. (A) Ketone test strip results for M. capitata, P. acuta, and P. compressa showing genotype-specific variation in response under the ambient (Amb) and high temperature (HT) treatments. (B) Leukocytes test strip results for M. capitata, P. acuta, and P. compressa showing genotype-specific variation in response under the ambient and high temperature (HT) treatments. These are standard box plots, with the boxes representing the first (Q1) to third (Q3) quartiles. The lines in the boxes are the median (Q2) values and lines (“whiskers”) extending beyond the boxes are the minimum and maximum values, excluding outliers. (C) PCA of the ketone and leukocytes test strip data for the Amb and HT treatments for the three Hawaiian coral species.Full size imageAnalysis of wild populationsTo assess natural variation, we collected apparently healthy M. capitata, P. acuta, and P. compressa nubbins (3 replicate nubbins per colony) from six reefs in Kāneʻohe Bay and from three sites near the Hawaiʻi Institute of Marine Biology on Coconut Island (Moku o Loʻe) (see Fig. 1) and analyzed these tissue extracts using the ketone and leukocytes tests. This analysis shows wide variation in the results with some interesting exceptions. Species identity was the most significant factor impacting the leukocytes and ketone REA scores (p-value = 0.002 and 0.001, respectively), but for leukocytes, colony identity, regardless of species, was also found to be a significant factor (p-value = 0.040). The M. capitata ketone test results are consistent among different reefs and within the same colony with the exception of some colonies (e.g., Colony 8 from Reef 9 and Colony 16 from Reef 43) that show wide intra-colony variation (Fig. 6A). Most of the ketone data for M. capitata fall between REAs of 10–20. In contrast, P. acuta shows more variation in the wild populations for the ketone test, suggesting that many of these coral colonies live under stressful conditions in the field (Fig. 6B). A similar situation to M. capitata, in terms of REAs, holds for the ketone test of P. compressa colonies that show more limited variation (Fig. 6C). The leukocytes test shows high variation for M. capitata (Fig. 6D), P. acuta (Fig. 6E), and P. compressa (Fig. 6F) colonies. These results again point out the complex nature of genome-environment interaction with respect to metabolic syndromes, both at the colony level and among different regions (replicates) of the same colony. For example, P. compressa Colonies 13–15 from Reef 9 show little to no intra-colony variation for the ketone test, yet another colony from this reef (Colony 16) shows high variation among replicates (Fig. 6C). In contrast, P. compressa Colonies 13–15 are far more variable when using the leukocytes test (Fig. 6F). On the basis of the more predictable lab-based results reported above, we interpret these “noisy” field data as evidence of the immense variation in the stress phenome of wild coral populations. Overall, the field results indicate that a starting set of test strip values, followed by repeated field sampling over time of wild colonies may be needed for accurate stress diagnosis, rather than the one-time measurement approach used here. Clearly, more work is needed with wild colony analysis, particularly under varying degrees of thermal stress and apparent bleaching, to fully realize the potential of the technique we present.Figure 6Test strip results from analysis of the 2021 collection of three Hawaiian coral species from the wild. (A and D) Ketone and leukocytes test strip results, respectively, for M. capitata showing intraindividual variation and among the different reefs (see legend) that were sampled (Fig. 1B). (B and E) Ketone and leukocytes test strip results, respectively, for P. acuta showing intraindividual variation and among the different reefs that were sampled. (C and F) Ketone and leukocytes test strip results, respectively, for P. compressa showing intraindividual variation and among the different reefs (see legend) that were sampled. These are standard box plots, with the boxes representing the first (Q1) to third (Q3) quartiles. The lines in the boxes are the median (Q2) values and lines (“whiskers”) extending beyond the boxes are the minimum and maximum values, excluding outliers.Full size imageAnalysis of transcriptomic dataTo identify pathways that may support the M. capitata leukocytes test strip results, which showed the most response in terms of change in REA at T1-HT (Fig. 4B), we analyzed existing transcriptomic (RNA-seq) data derived from the same coral nubbins. The RNA-seq and metabolomic data from these samples have been previously analyzed11,12. Here we searched for co-expression modules that contain genes that are up-regulated at the start of the thermal stress regime (T1-HT) when compared to the ambient treatment. It is at this timepoint that we find a strong cross-reaction with the leukocytes test, followed by loss of cross-reactivity at T3-HT and T5-HT (back down to T1-Amb levels), putatively indicating acclimation (Fig. 4B). As described above, the wound healing response in corals is complex and the (Urinalysis) leukocytes results need to be interpreted as a syndrome involving multiple pathways of stress and immune response. With these considerations in mind, we identified a module (Module 2; see Williams et al.12) of up-regulated genes that contains several markers associated with the coral stress response (Fig. 7). These include a tumor necrosis factor-activated receptor (TNFR)-Cys domain-containing protein (fold-change [FC] = 1.12) that is a well-known mediator of apoptosis and cell death that is functionally conserved in corals. Some members of the TNF family are associated with bleaching27. The most highly up-regulated gene in this module is C-type lysozyme 2 (FC = 2.42) that provides an anti-microbial function (e.g., digestion of peptidoglycan), and is likely expressed as a result of stress-induced dysbiosis in M. capitata. Other markers of stress that are up-regulated in Module 2 include E3 ubiquitin-protein ligase (FC = 1.03) involved in protein degradation, two protein disulfide-isomerase (FCs = 1.32, 1.09) involved in cellular defense against protein misfolding via chaperone activity28, and a metalloproteinase inhibitor 3 (FC = 1.42) which likely functions as a physiological anti-inflammatory molecule29. These data, although not directly substantiating the leukocytes test strip results, provide evidence that the wound healing and immune response pathways were up-regulated in Module 2 at T1-HT (albeit weakly, due to the stress resilience of M. capitata) as suggested by Fig. 4B.Figure 7Gene co-expression analysis. Module 2 representing significantly up-regulated genes in M. capitata from T1-HT in the 2019 differential gene expression analysis12. This module is enriched in genes involved in the wound healing and immune response. The legend for level of up-regulation is shown. Putative gene annotations are also shown.Full size image More

Erwin, D. H. & Tweedt, S. Ecological drivers of the Ediacaran-Cambrian diversification of Metazoa. Evol. Ecol. 26, 417–433 (2012).Article 

Google Scholar 
Laflamme, M., Darroch, S. A., Tweedt, S. M., Peterson, K. J. & Erwin, D. H. The end of the Ediacara biota: Extinction, biotic replacement, or Cheshire Cat?. Gondwana Res. 23, 558–573 (2013).ADS 
Article 

Google Scholar 
Mángano, M. G. & Buatois, L. A. Decoupling of body-plan diversification and ecological structuring during the Ediacaran-Cambrian transition: Evolutionary and geobiological feedbacks. Proc. R. Soc. B-Biol. Sci. 281, 20140038 (2014).Article 

Google Scholar 
Mángano, M. G. & Buatois, L. A. The Cambrian revolutions: Trace-fossil record, timing, links and geobiological impact. Earth-Sci. Rev. 173, 96–108 (2017).ADS 
Article 

Google Scholar 
Mángano, M. G. & Buatois, L. A. The rise and early evolution of animals: Where do we stand from a trace-fossil perspective?. Interface Focus 10, 20190103 (2020).PubMed 
PubMed Central 
Article 

Google Scholar 
Darroch, S. A. et al. Biotic replacement and mass extinction of the Ediacara biota. Proc. R. Soc. B 282, 20151003 (2015).PubMed 
PubMed Central 
Article 

Google Scholar 
Darroch, S. A., Smith, E. F., Laflamme, M. & Erwin, D. H. Ediacaran extinction and Cambrian explosion. Trends Ecol. Evol. 33, 653–663 (2018).PubMed 
Article 

Google Scholar 
Schiffbauer, J. D. et al. The latest Ediacaran Wormworld fauna: Setting the ecological stage for the Cambrian explosion. GSA Today 26, 4–11 (2016).Article 

Google Scholar 
Zamora, S., Deline, B., Javier Álvaro, J. & Rahman, I. A. The Cambrian Substrate Revolution and the early evolution of attachment in suspension-feeding echinoderms. Earth-Sci. Rev. 171, 478–491 (2017).ADS 
Article 

Google Scholar 
Hantsoo, K. G., Kaufman, A. J., Cui, H., Plummer, R. E. & Narbonne, G. M. Effects of bioturbation on carbon and sulfur cycling across the Ediacaran-Cambrian transition at the GSSP in Newfoundland, Canada. Can. J. Earth Sci. 55, 1240–1252 (2018).ADS 
CAS 
Article 

Google Scholar 
Boyle, R. A., Dahl, T. W., Bjerrum, C. J. & Canfield, D. E. Bioturbation and directionality in Earth’s carbon isotope record across the Neoproterozoic-Cambrian transition. Geobiology 16, 252–278 (2018).CAS 
PubMed 
Article 

Google Scholar 
van de Velde, S., Mills, B. J., Meysman, F. J., Lenton, T. M. & Poulton, S. W. Early Palaeozoic ocean anoxia and global warming driven by the evolution of shallow burrowing. Nat. Commun. 9, 2554 (2018).ADS 
PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Gougeon, R. C., Mangano, M. G., Buatois, L. A., Narbonne, G. M. & Laing, B. A. Early Cambrian origin of the shelf sediment mixed layer. Nat. Commun. 9, 1909 (2018).ADS 
PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Seilacher, A. & Pflüger, F. In Biostabilization of Sediments (eds. Krumbein W. E., Peterson D. M., & Stal L. J.) 97–105 (Bibliotheks und Informationsystem der Carl von Ossietzky Universität, 1994).Seilacher, A. Biomat-related lifestyles in the Precambrian. Palaios 14, 86–93 (1999).ADS 
Article 

Google Scholar 
Buatois, L. A. et al. Quantifying ecospace utilization and ecosystem engineering during the early Phanerozoic: The role of bioturbation and bioerosion. Sci. Adv. 6, eabb0618 (2020).ADS 
CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Hagadorn, J. W. & Bottjer, D. J. Restriction of a late Neoproterozoic biotope; suspect-microbial structures and trace fossils at the Vendian-Cambrian transition. Palaios 14, 73–85 (1999).ADS 
Article 

Google Scholar 
Buatois, L. A. & Mangano, M. G. Early colonization of the deep sea: Ichnologic evidence of deep-marine benthic ecology from the Early Cambrian of northwest Argentina. Palaios 18, 572–581 (2003).ADS 
Article 

Google Scholar 
Clausen, S., Álvaro, J. J. & Zamora, S. Replacement of benthic communities in two Neoproterozoic-Cambrian subtropical-to-temperate rift basins, High Atlas and Anti-Atlas, Morocco. J. Afr. Earth Sci. 98, 72–93 (2014).ADS 
Article 

Google Scholar 
Mángano, M. G. & Buatois, L. A. In The Trace-Fossil Record of Major Evolutionary Events: Volume 1: Precambrian and Paleozoic (eds. Mángano M. G. & Buatois L. A.) 73–126 (Springer Netherlands, 2016).Bayet-Goll, A., Buatois, L. A., Mangano, M. G. & Daraei, M. The interplay of environmental constraints and bioturbation on matground development along the marine depositional profile during the Ordovician Radiation. Geobiology 20, 33–270 (2022).Article 
CAS 

Google Scholar 
Bayet-Goll, A., Daraei, M., Geyer, G., Bahrami, N. & Bagheri, F. Environmental constraints on the distribution of matground and mixground ecosystems across the Cambrian Series 2–Miaolingian boundary interval in Iran: A case study for the central sector of northern Gondwana. J. Afr. Earth Sci. 176, 104120 (2021).Article 

Google Scholar 
Minter, N. J. et al. In The Trace-Fossil Record of Major Evolutionary Events: Volume 1: Precambrian and Paleozoic (eds. Mángano M. G. & Buatois L. A.) 157–204 (Springer Netherlands, 2016).Minter, N. J. et al. Early bursts of diversification defined the faunal colonization of land. Nat. Ecol. Evol. 1, 0175 (2017).Article 

Google Scholar 
Nemec, W. & Steel, R. J. In Fan Deltas: sedimentology and tectonic settings (eds. Nemec W. & Steel R. J.) 3–13 (Blackie and Son, 1988).Postma, G. An analysis of the variation in delta architecture. Terra Nova 2, 124–130 (1990).ADS 
Article 

Google Scholar 
Prior, D. B. & Bornhold, B. D. Submarine sedimentation on a developing Holocene fan delta. Sedimentology 36, 1053–1076 (1989).ADS 
Article 

Google Scholar 
Piper, D. J. W., Kontopoulos, N., Anagnostou, C., Chronis, G. & Panagos, A. G. Modern fan deltas in the western Gulf of Corinth, Greece. Geo-Mar. Lett. 10, 5–12 (1990).ADS 
Article 

Google Scholar 
Rasmussen, H. Nearshore and alluvial facies in the Sant Llorenç del Munt depositional system: Recognition and development. Sediment. Geol. 138, 71–98 (2000).ADS 
Article 

Google Scholar 
Steel, R., Rasmussen, H., Eide, S., Neuman, B. & Siggerud, E. Anatomy of high-sediment supply, transgressive tracts in the Vilomara composite sequence, Sant LlorencË del Munt, Ebro Basin, NE Spain. Sediment. Geol. 138, 125–142 (2000).ADS 
Article 

Google Scholar 
Zavala, C. et al. Deltas: A new classification expanding Bates’s concepts. J. Palaeogeogr. 10, 23 (2021).ADS 
Article 

Google Scholar 
Ekdale, A. A. & Lewis, D. W. Trace fossils and paleoenvironmental control of ichnofacies in a late Quaternary gravel and loess fan delta complex, New Zealand. Palaeogeogr. Palaeoclimatol. Palaeoecol. 81, 253–279 (1991).Article 

Google Scholar 
Buatois, L. A. & Mángano, M. G. Ichnology: Organism-substrate Interactions in Space and Time (Cambridge University Press, 2011).Book 

Google Scholar 
Hovikoski, J., Uchman, A., Alsen, P. & Ineson, J. Ichnological and sedimentological characteristics of submarine fan-delta deposits in a half-graben, Lower Cretaceous Palnatokes Bjerg Formation, NE Greenland. Ichnos 26, 28–57 (2019).Article 

Google Scholar 
Sendra, J., Reolid, M. & Reolid, J. Palaeoenvironmental interpretation of the Pliocene fan-delta system of the Vera Basin (SE Spain): Fossil assemblages, ichnology and taphonomy. Palaeoworld 29, 769–788 (2020).Article 

Google Scholar 
Kreis, L. K. et al. Lower Paleozoic map series: Saskatchewan. Miscellaneous Report 2004–8 (CD-ROM) (Saskatchewan Industry and Resources, 2004).
Google Scholar 
Marsh, A. & Love, M. In Saskatchewan Ministry of the Economy, Saskatchewan Geological Survey, Saskatchewan Geological Survey Vol. Open File 2014-1, set of 156 maps (2014).Sawatzky, H. B., Agarwal, R. G. & Wilson, W. Helium prospects of southwest Saskatchewan. 26 (Saskatchewan Department of Mineral Resources, 1960).Fyson, W. K. Deadwood and Winnipeg stratigraphy in southwestern Saskatchewan. Report 64, 37 (Saskatchewan Department of Mineral Resources, 1961).Kent, D. M. Paleotectonic controls on sedimentation in northern Williston Basin area, Saskatchewan. AAPG Bull. 67, 1345–1345 (1983).
Google Scholar 
Kent, D. M. & Haidl, F. M. The distribution of Ashern and Winnipegosis strata (Middle Devonian) on the Swift Current Platform, southern Saskatchewan. Summary of Investigations, Miscellaneous Report 93-4, 201–206 (Saskatchewan Geological Survey, Saskatchewan Energy and Mines, 1993).MacEachern, J. A., Zaitlin, B. A. & Pemberton, S. G. A sharp-based sandstone of the Viking Formation, Joffre Field, Alberta, Canada; criteria for recognition of transgressively incised shoreface complexes. J. Sediment. Res. 69, 876–892 (1999).ADS 
Article 

Google Scholar 
Pemberton, S. G., Frey, R. W. & Bromley, R. G. The ichnotaxonomy of Conostichus and other plug-shaped ichnofossils. Can. J. Earth Sci. 25, 866–892 (1988).ADS 
Article 

Google Scholar 
Hall, J. & Whitfield, R. P. Paleontology. In US Geol. Expl. 40th Par. Rept. 4, 197–302 (1877).Walcott, C. D. Cambrian and Lower Ozarkian trilobites. Smithson. Misc. Coll. 75, 53–60 (1924).
Google Scholar 
Meek, F. B. Preliminary paleontological report, consisting of lists and descriptions of fossils, with remarks on the ages of the rocks in which they were found. In U. S. Geol. Surv. Terr. 6th Ann. Rept., 429–518 (1873).Walcott, C. D. Cambrian geology and paleontology of Cambrian trilobites. Smithson. Misc. Coll. 64, 157–258 (1916).
Google Scholar 
Harding, S. C. & Ekdale, A. A. Trace fossils and glauconitic pellets provide insight into Cambrian siliciclastic marine environments. Palaios 33, 256–265 (2018).ADS 
Article 

Google Scholar 
Shillito, A. P. & Davies, N. S. The Tumblagooda Sandstone revisited: Exceptionally abundant trace fossils and geological outcrop provide a window onto Palaeozoic littoral habitats before invertebrate terrestrialization. Geol. Mag. 157, 1939–1970 (2020).ADS 
Article 

Google Scholar 
Shillito, A. P. & Davies, N. S. Archetypally Siluro-Devonian ichnofauna in the Cowie Formation, Scotland: Implications for the myriapod fossil record and Highland Boundary Fault Movement. Proc. Geol. Assoc. 128, 815–828 (2017).Article 

Google Scholar 
Buatois, L. A. et al. The invasion of the land in deep time: integrating Paleozoic records of paleobiology, ichnology, sedimentology, and geomorphology. Integr. Comp. Biol. 0, 1–35. https://doi.org/10.1093/icb/icac059 (2022).Davies, N. S. & Gibling, M. R. Paleozoic vegetation and the Siluro-Devonian rise of fluvial lateral accretion sets. Geology 38, 51–54 (2010).ADS 
Article 

Google Scholar 
McMahon, W. J. & Davies, N. S. Evolution of alluvial mudrock forced by early land plants. Science 359, 1022–1024 (2018).ADS 
CAS 
PubMed 
Article 

Google Scholar 
Fedo, C. M. & Cooper, J. D. Braided fluvial to marine transition; the basal Lower Cambrian Wood Canyon Formation, southern Marble Mountains, Mojave Desert, California. J. Sediment. Res. 60, 220–234 (1990).
Google Scholar 
Eyre, B. Early Cambrian alluvial fan-deltas in the Georgina Basin, Australia. Aust. J. Earth Sci. 41, 27–36 (1994).ADS 
Article 

Google Scholar 
MacNaughton, R. B., Dalrymple, R. & Narbonne, G. M. Early Cambrian braid-delta deposits, MacKenzie Mountains, north-western Canada. Sedimentology 44, 587–609 (1997).ADS 
Article 

Google Scholar 
Muhlbauer, J. G. & Fedo, C. M. Architecture of a river-dominated, wave- and tide-influenced, pre-vegetation braid delta: Cambrian middle member of the Wood Canyon Formation, southern Marble Mountains, California, U.S.A. J. Sediment. Res. 90, 1011–1036 (2020).ADS 
Article 

Google Scholar 
McMahon, W. J., Davies, N. S. & Went, D. J. Negligible microbial matground influence on pre-vegetation river functioning: Evidence from the Ediacaran-Lower Cambrian Series Rouge, France. Precambrian Res. 292, 13–34 (2017).ADS 
CAS 
Article 

Google Scholar 
Mikuláš, R. Trace fossils from the Paseky Shale (Early Cambrian, Czech Republic). J. Czech Geol. Soc. 40, 37–44 (1995).
Google Scholar 
MacNaughton, R. B. & Narbonne, G. M. Evolution and ecology of Neoproterozoic-Lower Cambrian trace fossils, NW Canada. Palaios 14, 97–115 (1999).ADS 
Article 

Google Scholar 
Buatois, L. A. et al. Colonization of brackish-water systems through time: Evidence from the trace-fossil record. Palaios 20, 321–347 (2005).ADS 
Article 

Google Scholar 
Hofmann, R., Mángano, M. G., Elicki, O. & Shinaq, R. Paleoecologic and biostratigraphic significance of trace fossils from shallow- to marginal-marine environments from the Middle Cambrian (Stage 5) of Jordan. J. Paleontol. 86, 931–955 (2012).Article 

Google Scholar 
Mángano, M. G., Buatois, L. A., Hofmann, R., Elicki, O. & Shinaq, R. Exploring the aftermath of the Cambrian explosion: The evolutionary significance of marginal- to shallow-marine ichnofaunas of Jordan. Palaeogeogr. Palaeoclimatol. Palaeoecol. 374, 1–15 (2013).Article 

Google Scholar 
Mángano, M. G. et al. Were all trilobites fully marine? Trilobite expansion into brackish water during the early Palaeozoic. Proc. R. Soc. Lond. B Biol. Sci. 288, 20202263 (2021).
Google Scholar 
Siggerud, E. I. H. & Steel, R. J. Architecture and trace-fossil characteristics of a 10000–20000 Year, fluvial-to-marine sequence, SE Ebro Basin, Spain. J. Sediment. Res. 69, 365–383 (1999).ADS 
Article 

Google Scholar 
Lockley, M. G., Rindsberg, A. K. & Zeiler, R. M. The paleoenvironmental significance of the nearshore Curvolithus ichnofacies. Palaios 2, 255–262 (1987).ADS 
Article 

Google Scholar 
Folk, R. L. Petrology of Sedimentary Rocks (Hemphill Publishing Company, 1980).
Google Scholar 
Wentworth, C. K. A scale grade and class terms for clastic sediments. J. Geol. 30, 377–392 (1922).ADS 
Article 

Google Scholar 
Pettijohn, F. J., Potter, P. E. & Siever, R. Sand and Sandstone 2nd edn. (Springer, 1987).Book 

Google Scholar 
Ingram, R. L. Terminology for the thickness of stratification and parting units in sedimentary rocks. Geol. Soc. Am. Bull. 65, 937–938 (1954).ADS 
Article 

Google Scholar 
Reineck, H.-E. Sedimentgefüge im Bereich der südliche Nordsee. Abh. Senckb. Naturforsch. Ges. 505, 1–138 (1963).
Google Scholar  More