Eco-augmentation can be controversial, but in many ways it is nothing new. People have shaped their environments in directions they find appropriate since time immemorial. More recently, there have been deliberate, if perhaps opportunistic, attempts to create artificial alternatives to ecosystem components — for example, by sinking oil platforms to create artificial reefs.
Now there is talk of assisting the migration of certain organisms to restore function where it is needed or to assist evolution in developing organisms that can provide such function. Such assisted evolution might be achieved through selective breeding, which is already familiar to most people, or through genetic engineering, which many people find strange and sometimes threatening.
At the extreme, we could go beyond promoting genes that natural selection has already established and design organisms which fill particular niches in an ecosystem. How can we decide whether this or any other intervention would be safe and justified? When we look at an ecosystem, how do we decide if it needs our help; and if it does, how can we understand what the underlying problem is, and then how we might be able to fix it?
The ecosystems that I spend most time studying are the vast underwater kelp forests off the US West Coast, which are hugely important in building up fish stocks, cleaning up fertiliser run-off and sequestering carbon dioxide. These ecosystem services are estimated to be worth $52 billion a year to the West Coast economy alone,1 and so it is in our interests that these forests thrive.
We focus on the sunflower sea star, a large starfish which seems to play a key role in the ecosystems of kelp forests by eating sea urchins, which in turn eat kelp. There is evidence of this animal’s importance from an outbreak of disease which killed nearly all the sunflower sea stars in the southern half of their range between 2013 and 2015: kelp-forest ecosystems collapsed, having already been weakened by a marine heatwave.
Such incidents are likely to grow more common in the near future, so with colleagues we are working on a breeding programme for sunflower sea stars. As part of that, we want to be sure that the brood stock we use represents the genetic diversity of this species, and in particular we are looking for genes that are involved in resilience to sea-star wasting disease. If we can find such genes, we can breed sea stars that have them and so are more resistant to the disease.
We could do that through selective breeding using sea stars that are evidently able to survive the disease. Or we could identify the genes that confer this resistance, then single out those sea stars that have these genes to use in our breeding programme, thus speeding up the process. Or we could modify the genes of many sea stars to make them resistant, which would speed things up even more. Which is the correct path?
These aren’t just decisions for researchers, but for all those involved with the kelp forests. At a minimum, we would need regulatory permission to introduce any kind of bred sea stars into the kelp forests, but the discussions should go much further.
My lab also studies the marine lake ecosystems of the western Pacific. These unique ecosystems have prospered evolutionarily, having been isolated from the ocean for thousands of years — but their dynamics appear to be shifting. In Palau, for example, we’ve seen changes over the past 25 years associated with warming lake temperatures, where previously dominant, unique, species are now often absent. That is of obvious environmental significance, but it is economically important too: large numbers of tourists pay a $100 fee to visit “Jellyfish Lake”, whose previously millions of golden jellyfish, Mastigias papua etpisoni, live nowhere else. These temperature-sensitive animals are at risk as the climate changes; could the lake be geoengineered to keep it cool? It is, relatively speaking, a small, simple and self-contained ecosystem, so monitoring and managing the environment may be feasible. If amelioration works there, perhaps it can provide principles for a more interventionist form of eco-augmentation.
In contrast, the kelp-forest ecosystem on the West Coast spans thousands of kilometres. The sunflower sea star has been a big part of our work but it is just one species. We need to understand how all the organisms in these ecosystems interact, and how they vary both between different forests and within individual forests. Our focus is on how this diversity responds to stresses or changes in the environment at a genetic level, so that we can better understand the ways that species or populations might adapt to the new conditions.
This is not a new approach, in principle, but what is new is the toolset now available to us — principally those that allow us to sequence the unique genetic codes of individual organisms cheaply, easily and quickly.
That means we can see how genetics varies between, say, healthy and degraded regions of kelp forest, and find those variants that help the organism survive in different environments. Genomes evolve over generations, and so we are also looking at genetic processes that change much more quickly, including epigenetic responses — chemical changes in DNA that occur during an organism’s life and affect how its genes are expressed. We could also look for genes that are important in, say, signalling or predation. Our ultimate aim is to make tools that can quickly assess the health and interconnectedness of different forests.
Our proposal now is to build up the number of species we have covered. At the moment we are looking at the ecological data on interactions. We will pick species that have known roles in the ecosystem: the predators and the prey, for instance, or the symbionts and the hosts. At the moment we’re targeting approximately 15 species per habitat, but that is a small fraction. There are lots of species out there that we don’t know yet, and many ways in which genetics varies even within the same species. These may be driven by small differences in their environment: their depth in water or exposure to sunlight, for example. Such differences may prove to be significant as those environments change incrementally in coming years.
We also need a much better understanding of how ecological interactions are linked to genomic adaptations. We might expect species that interact more frequently to co-evolve — that is, they should exhibit shared patterns of genetic variation through time as the environment changes.
The genetics of creatures that can only live in symbiosis, for example, might be expected to change in unison. Species that have nothing in common would evolve independently. Species that have some key interactions — predators and prey, for example — should be somewhere in between. There’s interest in doing this with sunflower sea stars and sea urchins, for example: putting them in aquaria together and see how they respond, what their biochemical and gene expression is and whether it is subsequently accompanied by epigenetic modification.
We can look for these relationships between organisms through statistical analysis of their genetic data. That can then be built upon through fieldwork that tracks the rise and fall of wild populations, or the changing presence of different organisms at different locations. Ultimately, we could build model ecosystems which include individual interactions between species. But that requires a commitment to collecting genetic and ecological data in great detail at large scales on an ongoing basis.
People once improved bridge construction by, in a manner of speaking, building them and seeing whether they fell down. Then scientists better worked out the physics of bridges. That means engineers can now design more complex and larger bridges and build them, confident that they will stay up. Eco-augmentation is still at the trial-and-error stage. But in 25 years’ time, if we have been able to collect sufficient data, we should be able to model the evolution of ecosystems under climate change and human pressures, and how they could evolve as a result — and how we can help them to recover and thrive.
- A.M. Eger et al., ‘The value of ecosystem services in global marine kelp forests’, Nature Communications, 14, 1894 (2023), https://doi.org/10.1038/s41467-023-37385-0
