Engineering Ecosystem Resilience
Responsible research
Engineering ecosystem resilience means working with living systems we only partly understand, and whose value people see in different ways. Working with them responsibly matters as much as the science. It also means starting from open questions rather than settled answers: whether the new approaches explored in this space are feasible at all, and under what conditions, if any, any of them could be safe. This page sets out how we approach it: why it is hard, how we have gone about listening and learning, what ARIA can and cannot fund, and the specific commitments that govern the Accelerated Adaptation programme.
Context
Why getting this right matters
Interventions in nature have a long record of unintended consequences. Countless ‘no-risk’ interventions have turned out to be harmful, and even well-intentioned conservation breeding programmes have raised population numbers while collapsing genetic diversity and reducing long-term resilience. Ecosystems are among the most complex systems we know, our ability to predict how they will respond is limited, and some changes cannot be undone.
Nor is there a single answer to what a good outcome looks like. People hold multiple, sometimes conflicting, values: the intrinsic worth of species, their cultural significance, the integrity of ecosystems, and the present and future dependence of humanity on what nature provides. Regulatory frameworks have boundaries, risk modelling has gaps, and public trust, once lost, is slow to rebuild.
At the same time, the balance of risks has shifted. Climate extremes, land-use change, pathogens and pollution put one in four assessed animal and plant species at risk of extinction within a century. Reducing those pressures remains the first-order response. But as we get better at pinpointing where collapse is likely, we have to face the possibility that new kinds of intervention become an essential part of the toolkit. The risks of intervening remain real; the risks of maintaining the status quo have grown larger still. That is a reason to find out, not a decision to intervene.
The wider conservation community is reconsidering too. The International Union for Conservation of Nature (IUCN) has adopted a policy framework for the responsible, science-based use of synthetic biology in nature conservation (Motion 87). The Accelerated Adaptation programme’s contained, evidence-first approach is aligned with it. We treat that framework as a starting point rather than a licence: it sets out how such work should be governed, and it is for the work itself, and its oversight, to earn trust.
Neither side of that balance can be waved away. The task is to hold both: to find out, with humility and rigour, whether new options for nature can be built safely, and to do so with an unwavering commitment to understanding risk. Getting that right matters as much as the science, and it is not something one team, one discipline or one funder can settle alone. That is why so much of the work described here is about listening, and why responsibility in this space is funded work rather than a preamble.
Process
Listening first
Before this space funded anything, and continuously since, we have talked to the people who work on, depend on, regulate or would use the kinds of tools it might produce. Hundreds of conversations and more than fifteen workshops, with conservation organisations, government departments, land managers, and researchers across ecology, evolution, biological engineering, conservation, ethics, robotics and AI, shaped our understanding of what is needed, what is feasible, and where the limits lie. Programme Director Yannick Wurm describes that process, and the change of mind it produced, in Expanding the conservation toolkit.
Those conversations shaped how the work is done and, just as important, what ARIA funds at all: knowing what other funders and organisations can already do is how we locate the gap that ARIA exists to fill. That is the subject of the next section.
We also funded 22 short discovery projects (November 2025 to January 2026) to test directions across the space. Several looked directly at the ethical, social, governance, policy and risk questions rather than at the technology:
- Ethics and governance framework for engineering ecosystem resilience Michael Gyapong Akrasi, Independent
- Socio-technical engineering for ecosystem resilience Lindsay Stringer, University of York
- Addressing nature-related risks to engineer ecosystem resilience Tom Oliver, University of Reading
- Risk, uncertainty and ecosystem resilience Michael O'Connor, Independent
- Understanding resilience for global biodiversity: a text‑mining investigation of national biodiversity strategies and action plans Jay Burns, University of Edinburgh
Their findings, alongside those of the projects on research directions and technical feasibility, informed how the Accelerated Adaptation programme was designed and which opportunity seeds we funded.
The listening has not stopped. We continue to talk to stakeholders of every kind as the work develops, and if you would like to be part of that, whether you work in a conservation organisation, a government department, on the land, or in research, contact ARIA.
ARIA’s role
What ARIA can fund, and what it will not
ARIA exists to fill a gap in the funding ecosystem. It complements other funding streams rather than adding to them: if a venture investor, a land manager, a research council or a conservation organisation could plausibly fund a piece of work, it is not ours to fund. What remains is the scientific and technological work that others are unable to pursue: research at the edge of the possible, too early or too uncertain for anyone else to back.
This is why we talk to so many different kinds of organisation: understanding what each of them can do, and cannot, is how we find that gap. It also means ARIA’s role is deliberately limited. We are not a substitute for the essential work of reducing the pressures driving nature’s decline, for established conservation practice, or for the funders who support them.
What will and will not be done is then set for each activity. The Accelerated Adaptation programme, whose purpose is to find out whether the adaptation of wild species can be measurably and responsibly accelerated, carries the most specific limits, set out below: contained settings only, no releases to the wild during the programme, and no sweeping approaches such as uncontrollable gene drives. Those limits are the programme’s. The opportunity seeds sit outside the programme but are likewise funded under strong ethical guardrails, and only where teams have demonstrated a substantial understanding of, and engagement with, the ethical and social responsibility dimensions of their work.
Programme: Accelerated Adaptation
How responsibility is built into the programme
Interventions in wild populations raise profound ethical questions that cannot be resolved through technical excellence alone. The programme does not presume the answer to them. It exists to find out whether accelerated adaptation is feasible, and under what conditions, if any, any of the new approaches could be safe; it may turn out that some should not be used at all. It recognises multiple, sometimes conflicting, value systems, and builds responsibility in at three levels: limits on what is done, oversight of how it is done, and funded research on whether and how it should be done.
Limits on what is done
-
Responsible by Design
The programme has established guardrails to limit risk from the outset. All research conducted within this programme’s timeframe will take place exclusively in contained settings — boundaries that allow us to develop and test interventions while limiting risks to real-world environments. No releases to the wild will occur during the programme funding period.
-
Deliberate limits
The programme deliberately avoids sweeping, broad-brush approaches — such as uncontrollable gene drives aimed at collapsing populations. Approaches under exploration include transient physiological or molecular treatments that enhance stress tolerance without permanent genetic change.
Oversight of how it is done
-
Ethics ownership in every team
Every funded team is required to engage with the social and ethical dimensions of its work from the outset — and to take ownership of the ethical and governance thinking for its particular study system, in addition to the programme-level ESR work.
-
Ethical + Social Responsibility (ESR) Advisory Committee
This committee, currently being formed, will provide advice and guidance to the Programme Director to support oversight of the ethical and social considerations of the programme and its funded projects — including advising on, and reviewing, outputs and significant milestones from funded teams. Its role and remit will evolve to meet the programme’s emerging ESR considerations.
Marie-Claire Cordonier SeggerESR Advisory Committee Chair
Marie-Claire holds the Chair in Sustainable Development Law and Policy at the University of Cambridge and is senior director of the Centre for International Sustainable Development Law. She has over 20 years’ experience advising UN treaty negotiations across more than 80 countries on climate, biodiversity and the Sustainable Development Goals.
Funded research on whether and how
Three of the programme’s five Technical Areas are funded specifically to keep the system-focused research honest, well-governed and clear-eyed about risk: an ethics and governance team, a modelling team, and an independent verification team.
- TA5 · Ethics + governance research Governing novel conservation through plural value frames and systemic risk assessment Marzia Briel (University of Reading) + Alexandra Davey (Royal Botanic Garden Edinburgh) Assessing whether, when, where and how there is a defensible hierarchy of intervention acceptability — bringing affected communities into defining what future ecosystems should look like, and identifying the systemic risks novel interventions could introduce.
- TA3 · Modelling risks + trade-offs MERGE: Modelling Evolutionary Resilience, from Genes to Ecosystems Katie Willis (Imperial College London) with Reading + LSHTM Quantifying the likely benefits, harms and uncertainty of candidate interventions — new approaches and well-established management alike, and the option of not intervening — so that decisions rest on explicit trade-off assessments rather than intuition.
- TA4 · Independent verification Shared data infrastructure for climate biodiversity research Terry Burke (University of Sheffield) + Felix Shaw (Earlham Institute) Independently checking the claims made by the system-focused teams, with common data standards and a verification process that makes results comparable, trusted and legible.
Read on
Further reading
For the fuller argument, why the balance of risks has shifted and why expanding the toolkit must be done with humility and rigour, read Expanding the conservation toolkit. ARIA’s own account of the programme, including its responsible-research commitments, is on aria.org.uk.
Explore
The programme’s four commitments and the Chair’s biography are ARIA’s wording, reproduced from its Accelerated Adaptation pages. The rest of this page is this site’s own account and, like the rest of the site, is not an official ARIA publication. See About.