An official website of the United States government
From out-competing grandmasters in chess to informing high-stakes healthcare decisions, emerging methods from artificial intelligence are increasingly capable of making complex and strategic decisions in diverse, high-dimensional and uncertain situations. But can these methods help us devise robust strategies for managing environmental systems under great uncertainty? Here we explore how reinforcement learning (RL), a subfield of artificial intelligence, approaches decision problems through a lens similar to adaptive environmental management: learning through experience to gradually improve decisions with updated knowledge. We review where RL holds promise for improving evidence-informed adaptive management decisions even when classical optimization methods are intractable and discuss technical and social issues that arise when applying RL to adaptive management problems in the environmental domain. Our synthesis suggests that environmental management and computer science can learn from one another about the practices, promises and perils of experience-based decision-making. This article is part of the theme issue ‘Detecting and attributing the causes of biodiversity change: needs, gaps and solutions’.
more »
« less
This content will become publicly available on March 18, 2026
Reinforcement learning–based adaptive strategies for climate change adaptation: An application for coastal flood risk management
Conventional computational models of climate adaptation frameworks inadequately consider decision-makers’ capacity to learn, update, and improve decisions. Here, we investigate the potential of reinforcement learning (RL), a machine learning technique that efficaciously acquires knowledge from the environment and systematically optimizes dynamic decisions, in modeling and informing adaptive climate decision-making. We consider coastal flood risk mitigations for Manhattan, New York City, USA (NYC), illustrating the benefit of continuously incorporating observations of sea-level rise into systematic designs of adaptive strategies. We find that when designing adaptive seawalls to protect NYC, the RL-derived strategy significantly reduces the expected net cost by 6 to 36% under the moderate emissions scenario SSP2-4.5 (9 to 77% under the high emissions scenario SSP5-8.5), compared to conventional methods. When considering multiple adaptive policies, including accomodation and retreat as well as protection, the RL approach leads to a further 5% (15%) cost reduction, showing RL’s flexibility in coordinatively addressing complex policy design problems. RL also outperforms conventional methods in controlling tail risk (i.e., low probability, high impact outcomes) and in avoiding losses induced by misinformation about the climate state (e.g., deep uncertainty), demonstrating the importance of systematic learning and updating in addressing extremes and uncertainties related to climate adaptation.
more »
« less
- Award ID(s):
- 2103754
- PAR ID:
- 10610303
- Publisher / Repository:
- Proceedings of the National Academy of Sciences
- Date Published:
- Journal Name:
- Proceedings of the National Academy of Sciences
- Volume:
- 122
- Issue:
- 12
- ISSN:
- 0027-8424
- Page Range / eLocation ID:
- e2402826122
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
na (Ed.)Nitrous oxide (N2O) emissions from agriculture are rising due to increased fertilizer use and intensive farming, posing a major challenge for climate mitigation. This study introduces a novel reinforcement learning (RL) framework to optimize farm management strategies that balance crop productivity with environmental impact, particularly N2O emissions. By modeling agricultural decision-making as a partially observable Markov decision process (POMDP), the framework accounts for uncertainties in environmental conditions and observational data. The approach integrates deep Q-learning with recurrent neural networks (RNNs) to train adaptive agents within a simulated farming environment. A Probabilistic Deep Learning (PDL) model was developed to estimate N2O emissions, achieving a high Prediction Interval Coverage Probability (PICP) of 0.937 within a 95% confidence interval on the available dataset. While the PDL model’s generalizability is currently constrained by the limited observational data, the RL framework itself is designed for broad applicability, capable of extending to diverse agricultural practices and environmental conditions. Results demonstrate that RL agents reduce N2O emissions without compromising yields, even under climatic variability. The framework’s flexibility allows for future integration of expanded datasets or alternative emission models, ensuring scalability as more field data becomes available. This work highlights the potential of artificial intelligence to advance climate-smart agriculture by simultaneously addressing productivity and sustainability goals in dynamic real-world settings.more » « less
-
Abstract As the global impact of climate change intensifies, there is an urgent need for equitable and efficient climate adaptation policies. Traditional approaches for allocating public resources for climate adaptation that are based on economic benefit-cost analysis often overlook the resulting distributional inequalities. In this study, we apply equity weightings to mitigate the distributional inequalities in two key building and household level adaptation strategies under changing coastal flood hazards: property buyouts and building retrofit in New York City (NYC). Under a mid-range emissions scenario, we find that unweighted benefit cost ratios applied to residential buildings are higher for richer and non-disadvantaged census tracts in NYC. The integration of income-based equity weights alters this correlation effect, which has the potential to shift investment in mitigation towards poorer and disadvantaged census tracts. This alteration is sensitive to the value of elasticity of marginal utility, the key parameter used to calculate the equity weight. Higher values of elasticity of marginal utility increase benefits for disadvantaged communities but reduce the overall economic benefits from investments, highlighting the trade-offs in incorporating equity into adaptation planning.more » « less
-
Integrating new sea‐level scenarios into coastal risk and adaptation assessments: An ongoing processAbstract The release of new and updated sea‐level rise (SLR) information, such as from the Intergovernmental Panel on Climate Change (IPCC) Assessment Reports, needs to be better anticipated in coastal risk and adaptation assessments. This requires risk and adaptation assessments to be regularly reviewed and updated as needed, reflecting the new information but retaining useful information from earlier assessments. In this paper, updated guidance on the types of SLR information available is presented, including for sea‐level extremes. An intercomparison of the evolution of the headline projected ranges across all the IPCC reports show an increase from the fourth and fifth assessments to the most recent “Special Report on the Ocean and Cryosphere in a Changing Climate” assessment. IPCC reports have begun to highlight the importance of potential high‐end sea‐level response, mainly reflecting uncertainties in the Greenland/Antarctic ice sheet components, and how this might be considered in scenarios. The methods that are developed here are practical and consider coastal risk assessment, adaptation planning, and long‐term decision‐making to be an ongoing process and ensure that despite the large uncertainties, pragmatic adaptation decisions can be made. It is concluded that new sea‐level information should not be seen as an automatic reason for abandoning existing assessments, but as an opportunity to review (i) the assessment's robustness in the light of new science and (ii) the utility of proactive adaptation and planning strategies, especially over the more uncertain longer term. This article is categorized under:Assessing Impacts of Climate Change > Scenario Development and Applicationmore » « less
-
Abstract Tropical cyclone (TC) track characteristics in a changing climate remain uncertain. Here, we investigate the genesis, tracks, and termination of >35,000 synthetic TCs traveling within 250 km of New York City (NYC) from the pre‐industrial era (850–1800 CE) to the modern era (1970–2005 CE) to the future (2080–2100 CE). Under a very high‐emissions scenario (RCP8.5), TCs are more likely to form closer to the United States (U.S.) southeast coast (>15% increase), terminate in the northeastern Atlantic (>6% increase), and move most slowly along the U.S. Atlantic coast (>15% increase) from the pre‐industrial to future. Under our modeled scenarios, TCs are more likely to travel within 100 km of Boston, MA, USA (p = 0.01) and Norfolk, VA, USA (p = 0.05) than within 100 km of NYC in the future. We identify reductions in the time between genesis and the time when TCs come within 100 km of NYC, Boston, or Norfolk, as well as increased duration of TC impacts from individual storms at all three cities in the future.more » « less
