Maximizing Nature-based Solutions using Artificial Intelligence to Address Climate Change
Nature-based Solutions (NbS) represent holistic pathways to reach sustainable biodiversity, climate, and water outcomes through conservation and restoration. Yet, existing prioritization frameworks rarely identify synergistic NbS that maximize these outcomes. Here, we present an AI-driven framework that prioritizes NbS locations, maximizing biodiversity protection with ecological intactness, carbon storage, and water surface stability.
We implement this framework through scenarios combining specific biodiversity and co-benefits outcomes to achieve Canada’s 30×30 conservation and restoration targets. Our results reveal inevitable environmental trade-offs and critical implications for equitable NbS and extractive activities. To achieve conservation targets, protecting threatened biodiversity and irrecoverable carbon storage in forests would effectively address trade-offs and enhance Protected Areas outcomes. However, minimizing trade-offs to achieve restoration targets will require targeted interventions in existing forested and agricultural lands.
These findings demonstrate that frameworks integrating AI and biodiversity-co-benefit scenarios reveal strategic land use planning, policies, and investments for holistic NbS. Global biodiversity loss is closely intertwined with climate change, water scarcity, and social inequality. Given these interconnections, many global targets emphasize the need for holistic pathways that simultaneously address environmental and social challenges.
Conceptual Framework
Centered around biodiversity, the Kunming-Montreal Framework aims to restore 30% of degraded land and establish 30% of area-based conservation by 2030 (i.e., 30×30 targets), ensuring enhanced biodiversity, ecosystems’ integrity, and related co-benefits. Similarly, the Sustainable Development Goals (SDGs) set targets to protect and restore ecosystems, including water-related ecosystems such as forests, wetlands, rivers, and lakes.
NbS may play a critical role in reaching these global targets. NbS encompass a wide set of interventions, many of which provide holistic pathways for countries to reach the 30×30 conservation and restoration targets in synergy with climate change mitigation, water management, and Sustainable Development Goals. Reaching these targets through NbS implies strategically aligning biodiversity outcomes with related co-benefits emerging from ecological integrity, carbon storage, and water.
Pathways to Reach 30×30 Targets
Over the past decade, analytical tools for NbS implementation have become increasingly available. State-of-the-art monitoring tools typically leverage field observations, remote sensing, and machine learning to predict spatial-temporal patterns of biodiversity, carbon stocks, and water quality. In parallel, priority frameworks have weighted the importance of biodiversity aspects such as irreplaceability and vulnerability. From an ecosystem services perspective, some frameworks have prioritized conservation co-benefits emerging from carbon storage, water supply, and tourism.
However, few studies leverage AI to develop common frameworks for conservation and restoration targets that maximize outcomes related to biodiversity as well as related co-benefits. The complexity of optimizing biodiversity and co-benefit outcomes is further challenged by the interplay of local and economic priorities. Governments and decision-makers have dominated conservation and restoration planning, relegating local priorities and capacities, including those of Indigenous Peoples. Economic priorities also challenge the feasibility of conservation and restoration targets, as some priority frameworks highlight the restoration potential of converted lands for pasture and agriculture.
AI-driven Prioritization Framework
To address this complexity, we present a framework that prioritizes NbS locations maximizing biodiversity and co-benefits outcomes to reach 30×30 conservation and restoration targets. Leveraging Reinforcement Learning, this framework trains an agent to learn from the state of a system (i.e., exploration) and cost-effectively select areas (i.e., exploitation) that maximize a reward.
We test this framework in Canada under a series of conservation and restoration scenarios that maximize the occurrence of threatened species in synergy with co-benefits from ecological intactness, irrecoverable carbon storage, and water surface stability. Additionally, we examine some local and economic implications of these scenarios.
Conservation and Restoration Scenarios
The conservation scenarios identified an additional ∼17.75% of Canadian land areas to be protected, achieving Canada’s 30×30 targets by maximizing the protection of threatened biodiversity and guaranteeing at least 10% of each threatened species range, while also prioritizing co-benefits of ecological intactness, irrecoverable carbon stocks, and water surface stability.
Conservation Scenario 2, prioritizing irrecoverable carbon stocks, led to higher outcomes for threatened species, species richness, water surface change, and carbon stocks compared to existing Protected Areas. In contrast, Conservation Scenario 1, prioritizing ecological intactness, resulted in lower biodiversity and co-benefit outcomes.
The restoration scenarios suggested priorities in 30% of degraded land (∼3.5% total surface) that would maximize the occurrence of threatened biodiversity in areas experiencing reduced ecological intactness, carbon stocks losses, and water surface instability. Restoration Scenario 2, prioritizing the recovery of carbon stock losses, exhibited the most effective outcomes for ecological intactness, carbon stocks, and water surface stability. However, we did not identify wide-scale improvements in multiple outcomes after prioritizing actions for threatened biodiversity and water surface instability.
Implications for Equitable and Sustainable NbS
Aligning biodiversity outcomes with other co-benefits across conservation and restoration scenarios results in varied environmental trade-offs and spatial alignments with Indigenous-led NbS and natural resource projects. To conserve 30% of lands, prioritizing the protection of threatened biodiversity and carbon stocks in forests can effectively address these trade-offs. Yet, minimizing trade-offs to restore 30% of degraded lands will require a combination of scenarios and interventions in existing forested and agricultural lands.
Our results indicate that a significant fraction of government-funded Indigenous-led NbS overlap or are near conservation and restoration priorities, especially those that align with climate change mitigation. Additionally, both conservation and restoration priorities share a wide alignment with Indigenous Lands that lack government funding.
Regarding major natural resource projects, conservation priorities that maximize biodiversity and carbon outcomes are in close proximity to energy and mining projects, while restoration priorities are more likely to be near these projects, especially in the energy sector. Some conservation priorities that maximize ecological intactness, irrecoverable carbon stocks, and overlapped co-benefits also exhibit a marked immediacy to major energy and mining projects.
Conclusions
This prioritization framework demonstrates that integrating AI, multiple co-benefits, and scenario analysis can reveal strategic land use planning, policies, and investments for holistic NbS. By maximizing biodiversity protection in synergy with ecological intactness, carbon storage, and water surface stability, our results highlight opportunities to reach 30×30 targets through proactive conservation in forested lands and targeted restoration in agricultural landscapes.
Importantly, these findings underscore the need to assert Indigenous land governance, balance extractive activities, and coordinate public-private investments to achieve equitable and sustainable NbS outcomes. As the world grapples with the intertwined challenges of biodiversity loss, climate change, and social inequality, AI-driven frameworks for NbS can guide impactful implementation toward a more resilient future.