Reconciling top-down and bottom-up conservation priorities with reinforcement learning from human feedback
Open Access
- 25 September 2026
- journal article
- Published by IOP Publishing in Environmental Research Letters
- Vol. 21 (18) , 181002
Abstract
To meet the Kunming–Montreal Global Biodiversity Framework, spatial planning must upscale area-based conservation, while ensuring equitable governance and local community integration. Historically, a deep divide has persisted between data-driven, top-down systematic conservation planning, often performed at large spatial scales, and participatory, bottom-up local initiatives. Here, we propose a novel framework that bridges this gap by harnessing the power of human and artificial intelligence. By deploying a technique called Reinforcement Learning with Human Feedback in our software CAPTAIN (conservation area prioritization through artificial intelligence), our approach can process vast biophysical and socioeconomic data to create fine-tuned conservation strategies that are both data-driven and sensitive to local realities. The methodology outlined here provides a promising avenue for scalable, effective, and equitable biodiversity conservation.Keywords
Funding Information
- Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung (31BD-0_233170)
- Eidgenössische Technische Hochschule Zürich
- Stiftelsen för Miljöstrategisk Forskning (MISTRA Biopath F 2022/1448)
- Vetenskapsrådet (2024-04303)
- Biodiversity and Ecosystem services in a Changing Climate
- Familjen Kamprads Stiftelse (20230142)
- Royal Botanical Gardens, Kew
This publication has 38 references indexed in Scilit:
- Large language models possess some ecological knowledge, but how much?Ecological Informatics, 2026
- Identifying conservation gaps: A framework for evaluating national contributions to the 30 × 30 targetOne Earth, 2026
- To increase or not to increase? The impact of terrestrial protected areas on tree cover and the risk of species extinctionJournal of Environmental Management, 2025
- Integrating artificial intelligence in biodiversity conservation: bridging classical and modern approachesBiodiversity and Conservation, 2024
- Using games for social learning to promote self-governanceCurrent Opinion in Environmental Sustainability, 2023
- Integrating Top-Down and Bottom-Up Approaches Improves Practicality and Efficiency of Large-Scale Ecological Restoration Planning: Insights from a Social–Ecological SystemEngineering, 2022
- Recognizing Indigenous peoples’ and local communities’ rights and agency in the post-2020 Biodiversity AgendaAMBIO, 2021
- The essential role of other effective area-based conservation measures in achieving big bold conservation targetsGlobal Ecology and Conservation, 2018
- Important Plant Areas: revised selection criteria for a global approach to plant conservationBiodiversity and Conservation, 2017
- Hard choices: Making trade-offs between biodiversity conservation and human well-beingBiological Conservation, 2011