30×30 Cannot Be an Area Count: Five Governance Cards for Indigenous Forest Co-Governance, Evidence, and Ecosystem Services

Taiwan's Forestry and Nature Conservation Agency held the “2026 Taiwan toward 30×30 Symposium” on October 6 and 7. On October 7, it announced a joint 30×30 declaration with the Ocean Conservation Administration and National Park Service. The strategy combines additional protected areas, state-forest areas with substantive conservation functions, other effective area-based conservation measures (OECMs), better habitat connectivity, and improved management. The same announcement identifies Saisiyat, Bunun, and Tsou participation in forest co-governance and says that future protected-area work should move away from purely top-down designation toward local communication and shared governance.

This is directly relevant to land, forests, production, culture, and public services across Taiwan's 55 Indigenous townships. The important question is not only whether Taiwan can exceed 30 percent. It is also which places are counted, who decides, whether management is effective, whether existing Indigenous governance and traditional ecological knowledge are recognized, and how restrictions, jobs, payments, and benefits are distributed. An area total alone can turn 30×30 into an administrative score. A framework that connects rights, knowledge, monitoring, resources, and accountability can make it a durable form of local governance.

The first card records “area, rights, and use.” Any proposed protected area, OECM, or ecological corridor should identify its geographic boundary, tenure, legal zoning, traditional territory or customary use, current farming, forestry, fishing, hunting, and gathering, seasonal routes, ceremonial responsibilities, restrictions, and unresolved claims. A single landscape may simultaneously be a watershed, hunting territory, gathering area, farm buffer, ancestral landscape, and monitoring zone. Disagreement must remain visible with its evidence, date, and resolution status instead of being flattened into a map that appears settled.

The second card covers “shared decision-making and responsibility.” It should identify how Indigenous representatives are chosen, which public bodies carry legal duties, which actions require consultation and consent, when documents must be supplied, how Indigenous-language interpretation is arranged, who may place an issue on the agenda, and how decisions, reservations, review, and withdrawal work. Co-governance is not a briefing held after a plan is finished. A meeting signature is not permanent authorization. Land use, access, species surveys, recording, tourism, research, and commercial collaboration each need their own decision level and permission scope.

The third card covers “knowledge, data, and sensitivity.” Scientific monitoring may use species, habitat, terrain, and time-series data. Community governance may use Indigenous place names, seasonal calendars, routes, indicator species, restrictions, and long-term observations. These systems can complement one another without making every cultural record public. Each field needs a source, knowledge holder, permitted audience, purpose, sensitivity level, storage location, update method, and correction or removal process. A public map may show management outcomes, while sensitive species locations, ceremonial sites, and restricted knowledge remain protected. Permission for a co-governance meeting is not permission for external research, product design, or AI training.

The fourth card covers “monitoring, outcomes, and versions.” The purpose of 30×30 is effective conservation, not designation alone. A site first needs a baseline for indicator species, habitat condition, hydrology, invasive species, disturbance, fire governance, resource pressure, and community concerns. Indigenous and public managers can then agree on frequency, method, responsibility, quality checks, and alerts. Remote sensing, soundscape sensors, cameras, and AI recognition can assist, but each result needs time, location, confidence, device, and human review. Low-confidence classifications or culturally sensitive locations must never trigger automatic punishment or restriction. Method changes must be versioned so technical differences are not mistaken for ecological change.

The fifth card covers “resources, benefits, and grievance.” Conservation can create ranger, monitoring, education, ecotourism, forest-economy, habitat-maintenance, and ecosystem-service work. It can also impose access limits, operating costs, or rights disputes. Funding sources, jobs, training, purchasing or partnership rules, payment conditions, ownership of results, local return, risk sharing, and grievance contacts must be visible. Ecosystem-service payments are not a purchase of permanent surrender of Indigenous rights, and a short project cannot replace sustained governance funding. If community monitoring supports corporate ESG or nature reporting, purpose, verification responsibility, and benefit return need prior agreement.

The cards work as a chain. The area card explains land, use, and disagreement. The decision card identifies authority. The knowledge card controls visibility and reuse. The monitoring card tests whether governance improves biodiversity. The resources and grievance card makes work, payment, restriction, and remedy traceable. Remove one card and familiar risks return: area without consent, data without permission, payment without long-term responsibility, or monitoring that never informs community decisions.

The agency reports that state-forest working circles cover about 42.5 percent of Taiwan. Nearly 90,000 hectares of protection forests and state forests with defined biodiversity targets have been identified. OECM certification began in 2025, with almost 3,000 hectares certified and another 107 sites applying in a second round. The figures show that 30×30 will involve state forest, private and community land, farms, schools, companies, and other places. A state-forest legal status does not erase historical relationships, customary use, or knowledge rights. Voluntary OECM participation must not become vague or unlimited land and data authorization.

Target 3 of the Kunming-Montreal Global Biodiversity Framework calls for representative, connected, equitably governed, and effectively managed protected areas and OECMs while recognizing the rights and territories of Indigenous Peoples and local communities. Taiwan's Indigenous Peoples Basic Law requires consultation and Indigenous or tribal consent or participation before governments or private actors undertake development, resource use, ecological conservation, or research on Indigenous lands or within affected surrounding areas, together with benefit sharing. Conservation cannot be used to bypass rights procedures; equitable governance is itself part of conservation effectiveness.

AI is useful here for version comparison, data catalogues, monitoring alerts, multilingual summaries, meeting follow-up, and source tracing. It should not determine traditional territory, decide which knowledge can be made public, infer who represents a community, or classify a violation from a single image. Automated output needs a source, date, method, uncertainty statement, human review, and appeal. Community-generated monitoring data also needs governance and benefit fields, not only copyright metadata.

A locality can test the approach in one small site over 90 days. The first month maps area, rights relationships, and existing data. The second month lets the community, managing agency, and monitoring team agree on three to five indicators, data permissions, and response procedures. The third month runs a complete patrol, upload, review, reporting, and grievance cycle. Success is not a larger database. It is whether communities can inspect and correct records, decisions reach field practice, sensitive data stays restricted, and monitoring supports real management.

The October 7 announcement sets a clear direction: connect mountains to the sea and make communities and Indigenous Peoples governance partners. The task on October 8 is to turn that direction into a testable process. When area, decisions, knowledge, monitoring, resources, and grievance all have dates, evidence, accountable roles, permissions, and next steps, 30×30 becomes more than a larger green patch on a map. It becomes a governance system that can sustain Indigenous rights, local work, and biodiversity together.

Verified sources

• Forestry and Nature Conservation Agency: building Taiwan's biodiversity network from mountains to the sea
• Convention on Biological Diversity: Kunming-Montreal Global Biodiversity Framework, Target 3
• Laws and Regulations Database of Taiwan: Indigenous Peoples Basic Law
• United Nations: Declaration on the Rights of Indigenous Peoples