Forest Data Partnership: Aligning and Innovating for a Shared Data Ecosystem
- 2. Data is vital to tackling deforestation & accelerating restoration Technology offers unprecedented opportunities to transform the way landscapes are monitored and managed
- 3. We need to harness the full potential of data and harmonize the current fragmented data landscape
- 4. The Forest Data Partnership Unites organizations, governments and private sector partners around trusted, transparent geospatial data solutions that enable credible monitoring, verification and disclosure of progress in reducing deforestation and restoring degraded lands.
- 5. Five action-oriented work streams
- 7. ● Focal areas ○ Cocoa in West Africa ○ Palm Oil in Southeast Asia ○ Cattle and Soy in the Amazon Basin ○ Restoration ● Who are we prioritizing? Mapping stakeholders based on: ○ Roles in how they interact with and/or produce data, ○ Their impact in affecting deforestation and restoration, ○ Their influence in the uptake and application of data ● Actors across supply chains and financial sectors, governments in producer and consumer countries, civil society, technology and service providers ○ Reaching beyond the usual suspects ○ Local and place-based affected stakeholders to ensure representation Align Who?
- 8. Align On What? Current data challenges and obstacles to uptake and application of data, such as: ● Gaps (the data doesn’t exist, or isn’t accessible) ● Conflicts (scales, methods) ● Accuracy ● Definitions ● Applications
- 9. Align - How and Why? ●Tiered approach ○ Information gathering from specific stakeholders already active in ongoing discussions ○ Targeted engagement strategies and business case for participation for tougher to reach stakeholders ● Tailored to the context and priorities of specific regions and supply chains ● Learning from existing MSAs to develop alignment and decision-making processes ○ Interaction between different stakeholders ○ Ensuring representation and buy-in for outputs (consultation vs. consensus)
- 10. Align on Foundational Data Gaps Develop Inclusive demand-driven Innovations Deploy Data Delivery Mechanisms Restoration Deforestation Continuous improvement and replication Continuous improvement and replication Impact Assessment Research Informs policy and project development How will this work? Informs policy and project development
- 11. Innovate process
- 12. Commodity mapping at scale: spatial prediction
- 13. Anonymized Data Raw Data Reference Data Existing maps Predicted Maps (Value + Uncertainty) Existing maps Existing maps Existing maps Map 1 Map 2 Map N ENSEMBLE REF DATA LCML Private Aligned Partners Public Commodity mapping at scale: harmonization
- 14. Which oil palm map should you believe? (Hint: trick question.) Meta-learner* * Wolpert, 1992, Stacked generalization.
- 15. Inclusive mapping at scale From the Ground to the Cloud and back If the harmonized map is wrong, users who are affected have an opportunity to correct the record by submitting more data to train the meta-learner. Cloud-based analytics and ML Section #5 Local communities, sustainability users, citizen scientists Local insights and knowledge, in- situ observations More accurate, timely information Ground measurements
- 16. A dynamic pathway for restoration monitoring & planning Reporting Monitoring Planning