Scenario-based assessment of payment for ecosystem services and forest carbon finance in Costa Rica
Description
This dataset supports a matched-baseline comparison of PES, GS4GG, and VCS/VERRA forest-carbon finance in Costa Rica for 2024–2060. The research question is how price assumptions and selected pathway fees affect revenue when all pathways receive an identical annual carbon-accounting quantity. The design isolates these financial effects without estimating differences in institutional performance or carbon delivery. The primary analysis assigns 1,000,000 Mg CO₂e to each pathway in every year. The secondary analysis applies the same retained annual harmonized Costa Rica CO₂-support quantity to all three pathways. This secondary quantity is a contextual support indicator, not observed removals, certified credits, eligible credits, or predicted issuance. The package contains retained ecological-support tables, PSA spatial records, source rasters and CSV extracts, processing scripts, assumptions, economic outputs, simulation summaries, and figure sources. Ecological context includes land cover, forest gain and loss, biomass, and soil/climate indicators. PSA areas were reconstructed from recorded area attributes and allocated using province and Indigenous-territory boundaries. Pre-2019 PSA totals represent incomplete spatial-record coverage. Coverage diagnostics retain surviving FONAFIFO comparison values, whose exact administrative definition remains unresolved. Four cases separate price and fee effects: A uses common prices without fees; B uses common prices with selected pathway fees; C uses pathway-specific scenario prices without fees; and D combines those prices with selected fees. Cases A and B are principal comparisons; C and D are structural sensitivities. The common-price grid is USD 7.5, 15, 25, 50, 100, and 147 per Mg CO₂e. Outputs include gross scenario revenue, fee-adjusted revenue, cumulative fee-adjusted revenue, and present values at 3%, 5%, and 7%. Selected fees do not represent complete project costs. Case A produces identical pathway results by construction. Subsequent differences reflect the selected fee schedules and price assumptions. Removing pre-2019 PSA records leaves the active financial outputs unchanged. Conditional price-level simulations use 10,000 mean-preserving lognormal draws, seed 37031, and price coefficients of variation of 0.05, 0.10, and 0.20. Each draw shares one multiplier across pathways and the complete price trajectory within a matched comparison. Quantities, fees, and discounting remain deterministic. Means and P10, P50, and P90 quantiles are provided; these are conditional simulation intervals, not comprehensive uncertainty bounds or market forecasts. The README, data dictionary, provenance records, and reproduction script explain reuse. Verified reproduction begins with the retained ecological CSVs. Historical Random Forest, GS4GG template, and policy-response outputs are separately labeled legacy diagnostics and excluded from the active analysis.
Files
Steps to reproduce
Extract the complete deposit and preserve the folder structure. The package contains retained ecological raster/CSV extracts, national and provincial support summaries, point-based PSA records, processing scripts, provenance files, and documented limitations. PSA hectares were reconstructed from recorded area attributes and spatially allocated using province and Indigenous-territory boundaries. Use Python 3.12. From the folder containing the README, verify file integrity with: python reproduce.py --verify-only Reproduce the active analysis with: python reproduce.py This creates a new reproduced directory without altering the deposited reference files. The runner executes the foundation, matched-economics, deterministic-interpretation, conditional-simulation, results-audit, and PSA exclusion-audit scripts in dependency order. Core calculations use only the Python standard library. Matched quantities cover 2024–2060. The primary baseline assigns 1,000,000 Mg CO₂e to PES, GS4GG, and VCS/VERRA in every year. The secondary baseline uses NatCO2_Sequestration_central from CO2Sequestration_AnnualNationalSummary.csv and applies the same annual value to all three pathways. Cases A/B use common prices of USD 7.5, 15, 25, 50, 100, and 147 per Mg CO₂e, without/with selected fees. Cases C/D use pathway-specific scenario prices, without/with selected fees. Gross revenue = quantity × price; fee-adjusted revenue = gross revenue − selected fees. Cumulative values cover 2024–2060. Present values use 3%, 5%, and 7% discount rates, with 2024 as t=0. Conditional price-level simulations use 10,000 draws, seed 37031, and CV values 0.05, 0.10, and 0.20. Lognormal multipliers use sigma = sqrt[ln(1 + CV²)] and mu = −sigma²/2. One multiplier is applied to the full price trajectory and shared across pathways within each matched comparison. Quantities, selected fees, and discounting remain deterministic. Outputs include simulated mean, P10, P50, P90, and P90−P10 width. Inspect reproduction_comparison.csv, reproduction_report.json, and the PSA exclusion-audit outputs. Deposited validation reported zero numerical differences across 23 regenerated CSVs. Excluding pre-2019 PSA records did not change active financial outputs. To reproduce Figures 1–5: python -m pip install -r requirements-figures.txt python reproduce.py --output reproduced_with_figures --figures all --dpi 600 Reproduction begins from retained ecological CSVs. It does not repeat satellite-data acquisition or execute legacy Random Forest, GS4GG template, or policy-response models; surviving legacy files are documented separately.
Institutions
- Chongqing UniversityChongqing, Chongqing