Whitepaper

Nature risk intelligence, made operational

LandPrint converts satellite data and peer-reviewed science into environmental risk assessments, so organisations can review nature exposure before it affects credit, sourcing, or disclosure decisions.

The Problem

Nature risk is the next climate risk

Regulators, investors, and lenders are rapidly converging on the same conclusion: exposure to nature loss is material, measurable, and in many jurisdictions now reportable. The EU Deforestation Regulation demands traceable proof that commodities are deforestation-free. The TNFD framework asks organisations to disclose nature-related dependencies, impacts, risks, and opportunities using a structured disclosure matrix.

Yet most teams still assess nature risk with spreadsheets, manual map lookups, or consultant reports that arrive months after the decision window has closed. The data exists. Hundreds of satellite constellations and scientific datasets already cover every hectare on earth, but stitching them into an actionable picture remains an engineering problem, not a data problem.

LandPrint closes that gap. We connect directly to authoritative geospatial and scientific data sources, run the analysis on demand, and deliver a scored, disclosure-ready report within minutes, not months.

What We Analyse

Twelve modules, one integrated picture

Each module draws on authoritative, peer-reviewed datasets and produces its own risk signal. Together they form a composite view that no single data layer can provide alone.

Forest Baseline & Deforestation

Landsat 7 ETM+ / Landsat 8 OLI30 m

Hansen Global Forest Change v1.11

Establishes a verified forest-cover baseline from the Hansen/UMD 2000 treecover layer minus cumulative loss through 2020, then detects post-cutoff tree-cover loss using the lossyear band. Forest is defined as pixels with >10% canopy cover. Year-over-year linear regression produces trend direction, annual loss rate (ha/yr), and acceleration percentage.

Protected Areas & Indigenous Lands

N/A (vector)Vector polygons

WDPA (UNEP-WCMC)

Spatial intersection with the World Database on Protected Areas and governance-type filtering for indigenous and community-conserved territories. Overlap area computed via geodesic vector intersection across overlapping geometries. IUCN management categories (Ia, Ib, II–VI) and governance types are extracted and reported.

Country Risk Classification

N/ACountry-level

EU Commission Classification

Country code derived from AOI centroid matched against the EU regulatory classification table (high / standard / low risk), surfacing expected audit rates and compliance thresholds. Updated per regulatory cycle.

Soil & Water Assessment

N/A (modelled)250 m / basin-level

SoilGrids 250m v2.0 + WRI Aqueduct 4.0

Extracts clay, silt, sand, SOC, bulk density, CEC, pH, and nitrogen at six ISRIC standard depths (0–5 cm through 100–200 cm). Hydraulic properties (available water capacity, field capacity, and permanent wilting point) are derived using Saxton–Rawls pedotransfer functions. Aqueduct 4.0 overlays basin-level water-stress indicators including baseline water stress and future projections.

Ecosystem Integrity Index

Multi-source composite300 m

Halpern et al. EII

Multi-band composite measuring structural, functional, and compositional integrity plus a combined EII score. Mean, min, max, and standard deviation computed for each band within the AOI boundary. Higher values indicate more intact ecosystems.

Land Cover & Vegetation Health

Sentinel-2A/2B MSI10 m

ESA WorldCover v200 + Sentinel-2 L2A Harmonized

Land use classified across 11 ESA WorldCover classes at 10 m. Natural vegetation share aggregates tree cover, shrubland, grassland, and wetland. Vegetation health uses NDVI = (B8 − B4) / (B8 + B4) from Sentinel-2, comparing a six-month current window against a five-year baseline to compute greenness anomaly percentage and condition classification (healthy > 0.6, moderate 0.4–0.6, stressed 0.2–0.4, degraded < 0.2).

Climate Risk & Projections

ERA5-Land reanalysis + CHIRPS + CMIP611 km / 5.5 km / 25 km

ECMWF ERA5-Land + UCSB CHIRPS + NASA NEX-GDDP-CMIP6

Five-year climate baseline from ERA5-Land (temperature) and CHIRPS (precipitation). Drought frequency = months with <50% of mean rainfall. Climate anomaly score (0–100) combines temperature range, precipitation coefficient of variation, and drought frequency. Forward projections from CMIP6 under SSP2-4.5 and SSP5-8.5 scenarios for 2040–2060, reporting temperature change (°C) and precipitation change (%).

Carbon Stock Estimation

Multi-source composite~300 m / 250 m

NASA/ORNL Biomass Carbon Density v1 + SoilGrids SOC

Above-ground biomass carbon density (Mg C/ha) from the NASA/ORNL global dataset (Spawn et al. 2020) averaged within the AOI. Soil organic carbon from SoilGrids integrated across profile depth bands (0–200 cm). Total carbon = AGB + SOC. CO₂e = total carbon × 3.67 (molecular mass ratio).

Fire Risk Intelligence

Suomi NPP / NOAA-20 VIIRS375 m

NASA FIRMS VIIRS Active Fires

Near-real-time active fire detections over the past 12 months via the FIRMS REST API. Points filtered to the AOI boundary using spatial containment. Fire density score (0–100) based on detections per km², with temporal clustering analysis to estimate burn recurrence intervals.

Terrain & Erosion Analysis

Copernicus Sentinel-1 (InSAR-derived)30 m

Copernicus DEM GLO-30

Mean elevation, slope (computed via GEE terrain algorithm), and aspect derived from the Copernicus GLO-30 DEM. Erosion risk scored from slope thresholds: moderate >15°, high >25°, critical >35°. Terrain variability measured as the standard deviation of elevation within the AOI.

Ecoregion & Biodiversity Screening

N/A (vector)Vector

RESOLVE Ecoregions 2017 + WDPA

RESOLVE Ecoregions intersected with the AOI to extract ecoregion name, biome, realm, and Nature Needs Half (NNH) priority score. Sensitivity classified as high when NNH ≥ 2. Biodiversity screening uses a 50 km buffer, counting WDPA polygons and strict IUCN categories (Ia, Ib, II) to produce a biodiversity priority score (0–100).

Ecosystem Service Dependencies

N/A (framework)Sector-level

ENCORE Framework v1 (UNEP-WCMC)

The ENCORE dependency matrix maps seven sector types to ecosystem service dependencies, including water flow regulation, pollination, soil quality maintenance, and flood and storm protection, and cross-references spatial findings from water risk, land cover, ecosystem integrity, climate, and terrain modules to flag dependencies where on-the-ground data indicates elevated risk.

Analysis Systems

Specialist runs, shared report anatomy

Depending on your regulatory context and business question, LandPrint offers specialist analysis configurations that all resolve into the same five-theme report anatomy across Water, Soil Health, Biodiversity, Carbon, Regenerative Practices.

Risk Assessment

The comprehensive analysis: the full module stack, sector-weighted scoring, financial impact estimation, and a five-theme report across Water, Soil Health, Biodiversity, Carbon, Regenerative Practices.

Climate & Deforestation Risk

Physical-risk focused: forest baseline verification, post-2020 deforestation detection and trend analysis, carbon stock and CO₂e liability estimation, ERA5/CMIP6 climate projections, FIRMS fire density scoring, terrain erosion risk, and Sentinel-2 vegetation condition monitoring.

Biodiversity & Land Impact

Location-sensitive biodiversity footprint: WDPA protected-area and IUCN-category overlap, indigenous territory interface via LandMark, RESOLVE ecoregion sensitivity ratings, ESA WorldCover land-use composition, biodiversity priority scoring, and ecosystem integrity assessment.

Nature Dependency Scan

Evaluates how operations depend on and impact ecosystem services: ENCORE sector-dependency mapping cross-referenced with spatial data on soil hydraulic properties, basin water stress, ecosystem integrity, land-cover context, vegetation health, and climate baseline.

Risk Scoring

One number with full traceability

The platform calculates the Landprint Risk Index (LRI) directly on the shared 1-5 scale as a weighted average across up to eleven scoring modules. That keeps the published score aligned with the real calculation while still disclosing which modules were available and how much each one contributed.

Each weight reflects regulatory materiality and scientific consensus. Deforestation carries the highest base weight given its direct regulatory and deforestation-free supply chain implications, while climate risk remains a lighter slow-onset context signal. Individual runs disclose which modules were actually scored so monitor-scoped analyses do not pretend to use missing inputs.

Native 1-5 Equation

S = Σi=111 wi · si
Weighted directly on the 1-5 scale

where S = weighted LRI score (1-5), si = individual module rating (1-5), wi = module weight (Σw = 1.0), and the weights sum to 1.0

When a sector context is provided (agriculture, forestry, fisheries, mining, or tourism), the base weights are adjusted using sector-specific multipliers and renormalized:

Sector-Weighted Adjustment

wieff = (wi · ai,sector) / Σj(wj · aj,sector)

where ai,sector = sector-specific multiplier for module i, ensuring effective weights always sum to 1.0

Published LRINative LRI BandInterpretation
1/5 · Very Low Risk0 - 1Minimal environmental exposure
2/5 · Low Risk1 - 2Localized concerns requiring monitoring
3/5 · Moderate Risk2 - 3Material exposure requiring active management
4/5 · High Risk3 - 4Significant exposure requiring near-term action
5/5 · Very High Risk4 - 5Severe exposure requiring escalation, redesign, or pause

Financial Impact Estimation

Revenue at Risk = areaha × revenueha × risk share

Regulatory Cost = areaha × $50/ha when country risk classification = High

Ecosystem Service Value = areaha × $4,000/ha/yr (TEEB global average)

Total Exposure = Revenue at Risk + Regulatory Cost + ESV

Financial estimates use a normalized risk share derived from the native 1-5 LRI score: risk share = max(0, (LRI - 1) / 4). They are indicative projections based on sector averages and published benchmarks, not financial advice.

Satellite Constellations & Data Sources

The instruments behind every pixel

LandPrint integrates data from eight satellite missions and five authoritative scientific databases. Each source is accessed at its native resolution. We never up-sample to fabricate detail.

Mission / SourceInstrument
Landsat 7 / 8ETM+ / OLI
Sentinel-2A / 2BMSI (13 bands)
Sentinel-1A / 1BC-SAR (InSAR)
Suomi NPP / NOAA-20VIIRS (I-band)
ISS (GEDI)Full-waveform LiDAR
ERA5-LandECMWF Reanalysis
CHIRPSGauge + IR satellite
NASA NEX-GDDP-CMIP6Multi-model ensemble
SoilGrids v2.0ISRIC modelled
Aqueduct 4.0WRI modelled
WDPAUNEP-WCMC
RESOLVE 2017Conservation Science
ESA WorldCoverSentinel-1/2 composite
Halpern et al.Multi-source composite
ENCORE v1UNEP-WCMC
EU CommissionRegulatory

Technical Framework

Key equations & derived indices

The following are representative equations used across the platform's analytical modules.

NDVI (Vegetation Health)

NDVI = (B8 − B4) / (B8 + B4)

Sentinel-2 Band 8 (NIR, 842 nm) and Band 4 (Red, 665 nm). Values range −1 to +1; healthy vegetation > 0.6.

Greenness Anomaly

ΔG = (NDVIₖm − NDVI₅yr) / NDVI₅yr × 100%

Six-month current window compared against a 5-year baseline mean. Negative anomalies flag degradation.

Carbon Stock (Total)

Cₜₒₜ = AGB + BGB + SOC₀₋₃₀

AGB and BGB from NASA/ORNL Biomass Carbon Density v1 (Spawn et al. 2020). SOC from SoilGrids 0–30 cm depth integration.

CO₂ Equivalent

CO₂e = Cₜₒₜ × 3.67

Molecular mass ratio of CO₂ (44) to C (12) = 3.667.

Terrain Variability

TV = σ(elevation)

Standard deviation of Copernicus GLO-30 elevation values within the AOI.

Erosion Risk

E = f(slope) | moderate > 15°, high > 25°, critical > 35°

Slope computed from DEM via GEE terrain algorithm. Thresholds aligned with FAO erosion susceptibility classes.

Drought Frequency

Dᶠ = count(Pₘ < 0.5 × P̄) / Nₘ

Fraction of months where precipitation (CHIRPS) falls below 50% of the long-term mean over the baseline period.

Fire Density Score

F = min(100, firesₐₒᵢ / areaₖₘ² × k)

VIIRS detections per km² over 12 months, scaled to 0–100.

Data Philosophy

Built on science, not black boxes

01

Peer-Reviewed Sources

Every dataset behind LandPrint comes from an established scientific institution or international standards body. We do not generate our own base data. We integrate, validate, and interpret the best available science.

02

Multi-Resolution Fusion

Analyses combine inputs ranging from 10 m optical imagery to basin-level hydrological models. Rather than defaulting to the coarsest layer, the platform preserves native resolution per module and aggregates only at the reporting boundary.

03

Transparent Scoring

The Landprint Risk Index is calculated directly on the shared 1-5 scale as a weighted average of up to eleven module ratings. Weights reflect regulatory materiality and scientific consensus, and every weight is disclosed. Sector-specific adjustments ensure a forestry concession is not scored the same way as a tourism site.

04

Continuous Updates

Satellite imagery, fire detections, and protected-area boundaries refresh at their native cadence, from near-real-time to monthly. Static layers like elevation and ecoregions are pinned to their publication version and clearly labeled.

TNFD Alignment

Designed around the LEAP approach

Every module maps to a phase of the TNFD's Locate, Evaluate, Assess, Prepare methodology, so your analysis doubles as disclosure preparation.

Locate

Identify where operations interface with nature: protected areas, indigenous territories, ecoregions, biodiversity hotspots, and land-use context.

Evaluate

Quantify dependencies and impacts: ecosystem integrity, vegetation trends, climate exposure, fire incidence, terrain constraints, carbon stocks, and water risk.

Assess

Determine material risks: standardized 1-5 LRI with sector context, regulatory classification, biodiversity priority rating, and financial impact estimates.

Prepare

Generate disclosure-ready outputs: TNFD pillar mapping, sector-specific recommendations, executive PDF reports, and an exportable evidence library.

Disclosure Matrix Coverage

The platform's outputs map into a TNFD disclosure matrix that rolls evidence up across the shared Landprint themes:

Disclosure CategoryMapped Evidence
A. Business Model & Value ChainNature interfaces across water, soil, biodiversity, climate, and regenerative management
B. Dependencies, Impacts, Risks & OpportunitiesMapped evidence for dependencies, impacts, risks, and opportunities across the five Landprint themes
C. Process & ResponsePrioritization, response planning, and rights-holder review signals drawn from the shared taxonomy
D. Metrics, Targets & TransitionQuantitative indicators, change-over-time signals, and regenerative transition opportunities

Outputs

From raw signals to boardroom-ready reports

Every analysis produces an interactive dashboard with module-level breakdowns, map overlays, and drill-down metrics. When it's time to share, one click generates a comprehensive PDF report containing:

Composite Risk Score

The overall score with tier classification, sector context, and module-level contributions, explained in plain language.

Financial Impact Estimates

Revenue at risk, regulatory exposure, and ecosystem service valuations translated from ecological indicators into monetary terms.

Module Deep Dives

Each module's key metrics, classification, and spatial findings: deforestation trends, fire density, soil profiles, biodiversity priority, and more.

TNFD Disclosure Matrix

A clear mapping of which indicators support each TNFD disclosure area, with supported, needs-review, and missing states carried into the report and PDF export.

Integrity

What we disclose about our own limits

No model captures reality perfectly, and we think being upfront about limitations builds more trust than claiming precision we don't have.

  • Temporal gaps: Some reference layers are static snapshots (e.g., ecoregion boundaries, digital elevation) and do not reflect recent changes.
  • Resolution mismatch: Modules operate at different spatial resolutions. Results are aggregated to the analysis boundary, which may mask fine-grained variation.
  • Financial estimates are indicative: Revenue and cost projections use sector averages and published benchmarks. Site-specific validation is essential.
  • Biodiversity proxy: We use protected-area density and conservation priority as biodiversity proxies, not species-level occurrence data.
  • Cloud cover sensitivity: Optical satellite indices may be affected by persistent cloud cover in tropical regions.
  • Regulatory lag: Country risk classifications update per regulatory cycle and may not reflect the latest policy changes.

Ready to see what's under the surface?

Run your first analysis in minutes. No setup, no consultants, no waiting.