Most restoration projects stall here. The ecological evidence funders require before committing exists in satellite and biodiversity data — vegetation trajectories from Sentinel-2, forty years of water history from the JRC, deforestation pressure from Global Forest Watch, biodiversity signals from GBIF. It's public, it's peer-reviewed, and it's already watching every restoration site on Earth.
Most project developers can't access it in a useable form. Field ecologists, camera traps, and eDNA surveys take months and cost money that hasn't been raised yet. The evidence that would unlock funding sits behind work the project can't afford until it's funded.
Archaster breaks that deadlock. Satellite and biodiversity data, structured for restoration work, available for any site in minutes.
Book a callBaseline evidence, assembled from sources funders and reviewers already accept.
Eight spectral indices from Copernicus Sentinel-2 at 10m resolution, available from 2015 to present. NDVI, EVI, SAVI for vegetation vigour. NDMI for moisture stress. NDRE for crop-specific nitrogen signal. NBR for burn severity. NDWI for surface water. BSI for bare soil. Pre-intervention vegetation condition, documented at pixel level, citable in baseline documentation.
Tree cover loss (2001–2024), tree cover gain (2000–2020), and tree cover density (2000 baseline) from Hansen/UMD. Near-real-time deforestation alerts from Global Forest Watch (GLAD, RADD, GLAD-S2), with buffer zone analysis at 500m and 2km.
Tree canopy height at 10m global resolution (ETH Zurich, 2020) — forest structure visible from space. Active fires from NASA FIRMS (VIIRS), near-real-time with temporal selector.
Forty years of surface water history from the European Commission Joint Research Centre: occurrence, seasonality, transitions (1984–2021 at 30m). Permanent water loss, seasonal shifts, new water bodies — the pre-intervention hydrology of a site as it existed before degradation, and the trajectory since.
Complemented by NDWI and NDMI from Sentinel-2 for current-state water detection at 10m with 5-day revisit cadence.
Species occurrence records from the Global Biodiversity Information Facility, research-grade and consensus-based observations only, from 2015 to present. Observations are categorised into functional ecological groups for restoration-relevant interpretation:
Functional gaps — groups with zero records or substantially below ecoregion expectation — are flagged. So are areas where GBIF coverage is genuinely sparse, so a missing functional group isn't misread as ecological absence when it might just reflect survey coverage.
Observations are computed for the site, plus 2km and 10km buffer zones, giving a picture of local presence and landscape connectivity.
Automatic screening against Natura 2000 protected areas (EU), potential wildlife and climate corridors (Global Safety Net), and indigenous and community territories (LandMark). Ecoregion and biome classification (RESOLVE). Global peatlands baseline (GFW/WRI, 2023). Land cover at 10m (ESA WorldCover, 2021).
These aren't add-ons. For restoration projects, landscape context is the frame the project exists within. A site adjacent to a climate corridor has different restoration value than one that isn't. A site overlapping indigenous territory carries due diligence obligations. Automatic screening means these aren't missed.
Funders contribute. Funders should be able to see what they contributed to.
Most restoration funding is pooled. A funder's contribution goes into the project, and they receive narrative updates, species counts, photos. They cannot see the specific hectares their money covers or what's happening to those hectares. The investment becomes an abstraction.
Archaster partitions restoration project areas using H3 — a hexagonal spatial indexing system — assigning identifiable cells to specific funders or cohorts. Each funder sees their cells: vegetation condition, water status, deforestation alerts in their area, updated as the data comes in.
Available now: H3 cell maps for any project area, exportable as images and tables, ready for presentations, pitch decks, or funder-facing landing pages. Cell resolution configurable to project size and funder tier.
Shipping in the platform shortly: Direct in-platform access with individual funder viewer seats, already working in our development environment.
Honest about where the platform ends and expert work begins.
Archaster provides ecological baseline evidence and continuous landscape monitoring inputs for restoration projects. It draws from peer-reviewed, openly licensed data sources that auditors, funders, and regulators already accept.
Professional judgment from ecologists, VVBs, and specialist consultants remains essential. Archaster shortens the path to having the evidence those professionals work with.
Nature data should work for the people trying to protect it.
Archaster is a founder-led, independent company. We have no conflicts of interest with land developers, offsetting schemes, or credit issuers. We don't rate credits, sell offsets, or take positions in projects.
We build tools that let restoration organisations, project developers, and funders see landscapes clearly — and that stay honest about what the data can and can't say. Every source is peer-reviewed and openly licensed. Every claim is cited. Every limitation is surfaced rather than hidden.
Book a callNo. MRV — Monitoring, Reporting, and Verification — is the three-part trust infrastructure for credible carbon and biodiversity credits. Verification, specifically, is done by independent accredited bodies (VVBs). Archaster produces baseline and monitoring data inputs that feed into MRV workflows. The methodology-specific calculations and the verification itself are done by specialist tools and VVBs.
No, and we're deliberate about this claim. These are methodologies, not data products — a project is compliant with VM0047, not a platform. What Archaster provides is the spatial evidence layer that projects and their verifiers draw from. Our data (Sentinel-2 indices, Hansen tree cover, JRC water, GBIF) is openly licensed and already accepted by Verra, Gold Standard, Plan Vivo, and other standards reviewers as legitimate source material. Methodology-specific calculations — like VM0047 stocking indices and performance benchmarks — we don't do.
Not directly. Forest age requires dendrochronology, forest inventory, or very high-resolution LiDAR analysis — none of which we do. What Archaster can do is build a structural and historical case that a forest is likely mature and undisturbed: canopy height from ETH Zurich (10m) as a structural proxy, Hansen tree cover loss records back to 2001 showing no recent clearing, stable vegetation signal over time. This gives you an argument about forest maturity, not a specific age.
Yes, and this is something the platform does well. Hansen tree cover loss at 30m from 2001–2024 shows any detected clearing events. GFW near-real-time deforestation alerts (GLAD, RADD, GLAD-S2) extend the picture to the present. Active fire detection via NASA FIRMS covers burn events. Buffer zone analysis at 500m and 2km shows encroaching pressure even when the site itself is stable. For projects where demonstrating absence of recent clearing matters, this combination is defensible in funder and reviewer conversations.
Not yet, but we'd love to develop it! Deadwood detection at individual-tree level requires LiDAR, very high-resolution optical imagery, or SAR (radar) backscatter analysis. Sentinel-2 at 10m doesn't resolve individual standing or fallen dead trees. Canopy height data might indirectly suggest gaps where large trees have fallen, but we don't treat this as a deadwood quantification method.
No, not yet. Archaster uses multispectral data from Copernicus Sentinel-2 (13 bands, 10m resolution, 5-day revisit) and derived products. We chose multispectral as the foundation because it combines global coverage, dense temporal cadence, and peer-reviewed open licensing in ways hyperspectral sources don't yet match. Hyperspectral is something we might add for specific analytical questions — species-level classification, some decay detection — where it earns its place alongside the multispectral baseline, but not as a replacement for it.
Yes. Archaster uses an open-source, global spatial indexing system to partition any project area into identifiable cells at configurable resolution. Each cell can be assigned to a specific funder or cohort. Cell maps are available now as image and table exports, ready for presentations and pitch decks. Direct in-platform access with individual funder viewer seats is shipping shortly.
There's no hard minimum, but the reliability of findings scales with area. For very small plots (under a few hectares), Sentinel-2's 10m pixels give you limited data density, GBIF records in the surrounding buffer will be thinner, and temporal signals are noisier. We work with plot sizes from small individual parcels up to full project areas of hundreds of square kilometres — but for plots under roughly 10 hectares, we'd be explicit about what the data can and can't tell you, and we'd expect you to pair it with site-level field observation.
In development. Cell-level partitioning and exportable cell maps are available now. Direct funder viewer seats — where each funder logs into Archaster and sees only their assigned cells, with vegetation, water, and deforestation signals updating continuously — are currently running in our development environment and shipping to the production platform shortly.
GFW is an excellent free platform focused on forest monitoring, and we use their deforestation alerts as one data source. Archaster adds multi-source synthesis — satellite vegetation indices, forty years of water history, GBIF biodiversity, ecoregion context, H3 partitioning, and role-adapted AI interpretation — structured specifically for restoration workflows. GFW is a monitoring platform; Archaster is a restoration project intelligence layer that incorporates GFW data alongside many others.