EarthDaily Constellation — AI-Ready Data for Unmatched Insights🔗
The EarthDaily Constellation (EDC) is a systematic, global change-detection mission delivering scientific-quality, AI-ready Earth observation data at scale. EDC provides daily access to ~92% of Earth's landmass plus coastal regions, captured at the same time and from the same angle every day — no tasking required.
EDC products are engineered for cross-time and cross-sensor consistency, supported by rigorous calibration and validation, and powered by an automated cloud-native ground segment that turns raw collections into analysis-ready imagery within hours.
Why EDC — The Four Pillars🔗
EDC is built around four principles that distinguish it from traditional commercial Earth observation:
- Consistency — Daily, nadir-looking acquisitions and a fixed local crossing time unblock programmatic analysis, predictive AI, and event triggers.
- Accuracy — Market-leading signal-to-noise ratio, stable radiometry, precise geolocation, and advanced cloud and atmospheric masking power reliable, AI-driven decisions.
- Completeness — 22 calibrated spectral bands across VNIR, SWIR, and TIR unlock coincident monitoring of vegetation, water, soil, snow/ice, surface temperature, and emissions.
- Interoperability — Low latency, standardized cloud-native formats, and compatibility with heritage missions like Sentinel-2 and Landsat streamline integration and extend your historical baseline by decades.
Designed for AI/ML pipelines
EDC's flagship AiRD (AI-Ready Data) product is processed, standardized, and structured to feed directly into machine learning models and automated analytics — eliminating the manual normalization and calibration that traditionally bottleneck Earth observation workflows.
Mission at a Glance🔗
| Parameter | Value |
|---|---|
| Satellites | 9 (+ 1 in-orbit spare) — 10 total |
| Orbit | Sun-synchronous, 608 km altitude |
| Local equatorial crossing time | ~10:30 a.m. |
| Coverage & revisit | ~92% of Earth's landmass daily (excluding Antarctica), plus coastal regions up to 100 km offshore |
| Swath width | 240 km (VNIR, SWIR, TIR) |
| Native resolution | 5 m (VNIR) · 95 m (SWIR) · 120 m (TIR) |
| Spectral bands | 22 (11 VNIR · 6 SWIR · 5 TIR) |
| Viewing angle | < 12° across track |
| Latency | Hours from acquisition to delivery |
Spectral Diversity🔗
EDC's 22-band design enables a wide range of analytics from a single acquisition:
- Visible (Coastal, Blue, Aqua, Green, Yellow, Red) — urban change detection, vegetation cover, water quality.
- Near-Infrared (Red Edge 1–3, NIR, Water Vapour 1–2) — plant health, biomass, soil moisture.
- Shortwave Infrared (SWIR 1, Cirrus, SWIR 2, Methane 1–2) — snow and ice, soil characterization, atmospheric correction, methane monitoring.
- Thermal Infrared (Wildfire, TIR 1–4) — land surface temperature, wildfire detection, emission monitoring.
Product Lines🔗
Several products will be produced using EDC, all tailored for user needs.
AiRD — AI-Ready Data (Level 2A)🔗
Processed, corrected, and standardized Bottom-of-Atmosphere (BOA) surface reflectance imagery with harmonized radiometry, consistent acquisition geometry, and embedded quality masks. Nine bands at 5 m / 10 m pixel sampling, delivered as 16-bit cloud-optimized GeoTIFFs. This data is precision corrected using ground control points, a digital elevation model, and rigorous physics-backed modelling.
Best for: AI/ML pipelines, operational monitoring, and time-series analytics across agriculture, energy, insurance, climate risk, and other industries. AiRD removes the typical preprocessing burden so teams can move directly to modeling and decision-making.
Science — Top-of-Atmosphere (Level 1C)🔗
Processed and scaled Top-of-Atmosphere (TOA) reflectance imagery which is ready for additional processing in your data processing and correction pipeline. This data is precision corrected using ground control points, a digital elevation model, and rigorous physics-backed modelling. The baseline includes 11 bands.
Best for: scientific and engineering users who need calibrated TOA reflectance prior to atmospheric correction — for example, sensor cross-calibration, radiative-transfer studies, or custom atmospheric retrieval pipelines.
Maritime🔗
Scaled Top-of-Atmosphere (TOA) reflectance imagery, processed with Systematic Correction. This product includes 4 bands — red, green, blue, and NIR — and can be delivered rapidly where speed is prioritized over high-precision geolocation.
Best for: maritime situational awareness, defense and intelligence workflows, and rapid-response monitoring at sea.
| Aspect | AiRD (L2A) | Science (L1C) | Maritime |
|---|---|---|---|
| Radiometry | Surface (BOA) reflectance | TOA reflectance | TOA reflectance |
| Geometric correction | Precision | Precision | Systematic |
| Bands (baseline) | 9 VNIR | 11 VNIR | 4 bands (RGBN) |
| Quality mask layers | Data, Cloud, Cloud Shadow, Thin Cirrus, Snow, Water | Data, View | Data, View |
| Typical user | AI/ML, operations, analytics | Scientific / engineering | Defense, maritime, gov |
Data Formats & Access🔗
EDC products are delivered as analysis-ready, cloud-optimized data, served through standardized APIs so you can query, subset, and stream pixels directly from object storage without bulk downloads.
- Imagery — Cloud-Optimized GeoTIFF (COG), one file per band, 16-bit unsigned integer, LZW-compressed, UTM / WGS84.
- Companion files — quality mask, per-band saturation masks, solar and view angle grids, atmospheric parameters (L2A), pixel processing flags (L2A), thumbnail (PNG), README, and license.
- Metadata — fully STAC 1.0.0 compliant, with standard extensions (
eo,view,raster,classification,processing,proj,file,sat,card4l) plus a customedaextension for mission-specific fields. - Delivery — STAC API for search and discovery; products framed to native acquisition footprints rather than fixed tiles.
Analysis-ready, cloud-optimized
EDC data combines two ideas: analysis-ready (radiometrically and geometrically standardized, with quality metadata) and cloud-optimized (chunked, indexed formats like COG that support partial reads over HTTP). The result: zero preprocessing, instant access, and pipelines that scale horizontally without moving terabytes.
A consistent file naming convention identifies the satellite, acquisition time, product type, processing level, and band, making it straightforward to organize and index large archives.
Use Cases🔗
EDC's combination of daily revisit, spectral breadth, and analysis-ready preprocessing supports a wide range of applications, including:
- Agriculture — crop identification, plant health, growth-stage monitoring, yield forecasting, parametric insurance.
- Energy & mining — methane and emission monitoring, mineral exploration, asset and operations monitoring.
- Insurance & climate risk — flood and wildfire modeling, disaster response, parametric triggers, exposure assessment.
- Forestry & environment — wildfire probability and burn severity, deforestation, water quality, snow and ice monitoring.
- Government, defense & maritime — persistent change detection, tip-and-cue workflows, food security, maritime domain awareness.
Building on a Heritage Archive🔗
EDC is compatible with Sentinel-2 and Landsat datasets, extending your historical baseline decades back while building forward from 2026 with consistent, daily AI-ready observations. Effective change detection requires not just current data but the ability to measure change against a stable reference — EDC's radiometric and geometric standardization is purpose-built for this.
Where to Go Next🔗
More documentation is on the way. Detailed pages covering product specifications, data formats, processing, and a quickstart guide will be released soon:
- Product Specifications (coming soon) — detailed band tables, sampling, and product components.
- Data Formats (coming soon) — file naming, masks, angles, metadata, and STAC schema.
- Processing Overview (coming soon) — auxiliary data, geometric refinement, orthorectification, and atmospheric correction.
- Quickstart Guide (coming soon) — open your first scene in minutes.