10
NASA Programme Assets Beyond SERVIR
SERVIR is one programme inside a much larger NASA applied-science estate, and most of that estate is open, free and already operational. This section inventories what the other programmes provide that these three platforms would otherwise build — data products, analysis platforms, foundation models, training curricula and funding mechanisms — and maps each to the matrix cell or platform layer it serves. The recurring finding: several cells currently marked "new build" in §04–06 have a NASA asset sitting behind them that nobody has claimed.
How to read this section. Every row is checked for availability, not announcement — an upcoming mission or a product still in development is labelled as such rather than counted as capability. Access route matters as much as existence: some of these are open APIs, some need an Earthdata login, and a few need a funded NASA relationship, which is a materially different proposition for an independently-led hub.
Global Risk — what NASA already runs
| Asset | Programme | What it gives the matrix | Clock | Access | Maturity |
|---|---|---|---|---|---|
| NEX-GDDP-CMIP6 | NASA Earth Exchange | Bias-corrected, statistically downscaled daily CMIP6, global, multiple SSPs. Replaces the generic "CMIP6" citation in every Clock 3 cell across drought, heat, cold and flood-precipitation at once. | Clock 3 | Open — AWS Open Data, GEE, NAS | Operational |
| ARIA Damage Proxy Maps | ARIA, JPL/Caltech | InSAR post-earthquake deformation and damage maps, hours to days after an event. Fills the matrix's most visible hole — earthquake had no NRT source. Decade-plus operational record across Mexico, Italy, Japan, Ridgecrest, Palu. | Clock 1 | aria.jpl.nasa.gov — licence terms not stated, confirm | Operational |
| Sea Level Change Portal incl. IPCC AR6 projection tool | NASA Sea Level Change Team (JPL) | Location-specific SLR projections on AR6 scenarios. Turns §04's generic "IPCC SLR scenarios" into a named, citable operational tool. | Clocks 2–3 | Open web tools | Operational |
| Black Marble (VNP46) | VIIRS Land / Earthdata | Moonlight-corrected nighttime lights, ~500 m, near-daily. The concrete source behind LitPop's nightlight input, which §04's exposure table currently cites generically — plus outage detection after an event. | All three · Assets | Open — LAADS DAAC | Operational |
| GRACE-FO drought products | GRACE-FO (JPL) + NDMC | Weekly global soil-moisture and groundwater wetness percentiles at ~13.7 km. An independent sub-surface drought signal alongside the NOAA VHI/ESI + IMERG-SPI stack. | Clock 1 | Open — nasagrace.unl.edu, GES DISC | Operational |
| OPERA DSWx | OPERA, JPL | Global 30 m surface-water inundation, several observations a week, from both optical (HLS) and SAR (S1). Cross-validation for HydraFloods rather than a replacement. | Clock 1 | Open — PO.DAAC | Operational; DSWx-NI pending |
| GPM / IMERG | Global Precipitation Measurement | Half-hourly 0.1° global precipitation, NRT and a 27-year research-grade record — the same product serves Clock 1 and the Clock 2 baseline. | Clocks 1–2 | Open — GES DISC | Operational |
| NASA POWER | Applied Sciences / Langley | Free REST API for meteorology, solar and agroclimatology, hourly to climatology. An alternative or complement to ERA5 for the heat and cold indices §04 marks as new build. | Clocks 1–2 | Open REST API | Operational |
| NRT Global Flood Product | LANCE + U. Maryland | NRT optical flood detection plus a 23-year historical flood archive — the archive is the interesting half, since Clock 2 needs an event set. | Clocks 1–2 | Open — LANCE | Operational |
| Disasters Mapping Portal | NASA Disasters Program | Event-activation hazard layers combined with exposure and vulnerability, free GIS downloads. A coordination surface rather than a data source. | Clock 1 | Open portal | Operational |
| FIRMS (LANCE) | LANCE / Earthdata | Already the platform's NRT wildfire backbone. Listed for completeness — no action beyond keeping the direct API integration. | Already used | Open API | Operational |
| NISAR | NASA–ISRO | Dual L/S-band SAR, 12-day global repeat, crustal deformation. Second flood-SAR source and an earthquake deformation time series. Launched 30 July 2025 — product maturity not yet confirmed. | Clocks 1–2 | Stated open; portal not located | Launched, products pending |
Food Security and NRM — what NASA already runs
| Asset | Programme | What it gives the matrix | Serves | Access | Maturity |
|---|---|---|---|---|---|
| MAAP Multi-Mission Algorithm & Analysis Platform | NASA–ESA joint | The single highest-value asset in this section. Cloud compute plus an algorithm development environment with harmonised biomass data across GEDI, ESA BIOMASS and NISAR. This is the carbon/MRV workbench §06's N5 would otherwise have to build. | NRM | Public sign-up | Operational since 2021 |
| NASA Harvest | Applied Sciences Agriculture | Already a named §05 partner — but the inventory adds detail: it runs the GEOGLAM Crop Monitor for Early Warning, holds the open Crop Harvest Database of ML training data, and its PIs already work in SERVIR regions. | Food Security | Public tools; consortium by partnership | Operational |
| NASA Acres | Applied Sciences Agriculture | Harvest's US counterpart. The transferable asset is ARYA, a county/national yield-forecasting model, and CONUS field-boundary extraction — both directly relevant to §05's Clock 2. | Food Security | portal.nasaacres.org | Operational; projects research-stage |
| GEDI | NASA / ISS lidar | Forest structure and above-ground biomass, footprint and gridded. Canonical structure input for §06's carbon stock row. Caveat: stowed March 2023, reinstalled April 2024 — that gap must be documented in any MRV baseline. | NRM | Open — LP DAAC / ORNL DAAC | Operational, with a data gap |
| ICESat-2 (ATL08) | NASA / NSIDC DAAC | Canopy and terrain height. The natural gap-filler for GEDI's stow period, and denser coverage. | NRM | Open — NSIDC | Operational |
| ORNL DAAC mangrove products | Carbon Monitoring System | Global mangrove distribution, above-ground biomass and canopy height — a ready-made input for §06's mangrove row, which currently reads as two regional tools. Caveat: a static 2000s/2010 snapshot; vintage acceptability is a real decision. | NRM | Open — daac.ornl.gov | Published, static |
| Carbon Monitoring System (CMS) | NASA programme since 2010 | The funding mechanism behind the mangrove and biomass-carbon products. The route to petition for a refresh of a dataset whose vintage does not work. | NRM | ROSES proposal | Operational |
| HLS Harmonized Landsat Sentinel-2 | NASA GSFC / LP DAAC | 30 m harmonised reflectance, ~3-day tropical revisit, 2–3 day latency. The practical EO backbone for crop type, land cover and deforestation alerts. HLS-LL (≤6 hr latency) is in development for ~2027 — worth planning for, not yet real. | Both | Open — LP DAAC | Operational; HLS-LL pending |
| SMAP | NASA / JPL | Global soil moisture, near-daily. Durable constraint: the radar failed in 2015, so resolution is radiometer-only ~40 km — a permanent limit for the field-level irrigation-timing row in §05, not a temporary one. | Food Security | Open — Earthdata | Operational, degraded |
| Landsat / Landsat Next | NASA–USGS | Fifty years of multispectral archive — the long baseline behind every land-cover-change claim. Landsat Next is in development with no confirmed launch date. | Both | Open | Operational; Next pending |
| CSDA Commercial SmallSat Data Acquisition | NASA Earth Science Division | Centrally licensed commercial imagery — Planet, Maxar, Spire, Airbus. A no-cost route to parcel-scale validation of deforestation alerts and crop boundaries. Two hard limits: ~30-day latency, and eligibility requires a funded NASA relationship — which independently-led hubs may not have. | Both | Gated — NASA-affiliated only | Operational since 2022 |
Shared infrastructure — what not to build
The §07 argument is that these are three domains over one shared system. This is the same argument one level down: several layers of that shared system already exist inside NASA, open-licensed and redeployable.
Data layerEarthdata +
Earthdata + earthaccess
Planet-scale archive with single sign-on, and an MIT-licensed Python SDK handling CMR search, auth and cloud streaming. Removes any need to build a catalogue-search or auth layer — the platform becomes thin config on top.
Compute + dashboards
VEDA
The one NASA asset explicitly documented to be forked and redeployed by another team — STAC API, raster API, JupyterHub and dashboard front end. For three platforms each needing exactly that stack, forking VEDA is a materially smaller lift than building it.
Model layer
Prithvi (IBM–NASA)
Confirmed Apache 2.0. Prithvi-EO-2.0 at 300M/600M with ready flood and burn-scar task heads; Prithvi-WxC-2300M for weather and climate. Fine-tuning a checkpoint avoids foundation-model pretraining entirely.
Capacity ladder
ARSET + Openscapes
ARSET has trained 100,000+ people across 183 countries with an open domain curriculum. Openscapes adds a CC BY 4.0 cohort-mentorship model. Together they cover both ends of §03's ladder without authoring a curriculum from zero.
Governance
SPD-41a + TOPS
NASA's open-source science policy, in force since December 2022, plus its training arm. An off-the-shelf open-science governance template for the FAIR commitments in §03 — adopt rather than draft.
Funding
ROSES A.13 and A.9
A.13 "Accelerating Earth Solutions" funds applied EO across every domain these platforms touch. A.9 funds applications of large geospatial foundation models specifically — which is precisely the GeoAI work in §03. Non-US organisations can participate on a no-exchange-of-funds basis.
One funding finding worth acting on, and one worth watching.
ROSES A.9 — "User-Centered Applications with Large Earth Foundation Models" — is a live funding line aimed exactly at what §03 describes, and international partners can participate. That is the closest match between this plan and an open NASA solicitation.
Meanwhile the SERVIR-specific Applied Sciences Team line was not solicited in ROSES 2024. Treat it as something to monitor annually rather than a standing open door — particularly now that hubs are independently led.
What this changes in the matrices
| Matrix cell | Was | Now |
|---|---|---|
| Earthquake × Clock 1 (§04) | Partner — USGS ShakeMap | Add ARIA Damage Proxy Maps — InSAR deformation, operational, purpose-built for this window |
| All Clock 3 cells (§04–06) | Generic "CMIP6 / ISIMIP" | NEX-GDDP-CMIP6 — named, downscaled, daily, open. One substitution fixes the whole column |
| Sea level rise × Clocks 2–3 (§04) | Generic "IPCC AR6 projections" | NASA Sea Level Change Portal — the actual operational tool |
| Heat / cold index (§04) | New build — nine cells depend on it | Still a build, but NASA POWER and NEX-GDDP-CMIP6 supply the inputs, shrinking it |
| Assets exposure stack (§04) | LitPop, nightlights generically | Black Marble VNP46 — the named nightlight product |
| Forest carbon stock × Clock 2 (§06) | CCI Biomass + GEDI | Add MAAP as the processing platform and ICESat-2 for the GEDI gap |
| Mangroves × Clock 2 (§06) | Built — regional (Ecuador, Guyana) | ORNL DAAC global mangrove products give a global baseline under the two regional tools |
| Shared capability layer (§07) | To be built | VEDA fork + earthaccess + Prithvi — an integration job rather than a build |
Confidence and caveats. Availability was checked separately from announcement throughout. Not yet real: NISAR products, OPERA DSWx-NI, HLS-LL (~2027), Landsat Next. Degraded rather than nominal: SMAP (radar lost 2015, ~40 km permanent), GEDI (2023–24 stow gap). Not independently confirmed and worth verifying before citing: ARIA's licensing terms, the IPCC AR6 sea-level tool page and Black Marble's product page (both failed direct fetch), MAAP's heavy-compute allocation process, and a single canonical ARSET materials licence. Several Applied Sciences programme areas publish no funding amounts or application procedures at all — converting those from "known to exist" into an actionable target needs a direct conversation with NASA HQ.