Space Innovation for Sustainable Earth

Research Question

Under what technical, institutional, and equity conditions can space research and Earth observation improve environmental decisions and public services on Earth?

Publication: Report
Published: September 23, 2026
Pillar: Space Exploration & Development
Evidence base: 9 primary sources
Review status: Foundation synthesis; not peer reviewed

Executive synthesis

Space innovation serves sustainable development when an observation changes a consequential decision and that decision improves a measured outcome. An image, an accurate model, an alert, a response, and a public benefit are different stages. This report proposes a chain of responsibility linking them: a defined problem; suitable measurements; validation; an accountable decision owner; resources for action; and independent review. It concentrates on terrestrial applications and capability sharing, complementing the Foundation’s separate work on orbital safety.

Existing public systems demonstrate useful parts of this chain. Landsat supports a wide range of environmental uses; Copernicus maintains operational mapping controls; UNEP connects methane observations to operators; and Digital Earth Africa organizes accessible data and local capability. Their existence does not prove that every application saves money, reduces emissions, or distributes benefits fairly. The Foundation proposes funding complete services and testing their additional value against realistic alternatives. Planetary science remains valuable research, but its indirect contribution should be assessed separately from an operational promise to improve conditions on Earth.

Findings

What the evidence supports.

Source markers link findings to the evidence below.

01

Open archives create an economic opportunity, not a universal return guarantee

USGS’s October 2024 account of its Landsat valuation estimated $25.6 billion in direct value for 2023. The calculation combined estimated use, expressed as scene equivalents, with a valuation per scene; it is not government revenue or an experimentally measured increase in national income. USGS also describes environmental and resource-management applications. The relevant lesson is to preserve accessible, continuous observations and examine how users employ them. It would be invalid to apply the headline value as a benefit multiplier to a new mission, country, or municipal contract. ^[1]

02

Calibration and local validation answer different questions

NASA’s NISAR documentation distinguishes instrument calibration from independent assessment of derived products, and requires sampling that represents global variability. Its radar guide explains why surface roughness, moisture, viewing geometry, and topography affect interpretation. A calibrated instrument therefore does not establish that a flood, biomass, or subsidence model works equally well everywhere. The Foundation’s inference is that procurement should require product-specific validation in the actual operating environment, including performance in difficult terrain and underserved locations, with continuing checks when algorithms or instruments change. ^{[2] [3]}

03

A blank map cannot establish the absence of harm

UNEP’s methane methodology describes minimum detection limits and constraints from sunlight, clouds, surface brightness, water, vegetation, and terrain. It also distinguishes broad-area instruments from imagery capable of identifying a facility. Undetected emissions are therefore not zero emissions, and a coarse plume does not necessarily identify a responsible operator. These are limits of the cited methane observing approach, not a claim that every satellite instrument requires daylight. A responsible dashboard should expose observation coverage, detection conditions, and attribution confidence alongside reported events. ^[4]

04

Detection and action remain separated by a large institutional gap

UNEP reported in October 2025 that responses to its methane alerts had increased from 1% to 12%, with more than 3,500 alerts across 33 countries and 25 documented mitigation cases in ten countries since MARS began. These are the report’s period-specific operational figures, not a live September 2026 rate. An acknowledgment is not a repair, and a repair is not independently established avoided emissions. The evidence supports investing in responsible contacts, investigation capacity, response funding, and verification alongside increasingly capable sensors. ^[5]

05

Operational mapping treats quality as a process

Copernicus Rapid Mapping subjects maps and geospatial products to provider and Joint Research Centre checks. Criteria include thematic and positional accuracy, completeness, readability, and consistency; rejected products are revised and versioned. This is a useful public-service model because emergency information must be timely while remaining correctable. It does not establish that every map is error-free or appropriate for every decision. Users need the observation time, delivery time, product version, and known limitations, particularly when decisions continue after an initial emergency picture has changed. ^[6]

06

A warning service needs capability beyond the satellite

The 2025 WMO/UNDRR global status publication reported 119 countries with a multi-hazard early warning system and persistent gaps, including only 43% of small island developing States reporting systems. Those country reports do not measure universal household reach or isolate a satellite’s contribution. WMO emphasizes governance, investment, and community co-development. The Foundation interprets this as a requirement to connect observation and forecasting with trusted communications and practical response arrangements. National system presence cannot stand in for evidence that people received, understood, and could act on a warning. ^[7]

07

Access includes institutions and skills, not just download rights

Digital Earth Africa describes an operational service combining free, open, analysis-ready data with African governance, partnerships, and in-country expertise. Its model makes capability development part of infrastructure. This is institutional evidence of an approach, not an independent estimate of livelihood or employment impact. A public archive may still be difficult to use without computation, local reference data, staff time, or a route into agency decisions. The Foundation therefore proposes judging access by recurring local use and decision ownership, with users able to challenge unsuitable products. ^[8]

08

Planetary research offers hypotheses rather than a ready-made Earth forecast

A NASA-hosted 2016 research abstract by Way and colleagues describes three-dimensional simulations of possible early Venus climates under explicit assumptions about atmosphere, oceans, topography, and rotation. It illustrates how comparative planetology can test physical understanding beyond a single planet. It does not show that Earth will follow Venus’s trajectory, nor supply a quantified benefit for an adaptation programme. The Foundation proposes evaluating this work through scientific contribution, reproducibility, and knowledge transfer while requiring separate evidence for any claimed near-term public-service application. ^[9]

Institutional Implications

What this changes in practice.

These are design implications drawn from the synthesis, not claims that one institutional model will fit every jurisdiction.

01

Begin with a decision that someone can actually make

The Foundation proposes a decision charter before acquiring additional data. It should name the agency or community partner, the choice to be made, the deadline, the consequences of acting or waiting, and the funds available to respond. For example, a water-stress signal might trigger a field inspection rather than an automatic restriction. If no authorized actor can use the information, the project should first resolve that institutional gap.

02

Purchase a maintained service with explicit failure modes

Public contracts should specify observation availability, usable coverage, delivery time, validation by relevant geography, correction procedures, and continuity arrangements. They should price staff, computation, ground measurements, maintenance, and user support as well as data. Where vendors provide proprietary models, agencies should retain enough documentation and audit access to evaluate performance. Procurement should compare the proposed service with existing ground networks and simpler methods before assuming an additional satellite product is the best investment.

03

Protect local authority over consequential interpretation

Affected communities and frontline agencies should help define useful outputs, acceptable errors, and appeal routes. A land classification must not automatically decide tenure, compensation, or enforcement. Publish aggregated information when precise locations could expose vulnerable people or sensitive ecological sites. Finance local analysts and reference-data stewardship, including long-term employment and fair credit. These are Foundation safeguards intended to make participation meaningful; the sources reviewed do not establish a universally optimal governance arrangement.

04

Make uncertainty visible at the point of use

A product should separate measured signals, modelled quantities, and human decisions. Display missing observations and stale data clearly. Present uncertainty in terms the decision owner can act on, such as whether an inspection is warranted, while retaining technical documentation for independent review. False alarms, missed events, and unresolved attribution should remain visible in evaluation. Avoid a single confidence score that conceals different uncertainties in location, timing, classification, and magnitude.

05

Protect a balanced public science portfolio

Operational services need predictable maintenance while exploratory research needs room to produce knowledge without guaranteed commercial returns. The Foundation proposes distinct budgets and evaluation criteria for those purposes, plus explicit routes for promising research to undergo application testing. Continuity, openness, and public capability should weigh alongside novelty. Environmental and orbital operating costs also belong in investment appraisal, although this report does not quantify launch impacts or repeat the Foundation’s separate orbital-governance analysis.

04 / Policy options

Choices and their tradeoffs.

Foundation analysis of the options. Evidence informs the choices; it does not remove their costs or uncertainty.

1

Foundation proposal: a public Earth-observation service standard

Pilot a common disclosure template covering decision purpose, accountable owner, measurement lineage, geographic validation, uncertainty, correction history, accessibility, and response arrangements. Copernicus’s documented quality process provides a practical reference rather than a universal certification. Start with voluntary procurement requirements in a small set of agencies; independently test whether the standard improves reliability without excluding capable smaller providers through unnecessary paperwork. ^{[6] [2]}

2

Foundation proposal: a response-and-verification fund

Pair observation spending with a protected budget for local investigation, corrective action, and follow-up. Methane applications could fund confirmed leak investigation and verification rather than paying simply per alert. Eligibility should require an identified decision owner and transparent additionality rules, while preserving obligations that operators already bear. Evaluate the fund against ordinary enforcement and maintenance spending; do not count all associated reductions as satellite-caused. ^[5]

3

Foundation proposal: regional capability compacts

Support public agencies, universities, and community organizations to maintain shared computing, local validation sites, and multilingual training. Require portable outputs and practical exit arrangements when a platform changes. Digital Earth Africa provides an institutional reference, but each region needs its own governance and financing design. Assess whether the compact expands independent problem-solving and recurring service use, rather than treating workshops or registered accounts as proof of equitable access. ^[8]

05 / Implementation

From a proposal to a working system.

A proposed sequence for local adaptation. These stages are not a claim of funded or completed Foundation activity.

Foundation proposal · first 90 days
Select two decisions and establish a baseline
Choose contrasting applications, such as flood-response planning and pollution investigation, with willing public partners. Document current decision times, costs, error consequences, ground information, and distribution of benefits. Agree access and privacy rules, a realistic comparison service, evaluation responsibilities, and conditions for stopping. Avoid selecting only settings with unusually strong staff or connectivity, which would conceal implementation barriers elsewhere.
Foundation proposal · months 4–9
Validate and rehearse before operational reliance
Run the service alongside current practice. Use independent reference observations and a prespecified validation sample that includes difficult locations and missed-event searches. Rehearse communication failure, model changes, and incorrect alerts. Record who receives information and whether they have authority and resources to respond. Move into limited operational use only when the partner accepts documented performance and remaining uncertainty.
Foundation proposal · months 10–18
Evaluate decisions, distribution, and full costs
Where ethical and feasible, compare a phased introduction with otherwise similar services or periods. Account for weather severity, policy changes, staffing, and other factors that could explain outcomes. Publish correction rates, unsuccessful cases, and cost per useful decision. Include community interviews and burdens imposed by erroneous classifications. An evaluator independent of the supplier should distinguish measured results from estimated avoided losses.
Foundation proposal · after evaluation
Scale selectively and preserve a fallback
Expand only applications with credible incremental benefit, acceptable error burdens, and funded maintenance. Retain usable archives and an alternative operating procedure if a mission, vendor, connection, or grant ends. Schedule reviews when products change and allow programmes to be redesigned or retired. Publish the reasons for expansion or closure so unsuccessful pilots still improve public learning.

06 / Measurement

What progress would mean.

Proposed measures for evaluation. Baselines, thresholds and reporting responsibilities must be agreed before implementation.

Usable observation coverage

Share of the agreed area and decision periods with observations that meet application quality criteria.

Interpretation: A nominal revisit or satellite footprint does not establish usable coverage.

Validated decision performance

False-positive and missed-event rates against independent reference observations, disaggregated by setting.

Interpretation: Reference data may also be incomplete; report sampling and uncertainty.

Time to verified action

Elapsed time from relevant observation to investigation, action, and separately verified outcome.

Interpretation: Keep acknowledgment, action, and outcome dates separate.

Incremental public value

Outcome difference against the comparison service and full recurring cost, with distribution across affected groups.

Interpretation: Do not attribute all programme benefits to satellite information.

Local operational capability

Share of routine interpretation and service maintenance performed by locally accountable teams.

Interpretation: Staff turnover and unpaid community work can conceal fragile capacity.

Research Programme

What the foundation should test next.

Preregister a flood-mapping decision trial with an emergency authority. Compare usable coverage, delivery time, field-verified errors, and decisions changed relative to existing practice; exclude scenarios where delayed information would endanger people.
Audit a sample of methane notifications from detection through response and verification. Include unresolved and cloud-obscured cases; estimate reductions only with a transparent baseline, uncertainty interval, and explanation of other contributing interventions.
Compare two access models for a low-resource partner: centrally delivered products and locally maintained analysis with training. Measure full cost, independent troubleshooting, staff retention, and usefulness after initial support ends.
Test a plain-language uncertainty and correction notice with frontline users and affected communities. Assess whether it changes interpretation and enables challenges, and revise it when users confuse missing observations with absence of risk.

Method & Limits

How to read this brief.

Method

This targeted evidence synthesis was completed on 23 September 2026 using nine publicly accessible primary-source pages from space agencies, service operators, and intergovernmental organizations. Searches covered Earth-observation valuation, calibration and validation, operational mapping, methane response, warning-system coverage, African data capability, and comparative planetary science. It is not a systematic review or a new economic model. Quantitative claims retain their source periods; institutional descriptions establish how a service is designed, not its causal impact. Implications, policy options, implementation phases, and proposed measures are Foundation judgments for testing.

Limitations

The Foundation has not reproduced underlying satellite retrievals, valuation models, or programme evaluations. Provider accounts may emphasize successful uses; missed events and unreported failures are harder to observe. Global statistics conceal local operating conditions, and reference measurements carry uncertainty. Reported methane response figures are a 2025 snapshot. The older Venus study is an illustrative modelling example, not a current consensus estimate of planetary history. No claim is made here that more satellites necessarily produce better outcomes or that a specific proposed policy already has demonstrated effects.

Unresolved evidence

Independent comparisons of complete observation-to-action services with realistic non-satellite alternatives remain insufficient in the reviewed material.
Distributional evidence is needed on who gains from improved information and who bears false classifications, enforcement burdens, and ongoing data costs.
Long-term service continuity, local staff retention, and verified outcomes after pilot funding ends need stronger public reporting.

Source Record

Read the official sources.

[1]

U.S. Geological Survey · 9 October 2024; valuation reference year 2023 · Accessed 2026-09-23

Landsat’s Economic Value increases to $25.6 Billion in 2023

Source-owned explanation of valuation inputs and use cases; economic estimate is not revenue or causal GDP gain.
[2]

NASA · Updated 4 August 2025 · Accessed 2026-09-23

NISAR: Calibration & Validation

Technical definitions and validation design; describes mission plans, not a claim that every local product has passed validation.
[3]

NASA · Updated 2 June 2026 · Accessed 2026-09-23

Get to Know SAR – Overview

Radar interpretation and geometric limitations; sensor capability is distinct from application accuracy.
[4]

UN Environment Programme, International Methane Emissions Observatory · Undated living methodology; accessed 23 September 2026 · Accessed 2026-09-23

Eye on Methane: Frequently asked questions

Detection limits, attribution, notification, and verification. Some descriptive programme text is older; not used for current satellite counts or sector coverage.
[5]

UN Environment Programme · 22 October 2025 · Accessed 2026-09-23

Better data driving action on methane emissions, but more work needed

Period-specific MARS response and documented mitigation figures; response is not equivalent to emissions reduction.
[6]

European Commission, Copernicus Emergency Management Service · Updated 31 January 2025 · Accessed 2026-09-23

Rapid Mapping Manual: Quality Control

Operational quality criteria and correction/versioning process.
[7]

World Meteorological Organization and UN Office for Disaster Risk Reduction · 12 November 2025 · Accessed 2026-09-23

Global Status of Multi-Hazard Early Warning Systems 2025

Country-reported system coverage in the 2025 publication; not individual reach or a satellite attribution study.
[8]

Digital Earth Africa · Undated programme page; accessed 23 September 2026 · Accessed 2026-09-23

About Us

First-party operating model and governance principles; impact aspirations are not treated as independently verified results.
[9]

NASA Goddard Institute for Space Studies; Way and colleagues · 2016 research article; source record updated May 2026 · Accessed 2026-09-23

Was Venus the first habitable world of our solar system?

Primary author abstract of model-dependent climate scenarios; illustrative scientific method, not an Earth forecast.

Suggested citation: True Tomorrow Foundation (2026). Space Innovation for Sustainable Earth. Report. Reviewed September 23, 2026.

Explore all research Contribute evidence

Space for Earth