Appearance
International Celestial Reference Frame Supply Chain
EGV Level: 3 — Global · Geometric | Date: September 2026 | Version: 5.1
Reference: EGV White Paper V7.0 (GGOS, January 2026) · Guiding Principles v4.6 · 1st Joint Development Plan
Major revisions:
- v5.1 — Replaced the raw
data/project/agy_research_findings_function_validation.mdfilename reference in the v5.0 entry below with a working link to the hosted portal (un-ggce.merchantsofmalta.com). Part of a sweep across the whole deliverable set for the same issue, per the user's request. - v3.6 — Renamed the per-capability "Phase 1/2/3" priority tag to "Horizon 1/2/3", matching the framework used throughout Work Assignment 1b.
- v3.8 — Added the Capability Evidence Trail (confirmed / asserted / considered / queued), extending the method piloted on the Satellite Orbits annex — Mandate is flagged as the strongest Queued case found across any annex, since Tier 3's own prose states no formally designated body holds accountability for ICRF combination.
- v4.1 — Incorporated previously-unaddressed Russia-Ukraine and NASA GSFC workforce/shutdown evidence into the Tier 0–2 narrative and Evidence Trail (Confirmed count 16 → 12 of 16); flagged for the Technical Product Working Group that GSFC's dual-hosting of the Washington Correlator and CDDIS may mean Tier 1's institutional concentration is broader than its Tier Risk Score currently reflects.
- v4.2 — Corrected the ⚠ ("below Level 2 floor") symbol, misapplied to four capabilities whose scores don't actually fall below the floor; corrected "IGS-level" to "IVS-level" in a sentence entirely about the VLBI network.
- v5.0 — Converted from the retired five-tier structure to eight real functions (F1...F8), following Work Assignment 2 v5.0's precedent and its own external validation pass (full findings). Tier 0 becomes F1 (Astrometric & Geodetic VLBI Observations) unchanged in scope. Tier 1 splits three ways into F2 (Raw Data Logistics & Telemetry — physical disk shipment and e-VLBI), F3 (Primary Archiving & Metadata Registration — CDDIS/BKG), and F4 (Interferometric Correlation & Fringe-Fitting) — the prior grouping bundled correlation into archiving, obscuring that correlation and archiving sit at two different NASA GSFC-hosted functions rather than one, a dual-hosting concentration worth reading as its own finding, not folded into a single "archiving and correlation" line. Tier 2 becomes F5 (Astrometric VLBI Analysis); Geodetic Software and Tools is now shown at both F4 and F5 — the same underlying capability score, since the same small-team-maintained software (Calc/Solve, VieVS) underpins both correlation and analysis, matching Work Assignment 2's own structure. Tier 3 becomes F6 (ICRS Multi-Catalogue Combination). Tier 4 splits into F7 (Astrophysical Source Structure & Validation — the Gaia optical cross-comparison) and F8 (Formal IAU Designation & Dissemination) — the prior bundling treated Gaia validation and IAU adoption/distribution as one step, when they are organisationally distinct (Bordeaux Observatory/USNO/IAU Commission A1/A2 for validation; IAU General Assembly/CDS Strasbourg/IERS ICRS Centre for designation and dissemination). F8 now also carries Data Distribution (2/2/1/3), a capability Work Assignment 2's own F8 structure includes that this annex had not previously scored — added here at the shared State of Geodesy 2026 baseline, raising the capability count this document tracks from 16 to 17 of 58. Matching Work Assignment 2 v4.0/v6.0's precedent, removed the retired Tier Risk Score / PPTD Gap × Step Criticality apparatus and the per-capability Horizon 1/2/3 tag throughout — PPTD scores are shown as descriptive context only. Standardised the ⚠ flag to "corrected average below 2.5," matching the convention already applied to the Satellite Orbits, ITRF, and EOP annexes — this supersedes this document's own v4.2 per-dimension "below the Level 2 floor" definition, which produced inconsistent flagging (Data Preservation, VLBI Data Processing and Analysis, and Geodetic Services all scored at or below 2.25 without the flag under the old rule; all three are flagged under the new one). Corrected the PPTD averaging convention to exclude not-applicable dimensions rather than scoring them 0, per Work Assignment 1b v7.0 Annex A.2 — Equipment Calibration and Maintenance's average moves from 2.50 to 3.33 (Data excluded) and Training and Education's from 1.25 to 1.67 (Data excluded, matching the "2/1/2/—" notation Work Assignment 2 already uses for this capability).
Subsequent versions applied editorial and terminology consistency fixes only unless otherwise noted above.
PPTD (People, Process, Technology, Data) scores: State of Geodesy 2026 — A baseline maturity assessment (UN-GGCE, 16 March 2026)[^sog2026]
Audience
This document serves two readerships: (1) UN-GGCE leadership and technical counterparts already familiar with this engagement; and (2) UN-GGCE staff and Member State stakeholders new to this specific analysis. Readers new to the engagement should read this Purpose & Reading Guide; the function-by-function detail that follows supports the summary findings reported in the Work Assignment 2 and Work Assignment 3 reports.
Purpose & Reading Guide
This is one of five technical workflow descriptions supporting Work Assignment 2 of the UN-GGCE Global Geodesy Supply Chain Assessment (Terms of Reference 2). It maps, function by function, how the International Celestial Reference Frame (ICRF) is produced — the institutions involved, the capabilities each step depends on, and where the production chain is most exposed to failure.
Companion Documents
Readers new to this engagement should start with the Critical Product Mapping Report (Work Assignment 2), which maps this product's functions to capabilities, organisations and information artifacts at summary level; the Investment Prioritization and Risk Rubric (Work Assignment 1b), which defines the risk-scoring method; and the EGV Data Levels and Supply Chain Tiers appendix (Work Assignment 1 Appendix), which explains the underlying data-level structure this document's functions are keyed to.
Overview
The International Celestial Reference Frame (ICRF) is the realisation of the International Celestial Reference System (ICRS) — the quasi-inertial background frame against which Earth's rotation and satellite orbits are measured. It is defined by the precise positions of extragalactic radio sources (quasars), which are so distant that they exhibit no measurable proper motion, making them geometrically stable reference points on the celestial sphere. The axes of the ICRF are aligned with the equatorial coordinate system and are consistent, at the sub-milliarcsecond level, with the dynamical reference frame defined by planetary motions.
Within the Essential Geodetic Variable (EGV) framework (V7.0, Global Geodetic Observing System (GGOS), January 2026), the ICRF is classified as a Level 3 EGV under the Global Reference Frames category. It is, however, structurally distinct from other Level 3 realisations: an episodically produced scientific catalogue, reissued when the quality and quantity of accumulated VLBI observations justify a new realisation. The most recent realisation, ICRF3, was adopted by the International Astronomical Union (IAU) in 2018 and contains positions for 4,536 compact radio sources, with 303 designated as "defining sources" that fix the orientation of the frame axes.
The ICRF is the fundamental celestial reference for all geodetic and astrometric science. It provides the inertial backdrop against which the terrestrial reference frame (ITRF) is oriented, and it is an indispensable input to the determination of Earth Orientation Parameters (EOP) — specifically UT1-UTC and the Celestial Pole Offsets — which quantify the rotational relationship between the two frames. Very Long Baseline Interferometry (VLBI) is the sole observational technique capable of simultaneously connecting the celestial and terrestrial reference frames, making the ICRF supply chain uniquely dependent on a single technique with no redundant observational pathway.
Co-Produced EGVs
| EGV | Level | Mechanism |
|---|---|---|
| Earth Orientation Parameters (UT1, CPO) | 3 | VLBI simultaneously estimates station positions, EOP, and quasar coordinates in each session |
Governance
| Institution | Role | Risk Level |
|---|---|---|
| International Astronomical Union (IAU) | Formally adopts and designates official ICRF realisations at General Assemblies | Low |
| International Earth Rotation and Reference Systems Service (IERS) ICRS Centre | Maintains and realises the International Celestial Reference System; coordinates combination work | Medium |
| International VLBI Service (IVS) | Executes observing programs, operates correlators, and coordinates analysis centre activities | High |
Supply Chain
Note on PPTD tables below: averages exclude not-applicable dimensions (marked N/A) rather than scoring them 0, per Work Assignment 1b v7.0 Annex A.2. Dimensionality (how many of the four dimensions apply) is shown alongside each average. ⚠ flags a capability whose corrected average falls below 2.5.
F1 — Astrometric & Geodetic VLBI Observations
Data level: Basis → Level 1 EGV (Geodetic Observations) — raw radio frequency signals from extragalactic sources recorded by radio telescopes and synchronised using hydrogen maser clocks.
The global VLBI network comprises approximately 40 geodetically capable radio telescopes operated by national agencies and research institutions. These observatories record broadband radio signals from distant quasars onto high-speed digital storage media, with each telescope's recording precisely synchronised using hydrogen maser atomic clocks to achieve the sub-nanosecond timing accuracy required for interferometric delay measurements. The fundamental data product of this function — the raw voltage-recorded signal from each antenna — is the irreplaceable observational input upon which the entire ICRF supply chain depends.
The geographic distribution of the VLBI network constitutes a structural vulnerability that is a direct gap against (Principle 1: Geographic Distribution for Technical Performance). The majority of geodetic VLBI telescopes are located in the northern hemisphere, concentrated in Europe and North America, with very sparse coverage across the southern hemisphere, Africa, South America, and the Pacific Ocean. This asymmetry weakens the geometric strength of the reference frame: the orientation of the celestial axes depends on an even distribution of observable sources across the sky, and the under-sampling of southern declinations introduces systematic errors in the estimated positions of southern radio sources, which in turn propagate into the orientation of the ICRF axes themselves. Southern hemisphere stations — including those in South Africa, Australia, and South America — are disproportionately important to the geometric integrity of the frame and carry correspondingly elevated criticality.
The network faces a compounding technological transition risk. The next-generation VLBI Global Observing System (VGOS) is designed to deliver broadband observations across the 2–14 GHz frequency range, enabling substantially improved precision in delay measurements and eventual continuous all-sky coverage. However, legacy astrometric VLBI observing programs, which underpin the historical continuity of the ICRF back to the 1970s, operate on different equipment and scheduling paradigms. The co-existence of VGOS broadband and legacy narrowband astrometric programs creates a risk of observational divergence: unless both observation types are systematically cross-calibrated and consistently included in combination solutions, the transition to VGOS could introduce reference frame discontinuities that compromise the integrity of the ICRF as a stable long-term reference.
The 2022 Russia-Ukraine conflict is the only documented test to date of the VLBI network's resilience to a geopolitical disruption event. Russian VLBI stations (Zelenchukskaya, Badary, Svetloe, operated by the Institute of Applied Astronomy) remained listed as active throughout 2022-2024, and no IVS-level network notice was issued regarding Ukrainian or Russian stations — in contrast to the Satellite Laser Ranging network, coordinated through the International Laser Ranging Service (ILRS), and the DORIS network, coordinated through the International DORIS Service (IDS), where member stations and IDS formally withdrew or decommissioned Russian-operated infrastructure over the same period. The International Association of Geodesy (IAG)'s position — that non-political cooperative international scientific organisations do not have the authority to dictate specific actions to their members — produced a member-driven response that varied by technique: significant for ILRS and IDS, minimal for IVS. (Source: UN-GGCE reference verification record, Section 4, "What Was NOT Disrupted".) This bears on (Principle 2: Political Resilience and Distributed Operational Control): the 2022 conflict is direct evidence that VLBI observational continuity remained intact through the most significant geopolitical stress test available in the historical record — a distinct consideration from the geographic-concentration gap against (Principle 1: Geographic Distribution for Technical Performance) identified above.
Risk: High — The VLBI network suffers from sparse geographic distribution, especially in the southern hemisphere, which weakens the geometric strength of the celestial frame. Aging legacy infrastructure and the divergence between VGOS broadband and traditional astrometric VLBI threatens observational continuity. UN-GGCE recommends formalised national backing for VGOS transition investment under (Principle 5: Minimum Capability Maturity) and JDP Objective 1.3.
| Capability | People | Process | Technology | Data | Dimensionality | Avg |
|---|---|---|---|---|---|---|
| VLBI Data Acquisition and Storage | 3.00 | 3.00 | 3.00 | 3.00 | 4 of 4 | 3.00 |
| Ground-Based Asset Management | 3.00 | 2.00 | 2.00 | 3.00 | 4 of 4 | 2.50 |
| Equipment Calibration and Maintenance | 4.00 | 3.00 | 3.00 | N/A | 3 of 4 | 3.33 |
(Source: State of Geodesy 2026, re-verified against the live capability-maturity graph 2026-08-29)
F2 — Raw Data Logistics & Telemetry
Data level: Basis → Level 1 EGV transition. The transport link between an observatory and its correlator is where a raw recording becomes a processable data stream.
Following observation, the raw recorded data volumes — which for a single VLBI session can amount to tens of terabytes per station — must be transported to specialised correlation facilities (F4). For legacy observing programs this typically involves the physical shipment of hard disks by courier, a logistical constraint that introduces latency of days to weeks into the processing pipeline. The e-VLBI approach, which transmits data in real time over high-speed fibre-optic networks, eliminates this latency but requires broadband connectivity that is not universally available at remote or developing-country stations.
Risk: Medium — Physical data transport for legacy programs introduces multi-day latency; e-VLBI eliminates this but is bandwidth-constrained at remote or developing-country stations.
| Capability | People | Process | Technology | Data | Dimensionality | Avg |
|---|---|---|---|---|---|---|
| Network Operations | 2.00 | 2.00 | 1.00 | 3.00 | 4 of 4 | 2.00 ⚠ |
(Source: State of Geodesy 2026)
F3 — Primary Archiving & Metadata Registration
Data level: Level 1 EGV (Geodetic Observations) — interferometric observables and auxiliary calibration data, archived to agreed formats.
The correlated delay data produced at F4, together with the auxiliary meteorological and calibration information required for geodetic analysis, are archived at the Global Data Centres: principally the Crustal Dynamics Data Information System (CDDIS) at NASA GSFC, and the Bundesamt für Kartographie und Geodäsie (BKG) in Germany. This function carries a structural single-point-of-failure risk. CDDIS holds the authoritative archive for the majority of IVS data products, and there is limited geographical redundancy: a prolonged outage at CDDIS — as documented during U.S. government funding lapses (Source: International GNSS Service (IGS) 2018 Technical Report; see the Satellite Orbits Workflow annex, F5, for detailed sourcing of this event) — would interrupt access to historical correlated data that analysis centres depend upon for reprocessing and combination activities. This concentration in a single political jurisdiction is inconsistent with (Principle 2: Political Resilience and Distributed Operational Control). The absence of Findable, Accessible, Interoperable, Reusable (FAIR)-compliant discovery mechanisms for alternative IVS data access paths is a gap against (Principle 4: Technical Interoperability and Data Accessibility).
CDDIS shares its host institution with F4's Washington Correlator: both are operated by NASA GSFC. GSFC lost approximately 32% of its civil servant workforce to DOGE-related reductions from February 2025 onward, and sustained a 42-day funding lapse during the October–November 2025 U.S. government shutdown (see the Satellite Orbits Workflow annex, F5, for detailed sourcing of this evidence). No product delivery failure attributable to either event has been documented in the IGS mail archive, consistent with the pattern observed in earlier shutdowns of automated data flows continuing while human oversight and curation capacity were degraded; whether archive curation at GSFC experienced comparable degradation during this period has not been independently confirmed for VLBI-specific products. This dual-hosting of F3 and F4 at the same institution, during the most sustained period of institutional disruption at that institution on record, is worth reading as its own finding, not folded into either function's risk statement alone.
Risk: High — Archiving is heavily concentrated at a single facility, CDDIS, with limited geographical redundancy; the same institution also hosts F4's primary correlator.
| Capability | People | Process | Technology | Data | Dimensionality | Avg |
|---|---|---|---|---|---|---|
| Data Quality Management | 2.00 | 2.00 | 1.00 | 3.00 | 4 of 4 | 2.00 ⚠ |
| Data Preservation | 2.00 | 2.00 | 2.00 | 2.00 | 4 of 4 | 2.00 ⚠ |
(Source: State of Geodesy 2026)
F4 — Interferometric Correlation & Fringe-Fitting
Data level: Level 1 (intermediate) — correlated interferometric observables (group delays and delay rates), not yet a station-position or catalogue solution.
The major correlation centres — including the Washington Correlator (operated by NASA GSFC), the Bonn correlator at the Max Planck Institute for Radio Astronomy, and the Haystack Observatory correlator — process the incoming station data transported at F2 to produce the fundamental interferometric observables: the group delay and delay rate that encode the geometric baseline between each pair of telescopes.
Correlation is a severe throughput bottleneck in the ICRF supply chain. The process requires purpose-built software correlators or dedicated hardware systems, and the availability of operational correlation capacity worldwide remains limited. The number of IVS-affiliated correlators is small compared to the volume of scheduled sessions, and any failure or capacity reduction at a major facility — through hardware failure, staffing loss, or funding interruption — directly halts the processing pipeline for the affected observing sessions, with no rapid fallback path.
Risk: High — Correlation requires specialised facilities available at only a small number of centres globally; failure at a major correlator halts the entire processing pipeline for affected observing sessions. This was the second-highest tier risk score across all five EGV workflows under the retired scoring method — the underlying structural exposure (small correlator inventory, no rapid fallback, NASA GSFC dual-hosting with F3) is unchanged by the terminology conversion.
| Capability | People | Process | Technology | Data | Dimensionality | Avg |
|---|---|---|---|---|---|---|
| Geodetic Software and Tools | 2.00 | 3.00 | 1.00 | 2.00 | 4 of 4 | 2.00 ⚠ |
Note on this table: Geodetic Software and Tools is shared with F5's table below — the same underlying score applies to both functions, not a duplicated assessment.
(Source: State of Geodesy 2026)
F5 — Astrometric VLBI Analysis
Data level: Level 1 → Level 2 (intermediate) — individual catalogues of radio source positions and EOP estimates from each analysis centre.
IVS Analysis Centres retrieve the correlated delay data from F3's archives and apply geodetic analysis software — principally Calc/Solve (NASA GSFC) and VieVS (TU Vienna) — to simultaneously estimate station positions, Earth Orientation Parameters, and the celestial coordinates (Right Ascension and Declination) of the observed radio sources. Each Analysis Centre produces an independent celestial catalogue reflecting its specific processing strategy, parameterisation choices, and software implementation. The diversity of these independent solutions is a deliberate design property of the combination architecture: systematic errors that are specific to a particular software package or analysis approach can be detected and mitigated when multiple independent catalogues are combined.
However, the VLBI analysis community is substantially smaller than its Global Navigation Satellite System (GNSS) counterpart. The IVS operates with fewer than ten regularly contributing Analysis Centres for astrometric and geodetic VLBI products, compared to the considerably larger IGS network for GNSS. This concentration creates methodological risk: if two or three major analysis centres share underlying software assumptions or parameterisation conventions, a systematic error in those conventions may not be detected through combination. The reliance on a small number of geodetic software packages — some of which are maintained by single research groups or individual scientists — amplifies the key-person dependency risk. Loss of a lead developer, combined with inadequate documentation or succession planning, can render a software tool operationally unavailable on timescales that affect the production of routine IVS products.
Calc/Solve's maintenance is a further point of institutional concentration at NASA GSFC. The DOGE-related workforce reductions affecting NASA's Space Geodesy Project from February 2025 onward (see F3/F4 above and the Satellite Orbits Workflow annex, F5, for sourcing) extend to the institution that develops and maintains Calc/Solve, one of the two geodetic analysis software packages most widely used across IVS Analysis Centres. No confirmed impact on Calc/Solve maintenance or availability has been documented, and the software continues to be used operationally; the exposure is the same small-team key-person risk already identified above for VLBI geodetic software generally, concentrated specifically at the institution experiencing the most sustained workforce disruption in the current assessment period.
The global community of VLBI analysts — scientists capable of operating geodetic VLBI software, interpreting results at the level of accuracy required for ICRF combination, and contributing to methodological development — is small and geographically concentrated, predominantly in Europe and North America. The evidence from the State of Geodesy 2026 baseline assessment identifies declining formal training programs in VLBI geodesy and an absence of centralised knowledge repositories, creating a fragility in which institutional knowledge is embedded in individuals approaching retirement without documented succession pathways. This directly engages (Principle 5: Minimum Capability Maturity), which recommends verifiable evidence of workforce resilience before entities are designated as critical-path providers, and (Principle 6: Transparency, Monitoring and Performance Assurance), which recommends that training and succession gaps be formally reported at the intergovernmental level.
Risk: Medium — Fewer analysis centres than GNSS and a small global analyst community create methodological concentration and knowledge-fragility risks. Software key-person dependencies present a latent but material threat to the continuity of VLBI analysis capacity.
| Capability | People | Process | Technology | Data | Dimensionality | Avg |
|---|---|---|---|---|---|---|
| VLBI Data Processing and Analysis | 2.00 | 2.00 | 2.00 | 3.00 | 4 of 4 | 2.25 ⚠ |
| Geodetic Software and Tools | 2.00 | 3.00 | 1.00 | 2.00 | 4 of 4 | 2.00 ⚠ |
| Training and Education | 2.00 | 1.00 | 2.00 | N/A | 3 of 4 | 1.67 ⚠ |
| Knowledge Management | 1.00 | 2.00 | 1.00 | 1.00 | 4 of 4 | 1.25 ⚠ |
Note on this table: Geodetic Software and Tools is shared with F4's table above — see note there.
(Source: State of Geodesy 2026, re-verified against the live capability-maturity graph 2026-08-29)
F6 — ICRS Multi-Catalogue Combination
Data level: Level 3 EGV (Global Reference Frames — ICRF realisation) — the official catalogue of extragalactic radio source positions defining the celestial reference frame axes.
To produce an official ICRF realisation, a joint IAU/IERS working group combines the independent celestial catalogues submitted by the F5 IVS Analysis Centres. The combination methodology involves the identification of a core set of "defining sources": radio sources with compact jet structure, high positional stability over multi-decade baselines, and good sky distribution sufficient to constrain the three orientation angles of the celestial frame. The axes of the ICRF are defined by enforcing that the combined catalogue introduces no net rotation with respect to the previous realisation — a no-net-rotation condition that ensures continuity between successive realisations. The positions of non-defining sources are determined in the same combination but with lower weight, providing an extended catalogue for users requiring densified sky coverage.
The episodic character of ICRF combination is the defining governance vulnerability of this supply chain. New ICRF realisations are produced on a cadence of roughly a decade — ICRF1 in 1995, ICRF2 in 2009, ICRF3 in 2018 — driven by the accumulation of sufficient new VLBI observations and the availability of expert working group capacity. There is no operational Service Level Agreement (SLA) governing the frequency of realisations, the timeline from data accumulation to publication, or the response protocol in the event that a systematic error is discovered in the current realisation. The working groups responsible for combination are convened on a best-effort basis by volunteers drawn from the IVS and IERS communities, without dedicated funding or staffing commitments. This governance model was adequate when VLBI accuracy requirements were less stringent, but is increasingly mismatched with the operational demands of geodetic infrastructure that downstream users — including ITRF combination and EOP series — depend upon continuously.
Risk: Medium — ICRF realisations are episodic scientific undertakings driven by best-effort working groups. The absence of a formal combination cadence, error-response protocol, or dedicated combination capacity represents a structural governance gap that would delay any correction of a systematic error in the current realisation for years. (Principle 3: Centralised Accountability through Multilateral Governance) recommends that a formally designated body hold accountability for combination cadence and product integrity.
| Capability | People | Process | Technology | Data | Dimensionality | Avg |
|---|---|---|---|---|---|---|
| Reference Frames | 3.00 | 3.00 | 3.00 | 3.00 | 4 of 4 | 3.00 |
| Geodetic Data Products | 2.00 | 3.00 | 3.00 | 3.00 | 4 of 4 | 2.75 |
| Research and Development Prototyping | 3.00 | 3.00 | 3.00 | 3.00 | 4 of 4 | 3.00 |
(Source: State of Geodesy 2026, re-verified against the live capability-maturity graph 2026-08-29)
F7 — Astrophysical Source Structure & Validation
Data level: Level 3 EGV (pre-designation) — independent optical cross-comparison used to verify the candidate ICRF before formal adoption.
Candidate ICRF realisations are validated against complementary independent reference frames before formal adoption. The primary external validation is cross-comparison with the Gaia optical reference frame, produced by the European Space Agency's Gaia astrometric satellite mission. Gaia provides positions for approximately 1.6 million quasars detected in the optical band, enabling a direct comparison of source positions derived from VLBI and optical interferometry at the sub-milliarcsecond level. Significant positional offsets between the ICRF and the Gaia-CRF for individual sources indicate either extended or variable jet structure in the radio band, optical variability, or systematic errors in one or both frames; these comparisons are used iteratively to refine the selection of defining sources and to assess the overall orientation of the celestial frame axes.
Risk: Low — The Gaia cross-comparison provides a genuinely independent, non-VLBI validation pathway — a structural strength most other functions in this workflow lack.
| Capability | People | Process | Technology | Data | Dimensionality | Avg |
|---|---|---|---|---|---|---|
| Standards Development and Promotion | 3.00 | 3.00 | 3.00 | 3.00 | 4 of 4 | 3.00 |
(Source: State of Geodesy 2026)
F8 — Formal IAU Designation & Dissemination
Data level: Level 3 EGV (validated Celestial Reference Frame) — formally adopted ICRF published by IERS.
Once the candidate realisation passes the F7 scientific validation, it is formally adopted by resolution of the IAU General Assembly, which convenes every three years. Following adoption, the official ICRF is published and maintained by the IERS ICRS Centre and distributed through IERS publications and data repositories, together with CDS Strasbourg. The slow cadence of updates is broadly appropriate for a stable, quasi-inertial reference frame — the positional stability of extragalactic radio sources means that annual updates are unnecessary for most geodetic applications. The principal limitation is that this slow cadence means any systematic error discovered in the current realisation cannot be corrected until the next full reanalysis and General Assembly adoption cycle, potentially leaving an erroneous frame in operational use for several years.
Risk: Low — The IAU General Assembly adoption process ensures broad scientific consensus and formal international endorsement. The main limitation is the slow cadence of updates; systematic errors, once identified, persist in the published frame until the next episodic realisation. (Principle 6: Transparency, Monitoring and Performance Assurance) recommends a defined error-response protocol enabling timely communication to downstream users if a systematic deficiency is identified in the current realisation before the next General Assembly.
| Capability | People | Process | Technology | Data | Dimensionality | Avg |
|---|---|---|---|---|---|---|
| Geodetic Services | 2.00 | 2.00 | 2.00 | 2.00 | 4 of 4 | 2.00 ⚠ |
| Regulatory Compliance | 3.00 | 3.00 | 3.00 | 3.00 | 4 of 4 | 3.00 |
| Data Distribution | 2.00 | 2.00 | 1.00 | 3.00 | 4 of 4 | 2.00 ⚠ |
Note on this table: Data Distribution is newly carried here per Work Assignment 2's own F8 structure — this annex had not previously scored it; see the v5.0 changelog note above.
(Source: State of Geodesy 2026)
Capability Evidence Trail
Applying the confirmed/asserted/considered/queued method piloted on the Satellite Orbits annex. Every capability that appears in this document's function tables — 17 of the 58-capability model — is classified below by how it got its score. This is the starting draft the Technical Product Working Group is expected to confirm or challenge.
Confirmed — cited directly in this annex's function tables, at the shared State of Geodesy 2026 baseline, with no product-specific departure asserted (13 of 17 capabilities currently in this document):
- F1 (2): VLBI Data Acquisition and Storage, Equipment Calibration and Maintenance
- F3 (1): Data Quality Management
- F5 (3): VLBI Data Processing and Analysis, Training and Education, Knowledge Management
- F6 (3): Reference Frames, Geodetic Data Products, Research and Development Prototyping
- F7/F8 (4): Standards Development and Promotion, Geodetic Services, Regulatory Compliance, Data Distribution
Asserted — none. No capability in this annex currently carries a product-specific numeric departure from the shared baseline.
Considered and not asserted — evidence exists but doesn't cleanly support a departure (4 of 17):
- Ground-Based Asset Management (F1, avg 2.50) — the 2022 Russia-Ukraine conflict is real, specific evidence that the physical VLBI network continued to be operated through a genuine geopolitical disruption event: Russian VLBI stations (Zelenchukskaya, Badary, Svetloe) remained listed as active throughout 2022-2024, in contrast to the significant withdrawals documented for the SLR and DORIS networks over the same period. A single stress test with no confirmed impact is not sufficient basis to raise this capability's score above baseline, and it is left there deliberately.
- Network Operations (F2, avg 2.00) — the Washington Correlator and CDDIS are both operated by NASA GSFC, which lost approximately 32% of its civil servant workforce to DOGE-related reductions from February 2025 onward and sustained a 42-day shutdown in October-November 2025 (see F3/F4 above and the Satellite Orbits Workflow annex, F5, for sourcing). This compounds the single-jurisdiction concentration already reflected in this function's already-low score, but no confirmed correlator throughput impact has been documented, so the score is left at baseline.
- Data Preservation (F3, avg 2.00) — the same GSFC workforce and shutdown evidence bears directly on CDDIS's archiving function, corroborating this capability's already-low score; no confirmed impact on VLBI archive availability during either event has been documented in the IGS mail archive.
- Geodetic Software and Tools (F4/F5, avg 2.00) — Calc/Solve is developed and maintained by NASA GSFC, the same institution experiencing the DOGE-related workforce reductions documented above — direct evidence for the software key-person risk this capability's score already reflects. No confirmed impact on Calc/Solve maintenance or availability has been documented.
Queued — evidenced in this annex's prose or Governance table, not yet added to any function table, not yet dimension-level rated (3 capabilities):
- Mandate — this remains the strongest Queued case across all five annexes. F6 states plainly that (Principle 3: Centralised Accountability through Multilateral Governance) recommends that a formally designated body hold accountability for combination cadence and product integrity — an explicit, prose-level finding that no such body currently exists. The Governance table names IAU (formal adoption authority) and IERS ICRS Centre ("maintains and realises... coordinates combination work") but neither is described as accountable for the combination process itself.
- Funding and Investment — F6: ICRF combination working groups are "convened on a best-effort basis by volunteers... without dedicated funding or staffing commitments."
- Legal Framework — F6: "no operational SLA governing the frequency of realisations, the timeline from data accumulation to publication, or the response protocol" if an error is found.
Training and Education is already Confirmed above (F5) — unlike ITRF, this annex already scores it (EOP does not).
The three above aren't scored yet for the same reason as in the Satellite Orbits annex — scoring them properly means the same per-dimension check the rest of this table went through.
Workflow Diagram
JDP Alignment
| Pipeline Element | Gap | JDP Objective |
|---|---|---|
| Southern Hemisphere VLBI (F1) | Sparse distribution weakens geometric strength of the celestial frame | 1.2 — Regional hubs in under-represented regions |
| VGOS vs. legacy astrometry (F1) | Diverging technological paths risk continuity of ICRF observations | 1.3 — Formalised national backing for critical capabilities |
| Correlator throughput (F4) | Limited specialised facilities create a processing bottleneck; dual-hosted with F3 at NASA GSFC | 1.2 — Modernised infrastructure |
| VLBI analyst community (F5) | Small global workforce with no succession planning or funding guarantees | 1.4 — Succession planning and knowledge transfer |
| ICRF combination (F6) | Episodic scientific working groups lack a continuous operational SLA | 1.1 — SLAs and Memoranda of Understanding (MoUs) for critical functions |
[^sog2026]: United Nations Global Geodetic Centre of Excellence, 2026, State of Geodesy 2026: A Baseline Maturity Assessment, Bonn, Germany, https://www.un.org/globalgeospatial/sites/default/files/2026-05/stateofgeodesy.pdf