Cross-Sectional Coverage Analysis

What transport data actually exists, where, and for when? This analysis cuts across all sources surveyed in the project — TDC, World Bank, OPSIS, PortWatch, OECD/ITF, and others — to map coverage by topic, geography, and time period.

The point isn't to catalogue every dataset. It's to answer a practical question: if you need transport data for a given country and topic, what's available and what's missing?

Data taxonomy

We classify transport data into ten categories. These emerged from the topics used by TDC and World Bank, cross-referenced with the transport modes taxonomy in docs/research/taxonomies/transport-modes.md.

Category	What it covers	Primary sources
Road & Vehicles	Road network length, paved %, vehicle registrations, traffic volumes	TDC (246), WB WDI, OECD/ITF, SUM4All
Road Safety	Fatalities, mortality rates, crash data by road user type	WHO GHO, WB WDI, OECD/ITF, SUM4All
Rail	Network length, passenger-km, freight-km, rolling stock	TDC (95), WB WDI, OECD/ITF, EUROSTAT
Aviation	Passengers, freight, departures, airport connectivity	TDC (76), WB WDI, WEF TTDI, IMF BOP
Maritime & Ports	Port traffic, container throughput, liner shipping connectivity, vessel calls	TDC (72), UNCTAD, PortWatch, WB WDI
Public Transit	Bus, metro, tram, BRT ridership and networks	TDC (35), WB WDI, GTFS feeds, Transitland
Logistics & Trade	LPI scores, customs clearance, freight corridors, trade facilitation	TDC (21), WB LPI, WB Enterprise Surveys
Transport Emissions	CO2 from transport, fuel prices, energy consumption	TDC (31), Climate Watch, WB WDI, Climate TRACE
Infrastructure (Geospatial)	Vector networks for roads, rail, ports, airports; climate risk overlays	OPSIS, African Transport DB, OSM, Overture
Transport Finance	ODA flows (DAC 210xx), private investment, project finance	IATI/DAC CRS, WB PPI, ieConnect

Numbers in parentheses are TDC dataset counts. World Bank contributes 264 indicators across 13+ databases; these overlap multiple categories.

Geographic coverage

The overall picture

Coverage is uneven. Southeast Asia and parts of Africa have the most datasets on TDC. World Bank indicators cover 200+ countries but at varying depth. OPSIS has global geospatial coverage. PortWatch covers 145+ port countries.

The table below shows TDC dataset counts for 30 selected countries across all categories. World Bank, OPSIS, and PortWatch add coverage on top of these numbers for almost every country listed.

Country	Total	Road	Rail	Avia	Mari	Transit	Logis	Emis
Vietnam	137	39	19	16	28	11	7	9
Indonesia	133	39	15	17	25	13	8	8
South Korea	122	35	14	17	25	10	8	5
Malawi	118	36	15	17	17	13	12	8
Myanmar	111	32	13	15	24	8	7	4
India	107	35	11	13	17	8	7	8
Kenya	100	32	15	13	11	15	3	11
Ghana	93	29	13	9	15	18	3	6
Pakistan	91	28	11	12	16	5	6	5
Bangladesh	90	27	11	12	16	5	6	5
Nepal	90	27	11	12	16	5	6	5
Zambia	86	22	15	11	18	9	6	5
South Africa	74	21	8	8	12	13	4	8
Germany	72	28	17	5	10	10	3	10
Nigeria	61	16	7	7	11	11	3	6
Uganda	61	15	8	7	11	11	3	6
Rwanda	60	14	8	7	11	11	3	6
UK	56	9	3	4	4	6	3	6
Ethiopia	—	—	—	—	—	—	—	—
Tanzania	—	—	—	—	—	—	—	—

Ethiopia and Tanzania fall outside the top 80 in TDC dataset count — a significant gap given their importance to FCDO transport programmes.

Regional patterns

Southeast Asia has the densest coverage. The Asian Transport Observatory contributes ~55 datasets per country across the region, covering most modes. Vietnam, Indonesia, Philippines, Malaysia, Thailand, and Laos all exceed 100 datasets.

East and Southern Africa is the next strongest region, driven by Transport for Cairo (Malawi, Kenya, Zambia, Ghana — 46–65 datasets each) and CCG (South Africa, Nigeria, Uganda, Rwanda — 18–26 datasets each). But this coverage is concentrated in a few countries. Ethiopia, Tanzania, and Mozambique are notably thin.

South Asia has moderate coverage. India sits at 107 datasets; Pakistan and Bangladesh around 90; Nepal at 90. Sri Lanka and Afghanistan are lower.

Europe is well covered by EUROSTAT and OECD/ITF but those sources sit outside TDC. TDC's European datasets are mostly EUROSTAT re-publications.

Americas are the weakest region on TDC (19 datasets for all of North America, 31 for South America). World Bank and OECD fill some of this gap.

Middle East has almost no dedicated coverage on any platform.

What each source adds geographically

Source	Geographic scope	Strength
TDC	246 countries, but very uneven	Best for SE Asia and selected African countries
World Bank Data360	200+ countries	Global statistical indicators, but 1–2 year lag
OPSIS	220+ countries	Only source for global geospatial infrastructure networks
PortWatch	145+ port countries, 1,985 ports	Only source for near-real-time maritime trade
OECD/ITF	66 member + partner countries	OECD and a few LMICs — strong for road safety and rail
African Transport DB	54 African countries	All of Africa, but static 2023 snapshot
IATI/DAC	140+ recipient countries	Transport ODA flows — global but financial data only
WHO GHO	194 WHO member states	Road safety only — broadest country coverage
SUM4All GTF	183 countries	Composite indicators, no API (web/PDF only)

Temporal coverage

Update frequency spectrum

Transport data ranges from real-time to static snapshots. Where data sits on this spectrum matters as much as whether it exists at all.

Frequency	Sources	Typical lag
Weekly / near-real-time	PortWatch (port calls, disruptions)	Days
Quarterly	IATI (aid disbursements)	Weeks to months
Annual	World Bank WDI, OECD/ITF, WHO GHO, TDC bulk	1–2 years
Biennial	WHO Global Status Report on Road Safety	2–3 years
Irregular / one-off	SUM4All GTF, WB LPI, enterprise surveys	Varies
Static snapshot	OPSIS, African Transport DB, Overture Maps	Point-in-time
Per-operator	GTFS feeds, GBFS bike-share	Varies (days to months)

TDC temporal depth by category

Most TDC datasets cluster in 2010–2024. Historical depth before 2000 is thin.

Category	Earliest	Latest	Peak decade	Datasets in 2010s	Datasets in 2020s
Road & Vehicles	1960	2030	2010s	1,142	445
Rail	1960	2025	2010s	591	182
Maritime & Ports	1960	2026	2010s	413	129
Aviation	1929	2026	2010s	375	115
Public Transit	1990	2030	2010s	160	91
Transport Emissions	1969	2030	2010s	149	95
Logistics & Trade	1990	2026	2010s	113	57

"Datasets" here means dataset-years — a dataset covering 2010–2020 counts once per year in the range. The 2020s numbers are lower partly because many datasets haven't been updated past 2022–2023.

Aviation has the longest historical tail (ICAO data back to 1929), but most of that is sparse. Rail and road data starts meaningfully in the 1990s. Public transit and logistics data barely exists before 2000.

World Bank temporal coverage

World Bank WDI indicators typically have annual data from the 1990s or 2000s onward, with a 1–2 year publication lag. Some series (air passengers, rail freight) go back to the 1970s. The WEF indices (TTDI, GCI) start around 2006. Enterprise surveys are cross-sectional — one year per country, no time series.

Gaps in temporal coverage

No source provides consistent time series before 1990 for most LMICs. Pre-2000 data exists for OECD countries (via ITF) and for aviation (ICAO), but not for African or South Asian road, rail, or transit metrics.

Real-time data is limited to maritime trade (PortWatch) and some transit feeds (GTFS where available). There's no equivalent real-time source for road traffic, rail operations, or aviation movements that's freely accessible.

The 2020s gap: many datasets stop at 2021–2022. COVID disrupted data collection in several countries, and some national statistical offices haven't caught up. The most current annual data available is typically 2022 or 2023.

Coverage matrix: topic vs. source

This matrix shows which sources contribute to each topic category, with rough data-quality indicators.

	TDC	WB Data360	OPSIS	PortWatch	OECD/ITF	WHO	IATI/DAC	SUM4All
Road & Vehicles	246 ds	28 ind	networks	—	stats	—	—	RAI
Road Safety	—	6 ind	—	—	fatalities	mortality	—	composite
Rail	95 ds	14 ind	networks	—	stats	—	—	—
Aviation	76 ds	17 ind	—	—	—	—	—	—
Maritime & Ports	72 ds	21 ind	ports	trade flows	—	—	—	—
Public Transit	35 ds	36 ind	—	—	—	—	—	—
Logistics & Trade	21 ds	41 ind	—	—	freight	—	—	composite
Emissions	31 ds	8 ind	hazards	—	—	—	—	composite
Infrastructure (Geo)	—	—	full	—	—	—	—	—
Transport Finance	—	10 ind	—	—	—	—	ODA flows	—

"ds" = datasets, "ind" = indicators. Dashes mean no meaningful coverage.

What this tells us

Road & Vehicles is the best-covered topic across sources. Every major platform has something. But the data is fragmented — TDC has registration counts, World Bank has network length and paved percentages, OPSIS has the actual road geometries, OECD/ITF has traffic volumes. No single source gives you the full picture for a given country.

Road Safety has a source-richness problem, not a data problem. WHO, World Bank, OECD/ITF, and SUM4All all publish road fatality data — but with different definitions, reference years, and estimation methods. TDC has no road safety topic at all.

Geospatial infrastructure depends entirely on OPSIS and the African Transport DB. If those sources don't cover your area of interest at the resolution you need, there's no fallback except raw OpenStreetMap.

Transport finance is only available through IATI/DAC (for ODA) and World Bank PPI (for private investment). Neither is on TDC.

Public transit looks data-rich in the indicator counts but is actually the weakest category for LMICs. Most WB and TDC transit data covers aggregate statistics. The granular route-level data that would make transit planning useful (GTFS feeds, schedules, stop locations) exists for perhaps 50–60 cities in Africa and South Asia, and most of those feeds are incomplete or stale.

FCDO-priority country profiles

For countries central to FCDO's transport programme (RIDE and related), here's what's actually available.

Kenya

Best-covered African country after Malawi. TDC has 100 datasets (Transport for Cairo and CCG are the main contributors). Road, rail, and public transit are all reasonably represented. OPSIS covers the road and rail network. PortWatch covers Mombasa. Digital Matatus provides GTFS for Nairobi. World Bank has full indicator coverage. Main gap: no high-frequency road traffic or safety data.

Nigeria

61 TDC datasets — lower than expected for Africa's largest economy. Road and maritime are the strongest categories. Lagos has some transit data but no official GTFS feed. PortWatch covers Lagos and other ports. OPSIS has the road network. World Bank indicator coverage is full. Gap: limited rail data (reflecting Nigeria's limited rail system), very little on informal transport.

India

107 TDC datasets, strong across all categories. World Bank and OECD provide additional indicator depth. OPSIS covers the full road and rail network. Transit GTFS exists for a handful of cities (Delhi, Mumbai, Bangalore, Kochi). Gap: India's state-level variation is enormous, and most sources only report national aggregates.

Ethiopia

Not in the top 80 TDC countries — a genuine blind spot. World Bank has standard indicators. OPSIS covers the road and rail network (including the Addis-Djibouti railway). No known GTFS feeds. No PortWatch coverage (landlocked). ODA data available via IATI. This country needs dedicated data work.

Bangladesh

90 TDC datasets with reasonable spread across categories. Dhaka has some transit data. PortWatch covers Chittagong and Mongla. OPSIS has road and inland waterway networks. World Bank coverage is full. Gap: inland waterway transport data is thin despite waterways being a major mode.

Gaps and priorities

Structural gaps

No unified country-level summary exists. You can't go to one source and ask "what transport data is available for Tanzania?" Each source has its own scope, coverage, and access pattern.
Geospatial and statistical data don't talk to each other. OPSIS gives you road geometries; World Bank gives you road-km by country. Joining them requires geographic reference translation that no source does automatically.
Road safety is fragmented across sources with incompatible definitions. WHO estimates, police-reported figures, and modelled rates give different numbers for the same country.
Transit data in LMICs is near-absent at the route level. Aggregate statistics exist, but the operational data (GTFS feeds, real-time positions) that's standard in OECD cities barely exists in Africa and South Asia.

Geographic gaps

Ethiopia, Tanzania, Mozambique — important FCDO countries with very thin data on TDC
Middle East — almost no dedicated transport data on any platform
Central Asia — minimal coverage beyond World Bank indicators
Pacific Islands — OPSIS covers infrastructure, but statistical data is sparse

Temporal gaps

Pre-2000 historical data for most LMICs simply doesn't exist in digital form
2023–2025 data hasn't appeared yet for many annual sources (publication lag)
Real-time road and rail data has no freely accessible global source

Method

This analysis was generated by a Python script (scripts/extract-coverage.py) that processes:

TDC: 460 datasets from all-datasets.json (fetched via TDC's tRPC API)
World Bank: 264 transport indicators from transport-indicators-filtered.json (fetched via Data360 API)
Additional sources: manually characterised from dataset profiles in docs/datasets/ and research docs in docs/research/

The taxonomy was built by mapping TDC's topics and sectors fields and World Bank's _transport_subtopics to ten broad categories, with keyword fallback matching for datasets that lack structured topic metadata.

Geographic counts are TDC-only (TDC is the only source with per-dataset country tagging). World Bank and other sources are described qualitatively. Temporal counts use TDC's temporal_coverage_start and temporal_coverage_end fields, expanded to dataset-years.