Flyovers and congestion

TL;DR

Infrastructure Malta says its Marsa junction cut travel time and air pollution, and that the Msida Creek flyover would cut delays, emissions and noise. The Marsa figures are the agency’s own, without published surveys. The Msida flyover opened on 18 December 2025; in the nine months since, nitrogen dioxide at the Msida roadside monitor rose to 25.2 µg/m³ (January–September), from 22.0 a year earlier. With one monitor, unvalidated data and works still under way, that does not show the flyover made air worse; it shows the promised improvement has not yet appeared where it is measured.

1. Marsa has reported figures, not a public dataset. 79% less travel time and up to 70% less particulate matter, 52% less NOx and 50% less CO2, without the surveys or calculations. No EEA-reported air monitor has operated within 1.4 km of the junction since 2016.
2. Msida’s statement was a forecast. IM wrote it in October 2024, before works began. The flyover opened to traffic on 18 December 2025; the wider project is due for completion in 2027.
3. NO2 at the Msida monitor rose after opening. January–September means: 19.2 (2024), 22.0 (2025), 25.2 µg/m³ (2026, unvalidated), up 14.3% on 2025. Attard and Żejtun rose about 4%; St Paul’s Bay fell 6% (Figures 1 and 2).
4. The series has limits. A new sampling point, about 300 m east of the old one and 410 m from the flyover, replaced it in January 2024, so the record is not continuous. No noise data were found for either junction.
5. Verdict: not substantiated. The benefits are plausible, but the public evidence does not verify them, and the one measured outcome at Msida does not yet show them.

1. Supported
2. Largely supported
3. Not substantiated
4. Misleading
5. Contradicted

IN THIS REPORT

1 Wording and project stages

Infrastructure Malta's 28 October 2024 Msida statement says the area has congestion that affects air quality and noise, and describes building a new flyover to replace the traffic-light-controlled junction as “a critical move to reduce delays, emissions, and noise pollution” [1]. The same page says initial work “will focus on strengthening the Msida waterfront quay” and gives 2027 for completion: works had not begun, so the stated reductions were expected effects, not measured results. The vehicle flyover opened to traffic on 18 December 2025, the day after its inauguration, but the wider project continued through 2026 [3].

The two cases have different timelines. Marsa is a completed, multi-level project. Msida's road flyover began operating while other elements, including public spaces and the canal, were still being delivered. A single combined claim therefore needs separate evidence for each location and outcome.

A planning document and an EIA-screening decision are not after-the-fact environmental measurements. The ERA file records the project process and planning description [4]. Its existence neither proves benefits nor shows that the impacts are absent.

2 What the public records show

Marsa's completion release gives four headline percentages. A Times of Malta report later quoted an Infrastructure Malta spokesperson saying periodic surveys confirmed time and air-quality benefits, even though junction users were reported to be more than 19% above the 2013 level [5]. This is a useful lead to the agency's monitoring, but the surveys and methods were not provided in the article.

For Msida, the Environmental Resources Authority project file includes the planning statement, which relies on 2019 traffic counts and anticipates reduced travel time and pollution [4]. The 2024 IM wording also forecasts lower noise. Those documents describe the case for a proposed project. They are not evidence of its operating performance.

3 What the Msida air monitor shows

Malta’s air-quality network has a station at Msida classified as a traffic station, sited to measure pollution from road traffic, and its hourly nitrogen dioxide (NO2) readings are reported to the European Environment Agency [7, 9]. It is the one public, independent measurement close to either junction. We downloaded the hourly data for Msida and three comparison stations from January 2022 to September 2026, kept only hours flagged valid, and averaged them by month (Figure 1) and over January to September of each year (Figure 2) [7]. For scale, the annual limit value is 40 µg/m³, falling to 20 µg/m³ from 2030 [8]; the new Msida point averaged 23.3 µg/m³ in 2024 (from 17 January) and 25.5 in 2025.

The monitor moved. The old roadside point (MT00005), 110 m from the flyover as now mapped and 5 m from the kerb, recorded NO2 from 2006; its annual mean fell from 37.5 µg/m³ in 2013 to 28.3 in 2023, and its last valid readings are from December 2023. A new point (MT00011) started on 17 January 2024, about 300 m east and 410 m from the flyover, 9 m from the kerb, according to the station metadata ERA reports to the EEA [9]. The two points never ran side by side with valid data, so the series is not continuous and only the new point can compare before and after the opening.

After the opening. In January–September 2026, the first nine full months after the flyover opened, the new Msida point averaged 25.2 µg/m³, 14.3% more than in the same months of 2025 (22.0) and above 2024 (19.2). At the urban background stations the rise was smaller: Attard 11.2 → 11.8 and Żejtun 13.8 → 14.3. At St Paul’s Bay, the other traffic station, NO2 fell from 21.1 to 19.9 (Figure 2). Msida’s excess over the background average grew from 9.5 µg/m³ in 2025 to 12.2 in 2026; it had already grown from 7.3 in 2024, before the opening.

4 Induced travel and what it means here

Duranton and Turner analysed US cities and found that added interstate lane-kilometres were associated with more vehicle-kilometres travelled [6]. This peer-reviewed evidence supports induced travel as a possible network-level response to added capacity. It is not a direct estimate for a junction in Malta.

The finding cautions against assuming that an initial local delay reduction will guarantee lasting network-wide congestion relief. It does not establish that Marsa had no local travel-time benefit, and it cannot tell us what happened at Msida. A junction-level analysis must distinguish new trips from route changes, time-of-day shifts, demographic growth and other network changes.

5 Verdict and evidence needed

Why. Infrastructure Malta reports large Marsa travel-time and emissions reductions, and a later news report says the agency cited periodic surveys. The underlying surveys and calculation methods were not located, and no EEA-reported monitor near Marsa has operated since 2016. For Msida, the statement was made before works began. Since the flyover opened, NO2 at the Msida monitor has risen, not fallen, more than at the comparison stations; no traffic or noise series was found. The combined claim is therefore stronger than the public evidence that can be checked. The Msida data support “not substantiated”, not “contradicted”: one site, a moved sampling point, unvalidated 2026 data and unfinished works cannot show the flyover made things worse. Confidence is moderate because agency claims and one independent series exist, but key data are missing.

What this verdict does not say. It does not say that the infrastructure had no local benefit or that the published figures are false. It says the cited public record does not let readers reproduce or independently confirm the outcomes, and that the one independent measurement does not yet show them.

Evidence that would settle the open questions

• Release Marsa's raw traffic counts and travel-time survey methodology, including routes, dates, time windows, baseline adjustments and uncertainty.
• Show how each emissions percentage was derived; report monitor-based ambient concentrations separately from estimated emissions.
• Publish Msida post-opening traffic, air-quality and noise measurements after the full road arrangement is operating, with defined comparison locations and periods.
• Explain why the Msida sampling point moved in January 2024 and publish any parallel readings from the two points.
• Track junction and wider-network outcomes for several years, accounting for fleet/population growth and route or time shifts.

Right of reply. Not sought in this draft, at the maintainer's direction. Handle before wider circulation.

6 Sources

1. [1] Infrastructure Malta, Msida Creek: enhancing connectivity and community spaces, 28 October 2024 (wording checked 5 October 2026). https://www.infrastructuremalta.com/news/msida-creek-enhancing-connectivity-and-community-spaces
2. [2] Infrastructure Malta, Infrastructure Malta completes the Marsa Junction Project, 15 April 2021. https://www.infrastructuremalta.com/news/infrastructure-malta-completes-marsa-junction-project
3. [3] Office of the Prime Minister, Inawgurata l-flyover tal-Imsida Creek, 17 December 2025 (Maltese). The flyover was to open to vehicles the following day. https://primeminister.gov.mt/latest-news/pr252279/
4. [4] ERA, PA/02053/20 and PA/06425/20 project record and documents. https://era.org.mt/era-project/pa02053-20_pa06425-20/
5. [5] Times of Malta, Roads agency insists Marsa junction cuts travel time by 70%, despite complaints, 29 November 2021 (secondary). https://timesofmalta.com/article/roads-agency-insists-marsa-junction-cuts-travel-time-by-70-despite.913440
6. [6] Duranton & Turner, The Fundamental Law of Road Congestion: Evidence from US Cities, American Economic Review 101(6), 2616–2652 (2011). https://doi.org/10.1257/aer.101.6.2616
7. [7] European Environment Agency, air-quality download service: hourly NO2 for Malta, sampling points MT00005, MT00011, MT00004, MT00008 and MT00009; validated (E1a) data to 2025 and up-to-date (E2a) data for 2026. Retrieved 5 October 2026; monthly, annual and January–September means in data/cc-008/. https://eeadmz1-downloads-webapp.azurewebsites.net/
8. [8] Directive (EU) 2024/2881 on ambient air quality and cleaner air for Europe (recast): NO2 annual limit of 20 µg/m³ from 2030; the limit in force is 40 µg/m³. https://eur-lex.europa.eu/eli/dir/2024/2881/oj
9. [9] Environment and Resources Authority, Air Quality e-Reporting dataset D (measurement configuration) for 2025, Eionet Central Data Repository, 3 February 2026: station and sampling-point locations, classification and kerb distance. https://cdr.eionet.europa.eu/mt/eu/aqd/d/envayh2iw/
10. [10] OpenStreetMap contributors, Msida Creek flyover carriageways (ways 1460775989 and 1460775991) and coastline, retrieved 5 October 2026 (ODbL). https://www.openstreetmap.org/way/1460775989

Version 1.2 · 5 October 2026 · Draft; right of reply not sought · Evidence register: data/cc-008/assessment.csv

Appendix B: revision log