Skip to main content

Spain October 2026 Flash Floods: EuroMeteo Gauges and OPERA Radar

EuroMeteo
Extreme Weather
Data Analysis
Extreme rainfall in eastern Spain, 30 September–5 October 2026: EuroMeteo gauge accumulations compared over matched UTC windows with OPERA radar rain-rate estimates.
Author

Climate Explorer Team

Published

October 5, 2026

Very heavy rainfall affected eastern and northeastern Spain from the end of September into the first days of October 2026. AEMET associated the unsettled conditions with cold air at middle and upper levels and Mediterranean moisture, while forecasting a cut-off low (depresión aislada en niveles altos, or DANA) west of the peninsula. The relative contributions of the large-scale circulation and local convective organization are not established by the gauge–radar comparison presented here. The Barcelona Airport gauge reports add up to 181.4 mm between 06:00 UTC on 30 September and 06:00 UTC on 5 October (120 hours), equivalent to 217.2% of the station’s 1991–2020 October precipitation normal of 83.5 mm (the window starts on 30 September, so this is a scale comparison with the October normal, not a like-for-like monthly figure).

All accumulations in this article are tied to explicit UTC windows. EuroMeteo stamps each gauge report at the end of its accumulation interval, so a PT24H report stamped 06:00 UTC on 4 October covers 06:00 UTC on 3 October to 06:00 UTC on 4 October. Each figure below is either reported directly by the station, calculated by summing non-overlapping reports, or derived by differencing longer reports; derived values are labelled, and a total is only given when the whole window is covered (see How accumulations are computed).

At Barcelona / Aeropuerto El Prat, the largest directly reported accumulation is 135.6 mm in the 24 hours to 06:00 UTC on 4 October. In AEMET’s provisional extreme-value listing for observing date 3 October, a daily total of 135.6 mm is given for October maximum daily precipitation; the AEMET observing window was not reconciled with the EuroMeteo report window here. According to AEMET’s provisional comparison, this value surpassed the station’s October maximum daily precipitation within the 1991–2020 climatological reference period (124.8 mm on 8 October 2002), though it remained below the full-series historical October record of 175.1 mm established on 3 October 1987. AEMET explicitly notes that these recent observations have undergone only automatic real-time quality checks and remain subject to final verification. Further up the coast at Arenys de Mar, a series of convective cloudbursts delivered 161.9 mm in the 24 hours to 06:00 UTC on 4 October, including 37.5 mm in a single hour (the hour ending 04:00 UTC).

This investigation pairs in-situ surface rain gauge observations from the EuroMeteo explorer with rain-rate time series extracted from the European OPERA weather radar composite network. Comparing point rainfall depths with radar-derived surface precipitation rate estimates based on reflectivity measurements aloft provides an exploratory look at how the two products behave during a convective episode, while highlighting essential observational differences between point ground measurements and volumetric remote sensing. In the pixels sampled here the radar estimates, integrated over the same windows, were well below most gauge reports; the reasons are not resolved by this analysis.

Lollipop chart of the largest reported 24-hour rainfall accumulation at twelve stations in eastern Spain between 30 September and 5 October 2026, each labelled with its UTC window. Murcia / San Javier 176.8 mm, Arenys de Mar 161.9 mm, Polinyà de Xúquer 138.4 mm and Barcelona Airport 135.6 mm lead the ranking, followed by Barcelona Port Olímpic with a hollow marker at 104.4 mm in 12 hours, then Talarn, Girona, Porqueres, Manresa, Valencia Viveros, Reus and Ripoll between 70.2 and 28.8 mm. Dotted ticks mark the 1991–2020 October normal where available.

Largest directly reported 24-hour gauge accumulation per station, 30 September–5 October 2026, with WMO 1991–2020 October normals marked where they exist in the archive. Windows are UTC and end at the stated hour: Murcia / San Javier 176.8 mm (24 hours to 06:00 UTC on 3 October; 554% of the October normal), Arenys de Mar 161.9 mm and Barcelona Airport 135.6 mm (both to 06:00 UTC on 4 October; the airport value is 162% of its October normal, and AEMET’s provisional daily listing assigns the same value to observing date 3 October), Polinyà de Xúquer 138.4 mm (to 06:00 UTC on 2 October). Barcelona Port Olímpic is shown with a hollow marker at 104.4 mm in 12 hours (to 06:00 UTC on 4 October) because no 24-hour report covers its burst. Stations peak in different windows, so the values are not simultaneous. Observations are compiled from EuroMeteo.

Atmospheric circulation and maritime moisture inflow

The heavy rainfall episode occurred within a complex, evolving synoptic and mesoscale environment:

  1. Upper-Tropospheric Cut-Off Low (DANA): In its special advisory bulletin issued on 3 October 2026 (Aviso especial de fenómenos adversos nº 34/2026, issued at 14:32 on 3 October 2026), AEMET forecast increasing atmospheric instability associated with cold air at middle and upper levels, anticipating the detachment and isolation of a cut-off low (DANA) centered west of the Iberian Peninsula. The bulletin noted uncertainty in the system’s evolution and in the timing and location of precipitation maxima. Contemporary event analyses presented differing interpretations of the episode’s primary driver: while AEMET framed the broader synoptic risk around the approaching upper-level cut-off low, meteorologist Alberto Alcántara for 3Cat argued that the extreme downpours in Catalonia were not driven by a DANA or a major surface depression, but rather by organized mesoscale convective systems developing within a persistent easterly maritime inflow. These sources describe the episode from different temporal and spatial perspectives. This article presents the observations without independently resolving the respective contributions of the large-scale circulation and local convective organization.

  2. Warm Mediterranean Waters and Boundary-Layer Moisture: Warm sea-surface conditions can support heat and moisture transfer to the atmosphere, depending on air–sea temperature and humidity contrasts, wind, and stability; the specific contribution of these fluxes to this event was not quantified here.

  3. Low-Level Inflow (Llevantada): AEMET’s 3 October bulletin forecast an easterly flow transporting Mediterranean moisture towards the eastern coast, including the Catalan coastal sector. This article presents the surface gauge and radar observations without an independent diagnostic analysis of the synoptic pressure gradient or vertical wind profile.

  4. Coastal Topography: The Catalan Coastal Ranges (Serralada Litoral and Serralada Prelitoral) rise sharply inland from the coast. In Mediterranean storm literature, coastal topography is recognized as a potential mechanism for mechanically lifting maritime inflow or anchoring convective lines. Establishing whether terrain helped anchor convection during this event would require event-specific analysis, potentially including radar observations, wind analyses, and model sensitivity experiments.

How accumulations are computed

The EuroMeteo archive carries Spanish precipitation as PT1H, PT3H, PT6H, PT12H and PT24H reports. Synoptic hours (00, 06, 12 and 18 UTC) are generally absent as PT1H and exist only inside the longer reports, and the data end at 06:00 UTC on 5 October. The annex therefore keeps three kinds of numbers apart:

  • Reported: a value exactly as stored in one report (for example the PT24H ending 06:00 UTC on 4 October).
  • Calculated: a sum of non-overlapping reports that tile the whole window, with no hour counted twice. A negative value in the source (97 reports of −0.1 mm in the source-report annex, across all durations) is treated as the BUFR trace code and counted as 0.0 mm.
  • Derived: an hour recovered by subtracting shorter reports from a longer one (for example PT3H minus the two hourly reports it contains). Each derived hour is flagged, and where a second longer report allows an independent derivation, the two are compared.

A total is stated only when every hour of the window is covered. Otherwise the annex gives an incomplete subtotal with the covered and missing hours, and no percentage of normal is computed. Missing data are never treated as zero.

In-situ gauges recorded severe accumulations across coastal Catalonia

Official surface synoptic stations compiled in EuroMeteo captured extreme accumulations across Catalonia and the Mediterranean coast:

  • Barcelona / Aeropuerto El Prat (WMO ID 08181, WIGOS 0-20000-0-08181): The five PT24H reports ending 06:00 UTC on 1–5 October sum to 181.4 mm over the 120 hours from 06:00 UTC on 30 September (217.2% of the 1991–2020 baseline of 83.5 mm; 199.3% of the 1981–2010 normal of 91.0 mm); 178.7 mm of it fell in the 72 hours from 06:00 UTC on 2 October to 06:00 UTC on 5 October (the three reports ending 3, 4 and 5 October). The largest report is the PT24H to 06:00 UTC on 4 October, 135.6 mm (162.4% of the 1991–2020 October normal), with 90.0 mm in the PT12H from 18:00 UTC on 3 October to 06:00 UTC on 4 October. The airport station reported hourly (PT1H) data for only one hour in the whole period, so its accumulation cannot be resolved below the 3-hour reports. In AEMET’s provisional extreme-value listing for observing date 3 October, a daily total of 135.6 mm was recorded for October maximum daily precipitation. AEMET’s provisional comparison confirms that this total exceeded the station’s October maximum daily precipitation within the 1991–2020 reference period (124.8 mm on 8 October 2002), while remaining below the historical October record of 175.1 mm set on 3 October 1987 (full 1924–2025 series). AEMET explicitly notes that these recent observations have undergone only automatic real-time quality checks and remain subject to final verification.
  • Arenys de Mar (WMO 08186, WIGOS 0-20000-0-08186, Maresme coast): Station records give 163.5 mm over the same 120 hours, with 161.9 mm in the PT24H from 06:00 UTC on 3 October to 06:00 UTC on 4 October (the 20 PT1H reports in that window sum to 144.4 mm; adding the derived 17.5 mm for the unreported hour ending 00:00 UTC, and 0.0 mm derived for each of the three other unreported synoptic hours, gives 161.9 mm). The hourly reports show an elongated sequence of severe convective pulses: 25.4 mm in the hour ending 23:00 UTC on 3 October, 25.8 mm ending 01:00 UTC on 4 October, 18.9 mm ending 02:00 UTC, and a peak of 37.5 mm in the hour ending 04:00 UTC. The hour ending 00:00 UTC is not reported as PT1H; differencing the PT3H, PT6H and PT12H reports gives 17.5 mm for it (derived). Across the seven hours ending 23:00 UTC on 3 October to 05:00 UTC on 4 October (22:00–05:00 UTC), the six reported hourly values add up to 120.4 mm, and with the derived hour the window total is 137.9 mm.
  • Barcelona, Port Olímpic (WMO 08180, WIGOS 0-20000-0-08180, central city coastline): The burst came in the evening of 3 October. The five hourly reports ending 19:00 to 23:00 UTC (4.8, 25.8, 6.6, 37.4 and 16.6 mm) sum to 91.2 mm over the five hours 18:00–23:00 UTC; the station’s own PT6H to 00:00 UTC gives 103.2 mm and its PT12H to 06:00 UTC on 4 October gives 104.4 mm. The station has no report covering 11:00–18:00 UTC on 3 October (the hours ending 12:00 to 18:00) and no PT24H report that contains the burst, so a 24-hour or whole-episode total cannot be stated; the sum of all available reports (69 of the 120 hours) is an incomplete subtotal of 125.8 mm.
  • Regional Southern & Inland Gauges: Further south along the moisture conveyor, Polinyà de Xúquer in Valencia (08287) recorded 185.8 mm across the 120 hours, with 138.4 mm in the PT24H to 06:00 UTC on 2 October and a largest reported PT1H of 31.4 mm. In Murcia, San Javier Airport (08433) registered 178.2 mm across the 120 hours (558.6% of its WMO 1991–2020 October normal of 31.9 mm; 456.9% of the 1981–2010 normal of 39.0 mm), with 176.8 mm in the PT24H from 06:00 UTC on 2 October to 06:00 UTC on 3 October (554.2% of the 1991–2020 normal; 453.3% of the 1981–2010 normal). Note that while monthly normal percentages illustrate rainfall magnitude relative to average conditions, they do not denote statistical return periods or recurrence intervals, which require localized extreme-value distribution modeling.

Regional spatial variability and localized reports

In its 3 October forecast advisory (Bulletin 34/2026), AEMET noted the possibility of localized 12-hour rainfall accumulations exceeding 200 mm in eastern coastal sectors. In the open EuroMeteo network, official synoptic and automated stations recorded 120-hour totals of 181.4 mm at Barcelona Airport (including 135.6 mm in the 24 hours to 06:00 UTC on 4 October; AEMET’s provisional daily listing for observing date 3 October also records 135.6 mm) and 163.5 mm at Arenys de Mar (161.9 mm in the 24 hours to 06:00 UTC on 4 October).

Protecció Civil activated the INUNCAT alert phase on 30 September 2026 ahead of forecast heavy rainfall. By 3 October, the plan was escalated to its emergency phase (fase d’emergència), with initially scheduled restrictions on mobility and outdoor activities applying from 20:00 on 3 October to 14:00 on 4 October local time (CEST). Official municipal updates on 4 October from affected municipalities (such as Montornès del Vallès) documented road closures, damage to municipal facilities, and emergency response operations, alongside widespread reports of localized inundations in coastal catchments.

These regional accumulations differ markedly from station to station, which is consistent with strong spatial variability in the rainfall. However, reports from other networks or adjacent municipalities do not serve as numerical confirmation or validation of the measurements recorded at Barcelona Airport or Arenys de Mar. Each gauge measures rainfall at its particular location. Furthermore, a largest reported hourly value between 35 and 45 mm (such as 37.5 mm at Arenys de Mar or 37.4 mm at Port Olímpic) indicates intense localized convective activity. Large accumulations depend on both rainfall intensity and duration and may involve persistent precipitation, repeated convective cells, or a combination of processes. A single hourly value alone does not establish the event total or the organization of the storm.

Two-panel chart for 3 and 4 October 2026. Top: hourly rainfall bars for Barcelona Port Olímpic and Arenys de Mar, with hatched bars for hours derived at 00:00 UTC and grey crosses for undeterminable hours before 18:00 UTC at Port Olímpic. Port Olímpic peaks at 37.4 mm in the hour ending 22:00 UTC; Arenys de Mar peaks at 37.5 mm in the hour ending 04:00 UTC. Bottom: running sums from 18:00 UTC, ending at 104.4 mm for Port Olímpic and 152.0 mm for Arenys de Mar at 06:00 UTC on 4 October.

Hourly rainfall at Barcelona Port Olímpic and Arenys de Mar, 11:00 UTC on 3 October to 12:00 UTC on 4 October 2026, with the running sum from 18:00 UTC on 3 October. Each bar covers the hour ending at its right edge. Solid bars are reported PT1H values, hatched bars are hours derived from PT3H/PT6H reports (Port Olímpic 00:00 UTC, 12.0 mm; Arenys de Mar 00:00 UTC, 17.5 mm), and grey crosses mark hours that cannot be determined from the archive (Port Olímpic, hours ending 12:00 to 18:00 UTC). Peaks: 37.4 mm in the hour ending 22:00 UTC at Port Olímpic and 37.5 mm in the hour ending 04:00 UTC at Arenys de Mar. The running sums reach 104.4 mm and 152.0 mm at 06:00 UTC on 4 October. Observations from EuroMeteo.

OPERA radar estimated instantaneous surface rain rates from reflectivity aloft

Weather radar provides continuous spatial and temporal tracking across Europe. Using the OPERA radar explorer, single-pixel time series were extracted from the permanent GeoZarr archive (RATE and ACRR composite products, 2 km Cartesian grid, 15-minute nominal cadence) for six stations, from 00:00 UTC on 1 October to 12:00 UTC on 5 October 2026 (433 nominal frames per product). Rather than spatial interpolation, extraction used the grid cell containing the station’s reference coordinates (for Barcelona Airport, the EuroMeteo catalogue lists 41.2969° N, 2.0783° E). The selected radar pixel centre is at 41.299° N, 2.086° E, approximately 0.7 km from those coordinates; the requested and pixel-centre coordinates for every station are in the annex. In the OPERA NIMBUS processing pipeline, the RATE composite represents a radar-derived estimate of instantaneous surface rainfall intensity (mm/h), calculated from volume reflectivity scans aloft using the standard Marshall-Palmer relation (\(Z = 200 R^{1.6}\)). These estimates are subject to observational uncertainties related to beam altitude, drop-size distribution shifts, and spatial-temporal sampling.

Every nominal frame is kept with its status: detected (a rate of at least 0.1 mm/h), undetect (below 0.1 mm/h or no echo, a valid dry sample) or nodata (missing, never treated as zero). Radar extractions were run only for Barcelona Airport, Arenys de Mar, Barcelona Port Olímpic, Talarn, San Javier and Polinyà de Xúquer; the other stations in the tables were not queried. The maxima below are the largest sampled 15-minute values, not the true storm peaks.

  • Barcelona Airport (Coastal Lowland): Up to the gauge cutoff at 06:00 UTC on 5 October, the largest sampled rate was 15.38 mm/h at 10:30 UTC on 2 October (within the PT24H of 26.8 mm that ended at 06:00 UTC on 3 October). During the 24 hours of the 135.6 mm report (06:00 UTC on 3 October to 06:00 UTC on 4 October) the largest sampled rate was 10.28 mm/h at 19:15 UTC on 3 October. After the gauge cutoff, the extract reaches 21.72 mm/h at 07:15 UTC on 5 October, with the one-hour accumulation product ACRR peaking at 6.90 mm at 07:30 UTC; this later pulse falls outside the gauge record available here and cannot be evaluated against it. The composite quality index (RATE_quality_qi_total) was 0.91 at these maxima. In EUMETNET’s published NIMBUS documentation, the total quality index is the minimum of three BALTRAD quality indicators associated with anomaly filtering (bropo), beam blockage (beamb), and satellite-based filtering (satfilter). The index should be interpreted according to the applicable processing version and should not be treated as an absolute measure of rainfall accuracy. EUMETNET cautions that the index can evaluate to 1.0 when none of those specific filters has been applied.
  • Barcelona Port Olímpic (Urban Coastal Fringe): Over central coastal Barcelona (pixel center 2.193° E, 41.398° N), the largest sampled rate was 43.34 mm/h at 19:30 UTC on 3 October (QI 0.91), inside the hour ending 20:00 UTC, for which the gauge reported 25.8 mm. The hour with the largest gauge amount, ending 22:00 UTC (37.4 mm), has a largest radar sample of only 11.53 mm/h.
  • Arenys de Mar (Maresme Convective Axis): Along the narrow coastal strip northeast of Barcelona (pixel center 2.551° E, 41.591° N), the largest sampled rate was 32.50 mm/h at 03:45 UTC on 4 October (QI 0.91), inside the hour ending 04:00 UTC, in which the gauge reported its peak hourly amount of 37.5 mm. In the following hour (ending 05:00 UTC), the gauge reported 9.7 mm while the radar samples integrate to only 0.07 mm.
  • Inland and Southern Stations: At Talarn in the Pre-Pyrenean foothills (08112), the largest sampled rate was 115.31 mm/h at 16:15 UTC on 1 October, during the earlier rainfall episode (gauge PT24H of 70.2 mm to 06:00 UTC on 2 October, with a 1-hour burst of 50.8 mm). At San Javier (08433), the largest sampled rate was 32.50 mm/h at 13:30 UTC on 2 October, inside the 176.8 mm PT24H; one nominal RATE frame (00:00 UTC on 2 October) and four ACRR frames are nodata. For Polinyà de Xúquer (08287) all 433 RATE and ACRR frames are nodata: the extraction returned nodata at that pixel for every frame, so no comparison is possible there (the cause was not investigated). Girona, Porqueres, Reus, Manresa, Valencia Viveros, Ripoll and Tortosa were not queried.

Matched accumulations: gauge reports vs. radar over the same windows

An instantaneous rate in mm/h and an accumulated depth in mm are different quantities, so they are compared here only after integrating the radar over exactly the gauge report’s window. For every hour ending in the window, the four nominal 15-minute RATE samples are converted to depth (\(\sum R_i \times 0.25\ \mathrm{h}\)) and the hours are added; an hour counts only if all four samples are valid (detected or undetect), and no window is totalled otherwise. Undetect samples count as 0 mm/h in the integral, and samples stamped at :15, :30, :45 and :00 are assigned to the hour ending at the next whole hour (the :00 sample closes its own hour), the same hour-ending convention as the gauge reports; a 15-minute shift in that assignment would change individual integrals. Gauge traces (−0.1 in the source) appear as 0.0 mm in the matched table. The hourly ACRR product reproduces these hourly integrals to within 0.1 mm at all 539 complete station-hours (correlation above 0.9999), which is consistent with ACRR being built from the same 15-minute rates, so it is not an independent check. Rolling overlapping accumulations are not summed.

Gauge reports compared with the radar RATE integral at the pixel containing each gauge, over identical UTC windows (all hours complete). Polinyà de Xúquer has no radar values; rows are a hand-picked selection of the largest reports. The annex lists all 281 gauge reports of three hours or more: 225 have a complete radar integral (radar ÷ gauge from 0.14 to 3.14, widest for small gauge amounts), 52 are at Polinyà de Xúquer, and 4 are incomplete at San Javier.
Station Gauge report Window (UTC, ends at) Gauge (mm) Radar RATE integral (mm) Radar ÷ gauge
Barcelona Airport PT24H 06:00 3 Oct → 06:00 4 Oct 135.6 24.45 0.18
Barcelona Airport PT12H 18:00 3 Oct → 06:00 4 Oct 90.0 16.68 0.19
Barcelona Airport PT24H 06:00 2 Oct → 06:00 3 Oct 26.8 9.44 0.35
Barcelona Port Olímpic PT12H 18:00 3 Oct → 06:00 4 Oct 104.4 39.72 0.38
Arenys de Mar PT24H 06:00 3 Oct → 06:00 4 Oct 161.9 70.08 0.43
Arenys de Mar PT12H 18:00 3 Oct → 06:00 4 Oct 152.0 63.34 0.42
Murcia / San Javier PT24H 06:00 2 Oct → 06:00 3 Oct 176.8 58.85 0.33
Talarn PT24H 06:00 1 Oct → 06:00 2 Oct 70.2 66.05 0.94
Talarn PT12H 06:00 1 Oct → 18:00 1 Oct 54.4 54.70 1.01

In the windows tabulated above, the radar integrals are 18–43% of the gauge reports at Barcelona Airport, Port Olímpic, Arenys de Mar and San Javier, but 94–101% at Talarn. The timing differs between stations: at Arenys de Mar the largest radar samples fall inside the hour of the gauge’s largest value, whereas at Port Olímpic the largest radar sample (in the hour ending 20:00 UTC) precedes the gauge’s largest hour (ending 22:00 UTC), in which the largest radar sample is 11.53 mm/h. This is a single-pixel, 15-minute-snapshot comparison; the pixel centre lies roughly 1 km from each gauge, 15-minute snapshots can miss short convective peaks between frames, the quality index does not establish accuracy, and rain gauges carry their own instrument and exposure uncertainties. This article does not attribute the differences to a specific cause.

Observational considerations: point rain gauges vs. radar surface estimates

The comparison between EuroMeteo ground gauges and OPERA radar curves highlights fundamental physical distinctions in how precipitation is observed:

  1. Point Collector Funnels vs. Volumetric Remote Sampling: A rain gauge provides a point measurement of accumulated rainfall at ground level. In contrast, OPERA composite RATE provides a radar-derived estimate of instantaneous surface rain rate based on volume backscatter reflectivity (\(Z\)) measured within a sampling volume aloft, with sampling height and spatial representativeness varying according to radar scan geometry and topography across a 2 km horizontal grid cell. While the 15-minute sampling interval of the archived composite captures snapshot intensity pulses, discrete product intervals can miss brief downpours between sweeps. The causes of any differences in peak timing or depth cannot be determined from this exploratory comparison.
  2. Quality Index: The total quality index in NIMBUS composite products reflects BALTRAD processing filters (bropo for anomaly filtering, beamb for beam-blockage analysis, and satfilter for satellite consistency). While the extracted pixels reported high QI values (0.91 at the maxima quoted here), these metrics should be interpreted according to published EUMETNET guidelines; they do not establish radar accuracy. Beam blockage and scan geometry for the contributing radars were not analysed here.
  3. Estimation Uncertainty: Radar rain-rate estimates rely on a reflectivity–rain-rate conversion (in NIMBUS the standard Marshall–Palmer relation, \(Z = 200 R^{1.6}\)) and on measurements made above the ground. Drop-size distributions and processes below the beam are recognized general sources of uncertainty in radar-derived surface rain rates; this analysis does not test them.

Two-panel chart for Barcelona Airport from 1 to 5 October 2026. Top: OPERA radar rain-rate samples with small pulses on 1 October, a 15.4 mm/h peak on 2 October, two pulses up to 10.3 mm/h during the shaded 24-hour window of the 135.6 mm gauge report, and a 21.7 mm/h pulse at 07:15 UTC on 5 October. Bottom: blue blocks for the airport's five daily gauge reports, the largest 135.6 mm for 06:00 UTC on 3 October to 06:00 UTC on 4 October, with a green running sum reaching 181.4 mm.

Barcelona Airport, 1–5 October 2026 (UTC). Top: OPERA radar RATE samples at the pixel containing the gauge, with undetect samples (below 0.1 mm/h or no echo) drawn as dry and missing frames shaded rather than drawn as zero. The largest sample up to the gauge cutoff is 15.38 mm/h on 2 October; within the shaded 24 hours of the 135.6 mm gauge report (06:00 UTC on 3 October to 06:00 UTC on 4 October) the radar integral is 24.45 mm, and the largest sample overall, 21.7 mm/h at 07:15 UTC on 5 October, falls after the last gauge report. Bottom: the airport’s five PT24H reports drawn over their true 24-hour windows, with their running sum (stepping at each report’s end) reaching 181.4 mm from 06:00 UTC on 30 September. AEMET’s provisional daily listing assigns 135.6 mm to observing date 3 October; that window was not reconciled with the EuroMeteo report window. The comparison is exploratory and does not resolve the gauge–radar difference.

Explore or verify the result: In the EuroMeteo Explorer, select Spain, choose stations BARCELONA/AEROPUERTO (08181) or ARENYS_DE_MAR (08186), and inspect hourly (PT1H) and daily (PT24H) precipitation values for 30 September to 5 October 2026. In the OPERA Radar Explorer, select the RATE product for 1–5 October 2026 to inspect animated surface rain-rate estimates and single-pixel rain rate time series across Catalonia. The EuroMeteo guide and OPERA radar guide explain parameter definitions, quality indicators, and export workflows.

Conclusion

Ground gauge observations from EuroMeteo demonstrate the severity of the 30 September–5 October 2026 rainfall across eastern Spain. At Barcelona Airport, the gauge reports sum to 181.4 mm over the 120 hours from 06:00 UTC on 30 September (178.7 mm in the 72 hours from 06:00 UTC on 2 October), and the largest report is 135.6 mm in the 24 hours to 06:00 UTC on 4 October, while AEMET’s provisional extreme-value listing for observing date 3 October records 135.6 mm for October maximum daily precipitation. This total surpassed the station’s October reference-period maximum within 1991–2020 (124.8 mm on 8 October 2002), though it remained below the historical October record of 175.1 mm set on 3 October 1987. Further north at Arenys de Mar, convective downpours yielded 161.9 mm in the 24 hours to 06:00 UTC on 4 October, with a largest reported hourly value of 37.5 mm. At Barcelona Port Olímpic, the available reports give 91.2 mm in the five hours 18:00–23:00 UTC and 104.4 mm in the 12 hours to 06:00 UTC on 4 October, but no complete 24-hour or episode total.

Rain gauges provide the principal in-situ reference for point rainfall, while retaining instrument, exposure, quality-control and sampling uncertainties, but their discrete reporting intervals can obscure rapid sub-hourly precipitation pulses. Weather radar composites provide continuous spatial monitoring and estimate instantaneous surface rain rates based on reflectivity measured aloft. In this exploratory comparison, the largest sampled OPERA rates were 43.34 mm/h at Port Olímpic, 32.50 mm/h at Arenys de Mar and 15.38 mm/h at Barcelona Airport before the gauge cutoff (21.72 mm/h after it). When the radar rates were integrated over the gauge windows tabulated above, the radar depths were 18–43% of the gauge amounts at the Barcelona, Arenys de Mar and San Javier pixels and 94–101% at Talarn; Polinyà de Xúquer returned no radar data. The pixels are single 2 km cells about 1 km from each gauge, the 15-minute snapshots can miss short peaks, and the quality index does not establish accuracy, so the reasons for the differences remain unresolved. Crucially, this exploratory radar–gauge comparison does not independently establish the specific dynamical mechanisms or atmospheric circulation responsible for the storm.

Data current as of 06:00 UTC on 5 October 2026 for gauges (the last report in the archive) and 12:00 UTC on 5 October 2026 for radar. Observations compiled from EuroMeteo and OPERA radar composites.

Data Annex

Every EuroMeteo gauge and OPERA radar number above can be traced to these files; figures from external sources (AEMET listings and bulletins, INUNCAT, NIMBUS documentation, press reports) are attributed to those sources in the text and are not in the annex. All times are UTC; gauge reports are stamped at the end of the interval they cover, (T − duration, T].

The CSV files and the figures were generated together from the same EuroMeteo reports and the same cached OPERA extraction (extract_data.py, then generate_figures.py); the radar extraction needs access to the OPERA archive and is cached rather than re-run each time.

Station metadata for the Spanish weather stations analyzed in the October 2026 flood study. Coordinates are the EuroMeteo reference coordinates used for radar extraction. Station elevations are omitted because the EuroMeteo catalogue lists none for several of these stations and no other source is used here.
Station Name WMO ID WIGOS Identifier Latitude (°N) Longitude (°E) Comarca / Region Observational Role
Barcelona / Aeropuerto 08181 0-20000-0-08181 41.2969 2.0783 Baix Llobregat, Barcelona Primary Synoptic Airport Gauge
Arenys de Mar 08186 0-20000-0-08186 41.5872 2.5400 Maresme, Barcelona Coastal Automated Weather Station
Barcelona, Port Olímpic 08180 0-20000-0-08180 41.3906 2.2000 Barcelonès, Barcelona Urban Coastal Automated Station
Manresa 08174 0-20000-0-08174 41.7199 1.8403 Bages, Barcelona Inland Foothills Station
Girona / Costa Brava 08184 0-20000-0-08184 41.9008 2.7606 Selva, Girona Synoptic Airport Station
Porqueres (Banyoles) 08120 0-20000-0-08120 42.1044 2.7636 Pla de l’Estany, Girona Pre-Pyrenean Automated Station
Talarn 08112 0-20000-0-08112 42.2047 0.8647 Pallars Jussà, Lleida Pre-Pyrenean Station
Reus / Aeropuerto 08175 0-20000-0-08175 41.1475 1.1672 Baix Camp, Tarragona Southern Catalonia Synoptic Airport
Polinyà de Xúquer 08287 0-20000-0-08287 39.1839 -0.3714 Ribera Baixa, Valencia Selected Mediterranean Coastal Station
Murcia / San Javier 08433 0-20000-0-08433 37.7750 -0.8125 Mar Menor, Murcia Southern Mediterranean Inflow Gauge
Summary of gauge accumulations, peak hourly gauge values, maximum sampled OPERA radar surface rain-rate estimates, and official WMO standard normal comparisons for 30 September–5 October 2026. Gauge totals are the sum of non-overlapping reports to the 06:00 UTC 5 October data cutoff and are shown only when all 120 hours are covered. Radar maxima are the largest 15-minute samples from 00:00 UTC on 1 October to the gauge cutoff, extracted only for the six stations named in the radar section; † Barcelona Airport’s sample of 21.7 mm/h occurred at 07:15 UTC on 5 October, after the gauge cutoff. ‡ Port Olímpic has no report covering 11:00–18:00 UTC on 3 October (the hours ending 12:00 to 18:00) and no PT24H report containing its burst, so it has no complete episode total and the largest report is a 12-hour value; 125.8 mm is the sum of the available reports and is not a total. Stations peak in different windows, so the largest-report column is not simultaneous. “Largest reported PT1H” is shown only for stations with at least 12 hourly reports. A ‘—’ marks a normal not available in the consulted archive or radar not applicable, in accordance with raw archive fidelity rules. Barcelona Airport’s 135.6 mm is the PT24H ending 06:00 UTC on 4 October; AEMET provisional extreme records assign 135.6 mm to observing date 3 October, and the two windows were not reconciled. Percentages compare each station’s 120-hour total with the monthly normal; they do not denote return periods.
Station Name WMO ID 120-h total (mm), 06:00 UTC 30 Sep → 06:00 UTC 5 Oct Largest reported accumulation (mm; window ends, UTC) Largest reported PT1H (mm) Max sampled radar RATE (mm/h) WMO 1991–2020 Oct Normal (mm) Total % of 1991–2020 Normal WMO 1981–2010 Oct Normal (mm) Total % of 1981–2010 Normal
Polinyà de Xúquer 08287 185.8 138.4 (24 h, 06:00 2 Oct) 31.4 no radar data — — — —
Barcelona / Aeropuerto 08181 181.4 135.6 (24 h, 06:00 4 Oct) not reported (1 PT1H report) 15.4 (21.7†) 83.5 217.2% 91.0 199.3%
Murcia / San Javier 08433 178.2 176.8 (24 h, 06:00 3 Oct) not reported 32.5 31.9 558.6% 39.0 456.9%
Arenys de Mar 08186 163.5 161.9 (24 h, 06:00 4 Oct) 37.5 32.5 — — — —
Barcelona, Port Olímpic 08180 incomplete: 125.8 over 69 of 120 h‡ 104.4 (12 h, 06:00 4 Oct)‡ 37.4 43.3 — — — —
Girona / Costa Brava 08184 85.7 67.2 (24 h, 06:00 4 Oct) not reported not queried 86.0 99.7% 87.9 97.5%
Talarn 08112 82.4 70.2 (24 h, 06:00 2 Oct) 50.8 115.3 — — — —
Porqueres (Banyoles) 08120 72.3 57.3 (24 h, 06:00 4 Oct) 18.9 not queried — — — —
Reus / Aeropuerto 08175 50.0 30.3 (24 h, 06:00 2 Oct) not reported not queried 72.8 68.7% 72.5 69.0%
Manresa 08174 49.2 39.6 (24 h, 06:00 4 Oct) 8.4 not queried — — — —

Data Sources