Extended Data Table 1 Input variables used in the PBREAK and CATHARE simulations
From: Methane emissions from the Nord Stream subsea pipeline leaks
- *Uncertainty bounds are only provided for variables that have not been explicitly reported in the literature which required us to estimate their values independently. †Initial pressures in the pipelines are taken directly from values reported by pipeline operators to the Danish Energy Agency (DEA) (Danish Energy Agency, personal communication). Pressure values for NS1 pipelines are corroborated by the steady state shut-in settle-out pressure reported in refs. 26,104. This pressure occurs when the pipelines are fully pressurized with gas but remain inactive, with no flow. The pressures used in our modelling of NS1A, NS1B and NS2A pipelines are consistent with natural gas volumes reported by pipeline operators to the DEA5,105,106 (see Supplementary Table 1). All pressures are reported in Absolute bar (barA). ‡Estimated using average seawater temperatures measured over 15 years in the Baltic Sea107. We used multi-year average temperatures over the entire Baltic Sea given the pipelines run over multiple oceanographic Basins. We have assigned an uncertainty of ±2.5 °C to this value to reflect the typical range experienced at the associated depths in the Baltic Sea. §Taken directly from measurements reported in ref. 26. Our simulations assume the orifice size at the rupture sites were equal to the reported inner diameter of the pipelines given the pipelines were reported to have completely ruptured31. ||Estimated by geospatially locating the coordinates of the leak sites to the nearest subsea terrain depths provided in the European Marine Observation and Data Network (EMODnet) Digital Terrain Model (DTM) database30. Reported depth errors in the EMODNet DTM used for the geolocated locations are below ±1 m. We conservatively assign ±2 m uncertainty to our depth estimates to exceed these values by a factor of two. ¶Note we do not vary the total pipeline length in our sensitivity analyses (see Methods), just the distances either side of the rupture points. #Given the rupture for NS1A occurred in the Swedish Economic Exclusion Zone (EEZ), pipeline lengths reported in ref. 26 indicate the rupture likely occurred <1004.9 km from Russian landfall. Based on the coordinates of the leaks, we estimate there was approximately 5 km distance (by surface) between the leak and the Swedish EEZ boundary. We therefore estimate the rupture occurred around 1000 km from Russian landfall. We assign an uncertainty of ±5 km to conservatively account for the fact the distances reported in ref. 26 may have changed due to final optimisation of the pipeline route. ☆Given the rupture for NS1B occurred in the Danish EEZ, pipeline lengths reported in ref. 26 indicate the rupture likely occurred >1004.9 km from Russian landfall. The coordinates provided in ref. 26 indicate there was approximately 6 km distance between the NS1A and NS1B leak sites. We therefore estimate the rupture occurred around 1006 km from Russian landfall. Like NS1A, we assign an uncertainty of ±5 km. **The exact route of the NS2A pipeline is not known. Previous attempts by ref. 3 to cross-verify surface-level distances between the rupture points and landfall (calculated to be 1210 km) with the reported pipeline length (1230 km) were unsuccessful. Accordingly, we assigned a higher uncertainty of ±20 km to this distance. ††The distances reported refer to the southernmost NS2A leak site. ‡‡Note the value reported here (93% by mass) differs to that reported in Extended Data Table 2 (96% molar) due to the different reported units. The uncertainty in this value (±2% by mass) is a conservative estimate of variations in Russian Natural gas reported for 202134 and 2022108, which differ by less than 1% molar.
