Abstract. Recent developments in atmospheric remote sensing from satellites have made it possible to resolve daily emission plumes from industrial point sources around the globe. Wind rotation aggregation coupled with statistical fitting is commonly used to extract emission estimates from these observations. These methods are used here to investigate how the Coriolis effect influences the trajectory of observed emission plumes as well as to assess the impact of this influence on satellite-derived emission estimates. Of the 16 industrial sites investigated, 9 showed the expected curvature for the hemisphere that they reside in, 5 showed no or negligible curvature, and 2 showed opposing or unusual curvature. The sites that showed conflicting curvature reside in topographically diverse regions, where strong meso-γ-scale (2–20 km) turbulence dominates over larger synoptic circulation patterns. For high-curvature cases, the assumption that the wind-rotated plume aggregate is symmetrically distributed across the downwind axis breaks down, which impairs the quality of statistical fitting procedures. Using annual NOx emissions from Matimba power station as a test case, not compensating for Coriolis curvature resulted in an underestimation of ∼ 9 % on average for the years 2018 to 2021. This study is the first formal observation of the Coriolis effect and its influence on satellite-derived emission estimates, and it highlights both the variability in the emission calculation methods and the need for a standardised scheme for these data to act as evidence for regulators. For high-curvature cases, the assumption that the wind-rotated plume aggregate is symmetrically distributed across the downwind axis breaks down, which impairs the quality of statistical fitting procedures. Using annual NO x emissions from Matimba power station as a test case, not compensating for Coriolis curvature resulted in an underestimation of ∼  9 % on average for the years 2018 to 2021. This study is the first formal observation of the Coriolis effect and its influence on satellite-derived emission estimates, and it highlights both the variability in the emission calculation methods and the need for a standardised scheme for these data to act as evidence for regulators. Here, we investigate the influence of the Coriolis effect on large industrial emission plumes using observations of nitrogen dioxide (NO 2 ) from TROPOMI and explore the impact of Coriolis-induced curvature, plume geometry, and wind fields on satellite-derived emission estimates from large point sources. 2  Data and methods 2.1  TROPOMI NO 2 TROPOMI was launched by the European Space Agency (ESA) in October 2017 aboard the Sentinel-5 Precursor (S5P) satellite. TROPOMI is a nadir-viewing (downward-facing) short-wave spectrometer, observing in the ultraviolet–visible (UV–Vis, 270–500 nm), near-infrared (NIR, 710–770 nm), and short-wave infrared (SWIR, 2314–2382 nm) ranges ( Veefkind et al. Panel  (a) presents a single overpass from TROPOMI for Belchatow power station on 3 June 2019. In panel  (b) , the plume is rotated so that its wind vector now points eastwards. This initial stage of the plume is well aligned, but Coriolis curvature causes the latter parts of the plume to deviate from the downwind x  axis. In panel  (c) , this rotational process is repeated for all quality observations and aggregated into a wind-rotated average. On average the Coriolis effect causes a clockwise deflection of the aggregate plume, increasing magnitude with distance. The deflection caused by this force is known as the Coriolis effect, and it manifests in the atmosphere as large-scale clockwise deflections in the Northern Hemisphere (NH) and anticlockwise deflections in the Southern Hemisphere (SH) (Fig. 3 ). The effect is greatest at the poles, negligible at the Equator, and greater for higher-velocity wind speeds. Figure 3 Illustration of the Coriolis effect on atmospheric circulation patterns. Panel  (b) is produced using an average of the ERA5 100 m winds for 2019 at 12:00 UTC. Figure 4 Schematic describing the process behind an Ekman spiral from the (a) side and (b) plan views. This diagram shows the spiral for the Southern Hemisphere. Furthermore, as the plume ascends due to its thermal buoyancy and the wind field's inherent vertical velocity, the plume will move to a degree with an Ekman spiral ( Ekman ,  1905 ; Åkerblom ,  1908 ) , which itself is a consequence of the Coriolis effect and is demonstrated in Fig. 4 . This can be explained by the topographical surroundings of Jorge Lacerda, as the power station sits between the South Atlantic and the Serra do Mar coastal mountain range, where there is an abrupt > 1200 m increase in altitude over a short distance. Onshore synoptic-scale winds and sea breeze penetrate inland and are “steered” by the topography. These local effects, over tens of kilometres, dominate over larger This confirms that rotations based on the wind vector near the source produce well-aligned and symmetric aggregate plumes in the near field; however, the Coriolis curvature becomes distinguishable from the initial alignment at greater distances, and it is increasingly deviated and asymmetric relative to the common axis as the plume progresses downwind. Aggregates using winds at 900–850 hPa are very similar, with 825–800 hPa showing better initial alignment with the axis of aggregation, within the first 10 km. Figure 10 Demonstration of the difference in aggregate when using wind products from different pressure levels, using data from Matimba power station for May 2018–November 2021. Download 3.3  Impact of the Coriolis curvature on the emission estimates From the wind-rotated aggregate, the typical next step is to take the integral of evenly spaced (1 km) across-wind ( ± 30 km) segments perpendicular to the x  axis, as shown in Fig. 11 a. Figure 11 Demonstration of the impact that the Coriolis effect has on the resulting emission estimate.