M2 Fluctuations of emission ratios in urban plumes

CO2MVS RESEARCH ON SUPPLEMENTARY OBSERVATIONS M2 Fluctuations of emission ratios in urban plumes Due date of deliverable 31/06/2024 Submission date 20/06/2024 File Name CORSO-M2-V1.1.docx Work Package /Task WP1 / T1.2 Organisation Responsible of Deliverable Barcelona Supercomputing Center Author name(s) Marc Guevara, Aitor Val Revision number V1.1 Status Dissemination Level / location www.corso-project.eu The CORSO project (grant agreement No 101082194) is funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or the Commission. Neither the European Union nor the granting authority can be held responsible for them. CORSO M2 2 Table of Contents 1 Introduction .................................................................................................................... 2 2 Methods ......................................................................................................................... 2 3 Results ........................................................................................................................... 6 4 Conclusions, limitations and future works..................................................................... 11 5 References .................................................................................................................. 12 1 Introduction Current satellite missions measuring CO2 concentrations such as the NASA Orbiting Carbon Observatories (OCO-2, OCO-3) are only capable of collecting a sparse and infrequent set of measurements over a selection of large cities and industries across the world (e.g., Chevallier et al, 2022). This limits the usage of these instruments to continuously monitor and assess emissions from urban and industrial hotspots. To overcome this limitation, co-emitted NO2 and CO emissions derived from Tropospheric Monitoring Instrument (TROPOMI) column measurements are being used to indirectly quantify CO2 emissions from fossil fuel combustion processes at a fine scale, relying on the co-emission of these species (e.g., Hakkarainen et al., 2021). The estimation of these emission ratios in urban plumes, which are typically derived from bottom-up emission inventories, largely depend on the mix activity sectors that are occurring in the hotspots and their spatiotemporal characterisation. This document shortly describes the methods and results of quantifying the temporal fluctuations of CO:CO2 and NOx:CO2 emission ratios in urban plumes. We focus on the monthly and weekly scales and on a selection of major European cities. To compute the fluctuations, we use as input the regional European CAMS-REG_v5.1 inventory and the collection of day-of-the-year (daily) emission temporal profiles constructed in CORSO Milestone 1, which have been refined to consider pollutant-dependencies for the road transport sector. Results indicate that temporal fluctuations of emission ratios can vary heterogeneously across months of the year, day of the week and cities. At the weekly level, a clear weekend effect is observed across almost all the urban study cases. The results presented in this work suggest that the temporal component of CO:CO2 and NOx:CO2 emission ratios needs to be taken into account when indirectly quantifying CO2 emissions from TROPOMI-based NOx or CO estimates. 2 Methods To estimate the temporal fluctuations of CO:CO2 and NOx:CO2 emission ratios in selected European urban plumes we combined gridded emissions provided by the CAMS-REG_v5.1 inventory (Kuenen et al., 2022) with the daily emission temporal profiles constructed in CORSO Milestone 1. We considered emissions from year 2019, being the most recent year of the CAMS-REG_v5.1 time series. For each city, we consider emissions from those grid cells intersecting with their boundaries as reported by ESRI (2021) plus a buffer of 5km. As a results, in some of the urban areas emissions from large airports (e.g., Madrid) and ports (e.g., Amsterdam) are also included, as illustrated in Figure 1. We selected the cities ranked under the category 1 out of 5 by ESRI (2021) and then filtered out those belonging to countries not included in the daily emission temporal profile dataset due to lack of data (see CORSO-M1-V2 report for more details). As a result, we ended up with CORSO M2 3 a total of 21 cities. We did not analyse cities belonging to lower ESRI rankings as we wanted to focus on major hotspots that can be easily detected by satellite instruments. For each city, emissions from selected grid cells are summed up per species and sector. Figure 2 shows the contribution of each sector (expressed following the Gridded Nomenclature For Reporting nomenclature, GNFR) to total CO2, NOx and CO emissions per city. For CO2 from public energy (GNFR_A), residential and commercial combustion (GNFR_C) and, to a lesser extent, road transport (GNFR_F) tend to dominate total emissions in the majority of the cities, with some exceptions such as Lisbon, Rome or Oslo, where the contribution of GNFR_A is much lower or even null and other sources are more important. In the case of NOx, road transport is the main contributor (average of 50%) in almost all cities, with the exception of port cities such as Barcelona, Hamburg or Amsterdam, where the contribution of shipping emissions (GNFR_G) is significant (between 25% and 45%). Finally, for CO two sources generally appear as main dominators: residential and commercial combustion and road transport, the contribution from industrial combustion and processes sources (GNFR_B) and other non-mobile sources (GNFR_I, which includes machinery used in industry, households and agriculture) being also significant in specific cities such as Lisbon and London, respectively, with contributions ranging between 40 and 50%. Figure 1 Centroids (cyan dots) of the CAMS-REG grid cells selected for the quantification of the CO:CO2 and NOx:CO2 emission ratios in the Madrid (left) and Amsterdam (right) urban areas. The red polygons represent the definition of the urban areas as reported by ESRI (2021) CORSO M2 4 Figure 2 Shares of total CO2, NOx and CO emissions per GNFR sector across European selected cities as reported by the CAMS-REG_v5.1 inventory for 2019 CORSO M2 5 Annual emissions reported by CAMS-REG_v5.1 were downscaled to the daily scale using the countryand sector-dependent daily temporal profiles reported in CORSO Milestone 1 (see CORSO-M1-V2 report for more details). The profiles assume that the temporal distribution of CO2, CO and NOx emissions from the combustion sectors included in the dataset (energy, aviation, residential, road transport, shipping and manufacturing industry) are very similar and therefore species independent. In this work we revised this assumption for the road transport sector and consider the pollutantdependency associated to meteorological influences on gasoline CO and diesel NOx road transport emissions, namely the cold start processes and the temperature penalty on diesel aftertreatment devices, respectively. The original daily profiles were adjusted to take into account these meteorological influences considering the methods and parametrisations reported by Guevara et al. (2021), which are also used for the development of the CAMSTEMPO dataset in the CAMS core emission service. The meteorological parametrisations considered take into account information on outdoor temperature, which was derived from ERA5. As a result, we developed specific CO/NOx temporal profiles for gasoline/diesel road transport emissions, which are different from the ones developed for CO2 road transport under Milestone 1. Figure 3 illustrates the results obtained in Germany. It is observed how the NOx and CO profiles report larger weight daily temporal factors during wintertime due to the influence of colds temperatures, while during summer emissions present a larger drop than the one reported for CO2, which is only related to changes in congestion levels. The resulting pollutant-dependent profiles were combined with the CAMS-REG_v5.1 inventory, which splits road transport emissions by fuel type. Figure 3 Daily temporal profiles for the road transport sector in Germany for CO gasoline, CO2 all and NOx diesel emissions. CORSO M2 6 3 Results The CAMS-REG_v5.1 city level emissions and collection of temporal profiles were combined to derive daily CO2, NOx and CO emissions per GNFR sector. Note that no hourly emissions were computed in the framework of this work, as this temporal resolution is out of the scope of the applications that are being developed under CORSO. Figure 4 shows an example of the results obtained for the cities of Paris and Warsaw. The daily emissions were aggregated to the monthly and day-of-the-week scale to derive monthly and weekly NOx:CO2 and CO:CO2 emission ratios per city. Figure 4 Day-of-the year CO2, NOx and CO emissions per GNFR sector estimated for the cities of Paris (left) and Warsaw (right) CORSO M2 7 Figure 5 and Figure 6 show the resulting monthly and weekly emission ratios computed for selected cities, including Amsterdam, London, Madrid, Oslo, Paris, Rome and Warsaw. The selection includes cities from different climatological and socio-economic regions and that also presents differences in the sectoral contributions to total emissions (Figure 2). The ratios are expressed as t CO2 / t NOx for the NOx:CO2 ratios and t CO2 / t CO, for the CO:CO2 ratios. Four clear patterns can be distinguished when looking at the NOx:CO2 monthly emission ratio (Figure 5): i) emission ratios being lower in winter than in summer (Warsaw), ii) emission ratios being lower in summer than in winter (Paris, London, Amsterdam), iii) same as ii) but with a peak occurring in summer (Rome, Madrid) and iv) emission ratios being mostly constant throughout the year (Oslo). These different behaviours are mainly associated to the different sectoral contributions to total CO2 and NOx reported in each city (Figure 2) as well as to the differences in the way these sectoral emissions are temporally distributed across countries. The seasonality obtained in Warsaw, which oppositely to most of the cities shows an increase in the ratio during summer when compared to winter (up to 8%), can be linked to the large contribution of the public power sector to total CO2 emissions and the fact that the daily distribution of emissions from this sector is rather constant throughout the year, as previously shown in Figure 4. The combination of these two facts, together with the fact that NOx emissions from road transport have a more marked winter to summer seasonality due to the temperature effect (i.e., larger drop when compared to road transport CO2 emissions), results in an increase of the NOx:CO2 emission ratios during summer time. In the case of Paris, which shows a completely opposite and more marked seasonality (ratios being up to -30% lower in summer than in winter), CO2 emissions are dominated by residential and commercial combustion sources, which mainly occurs during winter time (Figure 4), and therefore the drop of this species in summer is much larger than for NOx, which is dominated by road transport and does not present such a marked winter to summer drop. The specific increase in the NOx:CO2 emission ratios observed during August in Madrid and Rome (increases up to 15% when compared to May) can be attributed to a significant drop in road transport emissions due to summer holidays. For Oslo the resulting flat profile is mainly related to two facts. On the one hand, the sectoral contribution to total NOx and CO2 emissions for this city is quite similar (Figure 2), and on the other, for both species the contribution of off-road mobile machinery is quite important (approximately 20%), the temporal distribution assumed for this sector being flat due to the lack of more detailed information. Concerning the monthly patterns of the CO:CO2 emission ratios, these are similar to the ones reported for NOx:CO2, but slightly flatter, which is mainly related to the largest contribution of off-road mobile machinery to total CO emissions across cities, for which we propose a flat temporal profiles due to the lack of more detailed data. For this emission ratio also see how Warsaw presents a completely different seasonality to the ones reported by the other cities. Again, in this case the main reason is the much more marked winter to summer drop reported for CO emissions, which are dominated by residential and commercial combustion emissions, than for CO2, dominated by the electricity production sector, which has a much flatter distribution (Figure 4). Regarding the weekly NOx:CO2 emission ratio profiles, we see that in all cities except for Amsterdam and Oslo the results show a weekend effect, with the ratios increasing during Saturdays and Sundays up to 10% when compared to weekends. This pattern is related to the larger drop of road transport emissions during weekend when compared to other sectors, and the fat that NOx is generally dominated by this source. In the case of Amsterdam, the flatter profile can be attributed to the larger contribution from shipping to total NOx emissions (Figure 2), which present a rather flat weekly distribution, especially when compared to the CO2 emission sources, which show a larger weekend drop. For the case of Oslo, again we see a flat distribution, which can be linked to the large contribution of off-road mobile machinery emissions, as discussed before. In the case of the weekly CO:CO2 emission ratios, the weekend effect is less clear and profiles look in general much flatter across cities. CORSO M2 8 Figure 5 Monthly NOx:CO2 and CO:CO2 emission ratios constructed for selected cities. The emission ratios are expressed as t CO2 / t NOx and t CO2 / t CO, respectively. CORSO M2 9 Figure 6 Weekly NOx:CO2 and CO:CO2 emission ratios constructed for selected cities. The emission ratios are expressed as t CO2 / t NOx and t CO2 / t CO, respectively. CORSO M2 16 Naples 11 631.4 336.6 Naples 12 712.1 314.8 Oslo 1 297.4 38.6 Oslo 2 290.8 42.5 Oslo 3 280.2 46.7 Oslo 4 277.3 46.2 Oslo 5 277.3 52.8 Oslo 6 290.7 54.0 Oslo 7 273.0 38.1 Oslo 8 285.8 53.7 Oslo 9 285.3 54.2 Oslo 10 279.3 49.9 Oslo 11 282.2 48.2 Oslo 12 286.8 46.8 Paris 1 789.6 298.9 Paris 2 724.3 278.0 Paris 3 680.1 265.5 Paris 4 635.6 233.4 Paris 5 546.4 189.1 Paris 6 495.2 155.9 Paris 7 530.4 153.9 Paris 8 518.6 143.5 Paris 9 529.0 174.7 Paris 10 603.2 216.3 Paris 11 701.6 284.1 Paris 12 714.8 280.6 Prague 1 731.6 156.2 Prague 2 708.9 152.8 Prague 3 670.0 168.9 Prague 4 641.8 179.7 Prague 5 633.7 221.7 Prague 6 656.8 331.1 Prague 7 698.9 350.8 Prague 8 671.7 312.2 Prague 9 646.5 299.2 Prague 10 635.8 206.4 Prague 11 630.0 180.7 Prague 12 634.8 164.9 Rome 1 907.6 366.0 Rome 2 785.3 342.4 Rome 3 769.1 339.3 Rome 4 705.8 324.1 Rome 5 579.9 288.8 Rome 6 598.8 298.7 Rome 7 633.9 312.1 Rome 8 673.8 319.1 Rome 9 587.3 296.1 Rome 10 623.7 304.0 Rome 11 701.2 323.8 Rome 12 777.4 339.6 Stockholm 1 882.5 442.3 Stockholm 2 819.3 412.1 Stockholm 3 774.0 409.0 Stockholm 4 689.3 361.0 Stockholm 5 621.7 317.6 Stockholm 6 600.8 292.9 Stockholm 7 611.0 293.3 Stockholm 8 599.2 295.4 CORSO M2 17 Stockholm 9 631.8 323.5 Stockholm 10 698.8 365.0 Stockholm 11 730.1 381.7 Stockholm 12 772.3 386.8 Vienna 1 1116.6 271.1 Vienna 2 1066.0 275.7 Vienna 3 980.7 260.0 Vienna 4 874.7 237.7 Vienna 5 767.0 212.5 Vienna 6 763.8 207.0 Vienna 7 805.7 205.0 Vienna 8 797.7 209.4 Vienna 9 831.4 234.1 Vienna 10 909.7 256.2 Vienna 11 984.2 270.8 Vienna 12 1011.0 263.6 Warsaw 1 748.3 202.2 Warsaw 2 742.9 195.9 Warsaw 3 720.6 228.4 Warsaw 4 723.1 252.7 Warsaw 5 750.3 368.2 Warsaw 6 781.3 532.1 Warsaw 7 803.8 566.1 Warsaw 8 796.9 539.9 Warsaw 9 759.0 431.0 Warsaw 10 728.0 305.0 Warsaw 11 723.3 250.6 Warsaw 12 700.4 195.9 CORSO M2 18 City Day-of-the-week CO2_NOx CO2_CO Amsterdam Monday 536.8 431.2 Amsterdam Tuesday 547.1 422.2 Amsterdam Wednesday 544.2 431.4 Amsterdam Thursday 540.9 415.2 Amsterdam Friday 537.8 422.7 Amsterdam Saturday 505.8 434.9 Amsterdam Sunday 477.1 441.1 Barcelona Monday 474.7 476.2 Barcelona Tuesday 486.2 472.0 Barcelona Wednesday 482.9 467.9 Barcelona Thursday 482.1 468.4 Barcelona Friday 475.1 460.9 Barcelona Saturday 487.6 488.5 Barcelona Sunday 482.7 508.2 Berlin Monday 913.2 413.6 Berlin Tuesday 894.4 400.8 Berlin Wednesday 888.4 397.7 Berlin Thursday 884.6 395.2 Berlin Friday 898.3 403.6 Berlin Saturday 968.0 429.5 Berlin Sunday 1015.2 458.8 Brussels Monday 768.2 631.4 Brussels Tuesday 752.8 631.2 Brussels Wednesday 757.4 630.5 Brussels Thursday 754.5 631.6 Brussels Friday 757.0 631.9 Brussels Saturday 852.3 625.4 Brussels Sunday 920.9 617.6 Copenhagen Monday 563.1 229.9 Copenhagen Tuesday 564.7 231.3 Copenhagen Wednesday 560.7 230.4 Copenhagen Thursday 562.4 231.0 Copenhagen Friday 567.7 232.5 Copenhagen Saturday 583.4 233.7 Copenhagen Sunday 575.6 230.1 Dublin Monday 713.2 718.7 Dublin Tuesday 701.1 702.5 Dublin Wednesday 699.0 699.3 Dublin Thursday 693.1 695.0 Dublin Friday 698.1 698.6 Dublin Saturday 736.1 734.7 Dublin Sunday 749.0 752.8 Hamburg Monday 698.4 465.6 Hamburg Tuesday 712.2 444.6 Hamburg Wednesday 708.7 441.2 Hamburg Thursday 703.5 438.2 Hamburg Friday 702.3 441.0 Hamburg Saturday 681.7 440.6 Hamburg Sunday 663.9 462.9 Helsinki Monday 673.5 356.1 Helsinki Tuesday 675.4 354.4 Helsinki Wednesday 674.7 352.0 Helsinki Thursday 673.4 350.7 Helsinki Friday 670.9 349.4 Helsinki Saturday 673.9 336.0 Helsinki Sunday 674.6 336.7 Lisbon Monday 290.4 217.2 CORSO M2 19 Lisbon Tuesday 298.6 215.5 Lisbon Wednesday 298.7 214.9 Lisbon Thursday 297.6 214.7 Lisbon Friday 297.5 214.6 Lisbon Saturday 302.3 210.7 Lisbon Sunday 291.2 207.3 London Monday 620.2 403.6 London Tuesday 595.3 400.9 London Wednesday 594.0 400.8 London Thursday 598.7 403.7 London Friday 607.8 407.6 London Saturday 636.9 399.1 London Sunday 656.0 387.7 Madrid Monday 471.9 371.2 Madrid Tuesday 478.9 368.9 Madrid Wednesday 474.4 366.6 Madrid Thursday 476.1 368.0 Madrid Friday 470.0 366.9 Madrid Saturday 550.2 386.5 Madrid Sunday 593.7 401.6 Milan Monday 731.1 317.9 Milan Tuesday 730.5 325.0 Milan Wednesday 719.1 326.0 Milan Thursday 718.8 327.3 Milan Friday 716.7 326.3 Milan Saturday 794.0 330.4 Milan Sunday 847.5 326.0 Munich Monday 980.3 429.0 Munich Tuesday 950.1 416.3 Munich Wednesday 946.2 413.8 Munich Thursday 943.6 411.1 Munich Friday 964.0 418.9 Munich Saturday 1050.8 441.6 Munich Sunday 1143.4 472.0 Naples Monday 633.4 333.3 Naples Tuesday 634.2 337.4 Naples Wednesday 624.4 338.0 Naples Thursday 621.6 339.3 Naples Friday 618.9 337.8 Naples Saturday 648.3 338.0 Naples Sunday 668.6 335.1 Oslo Monday 281.9 51.7 Oslo Tuesday 285.5 52.0 Oslo Wednesday 286.7 51.0 Oslo Thursday 286.0 50.9 Oslo Friday 285.5 47.9 Oslo Saturday 281.8 38.5 Oslo Sunday 277.4 36.5 Paris Monday 635.4 243.6 Paris Tuesday 630.0 237.9 Paris Wednesday 631.5 232.8 Paris Thursday 622.3 233.1 Paris Friday 630.9 234.7 Paris Saturday 682.0 217.6 Paris Sunday 707.4 213.4 Prague Monday 627.8 187.9 Prague Tuesday 634.1 199.9 Prague Wednesday 630.6 205.3 CORSO M2 20 Prague Thursday 638.0 206.8 Prague Friday 647.2 202.7 Prague Saturday 762.7 183.8 Prague Sunday 776.5 171.8 Rome Monday 717.4 326.2 Rome Tuesday 709.5 325.8 Rome Wednesday 696.0 322.7 Rome Thursday 694.4 322.8 Rome Friday 693.7 322.7 Rome Saturday 773.8 339.9 Rome Sunday 832.1 350.8 Stockholm Monday 709.0 360.3 Stockholm Tuesday 704.4 360.4 Stockholm Wednesday 703.2 360.1 Stockholm Thursday 702.6 360.4 Stockholm Friday 711.0 364.7 Stockholm Saturday 720.4 360.3 Stockholm Sunday 732.6 361.4 Vienna Monday 898.2 258.0 Vienna Tuesday 922.0 253.7 Vienna Wednesday 921.7 253.5 Vienna Thursday 919.2 251.4 Vienna Friday 912.3 248.5 Vienna Saturday 955.4 231.5 Vienna Sunday 983.4 226.5 Warsaw Monday 727.8 269.5 Warsaw Tuesday 730.9 278.3 Warsaw Wednesday 727.1 289.0 Warsaw Thursday 728.5 290.7 Warsaw Friday 729.8 284.4 Warsaw Saturday 784.7 274.0 Warsaw Sunday 806.2 257.3 CORSO M2 21