Bild des Monats
Text: Rasmus Nüß, Nikos Daskalakis, Oliver Schneising, Michael Buchwitz and Mihalis Vrekoussis
Top-down fire CO emissions optimization using TM5-4dvar and TROPOMI S5P
This study aims at optimizing the carbon monoxide (CO) emissions of the two large wildfires in Camp and Woolsey, California, that started on November 8th, 2018 and raged for multiple weeks. All the above figures show aggregated data for the main burning period of ten days (starting on the 6th is an unfortunate necessity due to model internal discretization).
To this end, the TM5-4dvarinverse model has been used to optimize the CO emissions from biomass burning provided by the Global Fire Emissions Database Version 4, including small fire boost (GFEDv4.1s) as well as the CO production from VOCs and methane on a global scale for October to December 2018. The inversion was driven by TROPOMI CO satellite observations (a) and NOAA surface flasks stations (not shown).
Figure (b) shows the modeled total column concentrations after the initial (a priori) forward run as sampled at the times and locations of the satellite observations with the satellites averaging kernel applied, i.e., what the satellite would have seen if the true atmosphere was equal to the atmosphere in the model. The mismatch between the a priori concentrations and the satellite observations (c) shows a clear overestimation over Asia and the northern hemispheric background at large. This background discrepancy could lead to an underestimation of the biomass burning source in a local inversion. Considering the relatively long lifetime of CO of about one-month, long-range transport patterns become relevant for the background. Hence a global inversion was conducted. Additionally, most of the background discrepancy can likely be attributed to an overestimation of the chemical production of CO from VOCs and methane (instead of biomass burning), which is why this source needed to be optimized simultaneously as well. It is possible to tell this source apart from the biomass burning source because changes in the former have significantly smaller temporal and spatial frequencies. Yet, some aliasing remains in the optimized emissions from both sources, which will be addressed in a later study, including additional tracers.
With this input, the model then tried in an iterative process to find the emissions that lead to concentrations as close as possible to the observations while not differing too much from the initial emissions. Figure (d) shows the resulting biomass burning emission increment, i.e., the change from the initial (a priori) to the final (a posteriori) emissions. The change in production from VOCs and methane is not pictured. In this figure, the region of interest (highlighted in the inner box, with Camp near the upper left and Woolsey in the lower right corner) is embedded in a 1° by 1° zooming region over North America. In comparison, the rest of the globe is simulated at 6° by 4° to save computational capacities. Within this region, the total emission increment over the main burning period shows only a slight increase (~12%) compared to GFED4.1s; however, most of this increment is aggregated in the pixel that contains Camp, while the adjacent pixels, all the way down to Woolsey, show small negative increments.
Running the model again with the a-posteriori emissions leads to concentrations (e) much closer to the observations. Their difference (f) is a lot closer to zero, especially in the zooming region over North America.
This figure (Figure 3 of Gier et al. (2020)) shows the global mean timeseries of monthly mean column-averaged dry-air mole fraction of CO2 (XCO2) for satellite observations and model data from emission driven runs of Earth System Models (ESMs) participating in the Coupled Model Intercomparison Project Phase 6 (CMIP6, Eyring et al. (2016)) and CMIP5 (Taylor et al., 2012). The models are sampled as the satellite observations (Buchwitz et al., 2018), with the panels showing the timeseries (top), the monthly growth rate (middle), and the seasonal cycle (bottom). For CMIP6 all available ensemble members, i.e. runs started from different points of the pre-industrial control simulation, are used to show the intrinsic variability of the models. The multi-model mean includes only the first ensemble member of each model, to not skew the result towards models with many ensemble members. In most cases ensemble members show a similar behavior to one another. The spread in the total value of the global mean has not been reduced by much in CMIP6 compared to CMIP5, although the multi-model mean bias has been reduced from +10ppmv to +2ppmv. The multi-model mean has a slight positive offset for the growth rate, which quantifies how fast the CO2 concentration is rising. The timeseries shows a clear seasonal cycle due to the absorption of CO2 by plants for photosynthesis in the summer, and a release of CO2 through respiration in the winter, which are both dominated by larger vegetated areas in the northern hemisphere. The CMIP6 models show a large improvement compared to CMIP5 in their ability to reproduce the seasonal cycle seen in the satellite observations.
The artifact in the growth rate in 2009 is due to the start of data coming from a second satellite, whose timeseries shows a differently shaped seasonal cycle and thus influences the growth rate due to the computation method employed here. In the annual mean growth rate this artifact is averaged out. Further sampling characteristics are discussed in Gier et al. (2020), such as the introduction of a strong negative trend in seasonal cycle amplitude with increasing XCO2 due to the spatial sampling of the satellite data. This figure was made using the Earth System Model Evaluation Tool v2 (ESMValTool, (Eyring et al., 2020; Righi et al., 2020) and will be included in the v2.2 release. The ESMValTool includes many diagnostics to use with CMIP models and observations, as well as common preprocessor functions, such as the computation of multi-model means, area averages, and the derivation of custom variables, which were used for this work. Further analysis of an updated satellite dataset is part of the Climate-Carbon Interactions in the Current Century (4C) project, which is funded by the EU under the Horizon 2020 program.
Advanced Earth System Model Evaluation for CMIP (Eval4CMIP)
Climate-Carbon Interactions in the Current Century (4C)
Coupled Model Intercomparison Project (CMIP)
Buchwitz, M., Reuter, M., Schneising, O., Noel, S., Gier, B., Bovensmann, H., Burrows, J. P., Boesch, H., Anand, J., Parker, R. J., Somkuti, P., Detmers, R. G., Hasekamp, O. P., Aben, I., Butz, A., Kuze, A., Suto, H., Yoshida, Y., Crisp, D., and O'Dell, C.: Computation and analysis of atmospheric carbon dioxide annual mean growth rates from satellite observations during 2003-2016, Atmos Chem Phys, 18, 1-22, 10.5194/acp-18-17355-2018, 2018.
Eyring, V., Bony, S., Meehl, G. A., Senior, C. A., Stevens, B., Stouffer, R. J., and Taylor, K. E.: Overview of the Coupled Model Intercomparison Project Phase 6 (CMIP6) experimental design and organization, Geosci Model Dev, 9, 1937-1958, 10.5194/gmd-9-1937-2016, 2016.
Eyring, V., Bock, L., Lauer, A., Righi, M., Schlund, M., Andela, B., Arnone, E., Bellprat, O., Brötz, B., Caron, L. P., Carvalhais, N., Cionni, I., Cortesi, N., Crezee, B., Davin, E. L., Davini, P., Debeire, K., de Mora, L., Deser, C., Docquier, D., Earnshaw, P., Ehbrecht, C., Gier, B. K., Gonzalez-Reviriego, N., Goodman, P., Hagemann, S., Hardiman, S., Hassler, B., Hunter, A., Kadow, C., Kindermann, S., Koirala, S., Koldunov, N., Lejeune, Q., Lembo, V., Lovato, T., Lucarini, V., Massonnet, F., Müller, B., Pandde, A., Pérez-Zanón, N., Phillips, A., Predoi, V., Russell, J., Sellar, A., Serva, F., Stacke, T., Swaminathan, R., Torralba, V., Vegas-Regidor, J., von Hardenberg, J., Weigel, K., and Zimmermann, K.: Earth System Model Evaluation Tool (ESMValTool) v2.0 – an extended set of large-scale diagnostics for quasi-operational and comprehensive evaluation of Earth system models in CMIP, Geosci. Model Dev., 13, 3383-3438, 10.5194/gmd-13-3383-2020, 2020.
Gier, B. K., Buchwitz, M., Reuter, M., Cox, P. M., Friedlingstein, P., and Eyring, V.: Spatially resolved evaluation of Earth system models with satellite column averaged CO2, Biogeosciences Discuss., 2020, 1-40, 10.5194/bg-2020-170, accepted, 2020.
Righi, M., Andela, B., Eyring, V., Lauer, A., Predoi, V., Schlund, M., Vegas-Regidor, J., Bock, L., Brötz, B., de Mora, L., Diblen, F., Dreyer, L., Drost, N., Earnshaw, P., Hassler, B., Koldunov, N., Little, B., Loosveldt Tomas, S., and Zimmermann, K.: Earth System Model Evaluation Tool (ESMValTool) v2.0 – technical overview, Geosci. Model Dev., 13, 1179-1199, 10.5194/gmd-13-1179-2020, 2020.
Taylor, K. E., Stouffer, R. J., and Meehl, G. A.: An Overview of CMIP5 and the Experiment Design, Bulletin of the American Meteorological Society, 93, 485-498, 10.1175/bams-d-11-00094.1, 2012.
Record low ozone above the Arctic during spring 2020:
This year very low total ozone was observed in March over a large part of the Arctic. Compared to one year before total ozone levels were up to 200 DU lower this year. These low ozone levels were observed in a strong cyclone in the lower stratosphere (~20 km altitude), called the polar vortex. Temperatures inside such a cyclone are sufficiently low to form polar stratospheric clouds that activate halogens from their reservoir species and resulting in substantial chemical depletion (polar ozone loss) in late winter and early spring. Such conditions are very typical during Antarctic winter/spring (“Antarctic ozone hole”) but occur sporadically in the Arctic. A similar event in the northern hemisphere was last observed in March 2011 nine years ago. Total ozone columns regulate how much harmful UV radiation can reach the surface. We retrieved the total ozone using spectral measurements from the TROPOMI instrument aboard the Sentinel 5 Precursor satellite launched in October 2017. Our analysis showed that the minimum total ozone column north of 50°N were at a record low throughout March and April 2020. While global ozone levels are slowly recovering due to the slow decline in man-made stratospheric halogens following the phase-out of ozone-depleting substances regulated by the Montreal Protocol and Amendments, such extreme events will likely happen again, about once per decade in the future.
Mark Weber (email@example.com)
The sources and sinks of methane (CH4) in tropical rainforests are still a large source of uncertainty, and flux measurements are rare. In October 2018, INPA (National Institute for Amazonian Research, Manaus) and IUP started collaborative measurements, focusing on ecosystem CH4 flux measurements. An FTIR-analyzer, measuring CO2, CO, CH4, N2O and δ13CO2, was installed at a 50 m flux tower, measuring fluxes and concentrations of CH4. In addition, by use of a portable instrument, CH4 and CO2 flux measurements of soil, water, trees and termites were performed. Preliminary results show that the ecosystem is a small but constant source of methane, and reveal the possible large role termites play in this ecosystem.
Carbon monoxide emissions from industrial facilities
The TROPOspheric Monitoring Instrument (TROPOMI) onboard the Sentinel-5 Precursor satellite allows for the analysis of several atmospheric species, including carbon monoxide (CO) and methane (CH4), with an unprecedented level of detail by combining high precision and spatial resolution with daily global coverage.
The figure shows the CO distribution for November and December 2018 over China, India, and Southeast Asia highlighting emissions of congested urban areas and industrial facilities. The 2-month-average was chosen to get an overview of the complete region. Typically, large emission sources can even be detected in a single satellite overpass. The tracked facilities mainly belong
to the Chinese and Indian iron and steel industry. CO enhancements due to emissions from steel works are also detected in other regions of the world, for example in Turkey and Poland.
In steelmaking CO is formed during two processes. Firstly, it is an essential constituent of the blast furnace gas, which emerges when iron ore is reduced with coke to metallic pig iron. As the resulting pig iron has a relatively high carbon content, further processing is necessary to harden the metal. Therefore, the carbon-rich molten pig iron is converted to steel by lowering its carbon content via oxidation in the oxygen converter process (Linz-Donawitz-steelmaking). The resulting converter gas predominantly consists of CO.
Schneising, O., Buchwitz, M., Reuter, M., Bovensmann, H., Burrows, J. P., Borsdorff, T., Deutscher, N. M., Feist, D. G., Griffith, D. W. T., Hase, F., Hermans, C., Iraci, L. T., Kivi, R., Landgraf, J., Morino, I., Notholt, J., Petri, C., Pollard, D. F., Roche, S., Shiomi, K., Strong, K., Sussmann, R., Velazco, V. A., Warneke, T., and Wunch, D.: A scientific algorithm to simultaneously retrieve carbon monoxide and methane from TROPOMI onboard Sentinel-5 Precursor, Atmos. Meas. Tech., 12, 6771–6802, https://doi.org/10.5194/amt-12-6771-2019, 2019.
Stratospheric aerosols are an essential component of the climate system. Firstly, they change the radiative budget of the Earth, and secondly, they play an important role in ozone depletion. There is a background stratospheric aerosol layer build by the sulfuric gas emissions from the ocean surface; however, this layer is occasionally perturbed by volcanic eruptions which emit aerosol precursors directly in the stratosphere.
In the figure above, the changes in stratospheric aerosol loading after the 2018 eruption of Ambae are shown. Ambae, a volcanic island in South Pacific (Vanuatu), continuously erupted for over a year in 2017-2018; and it was the strongest volcanic eruption in the world for those two years. There were two eruption episodes when sulfuric gases reached the stratosphere, namely, on April 6 and on July 28, 2018. In the figure above, 10-day averaged aerosol extinction coefficient at 869 nm retrieved from OMPS-LP data is presented at 18.5 km (panel (a)) and at 20.5 km (panel (b)) as a function of latitude and time. The location and time of the eruption are marked by the black triangles. There is a slight increase in the aerosol extinction coefficient at 18.5 km after the fist, smaller eruption. At the same time, the increase after the second eruption is much larger; this is because the triple amount of precursor gases was emitted during the July episode. For both eruptions, the increase becomes noticeable right after the precursors were emitted, but the plume intensifies in about a month. This corresponds to the time needed for oxidation of the precursors to aerosols.
In particular after the second eruption, aerosols are not only localized in the tropics and are transported in the higher latitudes. Thus, the volcanic impact is seen at 18.5 km at 45°, both North and South, in late December 2018. In the tropics, at that time, the plume starts to dissipate; however, it still does not vanish completely. Such distribution of a volcanic plume is typical for a tropical eruption, as the Brewer-Dobson circulation moves the stratospheric air masses polewards from the equator.
Another resulting pattern of the Brewer-Dobson is aerosol distribution at 20.5 km. There, the increase after the first eruption is rather small; it appears in mid-May and lasts until the second eruption. The delay by approximately 1.5 months in the extinction increase at the higher altitudes is caused by convection. The lofting air in the tropics transports the aerosols and their precursors upwards. The time lag until the perturbation reaches higher altitudes is often referred to as a tape-recorder effect. After the second eruption, the plume appears at 20.5 km with 6-7 weeks delay as well. At this altitude, the maximal extension of the volcanic plume is achieved in November-December 2018. At the very end of 2018, the aerosols are transported to the higher latitudes, and the plume starts to dissipate.
This study shows the global impact of tropical volcanic eruptions. Even though their radiative forcing might be lower than for the extratropical eruptions of the same magnitude, the effect is seen all over the globe and does not stick to a particular location.
Reference: Malinina, E., Rozanov, A., Niemeier, U., Peglow, S., Arosio, C., Wrana, F., Timmreck, C., von Savigny, C. and Burrows, J. P. : Changes in stratospheric aerosol extinction coefficient after the 2018 Ambae eruption as seen by OMPS-LP and ECHAM-HAM, to be submitted in ACP in May 2020.
Wild fires are known to emit large quantities of trace gases and aerosols into the atmosphere, resulting in air pollution in the vicinity of the fires but often also at long distances downwind of the fires. In August 2018, unusually high temperatures caused severe drought in some areas of North America and resulted in the outbreak of many wildfires: the province of British Columbia (BC) in Canada was one of the most affected areas. Some of these fires were large enough to create enough heat to lift the emissions into the free troposphere where they were transported over long distances.
Measurements of the TROPOMI instrument on the European Sentinel 5 precursor satellite can be used to retrieve the column amounts of several trace gases emitted by wild fires (see figure). On August 10, 2018, a huge plume of particles and gases emitted from several big fires was covering large parts of North America, extending more than 1500 km from the location of the fires. While long-range transport of carbon monoxide (CO) has already been observed on many occasions, long-range transport of short-lived volatile organic compounds (VOCs) such as formaldehyde (HCHO) and glyoxal (CHO.CHO) is not expected as transport times are longer than the atmospheric residence time of these gases. However, the measurements show that also HCHO and CHO.CHO are present in the plume for at least a day, and comparison with FLEXPART transport modelling indicates effective lifetimes of the order of 30 hours. These unexpected observations are best explained by continuous production of HCHO and CHO.CHO from long-lived precursors in the plume, the nature of which still needs to be investigated.
This study highlights the importance of wild fires for air quality both locally and at a distance, an important aspect in a changing climate where hot and dry years such as 2018 are expected to become more frequent, resulting in more and bigger fires. It also demonstrates the power of S5P satellite observations which provide global trace gas column maps with high spatial resolution.
Alvarado, L. M. A., Richter, A., Vrekoussis, M., Hilboll, A., Kalisz Hedegaard, A. B., Schneising, O., and Burrows, J. P.: Unexpected long-range transport of glyoxal and formaldehyde observed from the Copernicus Sentinel-5 Precursor satellite during the 2018 Canadian wildfires, , 20, 2057–2072, https://doi.org/10.5194/acp-20-2057-2020, 2020.
Recent work at the Institute of Environmental Physics exploits the potential of the satellite instrument POLDER which was using multi-spectral and multi-viewing capabilities, for the retrieval of suspended liquid or solid particles in air (also known as aerosols), as well as surface properties.
The image shows the Aerosol Optical Depth (AOT) which is an important parameter describing the overall atmospheric aerosol burden. The study aims at retrieving AOT and surface reflectance at high quality which are important quantities for the modeling community as well as crucial parameters for air quality assessment from space. The agreement of the first results with ground-based measurements, as shown in the figure is very promissing: the color-coded points at the center of each station are close to the retrieved values based on satellite data.
This is of particular relevance because within the next decade a fleet of upcoming satellite instruments will provide measurements similar to POLDER which can be evaluated with the newly developed methodology.
Reference: Marco Vountas, Kristina Belinska, Vladimir V. Rozanov, Luca Lelli, Linlu Mei, Soheila Jafariserajehlou, John P. Burrows, "Retrieval of aerosol optical thickness and surfaceparameters based on multi-spectral and multi-viewing space-borne measurements" (2020).