Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “CORRELATION COEFFICIENT”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 199 records · Page 11
All sourcesNASA NTRSDOE OSTI
NASA NTRS2013

On the Nature of the mHz X-Ray QPOs from ULX M82 X-1: Evidence for Timing-Spectral (anti) Correlation

Using all the archival XMM-Newton X-ray (3-10 keV) observations of the ultraluminous X-ray source (ULX) M82 X-1 we searched for a correlation between its variable mHz quasi-periodic oscillation (QPO) frequency and its energy spectral power-law index. These quantities are known to correlate in stellar mass black holes (StMBHs) exhibiting Type-C QPOs (approx 0.2-15 Hz). The detection of such a correlation would strengthen the identification of its mHz QPOs as Type-C and enable a more reliable mass estimate by scaling its QPO frequencies to those of Type-C QPOs in StMBHs of known mass. We resolved the count rates of M82 X-1 and a nearby bright ULX (source 5/X42.3+59) through surface brightness modeling and identify observations in which M82 X-1 was at least as bright as source 5. Using only those observations, we detect QPOs in the frequency range of 36-210 mHz during which the energy spectral power-law index varied from 1.7-2.2. Interestingly, we find evidence for an anti-correlation (Pearsons correlation coefficient = -0.95) between the power-law index and the QPO centroid frequency. While such an anti-correlation is observed in StMBHs at high Type-C QPO frequencies (approx 5-15 Hz), the frequency range over which it holds in StMBHs is significantly smaller (factor of approx 1.5-3) than the QPO range reported here from M82 X-1 (factor of 6). However, it remains possible that contamination from source 5 can bias our result. Joint Chandra/XMM-Newton observations in the future can resolve this problem and confirm the timing-spectral anti-correlation reported here.

Pasham, Dheeraj R.↗
NASA NTRS1995

Performance of Residual Carrier Array Feed Combining in Correlated Noise

An array feed combining system for the recovery of SNR loss due to antenna reflector deformation has been implemented and is currently being evaluated on the Jet Propulsion Laboratory 34 meter DSS-13 antenna. In this system, the defocused signal field captured by a focal plane array feed is recovered using real-time signal processing and signal combining techniques. The current signal processing and signal combining algorithms are optimum under the assumption that the white Gaussian noise processes in the received signals from different array elements are mutually uncorrelated. Experimental data at DSS-13 indicate that these noise processes are indeed mutually correlated. The main result of this paper is an analytical derivation of the actual SNR performance of the current suboptimal signal combining algorithm in this correlated noises environment. The analysis here shows that the combined signal SNR can either be improved or degraded depending on the relation between the array signal and noise correlation coefficient phases. Further performance improvement will require the development of signal combining methods that take into account the correlated noises.

Residual↗
NASA NTRS1972

Evaluation of the pulse-contour method of determining stroke volume in man.

The pulse-contour method for determining stroke volume has been employed as a continuous rapid method of monitoring the cardiovascular status of patients. Twenty-one patients with ischemic heart disease and 21 patients with mitral valve disease were subjected to a variety of hemodynamic interventions. The pulse-contour estimations, using three different formulas derived by Warner, Kouchoukos, and Herd, were compared with indicator-dilution outputs. A comparison of the results of the two methods for determining stroke volume yielded correlation coefficients ranging from 0.59 to 0.84. The better performing Warner formula yielded a coefficient of variation of about 20%. The type of hemodynamic interventions employed did not significantly affect the results using the pulse-contour method. Although the correlation of the pulse-contour and indicator-dilution stroke volumes is high, the coefficient of variation is such that small changes in stroke volume cannot be accurately assessed by the pulse-contour method. However, the simplicity and rapidity of this method compared to determination of cardiac output by Fick or indicator-dilution methods makes it a potentially useful adjunct for monitoring critically ill patients.

Alderman, E. L.↗
NASA NTRS1991

Results of a sub-scale model rotor icing test

A heavily instrumented sub-scale model of a helicopter main rotor was tested in the NASA Lewis Research Center Icing Research Tunnel (IRT) in September and November 1989. The four-bladed main rotor had a diameter of 1.83 m (6.00 ft) and the 0.124 m (4.9 in) chord rotor blades were specially fabricated for this experiment. The instrumented rotor was mounted on a Sikorsky Aircraft Powered Force Model, which enclosed a rotor balance and other measurement systems. The model rotor was exposed to a range of icing conditions that included variations in temperature, liquid water content, and median droplet diameter, and was operated over ranges of advance ratio, shaft angle, tip Mach number (rotor speed) and weight coefficient to determine the effect of these parameters on ice accretion. In addition to strain gage and balance data, the test was documented with still, video, and high speed photography, ice profile tracings, and ice molds. The sensitivity of the model rotor to the test parameters, is given, and the result to theoretical predictions are compared. Test data quality was excellent, and ice accretion prediction methods and rotor performance prediction methods (using published icing lift and drag relationships) reproduced the performance trends observed in the test. Adjustments to the correlation coefficients to improve the level of correlation are suggested.

Flemming, Robert J.↗
NASA NTRS1991

Results of a sub-scale model rotor icing test

A heavily instrumented sub-scale model of a helicopter main rotor was tested in the NASA Lewis Research Center Icing Research Tunnel (IRT) in September and November 1989. The four-bladed main rotor had a diameter of 1.83 m (6.00 ft) and the 0.124 m (4.9 in) chord rotor blades were specially fabricated for this experiment. The instrumented rotor was mounted on a Sikorsky Aircraft Powered Force Model, which enclosed a rotor balance and other measurement systems. The model rotor was exposed to a range of icing conditions that included variations in temperature, liquid water content, and median droplet diameter, and was operated over ranges of advance ratio, shaft angle, tip Mach number (rotor speed) and weight coefficient to determine the effect of these parameters on ice accretion. In addition to strain gage and balance data, the test was documented with still, video, and high speed photography, ice profile tracings, and ice molds. The sensitivity of the model rotor to the test parameters is given, and the result to theoretical predictions are compared. Test data quality was excellent, and ice accretion prediction methods and rotor performance prediction methods (using published icing lift and drag relationships) reproduced the performance trends observed in the test. Adjustments to the correlation coefficients to improve the level of correlation are suggested.

Flemming, Robert J.↗
DOE OSTI2021

Nonlinear ultrasonic technique for the characterization of microstructure in additive materials

This study employs nonlinear ultrasonic techniques to track microstructural changes in additively manufactured metals. The second harmonic generation technique based on the transmission of Rayleigh surface waves is used to measure the acoustic nonlinearity parameter, β. Stainless steel specimens are made through three procedures: traditional wrought manufacturing, laser-powder bed fusion, and laser engineered net shaping. The β parameter is measured through successive steps of an annealing heat treatment intended to decrease dislocation density. Dislocation density is known to be sensitive to manufacturing variables. In agreement with fundamental material models for the dislocation-acoustic nonlinearity relationship in the second harmonic generation, β drops in each specimen throughout the heat treatment before recrystallization. Geometrically necessary dislocations (GNDs) are measured from electron back-scatter diffraction as a quantitative indicator of dislocations; average GND density and β are found to have a statistical correlation coefficient of 0.852 showing the sensitivity of β to dislocations in additively manufactured metals. Moreover, β shows an excellent correlation with hardness, which is a measure of the macroscopic effect of dislocations.

42 ENGINEERING↗
NASA NTRS1979

The prediction of AE, ap, and Dst at time lags between 0 and 30 hours

The magnetic activity indexes AE, ap, and Dst are correlated with approximately 35,000 hours of interplanetary plasma and magnetic field measurements acquired near the Earth (assembled by NASA/NSSDC into a composite data sat). Lag times between the indexes and solar wind parameters ranged from 0 to 30 hours. Correlations at lags less than 6 hours yield results in agreement with previous studies. At greater lags, the correlation coefficients between the solar wind parameters and AE and ap approach these parameters' autocorrelation (persistence) values. For Dst the correlation with solar wind parameters is lower than that with AE and ap in the 3 to 4 hour lag range whereas the autocorrelation of Dst is significantly higher over the entire 0 to 30 hour lag range. The implications of these differences between AE, ap, and Dst are discussed in terms of persistence of solar wind structure.

Smart, D. F.↗
NASA NTRS1982

Reduction in peak oxygen uptake after prolonged bed rest

The hypothesis that the magnitude of the reduction in peak oxygen uptake (VO2) after bed rest is directly proportional to the level of pre-bed rest peak VO2 is tested. Complete pre and post-bed rest working capacity and body weight data were obtained from studies involving 24 men (19-24 years old) and 8 women (23-34 years old) who underwent bed rest for 14-20 days with no remedial treatments. Results of regression analyses of the present change in post-bed rest peak VO2 on pre-bed rest peak VO2 with 32 subjects show correlation coefficients of -0.03 (NS) for data expressed in 1/min and -0.17 for data expressed in ml/min-kg. In addition, significant correlations are found that support the hypothesis only when peak VO2 data are analyzed separately from studies that utilized the cycle ergometer, particularly with subjects in the supine position, as opposed to data obtained from treadmill peak VO2 tests. It is concluded that orthostatic factors, associated with the upright body position and relatively high levels of physical fitness from endurance training, appear to increase the variability of pre and particularly post-bed rest peak VO2 data, which would lead to rejection of the hypothesis.

Greenleaf, J. E.↗
NASA NTRS1987

The correlation between far-IR and radio continuum emission from spiral galaxies

A sample of 30 galaxies selected for their intense IRAS flux at 60 and 100 micron using the Arecibo telescope at 21 cm to measure the continuum and HI line luminosities were observed. The centimeter wave continuum correlates very well with the far-infrared flux, with a correlation coefficient as high as that found for other samples, and the same ratio between FIR and radio luminosities. Weaker correlations are seen between the FIR and optical luminosity and between the FIR and radio continuum. There is very little correlation between the FIR and the HI mass deduced from the integral of the 21 cm line. The strength of the radio continuum correlation suggests that there is little contribution to either the radio and FIR from physical processes not affecting both. If they each reflect time integrals of the star formation rate then the time constants must be similar, or the star formation rate must change slowly in these galaxies.

Dickey, John M.↗
NASA NTRS1995

Performance of residual carrier array-feed combining in correlated noise

An array feed combining system for the recovery of signal-to-noise ratio (SNR) loss due to antenna reflector deformation has been implemented and is currently being evaluated on the Jet Propulsion Laboratory's 34-meter DSS-13 antenna. In this system, the defocused signal field captured by a focal plane array feed is recovered using real-time signal-processing and signal-combining techniques. The current signal-processing and signal-combining algorithms are optimum under the assumption that the white Gaussian noise processes in the received signals from different array elements are mutually uncorrelated. Experimental data at DSS 13 indicate that these noise processes are indeed mutually correlated. The main result of this article is an analytical derivation of the actual SNR performance of the current suboptimal signal-combining algorithm in this correlated-noise environment. The analysis here shows that the combined signal SNR can either be improved or degraded depending on the relation between the array signal and noise correlation coefficient phases. Further performance improvement will require the development of signal-combining methods that take into account the correlated noises.

Tan, H. H.↗
NASA NTRS2015

Comparisons of Characteristics of Magnetic Clouds and Cloud-Like Structures During 1995-2012

Using eighteen years (1995 - 2012) of solar wind plasma and magnetic field data (observed by the Wind spacecraft), solar activity (e.g. sunspot number: SSN), and the geomagnetic activity index (Dst), we have identified 168 magnetic clouds (MCs) and 197 magnetic cloud - like structures (MCLs), and we have made relevant comparisons. The following features are found during seven different periods (TP: Total period during 1995 - 2012, P1 and P2: first and second half period during 1995 - 2003 and 2004 - 2012, Q1 and Q2: quiet periods during 1995 - 1997 and 2007 - 2009, A1 and A2: active periods during 1998 - 2006 and 2010 - 2012). (1) During the total period the yearly occurrence frequency is 9.3 for MCs and 10.9 for MCLs. (2) In the quiet periods Q1 > Q1 and Q2 > Q2, but in the active periods A1 < A1 and A2 < A2. (3) The minimum Bz (Bzmin) inside of a MC is well correlated with the intensity of geomagnetic activity, Dstmin (minimum Dst found within a storm event) for MCs (with a Pearson correlation coefficient, c.c. = 0.75, and the fitting function is Dstmin = 0.90+7.78Bzmin), but Bzmin for MCLs is not well correlated with the Dst index (c.c. = 0.56, and the fitting function is Dstmin = -9.40+ 4.58 Bzmin). (4) MCs play a major role in producing geomagnetic storms: the absolute value of the average Dstmin ( MC = -70 nT) for MCs associated geomagnetic storms is two times stronger than that for MCLs ( MCL = -35 nT), due to the difference in the IMF (interplanetary magnetic field) strength. (5) The SSN is not correlated with MCs ( TP, c.c. = 0.27), but is well associated with MCLs ( TP, c.c. = 0.85). Note that the c.c. for SSN vs. P2 is higher than that for SSN vs. P2. (6) Averages of IMF, solar wind speed, and density inside of the MCs are higher than those inside of the MCLs. (7) The average of MC duration (approx. = 18.82 hours) is approx. = 20 % longer than the average of MCL duration (approx. = 15.69 hours). (8) There are more MCs than MCLs in the quiet solar period, and more MCLs than MCs in the active solar period, probably due to the interaction between a MC and another significant interplanetary disturbance (including another MC) which could obviously change the character of a MC, but we speculate that some MCLs are no doubt due to other factors such as complex birth conditions at the Sun.

CME↗
NASA NTRS2019

Scenario Complexity for Unmanned Aircraft System Traffic

This work introduces an approach to estimate the complexity of a low-altitude air traffic scenario involving multiple UASs using mathematical programming. Given a set of multi-point UAS flight trajectories, vehicle dynamics, and a conflict resolution algorithm, an abstract model is developed such that it can be solved quickly using a mathematical programming optimization software without running high-fidelity simulations that can be computationally expensive and may not suit real-time apA quick and accurate assessment of complexity for a given traffic scenario can help plan and schedule flights to alleviate traffic bottleneck and mitigate operation risks, especially for unmanned aerial system traffic management where high traffic density or complexity is expected. This work introduces a traffic scenario complexity metric that was constructed based on the number of potential conflicts weighted by the conflict resolution cost associated. The cost associated with a conflict is calculated based on the corresponding conflict resolution maneuvers. To obtain the conflict resolution maneuvers, a MILP-based optimization was formulated with the vehicle model and conflict management parameters incorporated. To evaluate the complexity metrics, an approach of using measurements from high-fidelity simulations was proposed. The scenario complexity measurements for 920 random-generated scenarios were obtained through high-fidelity simulations and treated as the ground truth. Two statistics methods: Pearson and Alternative Conditional Expectations were applied for analysis. The results showed that the number of flights has low correlation with the scenario complexity according to the correlation coefficients calculated by both methods. The Alternative Conditional Expectations method shows that the proposed scenario complexity metric has better correlation with the ground truth than the number of potential conflicts.plications. In the abstract model, each vehicle is represented by a time-varied vector associated with position, speed, and heading information. The total extra distance that aircraft need to divert from their original routes to avoid collisions is computed and used to setup a quadratic programming formula. The metrics including the number of conflicts and extra distances travelled by all vehicles are then utilized to estimate the complexity of a given UAS flight scenario. Results and verification against high-fidelity simulations will be provided in the final draft.

traffic complexity↗
NASA NTRS1975

Application of LANDSAT to the Surveillance and Control of Lake Eutrophication in the Great Lakes Basin

The author has identified the following significant results. Computer techniques developed for mapping water quality parameters from LANDSAT data were demonstrated, using ground truth collected in an ongoing survey of water quality in Saginaw Bay (Lake Huron), Michigan. Regression correlation coefficients varied from 0.99 for total phosphorus to 0.72 for chlorophyll-a corrected. Five water quality parameters are best correlated with LANDSAT band 6 alone. Temperature relates to band 5 alone and only two bands are justified for mapping the remaining six parameters. LANDSAT CCTs were used as a basis for inventorying land use within each of the Ohio-Kentucky-Indiana regional commissions, 225 drainage areas, and nine counties. Sixteen categories were merged into ten categories and mapped at a scale of 1 in. = 5,000 ft, with detail to 0.44 hectares for the 2,700 sq mi region. These products were produced in less than 90 days at a cost of one cent an acre. LANDSAT CCTS were also used as a basis for inventorying land cover within the Triangle J council of governments 1,750 sq mi study area. The complete inventory was accomplished within a period of 60 days at a cost of about one half cent per acre.

Rogers, R. H.↗
NASA NTRS2023

Heliocentric Distance and Solar Activity Dependence of Sustained Quasi-radial Interplanetary Magnetic Field Occurrence

Planets close to their stars experience an interplanetary magnetic field (IMF) that is dominantly quasi-radial. Our solar system serves as a laboratory to study how the occurrence of quasi-radial IMF varies away from the star and under different stellar activities. Furthermore, on time and spatial scales relevant to magnetospheric physics, solar wind variability prevails in the form of structures generated both at the Sun and locally in the interplanetary space. The stationary Parker spiral model only approximates the large-scale structure of the IMF. Deviations from the Parker spiral often result in strongly radial magnetic fields that give rise to kinetic foreshock turbulence, which in turn can impact planetary magnetospheres. The relative significance of this type of interaction can be estimated statistically based on the occurrence rate of cases where the IMF is directed along the radial direction, leading to the entire day-side magnetosphere being downstream of the ion foreshock. We use observations covering radial distances from 0.1 to 10 au and more than 2 solar cycles to quantify the prevalence of radial IMFs throughout the heliosphere. Near Earth's orbit, it is found that the occurrence rates of quasi-radial and southward IMF orientations are similar, and that the Pearson correlation coefficient is ${{ \mathcal R }}_{{xy}}\sim -0.7$ calculated between quasi-radial IMF occurrence rate and solar activity. A negative correlation is demonstrated for radial distances extending to at least Mars but not to Saturn.

Brandon L Burkholder↗
DOE OSTI2023

Functional Predictor Variables for the Leaching Potential of Arsenic and Selenium from Coal Fly Ash

The release of leachates from intact coal ash impoundments is a concern due to the enrichment and mobilization of toxic elements such as arsenic (As) and selenium (Se). This study aims to explore the intrinsic properties of coal fly ash that correlate with the relative leachability of As and Se. We performed leaching experiments with 52 fly ash samples collected from 15 different U.S. power plants and representing coal feedstocks from the three major domestic regions. We assessed the mobilization potential of As and Se in fly ash based on standardized leaching protocols and performed multivariate and lasso regression analyses to explore correlations of leachable As and Se contents with characteristics such as major element contents, loss on ignition, and pH. The results of regression models indicated that major elements (Fe, Ca, and Al) for a wide range of fly ashes can serve as predictor variables for the leaching potential of As but not for Se. LOI and pH were not important predictive variables in the models. Both regression approaches resulted in relatively strong fits for leachable As (correlation coefficient R 2 = 0.78 for both models) compared to models for leachable Se (R 2 = 0.49). Overall, these results suggest that correlation models combined with on-site elemental analysis with portable analyzers may enable a screening method for leachable As in coal ash.

01 COAL, LIGNITE, AND PEAT↗
DOE OSTI2025

Characterization of Arsenic and Selenium in Coal Fly Ash to Improve Evaluations for Disposal and Reuse Potential (Final Technical Report)

Coal fly ash is a high volume waste material that is discarded in landfills and surface water impoundments across the U.S. and is also widely recycled for a variety of applications. The leaching of potential of contaminants of concern, such as arsenic (As) and selenium (Se), is often the driver of risk assessments for coal ash disposal and reuse. The extent of leachable As and Se depends on several factors related to environmental conditions and fly ash characteristics. Previous studies employed various methods to delineate the concentration, chemical form, and distribution of As and Se in fly ash materials. However, few studies have attempted to directly correlate these properties to mobilization parameters relevant to disposal and reuse. Instead, the coal residuals industries often rely upon standardized leaching protocols that can be laborious or involve hazardous chemicals. The goals of the project were to: 1) Develop and evaluate a characterization protocol that can be used to screen fly ash samples for leachability of As and Se; 2) Characterize As, Se, and associated constituents of fly ash particles at multiple length scales (nanometer to micrometer) to determine if elemental associations differ as a function of the resolution of characterization; and 3) Establish a predictive model for the chemical composition of coal ash produced annually at major U.S. coal fired power facilities on 50-year national coal supply records. For the first objective, we performed leaching experiments with 52 fly ash samples collected from 15 different U.S. power plants and representing coal feedstocks from the three major domestic coal regions. For this work, we assessed the mobilization potential of As and Se in fly ash based on standardized leaching protocols and performed multivariate and lasso regression analyses to explore correlations of leachable As and Se contents with characteristics such as major element contents, loss on ignition (LOI) and pH. The results of regression models indicated that major elements (Fe, Ca, Al) for a wide range of fly ashes can serve as predictor variables for the leaching potential of As, but not for Se. LOI and pH were not important predictive variables in the models. Both regression approaches resulted in relatively strong fits for leachable As (correlation coefficient R 2 = 0.78 for both models) compared to models for leachable Se (R 2 = 0.49). Overall, these results suggest that correlation models combined with on-site elemental analysis with portable analyzers may enable a screening method for leachable As in coal ash. For the second objective, we utilized nanoscale 2-D imaging (30-50 nm spot size) with the Hard X-ray Nanoprobe (HXN) in combination with microprobe X-ray capabilities (~5 µm resolution) to determine As and Se elemental associations in fly ash particles. Speciation of As and Se was also measured at the nano- to microscale with X-ray absorption spectroscopy. The enhanced resolution of HXN showed As and Se that were diffusely located around or comingled with Ca- and Fe-rich particles. The results also showed nanoparticles of Se attached to the surface of fly ash grains. Overall, a comparison of As and Se species across scales highlights the heterogeneity and complexity of chemical associations for these trace elements of concern in coal fly ash. For the final objective, we developed a predictive model for major element composition of coal ash in reserve at disposal sites of major U.S. coal fired power plants. This model was constructed from coal purchase records of 705 power stations from 1973-2022 and was trained on coal ash composition data showing that coal ash elemental composition is strongly associated with the source of feedstock coal. The model showed regional shifts in the major element contents of ash produced by power plants in the last 50 years, particularly for calcium and iron (expressed as %CaO and %Fe 2 O 3 ), as coal-fired power stations changed their source of coal over this time frame. Our approach enables an estimation of coal ash chemical composition that is stored in waste impoundments at individual power stations. Such information can help delineate the regional market potential for material applications that would utilize coal ash harvested from disposal sites across the U.S.

01 COAL, LIGNITE, AND PEAT↗
NASA NTRS1982

A procedure for testing the quality of LANDSAT atmospheric correction algorithms

There are two basic methods for testing the quality of an algorithm to minimize atmospheric effects on LANDSAT imagery: (1) test the results a posteriori, using ground truth or control points; (2) use a method based on image data plus estimation of additional ground and/or atmospheric parameters. A procedure based on the second method is described. In order to select the parameters, initially the image contrast is examined for a series of parameter combinations. The contrast improves for better corrections. In addition the correlation coefficient between two subimages, taken at different times, of the same scene is used for parameter's selection. The regions to be correlated should not have changed considerably in time. A few examples using this proposed procedure are presented.

Dias, L. A. V.↗
NASA NTRS2022

Quantifying urban, industrial, and background changes in NO2 during the COVID-19 lockdown period based on TROPOMI satellite observations

The COVID-19 lockdown had a large impact on anthropogenic emissions of air pollutants and particularly on nitrogen dioxide (NO 2 ). While the overall NO 2 decline over some large cities is well-established, understanding the details remains a challenge since multiple source categories contribute. In this study, a new method of isolation of three components (background NO 2 , NO 2 from urban sources, and NO 2 from industrial point sources) is applied to estimate the impact of the COVID-19 lockdown on each of them. The approach is based on fitting satellite data by a statistical model with empirical plume dispersion functions driven by a meteorological reanalysis. Population density and surface elevation data as well as coordinates of industrial sources were used in the analysis. The tropospheric NO 2 vertical column density (VCD) values measured by the Tropospheric Monitoring Instrument (TROPOMI) on board the Sentinel-5 Precursor over 261 urban areas for the period from 16 March to 15 June 2020 were compared with the average VCD values for the same period in 2018 and 2019. While the background NO 2 component remained almost unchanged, the urban NO 2 component declined by −18 % to −28 % over most regions. India, South America, and a part of Europe (particularly, Italy, France, and Spain) demonstrated a −40 % to −50 % urban emission decline. In contrast, the decline over urban areas in China, where the lockdown was over during the analysed period, was, on average, only -4.4± 8 %. Emissions from large industrial sources in the analysed urban areas varied greatly from region to region from -4.8 ± 6 % for China to -40 ± 10 % for India. Estimated changes in urban emissions are correlated with changes in Google mobility data (the correlation coefficient is 0.62) confirming that changes in traffic were one of the key elements in the decline in urban NO2 emissions. No correlation was found between changes in background NO2 and Google mobility data. On the global scale, the background and urban components were remarkably stable in 2018, 2019, and 2021, with averages of all analysed areas all being within ±2.5 % and suggesting that there were no substantial drifts or shifts in TROPOMI data. The 2020 data are clearly an outlier: in 2020, the mean background component for all analysed areas (without China) was -6.0 % ± 1.2 % and the mean urban component was -26.7 ± 2.6 % or 20σ below the baseline level from the other years.

COVID-19↗