Methanol-CO correlations in Mexico City pollution outflow from aircraft and satellite during MILAGRO

Introduction Conclusions References


Introduction
Methanol (CH 3 OH) is one of the most abundant volatile organic compounds (VOCs).It is a significant source of tropospheric carbon monoxide (CO) and formaldehyde (CH 2 O) (Duncan et al., 2007;Millet et al., 2005;Hu et al., 2011).CH 3 OH can also have an important influence on air quality in and around polluted urban areas (Molina et al., 2010).CH 3 OH may also contribute to particulate pollution (Blando and Turpin, 2000).The first global detection of methanol was made by Singh et al. (1995).Globally the major sources of CH 3 OH are biogenic processes with ∼50 %-80 % of the estimated emissions involving plant growth (Harley et al., 2007;Karl et al., 2003) and, to a lesser extent, plant decay (Warneke et al., 1999).Other sources include biomass burning (Paton-Walsh et al., 2008;Dufour et al., 2006), anthropogenic sources from vehicles and industrial activities (de Gouw et al., 2005(de Gouw et al., , 2009)), as well as oxidation of methane and other volatile organic compounds (Tyndall et al., 2001;Madronich and Calvert, 1990).Although these other sources are much less important than the biogenic source on a global scale, they can be responsible for large CH 3 OH enhancements on regional or continental scales.The main sink of CH 3 OH is oxidation by OH (Heikes et al., 2002).Dry and wet deposition to the surface constitutes a minor sink (Karl et al., 2010).The resulting overall atmospheric lifetime of CH 3 OH is approximately 5-10 days (Jacob et al., 2005).
Estimates of the source strength, seasonality, and spatial distribution of CH 3 OH in the atmosphere have previously been obtained from field measurements (e.g.Heikes et al., 2002;Singh et al., 2000), chemical transport modeling in combination with ground and airborne measurements of CH 3 OH (e.g.Millet et al., 2008;Jacob et al., 2005;Tie et al., 2003;Galbally and Kirstine, 2002), and from remotely sensed measurements from ground-based infrared spectrometers (e.g.Rinsland et al., 2009; Introduction

Conclusions References
Tables Figures

Back Close
Full Satellite measurements offer the advantage of high spatial and temporal coverage, albeit with lower resolution than in situ observations.For example, CH 3 OH has been measured in biomass burning plumes in the upper troposphere from the Atmospheric Chemistry Experiment Fourier Transform Spectrometer (ACE-FTS), a limb-viewing instrument with relatively high vertical resolution and sensitivity to minor trace gases (Coheur et al., 2007;Dufour et al., 2006).More recently, it has been shown that nadir instruments can measure enhanced CH 3 OH in the lower troposphere.CH 3 OH retrievals have been performed using data from the Tropospheric Emission Spectrometer (TES) on the NASA Aura satellite (Cady-Pereira et al., 2012;Beer et al., 2008) and from the Infrared Atmospheric Sounding Instrument (IASI) onboard the polar-orbiting MetOp-A satellite (Razavi et al., 2011).In addition, TES and IASI observations have been employed to better constrain the seasonality of methanol emissions from northern midlatitude ecosystems (Wells et al., 2012) and to constrain biogenic and biomass burning emissions of methanol (Stavrakou et al., 2011).Both TES and IASI use the same spectral region (centered at 1033 cm −1 ) for CH 3 OH retrieval.TES has high spectral resolution of 0.06 cm −1 at nadir providing the ability to distinguish the target species from interferences, while the IASI instrument has larger spatial coverage providing global coverage twice per day due to the wide scans across its track.
Constraints on the importance of various CH 3 OH source types can be gained by examining the correlations between CH 3 OH and other measured species.The correlation between CH 3 OH and CO is of particular interest in source characterization (Warneke et al., 2007;de Gouw et al., 2005), and simultaneous measurements of CH 3 OH and CO are available from the satellite instruments listed above.CH 3 OH and CO share common anthropogenic and biomass burning sources.However, biogenic CH 3 OH emissions can modify the CH 3 OH-CO ratio to varying degrees depending on Introduction

Conclusions References
Tables Figures

Back Close
Full The Megacity Initiative: Local And Global Research Observations (MILAGRO) field campaign was conducted over the Mexico City Metropolitan Area (MCMA) and the Gulf of Mexico in March 2006, with the goal of examining the properties, evolution, and export of atmospheric emissions of trace gases and particles generated in the MCMA and to evaluate the regional and global impacts of these emissions (Molina et al., 2010;Singh et al., 2009).Major sources of pollutants in the MCMA include motor vehicles (Apel et al., 2010), pervasive incomplete combustion of fossil fuels including Liquid Petroleum Gas (LPG) for low-temperature household cooking and heating (Blake and Rowland, 1995), and biomass burning (Crounse et al., 2009;Yokelson et al., 2007).During MILAGRO, urban and regional measurements of a number of air pollutants were obtained by the National Center for Atmospheric Research (NCAR) C-130 and the National Aeronautics and Space Administration (NASA) DC-8 aircraft.In this analysis, we utilize MILAGRO aircraft measurements to determine correlations between CH 3 OH and CO, as well as between CH 3 OH and other trace gas pollutants, in order to qualitatively assess the influence of different sources on the CH 3 OH concentrations in the Mexico City outflow.
During MILAGRO TES also made a number of special observations over the MCMA.Retrievals of CH 3 OH and CO have been performed based on these observations.Direct comparisons of aircraft in situ profiles of CH 3 OH and CO with satellite retrievals during MILAGRO proved challenging as there were very few targeted profile underflights of the satellite track by either the C-130 or DC-8 aircraft over Mexico City, which limited sampling coincidence within reasonable spatial and temporal criteria.Instead, we have utilized CH 3 OH-CO correlations to evaluate the ability of the satellite data to capture the important source information offered by aircraft measurements, with the goal of assessing whether such space-borne data can be reliably applied to other regions and time periods where no aircraft measurements are available.In addition, in order to further explore the utility of CH 3 OH-CO correlations from TES, we have compared and contrasted satellite CH 3 OH-CO correlations from MILAGRO (dominated by anthropogenic emissions -e.g.Molina et al., 2010) with those over the Amazon Basin Introduction

Conclusions References
Tables Figures

Back Close
Full (dominated by biogenic emissions -e.g.Karl et al., 2007).Descriptions of the aircraft and TES satellite measurements are provided in Sect.2, and the analysis and results in Sect.3.

Aircraft measurements
Figure 1 shows the aircraft flight tracks (along with TES footprint locations) during MI-LAGRO.All the data used in this work were collected within 12 • N-30 • N and 102 • W-90 • W, covering the region downwind of Mexico City and the northern part of the Gulf of Mexico.The MCMA (19.43 • N, 99.12 • W) is located in the Valley of Mexico, a large basin 2.2 km a.s.l.The basin is surrounded on three sides by mountain ridges, with a broad opening to the north.The C-130 flight tracks are closer than those of the DC-8 to the urban area of Mexico City, though some of the flights were designed to fly over remote regions either to detect long-range plume transport (more than 1000 km from Mexico City) or to measure biomass burning plumes.The DC-8 flight tracks cover a larger geographical domain and the measurements are more representative of outflow at a distance from the sources.
Airborne measurements of CH 3 OH were made by the Total Organic Gas Analyzer (TOGA) (Apel et al., 2010) onboard the C-130, and the Peroxy Acetyl Nitrate/Aldehyde/Ketone Photo Ionization Detector (PANAK) (Singh et al., 2004) onboard the DC-8.The reported CH 3 OH uncertainty for both aircraft instruments is 20 % at 1σ level.However, intercomparisons of TOGA (C-130) and PANAK (DC-8) measurements during MILAGRO have shown that the CH 3 OH values from the C-130 are generally higher than those from the DC-8, by as much as a factor of 2. This suggests a possible inconsistency in calibration between the CH 3 OH measurements onboard the two aircraft (Kleb et al., 2011, more details at http://www-air.larc.nasa.gov/missions/intex-b/intexb-meas-comparison.htm).Introduction

Conclusions References
Tables Figures

Back Close
Full Volume mixing ratios of CO were measured by the Ultra-Violet Fluorescence (UVF) instrument on the C-130 (Holloway et al., 2000) with a quoted accuracy of 10 %, and by the Differential Absorption CO Measurement (DACOM) instrument (Sachse et al., 1987), a mid-IR diode laser spectrometer that measured CO, CH 4 , and N 2 O on the DC-8 aircraft.DACOM's quoted CO accuracy is 2 % or 2 ppb.A suite of simultaneous observations of other related hydrocarbons on the C-130 aircraft are also used in Sect. 3 for quantification of the source signature of CH 3 OH.All those species were measured with an estimated accuracy of ≤20 % (Singh et al., 2009).The MILAGRO data is supplied on a 1 min merge.However, the TOGA CH 3 OH measurements were actually made with 2.8 min temporal resolution.Therefore, in this analysis we have further merged the data to 3 min intervals for instances where CO differs by more than 10 ppb between time steps, but CH 3 OH values are constant.To reduce the influence of a small number of extreme outliers on the correlation analysis, we removed any data with CO concentrations greater than 500 ppb.These points were all located within 300 km of the city center and comprised around 3 % of the whole dataset.Any errors in the airborne CH 3 OH and CO measurements are assumed independent and not taken into account for the correlation analysis in Sect.3.

Satellite measurements from TES
TES is a Fourier transform spectrometer flying on the NASA Aura satellite.The instrument has high spectral resolution (0.06 cm −1 ) and a relatively small (5 × 8 km) nadir footprint.The TES instrument measures radiances in the spectral range 650 to 3050 cm −1 (Beer et al., 2001) that are very stable (Conner et al., 2011) with good radiometric calibration and signal-to-noise (Shephard et al., 2008).These instrument characteristics provide TES with the capability to provide information on the vertical profiles of temperature and numerous trace gases in the atmosphere.Profiles of temperature, water vapor (H 2 O and HDO), ozone (O 3 ), CO, methane (CH 4 ) and ammonia (NH 3 ) are currently produced routinely as operational products (Version 5).Other trace gas products, including CH 3 OH, are currently under development to be implemented 5711 Introduction

Conclusions References
Tables Figures

Back Close
Full operationally.CH 3 OH is scheduled to be processed routinely as part of the TES Version 6 algorithm release.TES conducted several special "step-and-stare" observations during the MILAGRO campaign.The TES nadir footprints in these observations are separated by ∼45 km along the Aura ground track.The Aura overpass times for Mexico City are 01:45 and 13:45 LT.
The first space-based nadir retrievals of CH 3 OH were reported by Beer et al. (2008) for a limited number of TES special observations over Beijing (Northeast China), and San Diego (California, USA).Since then, the TES CH 3 OH retrieval approach has been developed further and applied to larger volumes of data.The algorithm is based on an optimal estimation approach with a priori constraints.Details of the retrieval algorithm are reported in Cady-Pereira et al. (2012).The TES retrieval errors for CH 3 OH during MILAGRO range from 10 % to 50 %, with larger relative errors for small retrieved CH 3 OH values.The TES retrievals have been validated against vertical profiles from an ensemble of aircraft measurements over North America, where a 3-D chemical transport model was used as an intercomparison platform (Wells et al., 2012).
The TES operational CO product (Version 4) used in this analysis has been extensively validated against aircraft and other satellite measurements (Luo et al. 2007a(Luo et al. , 2007b;;Lopez et al., 2008), and showed a negative bias of <10 % in the lower and middle troposphere near Houston during INTEX-B and a positive bias of ∼5-10 % in the tropics.Here we assume no correlation between the TES measurement errors for CH 3 OH and CO as they are in retrieved in different spectral bands.Any measurement errors should thus have little impact on the derived CH 3 OH-CO enhancement ratios.
Figure 2 shows example averaging kernels for both CH 3 OH and CO.Both species have peak sensitivity in the 600-800 hPa region.The TES CO and CH 3 OH retrievals contain a limited amount of vertical information, with ∼1.4 degrees of freedom for signal (DOFS) for CO and typically <1 for CH 3 OH.The vertical resolution for TES CH 3 OH and CO retrievals (defined as the full width at half maximum of the averaging kernels) is about 5-6 km in the troposphere.Introduction

Conclusions References
Tables Figures

Back Close
Full For both species, the number of vertical levels used in the retrieval is significantly greater than the number of DOFS for the measurement.Performing the retrieval in this way allows variations in vertical sensitivity between different profiles to be characterized via the averaging kernels.However, it also means that the details of the result on any given retrieval level are highly sensitive to the shape of the chosen a priori profile.
In order to mitigate this issue, a post-processing step is performed to map the result from the relatively fine retrieval grid to a representation that better reflects the information available in the measurement.The exact details of the mapping depend on the sensitivity (averaging kernels) for each individual measurement.For CH 3 OH with ∼1 (or less) DOFS, we report a single "representative volume mixing ratio" (RVMR) value.
Further details of this approach can be found in Payne et al. (2009) and Shephard et al. (2011).The end result is that the RVMR value is less sensitive to assumptions about the a priori profile shape than a value at a single retrieval level would be.In order to assess the CH 3 OH-CO correlations, the CH 3 OH-sensitivity-based mapping was also applied to the CO retrieval in each case to produce a "pseudo CO RVMR".The same approach has been used to determine the emission ratio of ammonia and formic acid relative to CO using TES observations of boreal biomass plumes (Alvarado et al., 2011).In addition, an acceptance criterion of the CH 3 OH RVMR > 0.1 ppb was applied to exclude those retrievals with very weak CH 3 OH signals in the TES spectra.topography leading to a very efficient "air pump" exporting pollutants to the free troposphere (de Foy et al., 2006).

Source characterization of CH 3 OH in Mexico outflow
To examine the contributions of various source types to atmospheric CH 3 OH during MILAGRO, we examined a suite of gases indicative of anthropogenic, biogenic, and biomass burning sources (the presumed sources of CH 3 OH), and their correlations with CH 3 OH.These indicator tracers include CO, acetone ((CH 3 ) 2 CO), benzene (C 6 H 6 ), acetylene (C 2 H 2 ), hydrogen cyanide (HCN) and acetonitrile (CH 3 CN).CO is a general tracer for combustion sources (fossil fuel, biofuel, and biomass burning).Acetone can be a good tracer of biogenic sources due to its relatively long lifetime (Singh et al., 1995).There is also a large source of acetone from atmospheric oxidation of anthropogenic VOCs at northern midlatitudes (Fisher et al., 2012).Benzene can be regarded as an indicator of anthropogenic emissions involving solvent use, vehicle exhaust, and industrial processes (Karl et al., 2009).C 2 H 2 is a relatively inert tracer and comes mostly from automobile exhaust (Harley et al., 1992).HCN and CH 3 CN are tracers of biomass burning (Singh et al., 2010).Analysis of a subset of C-130 measurements near the MCMA during MILAGRO (Fig. 3) shows that CH 3 OH in this outflow region is correlated (1) strongly with CO, acetone and benzene, (2) moderately with C 2 H 2 , and (3) less with HCN and CH 3 CN, suggesting that the main sources of CH 3 OH during MILAGRO were anthropogenic and biogenic in nature.
In addition to primary anthropogenic and biogenic sources, photochemical formation could be a contributor to CH 3 OH during MILAGRO.As an oxygenated VOC, CH 3 OH is produced by reaction of the methylperoxy radical (CH 3 O 2 ) with itself and with other organic peroxy radicals (RO 2 ) (Tyndall et al., 2001;Madronich and Calvert, 1990).In general, reactions involving CH 3 O 2 and RO 2 are not expected to be a large source of CH 3 OH in urban environments under high NO x conditions (Molina et al., 2010).
However, the very active photochemical environment and the extremely high RO 2 concentrations in Mexico City could lead to reactions of CH 3 O 2 with other peroxy radicals, potentially leading to significant CH 3 OH production in this specific region.In addition, Jacob et al. (2005) previously inferred a larger photochemical CH 3 OH source than Introduction

Conclusions References
Tables Figures

Back Close
Full expected based on present understanding, though in the remote rather than urban polluted atmosphere.Any significant photochemical production of CH 3 OH during MI-LAGRO would have implications for the interpretation of CH 3 OH-CO correlations, as will be discussed in Sect.3.3.Figure 4c shows the corresponding variation in the TES CH 3 OH RVMR and pseudo CO RVMR.In contrast to the aircraft observations, the TES observations show no decreasing trend with distance downwind of the city.The limited spatial variability in TES CO for this area has been previously reported by Shim et al. (2009), who examined O 3 and CO from TES retrievals and MILAGRO aircraft measurements to characterize mega-city pollution outflow on a regional scale.They found that TES captures much of the spatial and day-to-day variability seen in the in situ data for O 3 , but not for CO.In particular, TES does not clearly resolve the CO pollution over the Mexico City Basin.We explored a number of different possible reasons for this lack of spatial variability in TES CO, including sampling issues, the vertical resolution of the TES measurements, and retrieval-related issues associated with the high altitude of Mexico City.Examination of these issues using 5 yr of TES Version 4 CO data over this region led to the conclusion that the handling of the TES Version 4 CO prior constraints over regions with high surface elevation is an important contributor to the inadequate spatial variability in TES CO during MILAGRO.Points within 300 km of the city center mostly have high surface elevation, while areas farther from the center are closer to sea level.For the Version 5 (and previous versions) TES retrievals, the CO constraints are not shifted in altitude for high surface elevation regions.This leads to low values in the constraint 5715 Introduction

Conclusions References
Tables Figures

Back Close
Full vector and tighter constraints from the prior than are desirable.Therefore, the TES CO values over Mexico City are generally constrained to lower values than they should be due to the a priori constraints.This issue will be considered in future TES algorithm updates.Note that this is less of an issue for the O 3 retrievals, which are far more loosely constrained in the upper troposphere than the CO retrievals.For the CH 3 OH retrievals the prior constraints are shifted vertically to account for variations in surface elevation (Cady-Pereira et al., 2012).
An additional contributor to the lack of spatial variability in CH 3 OH and CO during MILAGRO is the coarse vertical resolution of the measurements.In situ outflow profiles measured during MILAGRO show concentrations that peak in a relatively narrow range.
The TES CH 3 OH and CO retrievals are unable to distinguish between profiles with a sharp, strongly enhanced peak, and profiles where the trace gas enhancement has lower peak values but is spread over a wider vertical range.Therefore, TES does not reproduce the extreme high peak values observed by the aircraft, and would be expected to show lower average values than the plume-chasing aircraft, particularly in the region closest to the city center.However, this is not expected to bias the derived CH 3 OH-CO correlation.

Impact of vertical resolution of TES retrieval on the derived ∆CH 3 OH/∆CO ratio
In this study the CH 3 OH/CO enhancement ratio is defined as the slope of the regression between CH 3 OH and CO, and is denoted as ∆CH 3 OH/∆CO.The slope is calculated using the reduced major axis (RMA) method, which estimates the linear relationship between two variables by minimizing the residual variance in both x and y directions (Xiao et al., 2004;Hirsch and Gilroy, 1984).For the aircraft in-situ measurements, it is straightforward to derive ∆CH 3 OH/∆CO.For the TES observations, the ∆CH 3 OH/∆CO ratio is the regression slope between the CH 3 OH RVMR and the pseudo CO RVMR (see Sect. 2.2).Introduction

Conclusions References
Tables Figures

Back Close
Full The vertical resolution of the TES measurement should be taken into consideration for the interpretation of the retrieved CH 3 OH and CO and the associated spatial variability in Mexico City outflow.Simulated retrievals are utilized to examine whether the coarse vertical resolution of TES CH 3 OH and CO observations would be expected to affect the derived ∆CH 3 OH/∆CO ratios.
Starting with averaged C-130 aircraft profiles of CH 3 OH and CO near the Mexico City center (<300 km), we constructed a base profile for each species, which was then scaled by factors of 1.25, 1.5 and 1.75 to create four profiles in total.These four profiles (shown as dashed lines in Fig. 5a and b) were used as "truth" to generate four sets of simulated TES radiances.Realistic noise was included in the simulated radiances.
CH 3 OH retrievals were then performed on the simulated radiances.Simulated CO retrievals were generated from the assumed "true" profiles by applying the a priori and averaging kernel from a typical TES case with relatively strong sensitivity (from Version 4 of the TES operational algorithm).Retrieved profiles are shown as solid lines in Fig. 5a and b.Thus, for each set of CH 3 OH and CO retrieved profiles, we have a pair of CH 3 OH and CO RVMRs, resulting in the TES ∆CH 3 OH/∆CO ratio shown in Fig. 5d.Despite the coarse vertical resolution of the TES measurements, TES is able to recover the true ∆CH 3 OH/∆CO ratio when both CH 3 OH and CO retrievals show relatively high sensitivity.

Spatial variability in the ∆CH 3 OH/∆CO ratio
In order to explore the utility of CH 3 OH-CO correlations from TES observations, we evaluated the extent to which variability in the CH 3 OH/CO enhancement ratio is captured by TES based on in-situ aircraft observations during MILAGRO.The ∆CH 3 OH/∆CO ratio observed by TES reflects the bulk enhancements of CH 3 OH and CO in the Mexico City outflow, smoothed by the TES vertical sensitivity.For TES/aircraft comparisons of the ∆CH 3 OH/∆CO ratio, only aircraft data in the 600-800 hPa range, the region where the TES CH 3 OH and CO retrievals have the greatest sensitivity, are used.

Conclusions References
Tables Figures

Back Close
Full Figure 6a shows the CH 3 OH-CO correlations from the two aircraft.In contrast with the CH 3 OH and CO concentrations themselves, ∆CH 3 OH/∆CO ratios for air masses with significant source influence vary little with geographical location.Instead, slightly higher ratios are seen outside the 300 km range as compared to within 300 km around the city.This may be explained by photochemical production of CH 3 OH.The ∆CH 3 OH/∆CO ratios based on the C-130 aircraft measurements are 41-55 ppt ppb −1 , while the DC-8 observations show much lower values of 26-39 ppt ppb −1 .The different ∆CH 3 OH/∆CO ratios between the C-130 and DC-8 measurements appear to reflect a calibration offset (see Sect. 2.1) rather than the different sampling strategy for the two aircraft (Singh et al., 2009), since the discrepancy persists regardless of the distance from the MCMA (Fig. 6a).
Figure 6b shows the corresponding CH 3 OH-CO correlations as measured by TES.As with the aircraft observations (Fig. 6a), TES ∆CH 3 OH/∆CO ratios display little variation with distance from the city center (ranging from 18 to 24 ppt ppb −1 ), and are closer to the values derived from the DC-8 measurements during MILAGRO than to those derived from the C-130 measurements.Note that close to Mexico City center, we expect TES CO retrievals using the Version 4 algorithm to under-represent the true values due to the issues with the constraints discussed above.This particular issue should not be present in the TES CH 3 OH observations as our retrieval shifts the constraints.With more appropriate CO constraints, it is expected that the TES data will exhibit some spatial variation in the enhancement ratios, with slightly lower ratio values (increased CO values) closer to the city center, as seen in the aircraft data.
Any discrepancies between TES and aircraft ratios for the MILAGRO dataset may be due to aircraft instrument calibration issues, or to systematic errors in the TES results.Possible sources of systematic error in the TES retrievals include forward model errors (such as biases in spectroscopic line parameters or incorrect representation of interfering species), errors in the TES radiance calibration, and influence of the a priori assumptions in the retrieval (the effect of the a priori assumptions is mitigated, but not eliminated, by the RVMR representation).Given the present discrepancy between the Introduction

Conclusions References
Tables Figures

Back Close
Full aircraft measurements of CH 3 OH, it is difficult at this stage to use the aircraft observations as evaluation of the absolute TES enhancement ratios.
In contrast with Mexico City, a region dominated by strong anthropogenic emissions, we also examine the TES ∆CH 3 OH/∆CO ratios over the Amazon Basin, a region dominated by biogenic emissions (with an additional biomass burning source contribution) (Karl et al., 2007).Similar to the MILAGRO domain the Amazon region exhibits elevated CH 3 OH values, which provides a relatively high signal-to-noise ratio for the TES observations.Figure 7 shows TES ∆CH 3 OH/∆CO ratios over the Amazon Basin during August-September of 2005.The ratio of 42 ± 7 ppt ppb −1 is a factor of 2 higher than that observed near Mexico City during MILAGRO (18-24 ppt ppb −1 ).The significant ratio difference between Mexico City and the Amazon Basin indicates that the TES-derived ∆CH 3 OH/∆CO ratio has potential for differentiating source categories of CH 3 OH and other chemical species with multiple sources (e.g.CO 2 ).With future work, the TES-derived ratio could be applied globally to other regions and time periods where no in situ measurements are available.
It should be noted that connecting the observed ∆CH 3 OH/∆CO ratio from either aircraft or TES with the actual CH 3 OH/CO emission ratio from "urban" sources in Mexico City is complicated, since the ∆CH 3 OH/∆CO ratio could have already been enhanced due to photochemical production of CH 3 OH even in the plumes thought to be "fresh".Ground-based estimates of emission ratios tend to show lower values than the enhancement ratios measured aboard aircraft.For example, Bon et al. (2011) report CH 3 OH/CO emission ratios of 2.1 ± 0.5 ppt ppb −1 and 6.1 ± 2.1 ppt ppb −1 from boundary layer observations at urban sites during MILAGRO.The complicated relationship between emission ratios at the source and enhancement ratios in plumes does not apply just too urban sources.For instance, Holzinger et al. (2005) observed relatively high CH 3 OH and acetone enhancements in fire plumes over the Mediterranean, and concluded that secondary production of these species must have taken place.Caution is therefore needed when interpreting enhancement ratios as opposed to emission ratios.

Conclusions References
Tables Figures

Back Close
Full The ∆CH 3 OH/∆CO ratios derived in this analysis from the C-130 observations during MILAGRO (41-55 ppt ppb −1 ) are significantly higher than ratios from other aircraft studies in US cities (Table 1).Singh et al. (2010) derived a ratio of 21.0 ± 14.0 ppb ppb −1 in urban plumes mainly sampled over California.Warneke et al. (2007) derived a ratio of 9 ppt ppb −1 in fresh New York City and Boston plumes in July and August of 2004, and 8.4 ppt ppb −1 from aircraft observations over the Los Angeles Basin in the spring of 2002.The ratios from the C-130 measurements during MILAGRO are also much higher than the ratios of 4-11 ppt ppb −1 over Boulder, Colorado and Pittsburgh, Pennsylvania reported from ground-based winter measurements (Millet et al., 2005;Goldan et al., 1995).
To put these urban values into context, ∆CH 3 OH/∆CO ratios have also been reported for biomass burning plumes during the MILAGRO experiment.High ratios of 14-28 ppt ppb The aircraft ∆CH 3 OH/∆CO ratios from Mexico City are also much higher than those reported from biomass burning at northern mid/high latitudes.Simpson et al. (2011) and Singh et al. (2010) derived mean ratios of 9.6 ± 1.9 ppt ppb

Conclusions
The correlation between CH 3 OH and carbon monoxide (CO) is of interest for characterizing sources of CH 3 OH and other species.Coincident measurements of CH 3 OH and CO in the lower to middle troposphere from TES and aircraft observations during MILAGRO allowed us to assess the utility of satellite-based observations of the CH 3 OH-CO correlation.TES retrievals of CH 3 OH and CO show relatively high sensitivity in the 600-800 hPa range, where Mexico City pollution outflow peaks.In-situ aircraft observations of CH 3 OH and CO downwind of Mexico City center during MILAGRO in March 2006 were used to evaluate the extent to which variability in the ∆CH 3 OH/∆CO enhancement ratio is captured by TES special observations.During MILAGRO there are sharply peaked outflow profiles (from shallow plumes) that are particularly challenging for TES (or any nadir satellite measurements) to capture due to its vertical resolution (∼ 5 km).The analysis suggests that the TES operational CO retrieval algorithm (Version 5) maybe improved in regions of elevated topography (e.g.Mexico City) by updating algorithm CO constraints to better handle elevation changes.
∆CH 3 OH/∆CO ratios derived from the TES observations reflect bulk enhancements (smoothed according to the TES sensitivity) of CH 3 OH and CO in the Mexico City outflow.∆CH 3 OH/∆CO ratios derived from the TES observations (18-24 ppt ppb −1 ) are closer to those observed from DC-8 aircraft during MILAGRO (26-39 ppt ppb ratios for this analysis.However, the CH 3 OH/CO enhancement ratios derived from the TES and aircraft data during MILAGRO are both high relative to previous studies over US cities.This may be partly explained by photochemical production of CH 3 OH in fresh plumes as well as to the different types of fuel burned in the MCMA.TES derived ∆CH 3 OH/∆CO ratios show a significant difference between Mexico City and the Amazon Basin, with substantially higher ratios over the Amazon.Both regions feature strong CH 3 OH emissions: from anthropogenic sources in the first case and from biogenic sources in the second.This analysis shows that TES can clearly distinguish differences in the ∆CH 3 OH/∆CO ratio due to two different source categories of CH 3 OH, which indicates the potential of utilizing TES derived ratios globally as a diagnostic for emission sources in other regions and time periods.Introduction

Conclusions References
Tables Figures

Back Close
Full  Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Vertical profiles of CH 3 OH and CO mixing ratios from the DC-8 and C-130 during MI-LAGRO (not shown) indicate outflow mainly at 600-800 hPa near the city center, due to the high elevation of Mexico City.Downwind of the city center, enhanced CH 3 OH and CO were detected up to 600 hPa and down to the surface (∼1000 hPa), with the 5713 Discussion Paper | Discussion Paper | Discussion Paper | Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Figure4a, b shows the variation in CH 3 OH and CO concentrations at 600-800 hPa in the C-130 and DC-8 aircraft observations as a function of distance from the city center.The aircraft observations of both CH 3 OH and CO show a decreasing trend with increasing distance away from the urban source region, mainly due to dilution (mixing with background air) and chemical decay.
Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | −1 were reported byYokelson et al. (2009) based on aircraft measurements of fresh plumes from fires in the Yucatan region of Mexico.Comparable ratios of 19-32 ppt ppb −1 can be derived from the data presented by Yokelson et al. (2011), who reported emission factors for 25 open fires based on airborne measurements during March 2006.

Fig. 6 .
Fig. 6.CH 3 OH-CO correlations in Mexico City outflow during MILAGRO, binned by distance from Mexico City Center.Upper panels show aircraft measurements (black for C-130 and red for DC-8).Lower panels show TES measurements.

Table 1 .
The ∆CH 3 OH/∆CO ratio (in ppt ppb −1 ) from the aircraft and TES from this study versus those from the literature.