Peroxy radicals and ozone photochemistry in air masses undergoing long-range transport

A. E. Parker, P. S. Monks, M. J. Jacob, S. A. Penkett, A. C. Lewis, D. J. Stewart, L. K. Whalley, J. Methven, and A. Stohl Department of Chemistry, University of Leicester, Leicester, LE1 7RH, UK School of Environmental Sciences, University of East Anglia, Norwich, UK Department of Chemistry, University of York, York, UK School of Chemistry, National Centre for Atmospheric Science (NCAS), University of Leeds, Leeds, LS2 9JT, UK Department of Meteorology, University of Reading, UK Norsk Institutt for Luftforskning, P.O. Box 100, 2027 Kjeller, Norway


Figures Back Close Full Screen / Esc
Printer-friendly Version Interactive Discussion signatures, air-masses of biomass burning origin, and those that originated from the east coast of the United States.Enhanced peroxy radical concentrations have been observed within this range of air-masses indicating that long-range transported airmasses traversing the Atlantic show significant photochemical activity.The net ozone production at clear sky limit is in general negative, and as such the summer mid-Atlantic troposphere is at limit net ozone destructive.However, there is clear evidence of positive ozone production even at clear sky limit within air masses undergoing long-range transport, and during ITOP especially between 5 and 5.5 km, which in the main corresponds to a flight that extensively sampled air with a biomass burning signature.Ozone production was NO x limited throughout ITOP, as evidenced by a good correlation (r 2 =0.72) between P(O 3 ) and NO.Strong positive net ozone production has also been seen in varying source signature air-masses undergoing long-range transport, including but not limited to low-level export events, and export from the east coast of the United States.

Introduction
In recent years the importance of long-range transport of pollutants and precursor species to the composition of the troposphere remote to the source region has been recognised (Fehsenfeld et al., 1996;Penkett et al., 1998).Trace gases and aerosols can be transported over intercontinental distances (Duncan and Bey, 2004;Derwent et al., 2004;Stohl and Trickl, 1999;Wild and Akimoto, 2001).The relevance of ozone and its precursors to atmospheric pollution is well known, and progress has been made in the measurements of many trace gases associated with ozone, although there remain key uncertainties in the budgets of these species.Ozone has a lifetime of weeks to several months in the troposphere (Liu et al., 1987) and can be transported great distances in that time, whilst the emission of NO x and VOCs can lead to further ozone production downstream of the source with subsequent impact on the formulation of local air quality budgets (e.g., Li et al., 2002).Peroxy radicals (HO 2 +Σ i R i O 2 ) are key intermediates and chain carriers in the photochemical cycling of ozone in the troposphere (Monks, 2005).Peroxy radicals are formed via the oxidation of anthropogenic and biogenic species in the atmosphere such as CO, CH 4 and other organic compounds.Ozone is produced via the peroxy radical catalysed oxidation of NO to NO 2 and subsequent photolysis of NO 2 , whilst ozone can also be destroyed through reaction with HO 2 .Owing to the short lifetime of peroxy radicals (HO 2 has a lifetime on the order of a minute in clean air, much less than a minute in polluted air - Monks, 2005), they give a good indication in combination with NO of in-situ photochemical ozone production and loss.In addition, the selfand cross-reactions of peroxy radicals to form peroxides are a major sink for HO 2 and OH (Reeves and Penkett, 2003).The measurement of peroxy radicals thus remains of central importance in atmospheric chemistry, and rapid progress has been made in recent years with many deployments utilising various techniques for both ground and airborne studies (Mihele and Hastie, 2003;Cantrell et al., 2003a;Salisbury et al., 2001;Sommariva et al., 2004;Edwards et al., 2003;Green et al., 2003;Mihelcic et al., 2003;Introduction Conclusions References Tables Figures
The Intercontinental Transport of Ozone and Precursors (ITOP-UK) campaign took place during July/August 2004 and constituted the UK contribution to the International Consortium for Atmospheric Research on Transport and Transformation (ICARTT).The UK Meteorological Office/Natural Environment Research Council British Aerospace 146-300 atmospheric research aircraft was deployed from Horta Airport, Faial, Azores (38.58 • N, 28.72 • W) and flew a total of 12 science flights over a 22-day period.A full description of the flights and instrument payload of the BAe 146 has been given elsewhere (Fehsenfeld et al., 2006) whilst an overview of the ITOP campaign is given in Lewis et al. (Lewis et al., 2006).This paper details the measurements of the sum of peroxy radicals (HO 2 +Σ i R i O 2 ) using the PEroxy Radical Chemical Amplification (PERCA) technique.The data have been analysed to investigate the oxidant production in air masses undergoing long-range transport.

The PERCA instrument
The Chemical Amplification technique was introduced by Cantrell in the early 1980s (Cantrell and Stedman, 1982;Cantrell et al., 1984) and has been widely deployed since then (Cantrell et al., 1993;Mihele and Hastie, 2003;Monks et al., 1998;Zanis et al., 2000;Green et al., 2006), although there is only one previous publication of aircraft measurements using the technique (Green et al., 2003).The PERCA technique utilises the radical catalysed conversion of NO and CO into NO 2 and CO 2 respectively via addition of NO (3 ppmv) and CO (6% v/v) to the inlet region.NO 2 is subsequently detected via aqueous luminol (5-amino-2,3-dihydro-1,4-pthalazinedione) solution chemiluminescence at λ=424 nm with an improved LMA-3 detector as described by Green et al. (Green et al., 2006).
where CL is the chain length, i.e. the number of HO 2 /OH conversion cycles that occur before termination.A significant background NO 2 signal is also observed from other sources such as the reaction of ozone with the reagent NO.Assuming a chain length of 100 (i.e. each radical molecule produces 100 molecules of NO 2 ) and a radical mixing ratio of 20 pptv, the radical chain cycle would produce 2 ppbv of NO 2 .Under polluted conditions, ambient ozone could contribute up to 100 ppbv of NO 2 .Consequently, it is necessary to periodically measure only the background NO 2 produced by means other than peroxy radical conversion.This is achieved by injecting CO downstream of the NO injection point.OH produced as a result of Reaction (3) cannot be recycled into HO 2 as Reactions (4) and (5) do not take place, and instead some OH reacts with NO in a chain termination step (Reaction 7), whilst remaining radicals are lost to the walls of the inlet.
A flow of inert gas (nitrogen) is added in place of the CO so that in amplification mode NO and CO are injected upstream of N 2 , and in background mode NO and N 2 are injected upstream of CO.This ensures that the properties of the sample gas flow remain unchanged in both operation modes.It also helps reduce pressure pulsing in the detected signal and allows the flows to settle again more quickly after switching.The sensitivity of the PERCA instrument to humidity is well known (Mihele and Hastie, 1998;Mihele et al., 1999) and consequently a water correction as per Salisbury et al. (Salisbury et al., 2002) has been applied to all data in this study.This correction is relatively small out of the boundary layer as humidity is generally low.The average Introduction

Conclusions References
Tables Figures

Back Close
Full Screen / Esc

Printer-friendly Version
Interactive Discussion peroxy radical value increases from 18-29 pptv under 2.5 km with the average radical value over 2.5 km only increasing from 35-41 pptv when the correction is applied.

Aircraft Inlet System
In the previous aircraft deployment of a PERCA instrument (Green et al., 2003), a single inlet was connected to the air sample pipe on the former UK Meteorological Office Hercules C-130, whereas the instrument deployed during ITOP was a dual-channel instrument with the inlets sampling directly through the aircraft wall (for full details see Green et al., 2006).The dual-channel inlets directly sampling ambient air avoids two problems previously present in the single-inlet air sample pipe set-up, those of rapidly changing background signal and losses down the air sample pipe.
In an environment with rapidly changing background concentrations, as is the case for aircraft measurements, the background signal changes between two concurrent background cycles and can mask the amplified signal such that determination of the peroxy radical derived signal becomes less accurate.Cantrell et al. (Cantrell et al., 1996) thus developed a dual-channel PERCA where by use of two independent sampling and measurements systems operating out of phase both background and amplified signal are measured simultaneously and constantly.This gives the dual-channel configuration the ability to reduce the influence of atmospheric variability in background signals.Peroxy radical mixing ratios obtained by the dual-channel technique also have a higher temporal resolution than those from a single-channel, which requires the averaging of two background and one amplification period.Dual-channel measurements allow peroxy radical data to be taken on a 1 Hz timescale, although the data are then averaged to one minute.A more in depth discussion of the advantages of a dualchannel system is described elsewhere (Green et al., 2006).Introduction

Conclusions References
Tables Figures

Back Close
Full Screen / Esc Printer-friendly Version Interactive Discussion

Calibration
Ordinarily the LMA-3 detectors are calibrated by mixing the output from a VICI Metronics NO 2 wafer permeation device with varying flows of zero air, but owing to technical problems with the calibration system on-board the BAe-146, it was not possible to perform in-flight calibrations.An alternative method was developed for the BAe-146 by using the background signal on the LMA-3 detectors along with the in-situ ozone measurements made on-board the BAe-146 with a commercial Thermo Environmental Instruments Inc. 49C ozone analyser.The relative levels of ozone and NO 2 encountered during ITOP were such that the background signals measured on the LMA-3 detectors were almost entirely due to the oxidation of reagent NO by ozone with a negligible contribution from ambient NO 2 .Thus the NO 2 sensitivity of the detectors can be calculated for every point of the flight during which ozone measurements are available.
The precision of the ozone calibration method has been evaluated by taking the standard deviation of the ratio of ozone concentration divided by the background signal as a percentage of the mean of the same for a series of straight and level runs within a flight.The precision was determined to be around 6%.
It was also not possible to perform in-flight chain length calibrations, and consequently calibrations were performed on the ground only.It would be ideal to perform chain length calibrations at different altitudes, as the chain chemistry reactions in the inlets have temperature and pressure dependent rate constants.However, a series of in-flight calibrations were performed by (Green et al., 2003), where no clear chain length dependence on altitude was seen.All reported chain lengths were within error for chain length calculation from the chain length calibrations carried out on the ground.As such, although inflight chain length calibrations would be ideal, to a reasonable approximation using ground calibrations is acceptable.Introduction

Conclusions References
Tables Figures

Back Close
Full

Radicals and tracers
The average altitude profile of peroxy radicals measured during ITOP is shown in Fig. 1a, the data being binned into 500 m intervals.The full data set is also shown, coloured by ozone mixing ratio in parts per billion by volume.Figure 1b shows the temporal profile of median peroxy radicals over all altitudes.The campaign data set for other important species for the analysis in this paper (such as ozone and NO) can be found in the ITOP overview paper (Lewis et al., 2006).It should be noted that this groups together data from different latitudes and times of day, and thus one bin may be influenced heavily by a single air mass as will be discussed later.
The overall peroxy radical distribution increases with altitude from an average of 28 pptv at 0-0.5 km to 44 pptv at 5-5.5 km.There are very few points over 5.5 km owing to the inability of the installed pumping system to maintain a constant flow through the inlets at the lower ambient pressures found at higher attitudes and thus points over 5.5 km are not considered.Ordinarily the sample flow through the inlets is maintained at a constant rate for all altitudes.Any points where the flow rate is not constant have been removed.
Previous aircraft campaigns that have measured peroxy radicals in the lower/mid troposphere have focussed on the Pacific (e.g.Transport and Chemical Evolution over the Pacific (TRACE-P) campaign - Cantrell et al., 2003b) or over the United States (e.g.Tropospheric Ozone Production about the Spring Equinox (TOPSE) campaign - Cantrell et al., 2003a) with some of the earliest measurements over Europe (Reiner et al., 1997).Previous Atlantic campaigns have concentrated on the upper troposphere (e.g.Subsonic Assessment Ozone and Nitrogen Oxide Experiment (SONEX) - Jaegle et al., 2000), as opposed to the lower altitudes probed in this study.(Cantrell et al., 2003b) reported that measured values of HO 2 +Σ i R i O 2 during TRACE-P showed a small peroxy radical maximum at 2-5 km.A slight increase with altitude was also seen during TOPSE (Cantrell et al., 2003a).During ITOP an increase with altitude is Introduction

Conclusions References
Tables Figures

Back Close
Full seen to a maximum at 3-5.5 km.However, it is important to note that as the objective of ITOP was to intercept polluted air masses and most of these interceptions happened out of the boundary layer, the profile presented here for all flights is not necessarily that of the background mid-Atlantic.In order to distinguish the background Atlantic, a basic classification of air masses has been performed.Air masses have been divided into three groups comprising marine background air, air of Alaskan fire plume origin and all others, defined by ozone and carbon monoxide mixing ratios.Air with ozone and carbon monoxide mixing ratios of less than 40 ppbv and 90 ppbv respectively is designated marine, and air with a carbon monoxide mixing ratio of greater than 250 ppbv is designated of Alaskan fire plume origin.The levels of peroxy radical conditions found in differing air masses during ITOP are detailed in Table 1.Air characterised with a clean marine origin had a median peroxy radical mixing ratio of 16 pptv and an average mixing ratio of 19 pptv -less than half that of non-marine air. Figure 2 shows the altitude profile of marine background peroxy radicals, from which it can be seen that there is no discernable altitude trend.
An apparent anomaly in the altitude profile of peroxy radicals in marine air shows a distinct peak at 3.5-4 km.The air sampled at this altitude was almost exclusively from flight B036 on 29 July 2004.Flexpart (Stohl et al., 2005) emission tracers for the section of interest of flight B036 show very low concentrations of aged American emissions, and back trajectories (Methven et al., 2006) show air of Atlantic background as shown in Fig. 3.
The air masses classified as Alaskan fire plume origin show no increase in peroxy radicals over standard air masses but a minor decrease in both mean and median of 41 pptv to 36 pptv and 42 pptv to 37 pptv respectively.However, there are only 35 data points available for Alaskan fire plume air masses as the majority of the air masses associated with Alaskan fire plumes encountered were at high altitude where the PERCA had difficulty sampling as noted previously and it is therefore difficult to draw concrete conclusions.However, the fact that peroxy radicals were not elevated in Alaskan fire plume air-masses over the levels found in other non-marine air-masses whilst the me-  (Dlugokencky et al., 2003).CO and methane make a significant contribution as is expected, however the additional importance of non-methane hydrocarbons can be seen from Fig. 4. The decrease in OH reactivity attributed to methane with altitude even though the mixing ratio of methane is set to 1750 ppbv is due to the temperature dependence of the OH+CH 4 rate constant.The combined non-methane hydrocarbons contribute more to OH reactivity than methane and carbon monoxide combined, and are especially important under 3.5 km.From the measured non-methane hydrocarbons, it is the oxygenated VOCs that make the main contribution to OH loss as indicated in Fig. 5, with the most significant of those being formaldehyde and acetaldehyde.Indeed, acetaldehyde alone is responsible for almost as much OH reactivity as carbon monoxide and methane together.However, recent studies have demonstrated that acetaldehyde artefacts may be present in measured data leading to erroneously high concentrations of acetaldehyde being reported (Apel et al., 2003;Northway et al., 2004).It is worth noting that the artefact reported (Apel et al., 2003;Northway et al., 2004) is correlated with ozone, but acetaldehyde data measured during ITOP does not correlate with ozone.Furthermore, the veracity of the acetaldehyde data can be checked by comparing the concentration of peroxyacetyl radicals produced Introduction

Conclusions References
Tables Figures

Back Close
Full from reaction of OH with acetaldehyde to that produced from PAN thermal decomposition.Figure 6 shows that at altitudes up to approximately 5 km, there is no clear bias in production of peroxyacetyl radicals.Over approximately 5 km, the production from acetaldehyde oxidation begins to heavily outweigh that from PAN thermolysis.Consequently therefore, whilst acetaldehyde measurements should be treated with caution, it serves to illustrate the potential importance of oxygenated VOCs.
A steady state analysis of the production and loss of peroxy radicals during ITOP has been carried out using the method described by Mihele and Hastie (Mihele and Hastie, 2003).In this analysis it is assumed that peroxy radicals are in steady state, thus where is the rate of production of radicals, L SR is the rate of loss of peroxy radicals owing to self-reaction, and L NO x is the rate of loss of peroxy radicals owing to reaction with NO x .Assuming the major radical source is ozone photolysis and introducing γ as an additional fractional radical production, the radical production term becomes: Combining self-reaction rates of HO 2 and Σ i R i O 2 into a single rate constant k self , setting β=L NO x /L SR so as to describe the dominant loss mechanism and introducing parameter α to describe the partitioning between HO 2 and RO 2 , thus: gives the radical loss terms as:

Conclusions References
Tables Figures

Back Close
Full where the loss due to self-reaction is given by the following equations, which leads to the following self-reaction loss rate, Assuming that k 8 ≈k 9 , and that k 10 is slow enough that Reaction (10) can be discarded, Eq. (D) reduces to that found in Mihele and Hastie (Mihele and Hastie, 2003): However, the assumption in the derivation of the self-reaction loss rate that k 8 ≈k 9 only holds if RO 2 is solely in the form of CH 3 O 2 , neglecting the contribution of other species.The OH reactivity calculations earlier in this work have demonstrated that this assumption does not hold for the conditions encountered during ITOP where nonmethane hydrocarbons and especially acetaldehyde are very important in terms of OH reactivity.
If the acetylperoxy radical is introduced along with the ratio of then the self-reaction loss rate reduces thus: where If δ is set to 0.5, that is half of the RO 2 present is in the form of acetylperoxy (the remaining half being CH 3 O 2 ), then at 298 K, the combined peroxy radical self-reaction rate is approximately double the self-reaction rate when δ is zero.Further to this, the rate constant for the reaction of acetylperoxy with NO is approximately 2.6 times faster than that for methylperoxy with NO at 298 K. Consequently, the loss rate of peroxy radicals through reaction with NO x will also be greater than if all the RO 2 were CH 3 O 2 .
The importance of this relative to β and γ shall be shown in the following section.
Assuming that all RO 2 is in the form of CH 3 O 2 , the peroxy radical steady state can thus be re-written as where f is a measure of the proportion of O( 1 D) produced that consequently react with water vapour and is given by Eq. (H), where k 16 and k 17 are the reaction rates for Reactions ( 16) and ( 17) respectively.
The term j (O 1 D) is the photolysis rate of ozone to produce excited oxygen atoms (Reaction 18), and here is modelled data from the Tropospheric Ultraviolet and Visible radiation model (TUV) version 4 (Madronich and Flocke, 1998).Filter radiometers 18806 Introduction

Conclusions References
Tables Figures

Back Close
Full Screen / Esc Printer-friendly Version Interactive Discussion (Phillips, 2002;Volz-Thomas et al., 1996;Junkerman et al., 1989) to measure j (O 1 D) and j (NO 2 ) (the photolysis rate of Reaction 21) were fitted to the aircraft during ITOP, but unfortunately owing to technical problems with the data logging the data produced was unusable.Consequently, all photolysis frequencies should be considered clear sky maxima.The impact of this is dependent on whether measurements were being made above or below clouds.
Equation (G) can be rearranged to give the following expression for peroxy radical concentrations, β is therefore a measure of the dominant loss process for peroxy radicals.If β is less than one, then the dominant loss process is radical self-reaction rather than loss through reaction with NO x .Figure 7 is a time series of β for flights B031 to B034 and B036 to B039.B029 and B030 are omitted owing to a lack of NO data, and B035 because of a lack of PAN data.For the time series as a whole, β is greater than one for just over 35% of points, demonstrating that even in the low NO x conditions encountered during ITOP the loss of peroxy radicals owing to reaction with NO x is important.If acetylperoxy is considered in addition to methylperoxy with δ=0.5, then β is greater than one for just over 26% of points whilst the median value for β drops from 0.62 to 0.40, showing that whilst the peroxy radical self-reaction accounts for more peroxy radical loss, reaction with NO x remains at an appreciable level.The continued relevance of reaction with NO x is as the combined rate constant for [HO 2 +Σ i R i O 2 ]+NO is of a similar value to that for the peroxy radical self-reaction rate constant.Thus only low concentrations of NO x are needed for peroxy radical losses due to reaction with NO x to become significant.Introduction

Conclusions References
Tables Figures

Back Close
Full γ is a measure of additional peroxy radical production from sources other than ozone photolysis.Therefore a γ of zero indicates no excess production, whilst a γ of less than zero indicates that a higher concentration of peroxy radicals is calculated from ozone photolysis than is actually observed.Setting α equal to 0.5 (as HO 2 and Σ i R i O 2 were not measured discretely during ITOP) and δ equal to 0, and rearranging Eq. (I) results in an expression for gamma as per Mihele and Hastie (Mihele and Hastie, 2003), thus, As can be seen from Fig. 8 (γ+1 is shown so that the y-axis can be logarithmic), γ is less than zero for the ITOP time series for some 19% of points.This low γ could be due to either an overestimation of peroxy radical production from ozone photolysis or due to underestimation of peroxy radical losses.There is scope for overestimation of peroxy radical production in this work due to j (O 1 D) being from TUV modelled values rather than measured and thus a clear sky limit.Furthermore, γ is negative more often at lower altitudes, which would correspond to higher cloud cover at lower altitudes and thus a greater j (O 1 D) overestimation from TUV.There is also scope for an underestimation of peroxy radical losses both from self-reactions and reactions with NO x owing to the assumption that all RO 2 is present in the form of CH 3 O 2 .If δ is set to 0.5 such that half of the peroxy radicals present are CH 3 C(O)O 2 and half CH 3 O 2 and Eq.(J) is rewritten to take account of the expanded peroxy radical loss rates with the addition of acetylperoxy reactions, then γ is increased by a large amount over that given by Eq.
(J), with the median value for γ more than doubling.It should be noted that owing to the possible unreliability of acetaldehyde data, the use herein of acetylperoxy radicals as forming half the concentration of Σ i R i O 2 (with the remaining half being CH 3 O 2 ) is as a useful proxy for demonstrating the potential importance of fast reacting peroxy radicals.Introduction

Conclusions References
Tables Figures

Back Close
Full Screen / Esc

Printer-friendly Version
Interactive Discussion icals apart from ozone photolysis include alkene ozonolysis, PAN decomposition, and photolysis of carbonyls.During flight B032 Alaskan fire plumes were encountered with very high mixing ratios of PAN -the flight mean and median PAN mixing ratios were 1018 pptv and 785 pptv respectively, compared to a campaign mean and median of 316 pptv and 217 pptv respectively.NO was also enhanced over the campaign average, with the B032 median being 89 pptv compared to 21 pptv for the campaign, and the mean being 88 pptv compared to 36 pptv.

Overall photochemistry
The net photochemical ozone production or loss gives a measure of instantaneous insitu ozone production or loss within an air-mass and has been calculated in this work from measured peroxy radicals, thus, with k c representing the combined rate constant for oxidation of NO to NO 2 by Σ i R i O 2 and HO 2 (Reactions 1 and 3) and k 19 and k 20 the reaction rates for Reactions ( 19) and ( 20) respectively.
The first term in Eq. (K) is the ozone production term and is the only known way of producing ozone in the troposphere in excess of the photostationary steady state between ozone and NO x .NO 2 produced from the oxidation of NO by peroxy radicals in Reactions ( 1) and ( 3) is photolysed to produce oxygen atoms that then form ozone, viz,

Conclusions References
Tables Figures

Back Close
Full The other terms of Eq. (K) represent the loss processes of ozone in the troposphere, specifically the photolysis of ozone to form excited ozone atoms multiplied by f , the proportion of O( 1 D) that then react with water vapour, the reaction of ozone with OH (Reaction 19), and the reaction of ozone with HO 2 (Reaction 20).As stated previously, photolysis rates used in this paper are modelled values, and thus ozone loss rates should be considered clear sky maxima.The net ozone production or loss rate calculated by Eq. (K) contains a number of assumptions such as neglecting cloud processes, dry deposition, and reactions with alkenes and halogens.Nevertheless, Eq. (K) remains a good approximation for the free troposphere.
OH and HO 2 were not measured discretely during ITOP, and so assumptions have to be made about their concentrations.In this work α has been taken as 0.5 and the concentration of OH radicals has been taken at 1×10 6 molecules cm −3 .A sensitivity analysis with OH concentrations of 1×10 6 molecules cm −3 and 5×10 6 molecules cm −3 corresponding to a reasonable minimum and 90th percentile value respectively as derived from OH data recorded between ground level and 7 km on the NASA DC-8 during INTEX-NA demonstrated that the ozone loss term is relatively insensitive to OH concentration as increasing the OH concentration from 1×10 6 molecules cm −3 to 5×10 6 molecules cm −3 results in an increase of average L CS (O 3 ) of 0.04 ppbv h −1 from 0.51 ppbv h −1 to 0.55 ppbv h −1 .
Photochemical activity as characterised by enhanced peroxy radicals was observed within long-range transport air-masses with young east-coast North American air masses, biomass burning, and aged Asian emission signatures.Enhanced peroxy radicals (and photochemical ozone production) were also seen within low-level transported air masses, a larger than normal number of which occurred during the summer of 2004.A specific example of a low-level export event is presented later in this paper.
A time series of P(O 3 ), L CS (O 3 ) and N(O 3 ) for all flights is shown in Fig. 9.  3 shows net ozone production in different air masses defined as for peroxy radicals previously.It is interesting to note that the air designated of marine background has a net ozone loss rate of almost twice that of the remainder of the air.The same proviso is true for Alaskan Fire plume designated air masses with N(O 3 ) as previously with [HO 2 +Σ i R i O 2 ]; as there are only 30 points available for the entire campaign, it is not possible to draw any conclusions about conditions in these air masses.
In general, the summertime mid-latitude Atlantic is net ozone destructive at clear sky limit.This is true of air classified as both marine background air and non-marine background air.There were however many individual long-range transport events that showed enhanced ozone production, an example of one of which is presented later as a case study.A peak in overall net ozone production can be seen in the altitude profile within the 5-5.5 km band.This increase is driven by an increase in ozone production rate, rather than a reduction in loss rate.This altitude band is biased towards a single flight that is 48% of points are from flight B032 on 20 July 2004 and of the points with positive N(O 3 ), over 70% are from B032.During flight B032 long periods of biomass burning influenced air masses were encountered, with air masses characterised by high CO mixing ratios of consistently above 100 ppbv with peaks over 400 ppbv, enhanced NO mixing ratios with the median value being 88.7 pptv compared to 20.8 pptv campaign average, and enhanced PAN mixing ratios with a flight median of 785 pptv as opposed to the campaign median of 217 pptv.In contrast the median peroxy radical mixing ratio for flight B032 was 36.6 pptv compared to the campaign median of 34.4 pptv.The large enhancement in P(O 3 ) for flight B032 is thus driven by NO rather than peroxy radicals.
The P (O 3 ) being driven by NO is consistent for the campaign as a whole.Ozone production rate and NO are well correlated (r 2 =0.72) as shown in Fig. 11, and thus ozone production for the summer mid-Atlantic may be said to be NO x limited.Introduction

Conclusions References
Tables Figures

Back Close
Full In the period 14:45 to 15:05 LT a prolonged period of moderately to well aged air of American and Asian outflow was encountered with a biomass burning CO signature as shown by Flexpart analyses.The air mass was transported at low level, staying under 4.5 km for the majority of the time, as shown by back trajectories.The period 14:45 to 15:05 LT during flight B029 sees peroxy radicals of 30-40 pptv with higher peaks and enhanced ozone in addition to carbon monoxide as shown in Fig. 12.The Flexpart trajectories in Fig. 13a indicate air of mainly North American origin.However, the period directly following between 15:10 and 15:25 LT shows very low peroxy radical mixing ratios on the order of 5-10 pptv.Ozone is also at significantly lower mixing ratios during this period at 15-25 ppbv compared to the 70-80 ppbv of the preceding period.
The Flexpart trajectories for this period in Fig. 13b show air of Atlantic origin.This enhancement of peroxy radicals in an American origin air-mass is just one example of enhanced photochemical activity within low-level transport experienced during ITOP, a hitherto little characterised form of long-range transport.However, some case studies of low level export events seen over the Azores in July 2003 are presented in Owen et al. (Owen et al., 2006).A lack of NO measurements means that it is unfortunately not possible to calculate ozone production rates for this flight.) is over 0.8 ppbv h −1 , and is at least net neutral at clear sky limit.These two periods of ozone production are interspersed by a time of Atlantic origin air in which there is net ozone destruction, as demonstrated by the back trajectories in Fig. 15.This is clear evidence for ozone production in long-range transported air masses and indicates the impact that long-range transported air masses can have on regions remote from the source and the consequent implications for local pollution controls.The time period covered here also includes a lagrangian matched air-mass (case 5 in Methven et al. (Methven et al., 2006).

Conclusions
The sum of peroxy radicals has been measured with the Peroxy Radical Chemical Amplification technique onboard the FAAM BAe Systems 146-300 aircraft during the ITOP campaign in the summer of 2004 over the mid-Atlantic.The overall peroxy radical altitude profile displays an increase with altitude, but as the aim of ITOP was to target polluted air-masses this should not be taken as a summer-time mid-tropospheric mid-Atlantic background.The peroxy radical altitude profile for air classified as of marine origin (ozone and carbon monoxide mixing ratios of less than 40 ppbv and 90 ppbv respectively) shows no discernable altitude profile.A range of air-masses were intercepted with varying source signatures, including those with aged American and Asian signatures, air-masses of biomass burning origin, and those that originated from the 18813 Introduction

Conclusions References
Tables Figures

Back Close
Full east coast of the United States.Enhanced peroxy radical concentrations have been observed within this range of air-masses indicating that long-range transported airmasses traversing the Atlantic show significant photochemical activity.VOC reactivity was calculated as Σ i k i [VOC] i where k i is the rate of reaction of the relevant VOC with OH.The combined non-methane hydrocarbons contribute more to OH reactivity than methane and carbon monoxide combined, and are especially important under 3.5 km.Of the non-methane hydrocarbons, the oxygenated VOCs make the main contribution to OH loss, with the most significant of those being formaldehyde and acetaldehyde.Indeed, acetaldehyde alone is responsible for almost as much OH reactivity as carbon monoxide and methane together.However, owing to potential artefacts in the measured acetaldehyde, these data should be treated with some caution.The effect of acetylperoxy radicals on calculated β and γ values has shown that in environments where the dominant VOC is no longer methane, fast reacting peroxy radicals can have a large effect on any analysis carried out and cannot be ignored.
Instantaneous ozone production efficiencies have been calculated for all ITOP flights for which both peroxy radical and NO data are available.The net ozone production at clear sky limit is in general negative, and as such the summer mid-Atlantic troposphere is at limit net ozone destructive.However, there is clear evidence of positive ozone production even at clear sky limit within air masses undergoing long-range transport, and during ITOP especially between 5 and 5.5 km, which in the main corresponds to a flight that extensively sampled air with a biomass burning signature.Ozone production was NO x limited throughout ITOP, as evidenced by a good correlation (r 2 =0.72) between P(O 3 ) and NO.Strong positive net ozone production has also been seen in varying source signature air-masses undergoing long-range transport, including but not limited to low-level export events, and export from the east coast of the United States.[HO 2 + RO 2 ] /pptv Marine Air (O3 < 40 ppbv, CO < 90 ppbv)  Hantschz fluorescence UEA (Atkinson et al., 2006) d measurement techniques used to calculate VOC reactivity. 1 (Gerbig et al., 1999), o a constant 1750 ppbv, 3 rate for OH + 1,3-butadiene used.3 ) binned into differing air masses.
Correlation between ln(P(O 3 )) and ln(NO) Screen / EscPrinter-friendly Version Interactive Discussion Figure 10a, b and c show binned altitude profiles for P(O 3 ), L CS (O 3 ) and N(O 3 ) respectively, Fig. 10d shows median P(O 3 ), L CS (O 3 ) and N(O 3 ) split into 30 min time bins, with shown for α=0.25, 0.5 and 0.75.Table 15 July 2004 presented a good opportunity to sample American outflow.
Some flights allowed the opportunity to intercept long-range transported air-masses on multiple occasions.One such opportunity occurred on 31 July 2004 during flight B037.During this flight a prolonged period of US emissions of moderate age were encountered, as shown by Flexpart products.From the time series of ozone production, clear sky ozone loss and net ozone production in Fig.14, it can be seen that during the pe-00 to 13:20 LT there were multiple interceptions of US export interspersed with a period of Atlantic origin air.The first interception was between the times of approximately 11:00 and 11:30 LT, when a relatively young East Coast funnelled air mass was encountered showing enhanced ozone of up to 90 ppbv and CO exceeding 110 ppbv, peaking at 130 ppbv.During this period N(O 3 ) is enhanced to up to 0.5 ppbv h −1 even at clear sky limited L CS (O 3 ), clearly showing photochemical ozone production within air masses undergoing long-range transport.Another interception was made between 12:30 and 13:20 LT again showing strong photochemically active US emissions.Peak N(O 3

Figure 1a )
Figure 1a) -peroxy radicals binned by altitude and coloured by ozone levels.The box indicates 25 th percentile, median and 75 th percentile whilst the whiskers indicate 10 th and 90 th percentiles.Crosses indicate individual data points.Red triangles indicate mean CO binned by altitude and b) median peroxy radicals binned by time UTC averaged for all flights.

Fig. 1 .
Fig. 1.(a) peroxy radicals binned by altitude and coloured by ozone levels.The box indicates 25th percentile, median and 75th percentile whilst the whiskers indicate 10th and 90th percentiles.Crosses indicate individual data points.Red triangles indicate mean CO binned by altitude and (b) median peroxy radicals binned by time UTC averaged for all flights.
contributions of methane, carbon monoxide, and other non-methyl hydrocarbons to OH represent median values and error bars 25 th and 75 th percentiles.[X] is in molecules cm -3 .

Fig. 4 .Figure 4 -
Fig. 4. Relative contributions of methane, carbon monoxide, and other non-methyl hydrocarbons to OH reactivity.Points represent median values and error bars 25th and 75th percentiles.[X] is in molecules cm −3 .

Fig. 5 .
Fig. 5. (a) relative contributions of oxygenated and non-oxygenated C2-C4 non-methyl hydrocarbons to OH reactivity and (b) contribution of various oxygenates to OH reactivity.Points represent median values and error bars 25th and 75th percentiles.[X] is in molecules cm −3 .

Figure 10
Figure 10 -a) ozone production rate b) ozone loss rate c) net ozone production rate binned with altitude and d) median ozone production, loss and net production rates.In a, b, and c the box represents the 25 th , median and 75 th percentiles and the whiskers denote 10 th and 90 th percentiles.In d net ozone production is shown for α = 0.25, 0.5 and 0.75.

Fig. 10 .
Fig. 10.(a) ozone production rate (b) ozone loss rate (c) net ozone production rate binned with altitude and (d) median ozone production, loss and net production rates.In (a), (b), and (c) the box represents the 25th, median and 75th percentiles and the whiskers denote 10th and 90th percentiles.In (d) net ozone production is shown for α=0.25, 0.5 and 0.75.
concentrations of other species were greatly enhanced, e.g.median NO concentrations were enhanced by over a factor of 5, and median PAN concentrations were enhanced by more than a factor of 8, is of interest.Furthermore, within the Alaskan fire plume, airmasses with extremely high CO mixing ratios of up to 600 ppbv were observed with no associated enhancement of ozone.The non-enhancement of peroxy radicals within Alaskan fire plume air-masses is likely a result of the lack of enhanced ozone, combined with the low temperatures at high altitudes resulting in much NO 2 being sequestered into PAN or other reservoir species.A range of volatile organic compounds (VOCs) were also measured on the BAe-146.Compounds up to four carbon atoms as detailed in Table2have been used to calculate VOC reactivity as Σ i k i [VOC] i where k i is the rate of reaction of the relevant VOC with OH.Methane was not measured on board the BAe-146 and has been set to a constant 1750 ppbv dian