Observational Study of Surface O 3 , NOx , CH 4 and Total NMHCs at Kannur , India

This paper presents the results of measurements of the surface ozone (O3), oxides of nitrogen (NOx), methane (CH4) and total non-methane hydrocarbons (TNMHCs) in a rural coastal location at Kannur (11.9°N, 75.4°E, 5 m asl), India from November 2009 to December 2011. The diurnal cycle for surface O3 had a peak in the afternoon and declined during nighttime. The maximum and minimum mixing ratio of surface O3 was observed in winter and monsoon seasons respectively. NOx concentration was high during mid-night to early morning and low during noontime. The diurnal variations of mixing ratios for NOx and O3 were anti-correlated. Monthly average maximum (2.26 ± 0.44 ppmv) and minimum (0.43 ± 0.19 ppmv) CH4 concentrations were observed in December and August respectively. The diurnal variations of CH4 were similar to that of NOx. A CH4 buildup was observed during early morning hours throughout the observational period. On an annual basis, the maximum, minimum and average total NMHCs were (25.45 ± 6.58 ppbv), (13.84 ± 4.31 ppbv), and (19.23 ± 5.56 ppbv) respectively at the observational site. Analysis of O3, NO, NO2, CH4 and NMHC have been carried out and the correlation between O3 and its precursors is discussed detailed. Further, the diurnal variation of O3 over a free tropospheric region at Ooty, a hill station lying in the Western Ghats region of south India on clear sky days in February 2011 is also reported for a comparison.


INTRODUCTION
The concentrations of trace gases are low in the atmosphere, but they play a vital role in determining the ambient air quality over a location.Besides, trace gas chemistry can modulate radiative transfer by which it can affect the climate.In addition, the secondary species produced from pollutant trace gases impart environmental impacts.Ozone (O 3 ) produced on the surface level is such a secondary species which serves as an ideal tracer to investigate the chemistry of trace gases.Moreover, (O 3 ), produced on the ground is a secondary pollutant that influences human health and agricultural crop yield to a larger extent.O 3 is produced in the troposphere when methane (CH 4 ), non-methane hydrocarbons (NMHCs) and carbon monoxide (CO) are photo-chemically oxidized in the presence of nitrogen oxides (NO x ).The diurnal cycle of O 3 formation and destruction is driven by NO x and Volatile Organic Compounds (VOC) emissions (Peleg et al., 1997;Ryerson et al., 2001) and solar radiation.To retrieve O 3 chemistry, it is necessary to estimate the concentration and reactivity of the different VOCs involved in O 3 formation.Since surface O 3 is produced by the dissociation of NO 2 in the presence of sun light, the ratio of NO 2 /NO concentrations in the atmosphere is an indicator of O 3 production.However, the abundance of VOC further modifies O 3 chemistry more complex.Thus the O 3 production efficiency depends on the concentration of VOC and NO x in a locality.Subsequently, the ratio of organic compounds like NMHC and NO x indicates the influence of VOC on O 3 production at a site (Saito et al., 2002).Now a day, satellite based observations are available for atmospheric trace gases monitoring all over the globe.However, ground based observations are highly significant to validate these space based measurements to study the chemistry, transport and modeling of these green house gases (Wang et al., 2009).Surface O 3 in the atmosphere has been monitored at various sites around the globe for the past several decades (Oltmans, 1981;Volz and Kley, 1988).Many studies have been carried out on the photochemical characteristics of O 3 and its precursor gases in the United States (Aneja et al., 1991;Qin et al., 2004;Alvarez et al., 2008;Yarwood et al., 2008), Europe (Zerefos et al., 2002;Mazzeo et al., 2005;Bossioli et al., 2007;Ordóñez et al., 2007;Solberg et al., 2008), and Asia (Pudasainee et al., 2006;Zhu et al., 2006;Bonasoni et al., 2008;Shan et al., 2008;Moore and Semple, 2009;Yin et al., 2010;Han et al., 2011).In the free troposphere region over Northern Hemisphere, increasing ozone concentration has been recently reported based on aircraft and ozonesonde data (Proberaj et al., 2009).
In the Indian sub-continent, various groups have carried out long-term measurements of surface O 3 and precursors (Ghude et al., 2008;Kumar et al., 2010;Reddy et al., 2010;Singla et al., 2011;Mahapatra et al., 2012;Ojha et al., 2012;Sharma et al., 2013;Swamy et al., 2013) and the results were quite promising.Available observations and modeling studies in India show higher O 3 levels during winter and summer seasons extending until May.Such variations are due to high precursor concentrations and the availability of ample solar radiation.Nair et al., (2002) found that sea breeze and land breeze have a pronounced influence on the production and transport of O 3 at Thumba, a tropical coastal site in the south Indian city of Trivandrum.Measurements of surface O 3 , NO x and meteorological parameters at Kannur, another coastal city in south India by Nishanth et al. (2012b) reported that O 3 and NO x showed distinct diurnal and seasonal variabilities.The observations conducted during 2009 to 2010 revealed that O 3 production at Kannur is quite sensitive to the organic emissions.This study presents an account of the variability of surface O 3 and its prominent precursor NO x in the wake of measured concentrations of methane and total non-methane hydrocarbons at this site.

Location of the Study Area and General Meteorology
Kannur district in the State of Kerala, India lies between latitudes 11°40' to 12°48' North and longitudes74°52' to 76°07' east and covers an area of 2,996 km².Kannur can be geographically divided into highland, midland and lowland regions.Highlands are the mountainous region forming part of the Western Ghats and are covered by rainforests and plantations.The midland region comprises of undulating hills and valleys and the lowland is the narrow stretch containing rivers, deltas and the coastal region.The observations were carried out at the Kannur University Campus (KUC) (11.9°N, 75.4°E), situated 15 km north from Kannur town along the coastal belt of the Arabian Sea.This site is close to a National Highway (NH 17) and the Arabian Sea and is at a height of 5 m above mean sea level.Geographical location of observational site situated at Kannur University Campus, and the land use pattern of Kannur and its surroundings is shown in the Fig. 1.
In Kannur, the four major seasons are winter (December-February), summer (March-May), monsoon (June-August) and post-monsoon (September-November).The meteorological parameters such as wind speed, relative humidity, temperature and total rainfall experienced in the observational site were retrieved from the local automatic weather station operated by the Meteorological and Oceanographic Satellite Data Archival Centre (MOSDAC), established by the Indian Space Research Organization (ISRO).The total solar radiation was measured using a pyranometer installed at the local weather station of MOSDAC and the data were taken at 30 min time intervals.Fig. 2 shows the monthly average variations of meteorological parameters like wind speed, relative humidity, solar radiation, temperature and rainfall in Kannur.Generally, the wind speed is quite high during the period from June to September and low from December to April.The maximum and minimum humidity was observed during monsoon months and winter months respectively.Solar radiation is lower

Measurement Techniques
The concentrations of O 3 and NO x were started to monitor from November 2009 and CH 4 and total NMHCs were started to measure from October 2010 at Kannur University Campus (KUC) with the aid of ground-based gas analyzers obtained from Environment SA, France.Earlier, using this observational setup we studied the influence of solar eclipse on surface ozone (Nishanth et al., 2011) and air pollution due to fire cracking episode following festivals (Nishanth et al., 2012a) over this location and the results were quite promising.O 3 present in the ambient air is measured by the analyzer (O 342 M), which is working on the principle of strong absorption of UV radiation at 253.7 nm by O 3 molecules.The analyzer incorporates corrections due to changes in temperature, pressure in the absorption cell and drift in the intensity of the UV lamp.Calibration of the analyzer is performed using built-in O 3 generator.The lowest detection limit of the analyzer is 0.4 ppbv with minimum response time of 30 s.The analyzer is provided with a built-in correction facility for temperature and pressure variations as well as the intensity fluctuation of the UV lamp.
The NO x analyzer (AC 32 M) is working on the principle of chemiluminescence effect produced by the oxidation of NO by O 3 molecules, which peaks at 630 nm radiation.In the present study NO 2 is measured by converting it into NO using the thermal conversion method using a heated molybdenum converter.The molybdenum converter is found to have higher sensitivity and 100% conversion efficiency (Winer et al., 1974;Finlayson-Pitts and Pitts, 1986).It has been realized that molybdenum converter also converts other species such as Peroxy Acetyl Nitrate (PAN), Nitric acid (HNO 3 ) etc; however PAN is thermally unstable at temperature above 30-35°C and its concentration is may be very small at the surface level (Naja and Lal, 2002;Nishanth et al., 2012b).Since NO x analyzer does not have a blue light converter to distinguish between NO 2 present in the ambient air and NO 2 produced in the molybdenum converter by the oxidation of PAN, HNO 3 etc.; NO x represents the total NO x measured at this site.The analyzers have been calibrated by using sample gases on a regular basis.Calibration procedure is based on the photometric assay of O 3 concentration in a dynamic flow system.The concentration of O 3 in the absorption cell of the analyzer is determined from the estimation of the intensity of 253.7 nm radiation absorbed by the sample.In practice, a stable O 3 generator is used to produce O 3 concentrations over the required range.Calibration of NO x analyzers is performed by using permeation tube oven.In this gas analyzer, the flow volume is controlled at the specified level so that the pressure in the reaction chamber is maintained at a constant value.Generally, a standard gas of NO 2 is commonly used to check the performance of the converter efficiency as described by Rehme (1976).
CH 4 and total NMHCs were measured using HC 51 M analyzer from Environment SA, France.Flame Ionization Detection is used as a versatile tool for monitoring hydrocarbons present in the ambient air.Hydrocarbon analyzer works on the principle of Flame Ionization Detection (FID) scheme by igniting hydrogen gas with ambient air to reveal the signature of organic species.Its great sensitivity, wide scope of linearity makes it an ideal detector for the measurement of hydrocarbons.Facilities like auto calibration, auto zeroing, automatic adjustment of change in meteorological parameter and data acquisition are built-in in the instrument.The analyzer has been calibrated by using sample gases on a regular basis.Further, for measuring the concentrations of C 2 -C 5 hydrocarbons present in the ambient air, samples were collected in evacuated borosil glass flasks of 0.5 L capacity which are fitted with needle valves both its ends.The air samples were collected at a height of 2 m above the ground level from different locations in Kannur by using a battery operated pump for a period of 15 minutes.

Diurnal Variation and Rate of Change of O 3 during Different Seasons
During the day, surface O 3 concentration starts to increase gradually after sun rise, attains maximum value throughout noontime and then decreases to low value during night and early morning hours.The daytime increase in O 3 concentration is due to active photochemistry in the presence of intense sun light.Variation in O 3 concentration is also influenced by boundary layer processes, regional emission, long range transport and meteorology.Generally, the boundary layer height increases after sunrise, reaching a maximum during noontime and descends after sunset (Mahalakshmi et al., 2011).During the day, O 3 are characterized by high concentration (16-49 ppbv) in the noontime and low mixing ratio (3-15 ppbv) in the late evening and early morning hours.The minimum O 3 mixing ratio of about (4.1 ± 1.2 ppbv) is noticed during the early morning around 06:00-07:00 h in all months of the year.Table 1 depicts the annual surface O 3 variation, monthly average daytime, nighttime O 3 mixing ratios and the observed maximum-minimum values at KUC from November 2009 to December 2011.From this table, it is evident that the highest (49.83 ppbv) O 3 mixing ratio was observed in December while lowest (16.41 ppbv) mixing ratio was observed in July.The daytime (08:00-20:00 h) average high O 3 mixing ratio observed in December (33.71 ± 5.29 ppbv) may due to the shallow boundary layer, low relative humidity and clear sky days that usually occurred in winter.Devara and Raj (1993), Praveena and Kunhikrishnan (2004) have observed that the Atmospheric Boundary Layer (ABL) height is low during monsoon and winter period over the Indian subcontinent.In monsoon, the strong prevailing winds disperse the pollutants quickly and get them washed out by precipitation.In contrast to this, both winds and ABL depth are low over Indian subcontinent by which the pollutants get trapped during winter (Alappattu et al., 2009).The daytime average low O 3 mixing ratio (10.81 ± 3.84 ppbv) observed in July was mainly due to the intense south-west monsoon which washes out pollution over this location.
Fig. 3(a) shows the diurnal variations of mean O 3 with 1σ standard deviation during different seasons and this reveals that O 3 is high in winter and low during monsoon seasons.During winter, maximum (44.20 ± 5.70 ppbv) and minimum (4.75 ± 1.75 ppbv) concentrations of O 3 were observed at 15:00 h and at 07:00 h respectively.During summer, maximum (33.77 ± 6.29 ppbv) mixing ratio of O 3 was observed at 15:30 h and minimum (4.73 ± 1.78 ppbv) was at 06:30 h.During monsoon season, the highest O 3 mixing ratio (18.49± 4.98 ppbv) was observed at 14:00 h and the lowest observed (2.74 ± 1.56 ppbv) at 07:30 h.In monsoon season, peroxy radicals (HO 2 and RO 2 ) are washed away by wet deposition limiting the gas phase photochemistry of O 3 (Seinfeld and Pandis, 2006).During post monsoon season, the maximum O 3 mixing ratio (29.32 ± 5.54 ppbv) was observed at 15:00 h and minimum of (3.71 ± 1.67 ppbv) at 07:00 h.During winter, the northeasterly wind brings dirty continental air from the northeast parts of Kannur and this could lead to the buildup of O 3 precursors over this location during winter.On the other hand, at all seasons except in winter months, air mass comes over the Arabian Sea brings clean marine air over this coastal site.The seasonal pattern showed O 3 concentrations starts decreasing from December to June; this could be due to (i) active rainfall occurs from June (SW monsoon) at this site and (ii) change in the wind  It is clear that the rate of change observed was high during winter and its magnitude was of the order of 9.5 ppbv/hr at 10:00 h and 10.5 ppbv/hr at 11:00 h.The lower boundary layer heights during winter time may trigger higher pollutant concentrations in the presence of cloud free sky conditions and low relative humidity may enhance a maximum change in the O 3 concentration (Reddy et al., 2012).The rate of decrease of O 3 was high during evening and was 11.6 ppbv/hr and 12.3 ppbv/hr at 18:00 h and 19:00 h respectively in winter months.The maximum [d(O 3 )/dt] in summer was 7.1 ppbv/hr at 11.00 h.The [d(O 3 )/dt] was almost zero at 14:00 h both winter and summer season and increased to 1.8 ppbv/hr for winter and 1.6 ppbv/hr for summer after which it became negative.The lapse rate [d(O 3 )/dt] showed sharp changes during 18:00 to 19:00 h which can be attributed to the faster O 3 titration during this time.During midnight [d(O 3 )/dt] was found to be almost steady.During midnight, O 3 chemistry becomes less intensive and surface deposition likely becomes the main process for O 3 loss.A large fraction of O 3 loss by surface deposition occurs through direct uptake by vegetation through the stomatal pores (Wesely, 1989;Wesely and Hicks, 2000).Environmental factors such as temperature, humidity and solar flux are known to have important impacts in regulating stomatal uptake; however, they also play a role in non-stomatal uptake as well.Surface wetness also appears to play a crucial role in O 3 deposition (Altimir et al., 2006); however, this effect can be variable depending on the ecosystem.During monsoon and post monsoon seasons [d (O 3 )/dt] showed very little change of 5.2 ppbv/hr and 5.5 ppbv/hr at 11:00 and 12:00 h respectively, which was the lowest of all the seasons.There was also a negative rate of change during late evening and it was found to be steady over night hours.This is due to the absence of photochemical processes and the possibility of washout effect.
Rate of change of O 3 during morning and evening at KUC and other observational sites in India are shown in Table 2 for comparison.The value of average rate of production of O 3 during morning as well as in evening hours was observed high at KUC than other rural, semi arid-rural, coastal and sub urban locations in India.Rate of change of O 3 during morning (08:00-11:00 h) was 4.9, 4.5, 4.6, 3.1 ppbv/hr at KUC, Joharapur, Gadanki and Tranquebar which are considered to be rural sites in south India.The annual average rates of change of O 3 during the evening hours (17:00 to 19:00 h) at KUC was estimated to be -6.4 ppbv/hr which is higher than other rural sites Joharapur (-3.3 ppbv/hr), Gadanki (-2.6 ppbv/hr) and Tranquebar (-2.8 ppbv/hr), which indicating KUC is more polluted than these two sites.Thumba, a tropical coastal site in south India showed a higher production rate in the morning than KUC.The average rates of production of O 3 during morning hours and lapse rate during evening hours are quite higher in the Indo Gangetic region (Pantnagar and Mohal).This suggests that the NO x emissions could be higher over Indo Gangetic region compared with other observational sites in India (Ojha et al., 2012).The rates of change of O 3 during the evening hours at urban sites Delhi and Ahmadabad were higher than the morning hours, which are attributed to the higher NO x concentration from commuter vehicular emissions, responsible for fast titration of O 3 in the evening.Usually, NO levels are observed to be lower at rural sites than at urban sites, which lead to slower ozone titration by NO during evening hours at a rural site (Naja and Lal, 2002).

Diurnal Variations of Oxides of Nitrogen
NO is a primary air pollutant whereas NO 2 is a secondary pollutant produced through a set of complex reactions, which determines ambient O 3 concentrations (Jenkin and Clemitshaw, 2000;Seinfeld and Pandis, 2006).Monthly average NO, NO 2 , and NO x concentrations measured from 2009 to 2011 with one sigma standard deviation are shown in Fig. 4(a).NO concentration was found to be low in day time, high at night and early morning hours during most of the observational period.The maximum and minimum concentrations of NO observed at KUC were (2.65 ± 0.46 ppbv) and (0.85 ± 0.24 ppbv) in the month of February.Similar to NO, diurnal variations of NO 2 showed minimum concentrations during noontime and it peaked during late night and early morning hours.Maximum (2.75 ± 0.48 ppbv) and minimum (0.72 ± 0.21 ppbv) concentrations of NO 2 were observed in December and August respectively.During nighttime and daytime, the variations in NO were quite similar to that of NO 2 .
Diurnal variation of NO x for different seasons at KUC is shown in Fig. 4 was high during late night hours and early morning and low during noontime.Seasonal average maximum NO x concentration was observed in winter and minimum in monsoon.Average NO x concentration in winter, summer, monsoon and in post monsoon were (3.36 ± 0.50 ppbv), (3.16 ± 0.4 ppbv), (2.9 ± 0.36 ppbv), (3.02 ± 0.44 ppbv) respectively.The concentration of NO x did not show significant variation during the period of observation at this site but only had a small seasonal variation.During the period of observation, the maximum (4.65 ± 0.76 ppbv) and minimum (1.83 ± 0.28 ppbv) concentrations of NO x mixing ratios were observed in January and February respectively.Near to the observational site, traffic starts to buildup during morning hours, which causes a peak pollutant level around 06:00 to 09:00 h.During these hours, NO x concentrations started to increase due to hydrocarbons from vehicles and their reactions to produce ozone.Variations in NO x are caused by variations in boundary layer mixing processes, chemistry, anthropogenic emissions and local surface wind patterns (Rao et al., 2002;Mazzeo et al., 2005;Shan et al., 2009).During noontime, the higher boundary layer provided a larger mixing region and hence diluted the pollutants.On the other hand, during evening hours, pollutant emissions from vehicles get trapped in the descending boundary layer.Moreover the boundary layer remains low during winter season (Beig et al., 2007).The maximum rate of change of O 3 occurred in December, where observed NO x concentration was high and the minimum rate of change of O 3 was in July, during which NO x concentration was the lowest.During winter season, the convective activities and turbulent mixing were weak which lead to a lower boundary layer height (Ramana et al., 2004).As a result, the concentration of pollutants near the surface increases.In addition to this, biomass burning on the eastern side of the observation site was quite intense during winter season causing transport of NO, NO 2 , CO, NMHC over to this location.

Relationship between Surface O 3 and Oxides of Nitrogen
During the entire period of observations, we found that the diurnal variations of NO x mixing ratios showed opposite cycles to that of O 3 and thus these two are anti-correlated.Several studies have revealed that morning production and evening-nighttime lapse rates of O 3 are proportional to the NO x concentration (David and Nair, 2011;Song et al., 2011).Photochemical production of O 3 takes place during daytime initiated by oxidation of its precursors.As has been discussed, the NO x plays a critical role in the photochemical formation of O 3 and has been found to be a limiting factor in the atmosphere at rural and remote locations (Finlayson-Pitts and Pitts, 2000).O 3 production during day time is driven by the photochemical reaction between hydroxyl radicals (OH), organic peroxy radicals and NO, while it is removed at night by dry deposition and destruction by alkenes and NO.The conversion of NO to NO 2 by O 3 during the night is the primary reaction that increases NO 2 at night, with the reverse occurring during the day to increase O 3 and decrease NO 2 (Mazzeo et al., 2005).In addition, the relatively low air temperature near the ground at night could prevent the vertical dispersion of NO x , contributing to its accumulation and resulting in higher night-time concentrations.Surface O 3 is formed and destroyed in a series of reactions involving NO and NO 2. The O 3 and NO x concentrations show inverse relationship due to titration of O 3 during the daytime.This is similar to what is observed in most other parts of the world (Chameides and walker, 1973;Stephens et al., 2008;Song et al., 2011;Sun et al., 2011).
Fig. 5(a) represents the variation of O 3 concentration with NO mixing ratio within a sample interval of 15 minutes.It is evident that O 3 concentration diminishes as NO mixing ratio increases, which may be due to the removal of O 3 by titration with NO.The most appropriate fit between O 3 and NO levels observed in two years time is exponential and similar to our earlier observation (Nishanth et al., 2012c) and the empirical relation is A fast decay in O 3 concentration was observed at high NO mixing ratios from late evening to early morning.The daytime variation between O 3 and NO 2 observed on clear sky days available during winter mornings is shown in Fig 5 (b).A linear fit with excellent correlation with a significance level (p < 0.01) that occurs on clear sky days in winter season reveals the active photochemistry of O 3 production, which starts early morning.
The empirical relationship observed between [O 3 ] and O 3 and NO x shows a very strong inverse correlation during the period of observations suggesting the possible VOC sensitive characteristics of the study location as described by Seinfeld and Pandis (2006) and Song et al., (2011).
To reveal the photochemical production of O 3 from NO 2 , a correlation between variations of [NO 2 ]/[NO] and O 3 is made and is presented in Fig. 6 (3) Based on the two year data, a good correlation of 0.77 between O 3 and [NO 2 ]/[NO] is found and the correlation coefficient has a significance level of p = 0.01.O 3 has a tendency to stabilize beyond 44 ppbv mixing ratio and this shows that at higher levels of [NO 2 ]/[NO], O 3 concentration is close to reaching a photostationary state (Shao et al., 2009).Interestingly, at KUC, where there are small local emissions of NO x , its ambient concentrations are generally much lower than other suburban sites in India.Similar analyses were carried out with one year data from 2009 to 2010 between O 3 and NO x for KUC (Nishanth et al., 2012b) and it was found that the correlation pattern remained unchanged even after projecting their variations with two year data which confirmed the variations of O 3 and NO x at this site.

Diurnal Variation of CH 4 at KUC
One year observation of O 3 and NO x realized that organic compounds present in the ambient air over Kannur have a significant impact on the production of ground level O 3 at Kannur (Nishanth et al., 2012b).Realizing the crucial roles in photochemical production of O 3 and their large local emission sources, measurements of CH 4 and total NMHCs at KUC have been carried out since October 2010.Monthly 0.19) ppmv CH 4 concentrations were observed in December average maximum (2.26 ± 0.44) ppmv and minimum (0.43 ± and in August.The diurnal variations of CH 4 were similar to that of NO x and we found that CH 4 showed a buildup during early morning hours from the nearby wet lands, paddy fields and river basin.CH 4 was observed to be very low during noon time and thereafter started to increase during evening hours for all months.During evening hours, increase in the emission of hydrocarbons from the vehicles and trapping in the boundary layer caused an increase in NMHC.CH 4 concentrations were low during the southwest monsoon seasons (June-August) during which the observational site experienced the highest rainfall of the year.Thus, there was also the possibility of wet deposition removal of organic pollutants.This scenario continued in post monsoon season as well.Diurnal variation of CH 4 for seasonal average basis at KUC is shown in Fig. 7(a).On a seasonal basis, maximum CH 4 concentration was observed in winter and minimum in monsoon.The rate of change of CH 4 was found to be high during winter and minimum during the monsoon season and late evening and night hours for all seasons.After attaining a higher value in the morning, there was a decrease in the rate of change in noon time in all seasons.There are many anthropogenic sources which account for high CH 4 levels observed in winter and summer seasons at KUC such as swampy habitats, shallow ditches and puddles including low lying rice cultivation areas along the bank of Valapattanam River, near the observational site.Both biological and physical processes control the exchange of CH 4 between rice paddy fields and the atmosphere.This may be one of the major reasons for the enhanced CH 4 observed during winter seasons.
Non-methane hydrocarbons (NMHCs) play an important role in the chemistry of the troposphere as precursors of O 3 and PAN.In polluted environments, the photo-oxidation of NMHCs in the presence of NO x lead to the formation of O 3 .The importance of NMHCs as O 3 precursor depends largely on their reactivities and ambient concentrations.Measurements of biogenically emitted VOCs such as isoprene suggest that these compounds contribute to high O 3 concentrations in urban areas.Major urban sources of NMHCs are releases from chemical industries, refinery operations, solvent evaporation and vehicular exhaust (Pandit et al., 1990;Liu et al., 2008).Seasonal variations in NMHCs concentrations are mainly influenced by the photochemical removal, source strengths, dilution due to atmospheric mixing of air parcels and transport from source regions to the sampling site (Klonecki et al., 2003).
Diurnal variation of CH 4 and total NMHCs as annual hourly average are shown in Fig. 7(b).Clear seasonal variabilities were identified for CH 4 and total NMHCS with distinct maxima during winter season and minima in monsoon season.On an annual basis, maximum, minimum and average CH 4 were (1.35 ± 0.39 ppmv), (0.83 ± 0.23 ppmv), (1.12 ± 0.29 ppmv) and total NMHCs were (25.45 ± 6.58 ppbv), (13.84 ± 4.31 ppbv), (19.23 ± 5.56 ppbv) respectively.Backward air mass trajectory analysis using HYSPLIT model (Nishanth et al., 2012b) and CGER-METEX reanalysis data (Nishanth et al., 2012c) during the period of observation revealed that the movement of air mass during winter months appear to originate from the Time (IST) east of this observational site (continental air) that brought pollutants nearby industrialized region to the observational site.Low mixing ratios of CH 4 and total NMHCs observed during monsoon season were mainly due to the reduction in atmospheric hydrocarbons because of the reduced photochemical reactions and the substantial reduction in solar intensity.Jobson et al., (1994) reported that alkane and acetylene hydrocarbons in total NMHCs displayed a winter maximum and summer minimum at a remote site in Canada.Hydrocarbon abundance was fairly well mixed over the winter time when NMHCs mixing ratios were much higher than during summer (Hakola et al., 2006).Seasonal variation in strong hydrocarbon source, differences in atmospheric behavior such as increased convection, vertical mixing in the summer and differences in air mass climatology with season, also play a vital role in the seasonal variation of hydrocarbon (Jobson et al., 1994).Hov et al. (1991) found a seasonal trend for ethene and propene with a late January maxima and minima during July-August.A recent study conducted in Ahmedabad (Lal et al., 2008) and Delhi (Pandit et al., 2011) revealed that total NMHC has a significant seasonal variation; with maximum concentrations in winter and low in late summer and in monsoon seasons.

C 2 -C 5 Hydrocarbons Measured in the Ambient Air
In order to retrieve the concentrations of C 2 -C 5 hydrocarbons in ambient air, air samples were collected at five different locations in Kannur so as to represent different characteristics, such as rural/semi urban, paddy field, forest area, riverbank and traffic junctions.Table 3  Acetylene (C 2 H 2 ) has common sources from combustion with mean lifetimes of about two weeks and it is removed from the atmosphere by reaction with the OH radicals.The air samples collected from KUC contained higher concentration of acetylene (139.84 ppbv) and that collected from the nearby river basin had a lower concentration of acetylene (4.08 ppbv).Surprisingly, air samples collected from all other locations did not exhibit the presence of acetylene.The enhanced concentration of acetylene observed during early morning at KUC is mainly due to its emission from the adjacent ceramic powder production unit which employs a high temperature furnace in which acetylene is used as gas to process the solid ceramic into its powder form.The furnace usually operates in the nighttime to get uninterrupted electric power in the unit.In the early morning, they used to flush out the gas for safety.But at the river basin, ) was the second most abundant organic trace gas in the background troposphere and it plays a crucial role in the chemistry of the lower atmosphere, chiefly through reactions with the OH radical.Ethane is primarily emitted to the atmosphere during mining, processing, transport and consumption of fossil fuels, during use of biofuels and during biomass burning.Further, it is mainly emitted from paddy fields where organic fertilizers are commonly used and has a seasonally varying lifetime (Blake and Rowland, 1986;Rudolph, 1995).In this analysis, the concentrations of ethane in the ambient air collected at traffic junctions (110.49ppbv) and paddy field (87.27 ppbv) and were below the detectable limit in the other air samples.The enhanced emission at traffic junction is due to the intense vehicular activities during the peak hour at 5 pm.
Presence of propylene (C 3 H 6 ) in the ambient air was found only at paddy field with a concentration of 2.5 ppbv in morning and propane (C 3 H 8 ) concentrations was observed merely at paddy field (9.54 ppbv), traffic junctions (17.30ppbv) and the existence of i-Butane was limited to paddy field alone in this investigation.N-Butane was found high (48.84ppbv) in the paddy field at 17:00 h and minimum (2.49ppbv) in KUC at 07:00 h.I-Pentane and n-Pentane were observed in all the air samples collected in different locations and it was found that concentration of i-pentane was found to be maximum (78.74 ppbv) at KUC and npentane was maximum (172.51 ppbv) in the paddy field.The quantification of these prominent organic species validates the abundance of VOCs that induce the production of surface O 3 in the presence of NO 2 .

Correlation between O 3 , CH 4 , NMHC and NMHC/NO x
A precise characterization of O 3 and its precursors are useful for the analysis of O 3 chemistry.The burning of agricultural crop waste emits precursor gases such CH 4 , NO 2 , CO and hydrocarbons in this location.Fig. 8(a) represents the variation of O 3 concentration with CH 4 mixing ratio during the morning periods of winter season at the observational site.An excellent linear fit (p < 0.01) revealed the active photochemistry of O 3 production associated with CH 4 .The empirical relationship is However, by noontime, with the active presence of sea breeze influencing the site from the Arabian Sea, major part of CH 4 is lost through reaction with OH as Non-methane hydrocarbons (NMHCs) are abundant in smoke resulting from biomass burning and have short atmospheric lifetimes.Total Nonmethane hydrocarbons (TNMHCs) in the atmosphere influence the climate through their production of organic aerosols and their involvement in O 3 photochemistry.Mean concentrations of TNMHCs were relatively higher in winter than summer and monsoon.Though there was not much seasonal variability in emission sources, the variation in TNMHCs concentration during different seasons can be attributed to the OH radical chemistry in the atmosphere.TNMHC/NO x ratio was higher during daytime hours, which was mainly due to the increased emissions of hydrocarbons at the site.Oxidation of hydrocarbons in the atmosphere by OH leads to alkyl peroxy radicals and hydroperoxy radicals.The impact on O 3 production is dependent not only ambient NO x , but more so on hydrocarbon to NO x ratio (TNMHC/NO x ).
From the Fig.    proposes that locally emitted TNMHCs and CH 4 at KUC has a major role in active photochemistry during winter season.

Comparitive Obervations of Surface O 3 at Ooty, a Free Tropospheric Region in Western Ghats
An attempt was made to study the diurnal variation of O 3 in a free tropospheric region at Ooty lying in the Western Ghats region on clear sky days in February 2011 in order to get a glimpse of O 3 variation along the east of Kannur.The location of observation site at Ooty is in the Nilagiris, the queen of hills at 11.3°N latitude, 74.4°E longitude and at an altitude of 2240 m above mean sea level which is the same latitude of Kannur.Ooty is located along east of Kannur with an air distance of 140 km.The geography of Ooty is quite unique owing to its location nestled amidst the district of the Nilgiris mountain range which is home to Doddabetta peak, considered as the highest peak in south India.Being a hillock region, nighttime is typically chilly in the months of January and February.Temperatures are relatively consistent throughout the year; with average high temperatures ranging from about 17-20°C and average low temperatures in the range 5-12°C with annual average precipitation of 1250 mm.Measurements of surface O 3 concentration at Doddabetta peak in Ooty were carried out on a few clear sky days in February 2011 and the diurnal variations of O 3 is shown in Fig. 9. Lower O 3 concentrations are observed during early morning and mid day time and they gradually increase to a maximum concentration till late evening and midnight hours.The minimum mixing ratio of O 3 is 26.92 ppbv and maximum of 64.06 ppbv O 3 concentrations are observed at 08:50 h and 19:00 h respectively.The average O 3 concentration during day time (08:00 to 17:00 h) is 48.12 ppbv and night time (1700-0800 h) is 52.35 ppbv.Since Ooty is situated at a higher altitude, the height of the boundary layer plays a significant role on the mixing and transport of O 3 .After the sun rise, solar heating warms the air in the valley producing warm upslope winds.Hence, air reaching the mountain top may have had a long path of travel and thereby O 3 loss can occur due to surface deposition.
After sunset, radiative cooling of the mountain surface cools the air adjacent to the surface, resulting in cold down slope winds.These winds bring air rich in O 3 from aloft.Hence O 3 concentrations gradually build up during late evening and mid night hours.The other tropical high altitude site Mt Abu (1680 m asl) also reported the strong effects of mountain induced wind flow patterns on the diurnal variations of O 3 (Naja et al., 2003).Another high altitude site at Mauna Loa (3397 m asl) also reported the same outline (Oltmans and Kohhyr, 1986).Recent observations Time (IST) near Mount Everest by Moore and Semple (2009) reveal the presence of high levels of O 3 at elevations from 5000-9000 m.They suggest that this mounting level of O 3 is due to the descent of O 3 rich stratospheric air as well as the long range transport of tropospheric pollutants from Southeast Asia.There may be a possibility of O 3 transport from the regions of Western Ghats to the coastal locations in winter months because of the north easterly winds found to prevail over Kannur during the winter season.

CONCLUSION
There have been several studies of surface O 3 and its precursors for different regions around the world and there is good evidence for an increase in the global surface level of O 3 in the past century.This work was focused on the study of variability of surface O 3 and its precursors (NO x, CH 4 and total NMHC) that influenced the air quality over Kannur a tropical site confined between the coastal belt of Arabian Sea and Western Ghats.Diurnal variations of surface O 3 and its precursor NO x were investigated since November 2009.Significant seasonal variations for O 3 and NO x mixing ratios were observed at this site.The daily average diurnal variation of O 3 showed a maximum mixing ratio in the late afternoon during all seasons due to its enhanced production by the photolysis of NO 2 and minimum during early morning over a period of two years from November 2009 to December 2011.The average seasonal variations of O 3 were observed to be maximum during winter and minimum during the monsoon period.During winter, both winds and atmospheric boundary layer depth are low and pollution hazards occurring during this season may contribute the observed high O 3 mixing ratios at this location.However, during summer, in spite of higher solar flux, O 3 concentration was low at this site.This may be due to cloud cover and the higher humidity near the Arabian Sea.The rate of production of O 3 during morning was found to be higher in December and minimum in July.The rate of decrease of O 3 during evening was found to be maximum in December and minimum in June.A good correlation was found between O 3 , NO and NO 2 which revealed the chemistry of O 3 production from NO 2 and titration with NO at this location.The net effect of NO x on O 3 concentrations was negative with a decaying exponential correlation indicating a possible VOC sensitive location, which showed the prominent role of abundant biogenic VOCs on the production of O 3 even at lower levels of NO 2 .These observations were valuable in understanding the ambient air quality at a location with a high population density but in an industrially weak area.There is a need for a detailed chemistry transport model for understanding various physical, chemical and dynamical processes.The results of this study are preliminary and need confirmation with more detailed, continuous and systematic measurements of ozone and its precursors over this region to explore the O 3 chemistry to improve the understanding of the chemical and dynamical processes and transport over this location.This will be further ascertained through a precise analysis of both indoor and outdoor air quality in this area.

Fig. 1 .
Fig. 1.Geographical location of observational site situated at Kannur University Campus (black spot), and the land use pattern of Kannur and its surroundings.

Fig. 2 .
Fig. 2. Monthly average variations of (a) wind speed, relative humidity, solar radiation (b) temperature, rainfall during the period of study at KUC.

Fig. 3 .
Fig. 3. Diurnal variation of (a) surface O 3 in different season (b) rate of change of O 3 in different season from November 2009 to December 2011.

Fig. 4 .
Fig. 4. (a) Monthly average variations of NO, NO 2 , NO x (b) Diurnal variation of NO x for different seasons at KUC from November 2009 to December 2011.

Fig. 5 .
Fig. 5. Scatter plot showing the variation of O 3 concentration with NO (b) correlation between O 3 and NO 2 in winter morning.
. A higher cluster of low O 3 values were found at lower values of [NO 2 ]/[NO] and only few data points representing higher O 3 values were found at higher values of [NO 2 ]/[NO].A logarithmic fit represents this behavior and an empirical relationship was obtained and this can be used to project the O 3 concentration at this site during daytime as [O 3 ] = 11.89ln([NO 2 ]/[NO]) + 18.30
8 (b), it is clear that surface O 3 and [TNMHC]/[NO x ] during daytime was linearly correlated with a correlation coefficient of 0.82.Non-methane hydrocarbons do not have secondary sources in the atmosphere; the atmospheric sources of these species are linked to the emission processes near the source regions.The correlation between CH 4 and NMHCs are useful for unfolding their chemistry.Fig 8(c) shows the linear correlation plot of TNMHCs and CH 4 with a correlation coefficient of 0.71 (p < 0.01) during winter season.This high correlation coefficient

Fig. 9 .
Fig. 9.Diurnal variation of surface O 3 at Ooty observed on clear sky days.

Table 1 .
Monthly daytime and night average, observed maximum and minimum O 3 , and the total days of observations from 2009 November to 2011 December.

Table 2 .
Rate of change of O 3 concentration at KUC and other sites in India.

Table 3 .
Concentrations of non-methane hydrocarbon in the ambient air at KUC and other locations in Kannur.