Study of surface ozone (O 3 ) and its relationship with NO, NO 2 , NO x , OX and CO at ve different locations in New Delhi, India from 2013 to 2019

In the present study, continuous measurements of Surface Ozone (O 3 ), Oxides of Nitrogen (NOx (NO + NO 2 )), and carbon monoxide (CO), monitored at ve different locations in Delhi National Capital Region have been studied for the period 2013–2019. The ve monitoring locations used are namely IMD Lodi Road, IGI Airport Palam, CV Raman Dheerpur, CRRI Mathura Road, and NCMRWF Noida. The average hourly concentration of O 3 , NO, NO 2 , CO, NOx (NO + NO 2 ), and OX(NO 2 + O 3 ) are found in the range of 32.44 ppb to 36.57 ppb, 19.46 to 28.09 ppb, 20.83 to 26.89 ppb,1.67 to 1.89 ppm,43.04 to 54.99 ppb, and 54.06 to 60.99 ppb respectively during the study period. Diurnal variation of NOx and CO Concentrations show higher values during the morning (0600-0900h) and late evening (1900-2400h) hours while the highest concentrations of ozone have been observed during afternoon hours. The relationship between NO, NO 2 , and surface O 3 as a function of NOx has also been examined during daylight hours (0500hrs IST to 1900 hrs IST) and chemical coupling of the three species i.e. NO, NO 2 and O 3 have been studied. The ground-level concentration of Ozone have been found to decrease with increasing NOx concentration during daytime. The variations in concentrations of oxidants (NO 2 + O 3 ) with the concentration of [NOx] have been studied to examine the contributing pollution sources of oxidants at all the study sites. The average rate of change of O 3 concentrations (dO 3 /dt) has been examined at all ve locations. The monthly and diurnal variation of oxidants [OX] at all the study locations has shown a strong positive correlation with temperature whereas a negative correlation with humidity.


Introduction
Atmospheric trace gases are one of the challenging environmental issues in urban and industrial areas in India. It has demanded an increased awareness among the research communities as greenhouse gases and trace gases both alter the energy balance of the climate system. For example, the radiative forcing for the tropospheric ozone (O 3 ) changes due to the emission of O 3 precursors (e.g. oxides of nitrogen, NOx, and carbon monoxide, CO) and has been found between + 0.25 to + 0.65 W m − 2 (with a mean value of + 0.35 W m − 2 ). [1] Tropospheric O 3 has two main sources, namely stratosphere intrusion affecting surface O 3 [2,3] and insitu production via photochemical oxidation of carbon-link compounds (i.e. CO, CH 4, and VOCs), in the presence of NOx. [3,4] Ozone production e ciency has been observed to depend on meteorological conditions such as solar radiation ux, wind speed, temperature and pressure. The daytime increase in ozone concentration is a pronounced feature of urban polluted sites, which is basically due to photooxidation of the precursor gases such as CO, CH4, NMHCs, etc. in the presence of su cient amounts of NOx. [5] It has been found that photochemical ozone production in urban areas is more sensitive to NOx emissions and is less sensitive to VOC emissions. [7] In areas having moderate air pollution, ozone sensitivity to the emission of nitrogen oxides depends on the meteorological conditions and the emission rates of nitrogen oxides. [6] Due to the chemical coupling of ozone and nitrogen oxides, any resultant reduction in the level of nitrogen dioxide is invariably accompanied by an increase in the air concentration of ozone. It has become important to understand the chemical pathways leading to the formation of ozone in the atmosphere to devise effective strategies to reduce surface ozone levels.
NOx plays a major role in the oxidizing capacity of the lower atmosphere due to its signi cant effects on the partitioning, formation and loss of free radical species (OH, HO 2 and RO 2 ). [7,8] CO also plays an important role in controlling the oxidizing capacity of the troposphere, by acting as a sink for a larger fraction of reactive hydroxyl (OH) radicals available in the lower atmosphere. [9] The oxidation begins with the reaction of Carbon monoxide (CO) released from various sources present in the lower troposphere with the hydroxyl radical ( • OH), which is abundantly present near the surface of the earth.
The radical intermediate formed by this reaction reacts rapidly with oxygen to give a peroxy radical HO 2 • .
Peroxy radicals then go on to react with NO to give NO 2 which undergoes photolysis to give atomic oxygen and through reaction with oxygen (O 2 ) a molecule of ozone is produced in the lower troposphere.
India with a population of 1210 million [10] is a developing country with a very fast growing economy.

Materials And Methods
Delhi is geographically located in the Northern part of India within the latitude 28°24'17" to 28°53'00''N, and longitude 77°45'30'' to 77°21'30''E which is a subtropical belt. The summers in Delhi are very hot and winters are very cold. The temperature range varies from 3°C in the winter season to 45°C in the summer season. The winters are marked by mist and fog in the mornings and often sun is seen in the afternoons. The cold wave from the Himalayan region makes winters very chilly. Figure 1 shows the variation of monthly average temperature and relative humidity over Delhi (5 sites average). In this part of India, the dry summer season runs from late March to June, the monsoon season from late June to September, the autumn season/post-monsoon season from October to November and the winter season from December to February. In the monsoon season, the southwest monsoon prevails over India, and precipitation, occasionally heavy rainfall and high humidity are common throughout the country. After mid-September, the shift of the southwest monsoon to the northeast monsoon generally occurs, which brings air masses from the north and northeast. The dry season can subsequently be classi ed into two periods, the months of strong solar radiation from April to June and another from October to November. Both NOx and CO have shown build-up during the morning (0700-0900 h) and late evening/night hours (1900-2400 h) during the study period which is different from the variations in ozone (Fig. 3). Higher levels of NOx and CO during morning and late evening hours at all the sites used in this study are due to the combinations of anthropogenic emissions, boundary layer processes, chemistry as well as local surface wind patterns. During night hours, the boundary layer descends and remains low till early morning, thereby resisting the mixing of the anthropogenic emissions with the upper layer. Hence, pollutants get trapped due to shallow nocturnal boundary layer depth resisting the mixing of local emissions with the free tropospheric air. It is important to note that the major anthropogenic source for CO and NOx in an urban region like New Delhi (mainly NO) is fossil fuel burning (combustion in motor vehicles). [21] Fig. 4 shows the monthly variation of NO, NO 2 , O 3 , and CO at all the monitoring stations during the study period. The concentrations of NOx have been found highest in post-monsoon months (October, November and December) and the lowest value in monsoon months (July, August and September). The concentrations of CO have been found highest in winter months (December, January and February) and lowest value in monsoon months (July, August and September) ( Table 2 and Table 3). This seasonal pattern may be due to a combined effect of large near-surface anthropogenic emissions, boundary layer processes, retarded photochemical loss owing to lower solar intensity, as well as local surface wind patterns. In contrast, O 3 peaked during the summer months (March, April, May and June) ( Fig. 4, Table 2 and Table 3), clearly due to its direct linear relationship with incoming solar radiation.    Table 4). The mean rate of change of O 3 at all the sites used in the present study was found similar to that of other urban locations like Agra, Kanpur, Ahmedabad, Pune and also at an urban site in New Delhi India. [15,[30][31][32][33][34]The night-time rate of change was found almost steady and slightly negative, perhaps due to O 3 loss to surface deposition and also due to fast titration of O 3 in the evening at these experimental sites.

Chemical coupling of O 3 and its precursors:
Leighton [39] has demonstrated the fact that in the troposphere the photochemical inter-conversion of O 3 , NO and NO 2 is generally controlled by the following reactions: and O * is the active mono-atomic oxygen. The above reactions constitute a reversible cycle i.e. the overall effect of Eqs. (2) and (3) are exactly opposite and canceled by the effect of Eq. 4. It represents a null cycle where there is no net production of O 3 . [40,41] Therefore, these reactions as described in Leighton [40] represents a closed energy system in which oxides of nitrogen (NOx) is partitioned between its constituents, NO and NO 2 (primary NO 2 ), and oxidant (OX) is partitioned between its constituents NO 2 and O 3 .This state is also de ned as a photo stationary state (PSS). During daylight hours (as these reactions are driven by sunlight) NO, NO 2 and O 3 are at an equilibrium state for a few minutes. [40,41] Hence the concentrations of the above species during PSS can be de ned by the following equation: [39] Where j 1 is the NO 2 photolysis rate and k 3 is the rate coe cient for the reaction between NO and O 3 ,  and O 3 were drawn that adequately described the interaction between the three species.
1. As seen in Fig. 7  . Similar behavior has been reported earlier by Clapp and Jenkin [43] and Mazzeo et al. [40] for urban and rural sites in the UK and Buenos Aires, respectively. Similar behavior has also been reported by Tiwari et al. [41] for an urban site in New Delhi, India   Figure 10 depicts the relationship between daylight averaged (60 min average) concentrations of [OX] with respect to [NOx] at all sites during the whole study period. It is seen from the gure that the concentrations of [OX] increase with the increasing concentration of [NOx] at all the sites of this study. As seen from Fig. 10, at all the sites the curve ts the linear regression curve[y = mx + c]. A non-signi cant positive correlation has been observed at all the sites during the study period (r 2 value ranges between 0.08 to 0.16 at P = 0.05). With help of the slope of the curve obtained for all sites, we can divide it into an [NOx] independent contribution and [NOx] dependent contribution for the concentration of oxidants at all the sites of an urban location. [40,41,43] The NOx-dependent contribution can be contributed to local contribution sources of oxidants and can also be correlated to primary pollutant sources. [40,41,43] It is called as [NOx] dependant contribution because an increase/decrease in [NOx] affects the concentration of oxidant sources at all the sites. [40,41] On the other hand, we can say that NOx independent contributing sources of oxidants at all the sites can be called as more regional contributing sources of oxidants as an increase/decrease in [NOx] values does not cause an effect on oxidant concentration at all sites. The local sources depict the prevalent photochemistry at the urban site. As reported by earlier studies at urban sites the major contributor to NOx is NO 2 , and in turn, the major contributor of NO 2 is the process of combustion of fuels (diesel trucks, cars, motor-generator). Nagpure et al. [45] and Badrinath et al.[46] have rst reported that NO 2 concentrations in Delhi are unusually high due to increased tra c emissions by diesel trucks, even at night and also due to crop residue burning in adjoining areas of New Delhi during post monsoon and winter season (October to January). NO being a highly unstable compound gets converted to NO 2 as soon as it is produced in the troposphere. [47][48] VOC's and CO initiated chain reactions with hydroxyl and peroxy radicals present in an urban atmosphere are also the major contributors of oxidant and ozone formation at an urban site like New Delhi. These chain reactions in the troposphere catalyzes the conversion of NO and NO 2 (Eqs. (6) and (7)) and act as major contributors to the accumulation of oxidants at the site. Due to a lack in the consistency of VOC's data we were not able to further investigate the above relationship. However, we aim to further analyze this relationship in our next study.

Diurnal and monthly variation of Oxidants [OX]
The above sections of this study have established the fact that the production and accumulation of oxidants at the site are generally governed by photochemical processes. To investigate further into this relationship we have plotted averaged diurnal and monthly variation of the oxidants at all the locations of this study during the period of 2013-2019 ( Fig. 10 and Fig. 11). Large variability was seen in the diurnal curve during the study period at all the sites of the study with concentrations of oxidants [OX] increasing after 0800h in the morning as the sun rises and attaining its maximum peak at noon (1200h to 1400h), and decreasing thereafter as the sun sets (after 1700hrs) (Fig. 11). This is because during noon rate of photochemical production is high as the intensity of sunlight is maximum during the noon hours in Delhi (1200h to 1400hr). Figure 12 shows the average monthly variation of oxidants at all the sites during the whole study period. It clearly shows that the maximum concentration of [OX] is observed during summer months (71.88 ppb during the month of June at CV Raman Dheerpur station) and minimum concentration is observed during winter months (37.86 ppb during the month of January at IMD Lodi Road station) (Fig. 12). Maximum values during the summer season depict enhanced photochemical activity because of high-temperature values at all the sites due to intense solar radiation and hence an increase in the number of sunny days. Minimum values during the winter season may be due to fewer sunny days which may be attributed to cloudy skies and high humidity because of frequent rainfall that may be due to frequent western disturbances occurring during this season in New Delhi during the study period, hence resulting in washout of pollutants (Fig. 16). Tiwari et al. [41] have reported similar behavior based on an observational study (SAFAR data) at a single site in New Delhi which showed maximum ozone concentrations during the summer in Delhi due to favorable meteorological conditions such as high solar intensity, clear skies, and low relative humidity. Ghude et al. [16] also reported similar behavior based on a model study. Earlier studies and this study have established that the regional and global contribution of oxidants at a site is governed by photochemistry as well as the prevailing meteorological conditions. In lieu of the above, the relationship of oxidants with temperature, humidity, wind speed and wind direction have been studied during the year 2018 and 2019 for New Delhi (5 station average) (Fig.  13, Fig. 14 and Fig. 15). Figure 13  and -4.48 ppbh -1 , respectively. This can be because, in morning time surface O 3 formation is strongly dependent on the available amount of precursors emitted from morning vehicular tra c (0700h-1100h) and sudden change in boundary layer height with the sunrise, while the evening time loss rate (1700h-2200h) largely depends on nitrous oxide (NO) (conversion of NO 2 to NO in the evening) (Reaction 1) concentration which participates in O 3 titration processes.
3. Polynomial t curves for NO, NO 2, and O 3 were drawn to describe the interaction and chemical coupling of the three species. (Fig 7)  and [NO 2 ] for all the sites used in the study as can be seen in Fig. 7. A polynomial relationship was drawn and calculated between hourly averaged NO, NO 2 and NOx concentrations. (Fig 7) This relationship can be used in air pollution control strategies. According to Fig. 8 Fig. 10 shows that the NOx-dependent contribution can be equated to the local ozone     Frequency distribution curve of hourly average surface ozone concentration at New Delhi.

Figure 18
Variation of Surface Ozone during daylight hours with Temperature and NOx at New Delhi.