Variations in surface ozone and carbon monoxide in the Kathmandu Valley and surrounding broader regions during SusKat-ABC field campaign: Role of local and regional sources

Air pollution resulting from rapid urbanization and associated human activities in the Kathmandu Valley of Nepal has been leading to serious public health concerns over the past 2 decades. These concerns led to a multinational field campaign SusKat-ABC (Sustainable atmosphere for the Kathmandu Valley - Atmospheric Brown Clouds) that measured different trace gases, aerosols and meteorological parameters in the Kathmandu Valley and surrounding regions during December 2012 to June 2013 to understand local- to regional-scale processes influencing air quality of the Kathmandu Valley. This study provides information about the regional distribution of ozone and some precursor gases using simultaneous in situ measurements from a SusKat-ABC supersite at Bode, Nepal, and two Indian sites: a high-altitude site, Nainital, located in the central Himalayan region and a low-altitude site, Pantnagar, located in the Indo-Gangetic Plain (IGP). The diurnal variations at Bode showed a daytime buildup in O3 while CO shows morning and evening peaks. Similar variations (with lower levels) were also observed at Pantnagar but not at Nainital. Several events of hourly ozone levels exceeding 80 ppbv were also observed at Bode. The CO levels showed a decrease from their peak level of about 2000 ppbv in January to about 680 ppbv in June at Bode. The hourly mean ozone and CO levels showed a strong negative correlation during winter (r2 Combining double low line 0.82 in January and r2 Combining double low line 0.71 in February), but this negative correlation gradually becomes weaker, with the lowest value in May (r2 Combining double low line 0.12). The background O3 and CO mixing ratios at Bode were estimated to be about 14 and 325 ppbv, respectively. The rate of change of ozone at Bode showed a more rapid increase ( ĝ1/4 17 ppbv hĝ'1) during morning than the decrease in the evening (5-6 ppbv hĝ'1), suggesting the prevalence of a semi-urban environ. The lower CO levels during spring suggest that regional transport also contributes appreciably to springtime ozone enhancement in the Kathmandu Valley on top of the local in situ ozone production. We show that regional pollution resulting from agricultural crop residue burning in northwestern IGP led to simultaneous increases in O3 and CO levels at Bode and Nainital during the first week of May 2013. A biomass-burning-induced increase in ozone and related gases was also confirmed by a global model and balloon-borne observations over Nainital. A comparison of surface ozone variations and composition of light non-methane hydrocarbons among different sites indicated the differences in emission sources of the Kathmandu Valley and the IGP. These results highlight that it is important to consider regional sources in air quality management of the Kathmandu Valley.