As mentioned in Section nbsp the underlying

2.2.2. Data sources
The analyses presented here VE-822 based on four years (2009–2012) of publicly available data from routine sampling programs.
2.2.2.1. IMPROVE network
Details pertinent to IMPROVE sample collection and analyses are described elsewhere (Malm et al., 1994 and Huffman, 1996; http://vista.cira.colostate.edu/improve).
2.2.2.2. Ozone
Hourly O3 concentration data for GRBA were obtained from the National Park Service Gaseous Pollutant Monitoring Program Database (http://ard-request.air-resource.com/), and data for LAVO downloaded from the USEPA Air Data web site (http://www.epa.gov/airdata/). Daily 8-hr maximum concentrations were computed for local time (hour 00 to hour 23) and synchronized with the IMPROVE sampling dates. Daily maximum 8-hr O3 average concentrations were only calculated and used when 75% of a day's data were available for the 24-hour period.
CASTNET O3 sites were not originally intended for regulatory uses, and allowances were made in the past for the difficulty of meeting full EPA monitoring criteria at remote sites. Updates were put in place during 2010, with all sites meeting EPA criteria beginning in 2011 (USEPA, 2013). Current CASTNET ozone instrument audit criteria allow an error of no more than 15% or 1.5 ppbv, whichever is greater, for values from 4 to 260 ppbv (USEPA, 2014b).