Musings on Monitoring: The Who, What, When, Why, and How of Atmospheric Measurements

The installation and operation of ambient pollutant and meteorological monitors has never traditionally been a part of the laundry list of environmental tasks at most facilities.  Ambient monitoring installations by facilities have been limited to consent decrees and large Prevention of Significant Deterioration (PSD) permitting projects that required determining what a specific source’s impact on air quality would be or assessing the local background ambient air concentration levels.  In general, all other ambient monitors are operated by state agencies as a way to measure air quality against the National Ambient Air Quality Standards (NAAQS).  This reliance on state run monitoring programs to determine ambient air concentration levels could all change with U.S. EPA’s implementation of the 1-hour sulfur dioxide (SO2) NAAQS.  Similarly, the installation of ambient meteorological monitoring systems have also been associated with large PSD projects where air quality modeling is required and representative National Weather Service (NWS) data are unavailable; or a facility may have a small scale meteorological monitoring station for facility activities that could range from collecting a couple of meteorological variables to having a simple windsock.   The following sections discuss the basics of ambient air and meteorological monitoring and describe when they may be required and why they may be beneficial to consider right now.

As documented in ALL4’s previous blog posts and articles, U.S. EPA’s approach to implementing the 1-hour SO2 NAAQS has taken a long and winding path.  While historic NAAQS have been evaluated through the installation and operation of a network of ambient pollutant monitors, U.S. EPA reversed course with the 1-hour SO2 NAAQS and initially implied that air dispersion modeling would be used for attainment/non-attainment designation purposes in the absence of ambient monitoring data.  U.S. EPA reconsidered the dispersion modeling approach after receiving unfavorable comments, and has since held three (3) stakeholder workgroups to discuss the NAAQS implementation approach with industry representatives, environmental groups, and state and local representatives.  The main takeaways from the stakeholder meeting with industry groups: an enhanced ambient monitoring network for SO2 is likely and the installation and operation of those monitors may ultimately be the responsibility of industrial facilities, not the state agencies as it has been in the past.

A decision to place a monitor in an area (or near a facility) is likely to depend on a facility emissions threshold or population weighted emissions threshold for a given area.  Due to the staffing and budgetary restrictions being faced by state agencies, the possibility exists that monitors will need to be paid for, installed, and operated by facilities or groups of facilities under the supervision of the states.  While this monitoring arrangement is far from being final (and we don’t anticipate final rulemaking from U.S. EPA before mid-2013), it is important enough to warrant increased awareness by facilities of the technical aspects of siting, operating, and maintaining ambient pollutant monitors.  The following questions and answers highlight important aspects to consider in anticipation of future ambient monitoring programs:

• Where should monitors be located?  U.S. EPA should be encouraged to locate ambient pollutant monitors in areas where sensitive populations (e.g., schools) could be exposed to elevated SO2 concentrations, which would generally correspond to where people live and work.  The likelihood of sensitive populations being exposed to elevated concentrations at the fenceline of a major facility is relatively low since these populations are not likely to be located along a facility fenceline during worst case dispersion and worst case emission conditions.  Apart from that debate, dispersion modeling can and has been used to determine where peak ambient concentrations can reasonably be expected to occur and therefore where an ambient monitor should be placed.  Dispersion modeling results, in combination with a common sense approach to account for where sensitive populations may be exposed to elevated concentrations, should serve as a good starting point for considering monitor placement.  However, there is a myriad of factors that will vary by location, including meteorology, topography, power access, monitor security, presence of local obstructions (e.g., trees), and building downwash that can influence the decision of where to locate a monitor, resulting in the likely need for a case by case assessment.  These factors will all need to be evaluated by states and facilities when considering the placement of ambient monitors.
• Should I collocate a meteorological monitor with an ambient pollutant monitor? Yes, yes, and yes. A meteorological monitor should almost always be collocated with ambient pollutant monitors to understand the meteorological conditions that existed during each hour of data collected.  This will assist facilities in determining who and what is impacting an ambient monitor.  The meteorological information can be especially beneficial in areas with multiple facilities located nearby to determine which facilities might be contributing to elevated concentrations at the ambient pollutant monitor.  Obviously it will become more and more difficult to determine which source or facility is contributing to an ambient monitor in areas with a large number of local facilities.  (See additional discussion of meteorological monitoring in the following section.)
• What are the limitations on where I can place an ambient pollutant or meteorological monitor?  You need adequate physical space to install an ambient pollutant monitor and a monitoring shelter (typically the size of a small trailer on the order of 15 by 15 feet), and for a meteorological monitor you need an open space void of major objects that may influence meteorological conditions in order to erect a 10 meter tower (at a minimum).  Options exist to construct smaller enclosures that limit mobility inside the shelter but may be necessary where space limitations are a problem.  You also need availability of power, phone lines (or cell phones), or an internet connection to run the equipment and access the monitoring data.  If there is no infrastructure, there will be no way to install and operate an ambient pollutant monitor that meets U.S. EPA guidelines.
• How much does it cost to install and operate an ambient monitor?  The capital cost of an ambient monitoring station typically ranges from $100,000 to $200,000.  Labor costs to install and maintain an ambient monitor typically range from $50,000 to $75,000 on annual basis.  In order for the ambient monitor to be used for NAAQS designation purposes, it must collect three (3) years of ambient concentration data.  Note that although three (3) years is a typical data collection requirement, nonattainment designations can also be made on the basis of a shorter duration of data collection as soon as a NAAQS violation is observed.
• Are there disadvantages to installing an ambient monitor now in anticipation of the SO2 NAAQS implementation process?  An ambient monitor generates ambient concentration data that could potentially indicate an exceedance of the 1-hour SO2 NAAQS design value, so there are always risks in making that discovery, particularly if there is no guarantee that an ambient monitor will eventually be required as part of the NAAQS implementation process.  That being said, it is important to understand your facility’s emissions profile and its potential impacts on measured concentrations prior to deciding to proactively (or by requirement) install an ambient monitor.  Since ambient monitors reflect real concentration levels, a well-controlled facility with emissions well below permitted emission rates could reasonably be expected to result in measured concentrations below the applicable NAAQS level in certain cases, thus reducing the risks (and enhancing the benefits discussed below) of installing an ambient monitor.  Consult with a dispersion modeler to gain a better understanding of how your emissions profile and other facility characteristics (such as building downwash) may impact monitored concentrations and whether there are any discernible risks associated with installing an ambient monitor.
• Should I consider installing an ambient monitor in advance of the NAAQS rulemaking being finalized?  If air dispersion modeling will be required as part of the NAAQS implementation process, and that possibility still exists, then ambient monitoring information could be useful to “calibrate” and/or refute the results of the air dispersion modeling study, which in many cases results in modeled concentrations well in excess of actual measured concentrations.

As mentioned previously, air quality modeling is a requirement for projects that trigger PSD permitting requirements.  Although there is some uncertainty around air dispersion modeling requirements for the 1-hour SO2 NAAQS implementation process, neither the NAAQS level for SO2 of 75 parts per billion (ppb) nor the associated PSD modeling requirement are in question and neither is likely to change in the near future (recent Court decisions have rejected lawsuits seeking to remove the promulgated 1-hour health based standards).  The stringency of the 1-hour SO2 and NO2 NAAQS in the context of air dispersion modeling studies places increased importance on having the best meteorological data available (i.e., site-specific data).

A dispersion model is only as good as the information you supply it with.  In most cases facilities are using NWS data which only has the bare minimum meteorological variables such as one level of wind speed, wind direction, temperature, and solar radiation data.  AERMOD, the approved U.S. EPA dispersion model, is capable of using meteorological data collected from multiple levels.  Such data can be collected from either a tall tower (>10 meters) and supplemented with a Sonic Detection and Ranging (SODAR) system to levels of 200 meters or more.  The collection of multiple levels of meteorological data allows AERMOD to more accurately simulate the atmospheric profile and provide a better estimate of the transport and dispersion of pollutants.  Furthermore, the simple addition of a vertical wind speed instrument can be used to calculate more accurate atmospheric turbulence data.

In addition to the beneficial refinements that can result from using site-specific meteorological data in dispersion modeling, the short-term nature of the NAAQS levels has caused states to place increased importance on having representative meteorological data available for conducting such modeling. As a result, justifying the use of off-site meteorological data will become more difficult in certain cases.  For all of the reasons discussed, it would appear to be in the best interest of facilities considering projects now or as part of a long-term plan, to consider collecting on-site meteorological data in order to have the best data available if and when air quality modeling is required as part the PSD permit program.  Even if there are no current plans for projects, collecting meteorological data now will enable facilities to potentially avoid delaying a future project by as much as a year in order to collect on-site meteorological data if required.

Here are some things you need to know about setting up a meteorological monitoring station for air quality modeling:

• One (1) year of on-site meteorological data that is at least 90% complete per quarter is required for an air quality modeling evaluation.  Once these data are collected they can be used for all future air quality modeling efforts at a facility, including 1-hour SO2 NAAQS implementation modeling, PSD modeling, and air toxics modeling.
• The height of the meteorological tower depends on the geographic setting.  Facilities located in flat terrain may only require a 10 meter tower while facilities in more mountainous terrain may require a tall tower (up to 100 meters) and/or a SODAR.  Again, the more robust the meteorological dataset the more accurate AERMOD predicted concentrations will be.
• Meteorological data collection must be done in accordance with guidance contained in U.S. EPA’s “Meteorological Monitoring Guidance for Regulation Modeling Applications.”
• Semi-annual quality assurance audits must be completed in accordance with Volume II Part 1 of the U.S. EPA “Quality Assurance Handbook for Air Pollution Measurement Systems.”

Once a one (1) year period of meteorological data has been collected for air quality modeling purposes, a less intensive meteorological monitoring program can be implemented in order to continue to collect data for other end uses at your facility.  Meteorological monitoring systems can be set up for end users to access the data in a wide range of ways including websites, voice modems, programmable logic controllers (PLC), and even as an app on your smart phone.

Summary

Stay tuned as U.S. EPA works to finalize the 1-hour SO2 NAAQS implementation process.  While no immediate action is required for ambient pollutant monitors, ambient SO2 monitoring is likely to become a much more familiar concept to major sources of SO2 within the next 12 months.  Consider the benefits of meteorological monitoring right now and think about how that data could help your facility in future air permitting efforts.  Also, engage in proactive conversations with your state agency to get their take on the pending ambient monitoring network and how they envision it being implemented.  Contact ALL4’s Colin McCall or Dan Dix with questions related to ambient and meteorological monitoring.