April 22, 2021

Spring 2021 Newsletter

In this issue, we give a brief report of ozone in the Uinta Basin during the past winter and discuss a few ongoing projects, including measurement of emissions from engines, recent modeling work, and thoughts on NOx versus VOC control of winter ozone.  And we introduce Dr. Liji David, a photochemical modeler who has recently joined our research team.  Welcome, Liji!

Kenzy with Pods

Two Exceedances of the EPA Ozone Standard During Winter 2020-21

Seth Lyman
December and January storms blanketed most of the Uinta Basin with snow.  Because snow cover reflects sunlight and keeps the earth's surface cold, inversion conditions (inversion = colder air at the surface than aloft) existed in the lowermost parts of the Basin through most of winter 2020-21 .  The graph at right shows that ozone at Ouray increased during these inversion periods, as is common during snow-covered inversion conditions in the Uinta Basin.  Ozone only exceeded the EPA ozone standard twice during the winter, in spite of these sustained inversion conditions.  Also, preliminary data show that ozone at other regulatory stations in the Basin did not exceed the EPA standard. 

Marc Mansfield recently published a paper in the journal Atmosphere that shows a reduction in Uinta Basin wintertime ozone has occurred over the past decade.  Much remains to be learned about the reason for this decline, but it appears to be driven by a reduction in ozone-forming pollution, rather than changes in meteorology.  Fortunately, Marc's finding means that less ozone is forming now when we have snow-covered inversion conditions than occurred in the past.  We caution, however, that winters with many ozone exceedance days are still possible, and that winter ozone depends on pollutant emissions, so reducing pollution in the Basin is still paramount, especially during winter months.
Ouray monitoring station dataTime series of daily maximum 8-hr average ozone at the Ouray monitoring station from 1 December 2020 through 15 March 2021.  Total (incoming and reflected) UV radiation and snow depth at the Horsepool station are also shown, and periods with inversion conditions are indicated by blue shading.  "Inversion conditions" were determined as days during which the daily average temperature at Ouray was at least 5 C lower than the daily average temperature at Fruitland (Fruitland is at high elevation and above the inversion layer).  Ouray data were collected by the Ute Indian Tribe and accessed from EPA's website.

Team News


NOx vs. VOC Control of Winter Ozone In the Uinta Basin

Marc Mansfield

figImagine you want to host a tea party.  You have eight teacups and twelve saucers.  Adding or subtracting a few saucers will not impact the number of guests you can serve.  To do that, you have to add or subtract a few teacups.  We could say that the number of guests you can serve is under the control of the teacups.  Similarly, ozone formation requires both oxides of nitrogen (NOx) and volatile organic compounds (VOC) to be present in the atmosphere, but there is a critical VOC:NOx ratio above which VOCs are in excess and changing their concentration has only a weak effect on ozone formation.  Above that ratio, we would say that ozone formation is “NOx limited” or is under “NOx control.”  Below the critical ratio, we would say that the airshed is “VOC limited” or is under “VOC control.”  But unlike the tea service analogy, the critical VOC:NOx ratio is not 1:1.  Rather it depends on local conditions within the airshed such as temperature, absolute humidity, and the types of VOCs present.  Extensive study, usually including modeling, is needed to determine the critical ratio in any given region.  A knowledge of the ratio can be used to guide policy and regulatory decisions.  For example, pollution controls targeting VOC in a NOx-limited region would be less effective.

Some preliminary evidence now suggests that winter ozone in the Uinta Basin may be under NOx control, though most pollution controls that have been applied over the last decade to control ozone have targeted VOCs.  There are three pieces of evidence, presented below.  Over the past decade, winter ozone concentrations seem to be more strongly correlated with NOx than with VOC (see the figure at right).  NOx and ozone concentrations have both trended downward over the previous decade.  VOC concentrations have also trended downward, but the available data show a recent upturn.  No corresponding upturn is seen in ozone concentrations, which might be interpreted as insensitivity to VOC.  However, with VOC recorded at only two sites, the VOC traces in Figure 1 cannot be assumed to reflect basin-wide trends. 

  1. Researchers in Mexico City described a data analysis technique that computes sensitivity of ozone to NOx.  [Wood et al., Atmospheric Chemistry and Physics, 9, 2499-2516, 2009]  We have applied the technique to Uinta Basin data.  [https://binghamresearch.usu.edu/files/reports/UBAQR_2018_AnnualReport.pdf] The analysis indicates relatively strong NOx sensitivity, although it gives no information about VOC sensitivity.  Interestingly, it indicates that the critical VOC:NOx concentration may be changing over the course of the winter. 
  2. NOAA researchers studied a model of Uinta Basin ozone, and reported that sensitivity to NOx is stronger than sensitivity to VOC.  [Edwards et al., Nature, 514, 351-354 (2014)] They only modeled conditions in early February 2013, and so their work cannot confirm the early-to-late-winter trend cited in the previous paragraph, and atmospheric chemical conditions may have changed since that time. 

We are currently seeking funding to study the question in greater detail.  The single, most significant NOx sources in the basin are the pump jack engines used to lift oil.  We hope to study the impact of converting pump jacks from engines to electric motors.  We also want to examine the impact of the proposed Uinta Basin Railway, since it is expected to open up new markets for Uinta Basin oil.  More oil will likely mean more pump jacks and more NOx.


Engine Emissions Measurement Study

Seth Lyman
With funding from the Utah Division of Air Quality, the Utah Legislature, and the Uintah Impact Mitigation Special Service District, we have been measuring emissions from engines in the Uinta Basin since January.  Our study focuses on emissions from natural gas-fueled lift engines, or the engines that are used to power pumps that pull oil up from subsurface reservoirs.  There are two reasons that this study is important for understanding and improving the winter ozone situation in the Uinta Basin:
1)Previous measurements in ambient air have shown higher concentrations of reactive organics (i.e., organic compounds with a high potential to form ozone), including aldehydes and alkenes (e.g. ethylene, propylene) in areas with more of these types of engines.
2) Currently, very little detailed data about the composition of emissions from these types of engines is available.  The preferred dataset modelers use to simulate their emissions composition is based on measurements from two engines in California in the 1980's.  An update to this knowledge base is clearly needed.

We are measuring exhaust temperature, dewpoint, gas flow rate, and concentrations of CO2, CO, NO2, NO, O2, methane, and a suite of 70 individual non-methane organic compounds, including alkanes, alkenes, aromatics, alcohols, and aldehydes.  So far, we have measured emissions from 30 engines, and we are gearing up to measure more.  A very preliminary synopsis of our early findings (subject to change) is available in a poster we produced for the Air Quality: Science for Solutions Conference in March.
engine emissions measurements
USU team measuring emissions from a lift engine in the Uinta Basin.

Photochemical Modeling Improvements

Huy Tran

We evaluated photochemical model performance during a high-ozone episode during winter 2019 with the CAMx modeling platform and Utah Division Air Quality’s 2017 oil and gas emissions inventory (OGEI2017). The OGEI2017 has been updated recently and shows a significant increase in Volatile Organic Compound (VOC) emissions from oil and gas activities in Duchesne and Uintah counties in comparison to its first release in late 2019 (103,723 tpy vs.  66,185). The largest VOC emissions increases in the revision are from tanks, water disposal facilities, and tank emissions control effectiveness. Interestingly, these increases led to simulated ozone that was much higher than that simulated with the original OGEI2017 (see figure below).  Simulated ozone at several stations, including Castle Peak, was even higher than observed ozone. Poor ozone performance in the model on February 26, 2019 and later days (shown in the figure above) was due to discrepancies in the meteorological model. 
camxperformance2019
Observed (OBS) vs. simulated ozone concentrations at the Ouray and Seven Sisters monitoring stations as simulated by the CAMx model with the Updated and Original OGEI2017.

Total VOC was well approximated at Roosevelt but underestimated at the Horsepool station. Modeled highly reactive VOC compounds, such as alkenes, aromatics, and carbonyls, were also highly underestimated at Horsepool but had better agreement with observed values at Roosevelt. Comparison of simulated VOC composition to observed values from the distributed canisters that were deployed in the past winters also showed lower concentrations of reactive compounds in the simulated VOC (see figure below). We are improving meteorological model performance to improve ozone performance for this winter episode. Further analyses with observed VOC from the distributed canisters to identify potential improvement in the VOC composition of oil and gas emissions are also underway.
pods comparison
USU team measuring emissions from a lift engine in the Uinta Basin.


Coming Up Next

  • Over the next few months, we will be continuing with emissions measurements from engines, continuing modeling work, and finalizing all data collected this winter.  Watch for more updates in our summer 2021 newsletter and our 2021 Annual Report, to be released in November.
  • We received lots of great suggestions from stakeholders in response to our survey last year.  Things get pretty busy for us in the winter months, but we plan to connect with stakeholders this spring to ask for more information about what we can do to improve.  If you'd like to be part of that process, email Seth Lyman at seth.lyman@usu.edu