2013 Progress Report: Assessing The Synergistic Impact Of Anthropogenic And Biogenic Emissions On Air Pollution Using Novel High-Sensitivity, Real-Time Monitors For Fundamental Carbonyls

EPA Grant Number: R835138
Title: Assessing The Synergistic Impact Of Anthropogenic And Biogenic Emissions On Air Pollution Using Novel High-Sensitivity, Real-Time Monitors For Fundamental Carbonyls
Investigators: Keutsch, Frank N
Institution: University of Wisconsin - Madison
EPA Project Officer: Peterson, Todd
Project Period: February 1, 2012 through January 31, 2015 (Extended to January 31, 2016)
Project Period Covered by this Report: February 1, 2013 through January 31,2014
Project Amount: $250,000
RFA: Developing the Next Generation of Air Quality Measurement Technology (2011) RFA Text |  Recipients Lists
Research Category: Air Quality and Air Toxics , Air


The goal of this project is to demonstrate instrumentation that will allow the use of novel high time resolution monitoring data sets as new metrics to determine the contribution of anthropogenic and biogenic emissions to ozone and organic aerosol (PM). We will demonstrate that the proposed novel instrumentation can obtain long-term, low-maintenance and accurate measurements of key carbonyl-containing compounds such as formaldehyde and glyoxal, consistent with the requirements for instrumentation employed in monitoring networks. The project will validate this approach by collecting a year-long data set at the Horicon National Core Monitoring Station, a rural site in Wisconsin that is part of the EPA Region 5 Ambient Monitoring Network. An objective of this work is to develop the ratio and absolute concentration of monitored glyoxal/formaldehyde as a metric of the contribution of biogenic and anthropogenic volatile organic compounds (VOCs) to atmospheric oxidation. This metric will allow distinction between the direct contribution of anthropogenic VOCs and the anthropogenic impact on biogenic VOC oxidation (resulting O3 and PM via NOx), an emerging issue in air quality control, in particular for PM. In addition, these data will be compared with the WRF-CMAQ model output to evaluate and improve the representation of atmospheric oxidative chemistry in models, thereby helping to provide strategies to control air quality related to O3 and PM. 

Progress Summary:

In year 2 of the project, instrumentation was deployed and installed at the Horicon site and first long-term measurements were conducted. In addition the 1-D box model was extended from the previous steady-state version of the model to one that allows modeling of diurnal cycles and extended data sets. During deployment, the novel laser systems encountered problems that were diagnosed in collaboration with the companies that developed the laser systems. These problems were solved and deployment was started again to obtain year-long measurements. The activities conducted in year 2 of the project can be summarized into three major thrust that focus on (1) deployment and measurements at the Horicon site, (2) diagnosing laser performance under long-term operation conditions, and (3) advancing the detailed chemical model for determination of the anthropogenic influence on rural secondary pollutant formation. In addition, the quality control of the formaldehyde instrument from the intercomparison started in year 1 was completed. 
(1) Installation of instrumentation at the Horicon:
A major focus of the work conducted in year 2 was deployment of instrumentation for long-term monitoring at the Horicon site. In coordination with site personnel, we arranged space and electrical power for the formaldehyde and glyoxal instruments. In addition, calibration units and gas supplies were deployed. The formaldehyde instrument was installed (Figure 1) and measurements conducted in April and May of 2013, the results of which are shown and discussed below (Figures 2, 3). As a result of intermittent laser behavior, our focus then shifted from the measurements to diagnosis and improvements of the formaldehyde laser as described in section (2). The glyoxal laser was installed and the instrument tested. Testing highlighted that the laser did not meet the specifications provided by the company, which resulted in extended discussions with the company to determine a solution as described in section (2). Figure 1 shows the formaldehyde instrument deployed at the Horicon site. Figure 2 shows a part of the measured data, 1 week of measurements in May. The week was particularly interesting as a controlled burn took place in the area adjacent to the measurement station. The figure shows that formaldehyde concentrations in the biomass-burning event were about 40 times the maximum values otherwise observed. The extremely high formaldehyde concentrations lasted about 40 minutes. The time integrated amount of formaldehyde in those 40 minutes was equivalent or larger than the amount normally observed in 24 hours at the site. The formaldehyde concentration is 4-5 times the 8 hr NIOSH recommended exposure limit (REL) of 16 ppb and only 4 times lower than the ACGIH short-term exposure limit (STEL) threshold of 320 ppb that should never be exceeded during a work day. Formaldehyde is an important contributor to ozone production as it is an important photochemical source of HO2, which produces ozone via conversion of NO to NO2 followed by NO2 photolysis. We plan to analyze the contribution of formaldehyde from biomass-burning to local ozone production: In combination with additional controlled burns, for which information is available, being conducted in year 3 and the complete suite of measured VOCs and oxidants, this will provide valuable information on the impact of biomass burning on secondary pollutant formation. 
Figure 2 also shows formaldehyde concentrations measured with our novel formaldehyde monitor excluding the biomass-burning event. The only clear diurnal feature is that formaldehyde always decreases before sunrise, although it never goes to zero. The temporal behavior of daytime formaldehyde concentrations is highly variable. This strongly supports that the site is exposed to varying regional transport and we are using back trajectories to evaluate regional source regions (e.g., Milwaukee or Madison). Combined with observations of biomass-burning events, this will allow comparison of the measurements with our model as well as the regional models of our collaborators. Specific events likely cannot be compared with the latter models. However, we will compare model output for transport from different source regions with measurements for events from the same source regions based on back trajectories. This analysis will utilize the full set of VOCs and oxidants (NOx, ozone) as well as tracers (CO, SO2) available, which is why we chose Horicon, as the most instrumented site in Wisconsin for this project. The importance of our new fast monitors becomes apparent as slower measurements average of the changes in formaldehyde that result from changes in regional transport making comparison with models much more difficult. In contrast, our high time resolution measurements are ideally suited for identifying source regions and testing the chemistry in the transport from this source regions with models. 
(2) Long-term testing of lasers and diagnosis of problems:
An important aspect of our work using the novel monitors is evaluation and improving the new technology. The initial deployment of new technology often reveals technical issues that need to be addressed and these usually are uncovered only during deployment under realistic operating conditions. Thus, thorough characterization and diagnosing of problems is a central part of the work under this grant. Two unforeseen problems occurred with the novel lasers being used: During operation the fiber laser used for the formaldehyde instrument started having intermittent errors in which the laser would turn on and off sporadically. This resulted in data gaps and also made data analysis more challenging. We spent considerable time diagnosing this problem. The laser was also returned to the company. We have determined several causes, which only became apparent during long-term operation. A primary reason was that the power supplies in the laser were of low quality, which resulted in instability of the supplied voltage. As a result, operating thresholds were not met, turning the laser off. A second problem was degradation of some optical components in the laser resulting in degraded performance. This in turn prevented thresholds to be reached, shutting off the laser. Diagnosing and addressing these problems was time consuming as there is only one company that provides this laser, and initial discussion was slowed down by purchase of the company (and its products, including the laser) by ThermoFisher. However, we now have gained in-depth understanding of the laser, initially not provided by the company. We now have a fully operational laser that has been redeployed to Horicon and is operating well.

The laser for the glyoxal instrument did not meet the specifications with respect to the fall time of the laser pulse after turning off the laser, despite direct communication with the laser company prior to grant submission and in the purchase order. The slow fall time prevented the acquisition of phosphorescence signal as it was overwhelmed by laser scatter. Consultation with the laser company resulted in a new solution for generating pulsed signals consisting of adding an acousto-optic modulator (at no cost to the grant). This system allows acquisition of a laser-induced phosphorescence signal. 

Although the laser problems delayed measurements, they are part of the development and testing of novel monitors based on new laser technology. Troubleshooting and eliminating the problems, in particular for the formaldehyde system, were only feasible through the long-term measurements under this grant. Short duration measurements would have not allowed discovery or diagnosing the issues. The work on this project thus was central to improved laser systems and monitoring instrumentation. As a result of this work, greatly improved lasers, resulting in a new class of monitoring instruments for small carbonyls will be available. 

(3) Advances in box modeling: 
We have further extended the box model box model being used in the Keutsch Group, which was based on the Chemistry of Atmosphere Forest Exchange (CAFE) model, developed by Glenn Wolfe, a former Postdoctoral Researcher in the Keutsch Group. The model is a 1-D (vertical) box model that uses the Leeds Master Chemical Mechanism v3.2 with vertical transport modeled via eddy diffusivity. In the first year, we upgraded the mechanism to allow integration of a number of recently proposed reaction schemes involving OH radical chemistry and glyoxal production. We have added a critical extension: extension of the previous steady-state model to a time dependent model that allows modeling of diurnal cycle and extended datasets. In addition in using back trajectories, we can run Lagrangian models for transport events from those source regions for which input constraints are available. Although the model is substantially more demanding on computer time, it is still is very fast. This allows evaluation of the impact of different mechanisms in the model, which is one of the goals of our work. We have tested the model and it is now ready for application to the Horicon data to provide complementary information to the CMAQ models provided by our collaborators. 
Minor task: Completion of the quality control for formaldehyde instrument.
As discussed in the year 1 report, we conducted an intercomparison, directly relevant to the quality control of this project, between a state-of-the-art formaldehyde Hantzsch instrument with the formaldehyde fiber-laser induced fluorescence method. Completion of the analysis of the intercomparison confirms the excellent performance of our autonomous instrument. Results show that under all conditions the two techniques are well correlated (R2 ≥ 0.997), and linear regression statistics show measurements agree within stated uncertainty (15% FILIF + 5% Hantzsch). Importantly, no water or ozone artifacts are identified, which have been found for formaldehyde instruments due to inlet interactions. While a slight curvature is observed in some Hantzsch vs. FILIF regressions, the potential for variable instrument sensitivity cannot be attributed to a single instrument at this time, although Figure 4 shows that this likely results form a drift in the Hantzsch baseline or calibration factor. Measurements at low concentrations (< 100 ppt) highlight the need for a secondary method for testing the purity of air used in instrument zeroing. This concentration range is not important for measurements at Horicon, but we are still addressing this and designing a new catalytic converter that will remove formaldehyde without affecting humidity. This is to eliminate the use of zero air, which has very low water content and any potential effects that may result from conducting zero measurements with different humidity than the ambient measurements. The results of the intercomparison also support the use of the new single-pass detection system, which has more stabile alignment as well as a much smaller internal surface area and thus less interaction between the sampled gas and the instrument. Overall, the results confirm that our instrument is ideally suited for the first long-term, continuous and high-time resolution formaldehyde measurements now starting at the Horicon site. 

Future Activities:

The focus of year 3 will be continuation of the long-term measurements of fast (high time-resolution) formaldehyde and glyoxal at the Horicon site. This effort will continue extensive evaluation of the long-term behavior of the instruments with respect to sensitivity, accuracy and overall instrument performance. In the second half of year 3, we expect to start addressing central scientific questions of evaluating the influence of anthropogenic influence on rural photochemistry using the novel data sets. We are finalizing the plan for collaboration with our CMAQ modeling partners (Annmarie Carlton, Rutgers). We will also make the data available to any other interested partners and plan to implement real-time access to preliminary formaldehyde and glyoxal data measurements at Horicon. 

Journal Articles:

No journal articles submitted with this report: View all 2 publications for this project

Supplemental Keywords:

Formaldehyde, glyoxal, secondary pollutant formation, ozone, anthropogenic influence, anthropogenic emissions, biogenic emissions, air pollution, carbonyls

Progress and Final Reports:

Original Abstract
2012 Progress Report
2014 Progress Report
Final Report