LaPorte Ground Site Measurements

Statement of Work

Measurement of OH and HO2 mixing ratios, the total OH loss rate, and HO2 fluxes

William Brune, Principal Investigator, Pennsylvania State University, Department of Meteorology, 503 Walker, University Park, PA 16802
Phone (814) 865-3286, Fax (814) 865-3663. Email


The hydroxyl radical (OH) and the hydroperoxyl radical (HO2) play central roles in urban and regional air quality. OH initiates most reaction sequences that cycle surface emissions through the atmosphere. HO2 is readily exchanged with OH. In the exchange process, HO2 reacts with NO to form OH and NO2, leading to ozone formation. OH and HO2 together are called HOx. Predictive capability for ozone and its response to regulatory action requires a firm understanding of HOx sources, sinks, and interactions with anthropogenic hydrocarbons and nitrogen oxides.

Recent field studies have attempted to validate HOx chemistry by measuring OH (and HO2) simultaneously with all known HOx sources and sinks. Generally, models constrained by measurements of NO, hydrocarbons, and meteorology often predict more OH than is observed by 10-40% and predict more HO2 than is observed by a factor of 3-6. Interestingly, in the free troposphere, models often underpredict OH and HO2. The observed dependence of HOx on factors such as solar zenith angle and NO seems to be fairly well described in some environments, but not all. Problems occur in moderately clean near-surface, some very clean and some polluted free tropospheric environments, and at high solar zenith angles. Thus, while some of the basic characteristics of tropospheric HOx chemistry appear to be well understood, some significant amount of its observed behavior cannot yet be explained.


The overall objective is to improve the understanding of atmospheric oxidation and ozone formation processes. Houston's urban plume will provide a contrast to the rural forest of Michigan (PROPHET, 1998) and the free troposphere (aircraft measurements, SUCCESS, 1996; SONEX, 1997). To our knowledge, these will be the first successful measurements of OH and HO2 mixing ratios in an urban environment and the first ever measurements of total OH loss rate and HO2 flux. The previous SOS studies have shown the importance of both anthropogenic and natural hydrocarbons in this environment. Knowledge of these HOx sources and sinks will be important for defining the oxidation and ozone formation processes in this urban plume.

We have some specific objectives:

  • identify the variation of HOx and HO2/OH with NOx and VOCs: Do they vary as expected?
  • quantify OH loss rates: Are some sink species missing?
  • examine the suggestion of nighttime OH chemistry: Is O3+alkenes are significant source?
  • measure HO2 surface deposition: Are current theories correct?

Answering these questions will aid the understanding of HOx behavior and may point to reasons for current discrepancies.


GTHOS (Ground-based Tropospheric Hydrogen Oxides Sensor) uses laser-induced fluorescence (LIF) to measure OH and HO2 simultaneously. OH is both excited and detected with the A2( 2\Sigma+ (v'=0)(X2( (v"=0) transition near 308 nm. HO2 is first reacted with reagent NO to form OH and is then detected with LIF. The ambient air is pulled by a vacuum pump through a small inlet (1mm diameter), which faces up, down a sampling tube, and into two low-pressure multipass White cell detection cells. The first cell is for OH and the second for HO2. Detection occurs in each detection cell at the intersection of the airflow, the laser beam, and the detector field-of-view. The pulsed laser has a 3 kHz repetition frequency, 28 ns long pulses, and produces about 10-20 mW of tunable UV near 308 nm. The laser is rapidly tuned on and off resonance with the OH transition; the OH fluorescence is the difference between the signal on resonance and the signal off resonance. The detector is gated to detect the OH fluorescence after each laser pulse has cleared the detection cell. A reference cell containing OH indicates when the laser is on and off resonance with the OH transition.

The instrument's ground-based configuration has the detection cells mounted in a weatherproof box at the top of a 10-meter tower. The laser and detector electronics are housed in a trailer at the base of the tower, and the vacuum pump is located in a housing within 30 meters of the tower. The detection cells and equipment in the trailer laboratory are connected by electrical cables and a fiber optic cable for the UV laser light. The vacuum pump is connected to the detection cells with a long vacuum hose. We used this configuration during a PROPHET summertime intensive in 1998 with considerable success.

The absolute uncertainty, which is determined in the laboratory and maintained with power and signal monitors and weekly in situ calibrations is (40%. The minimum detectable mixing ratio (S/N =2, 60 seconds) is 0.015 pptv (3.5x105 cm-3) for OH and 0.06 pptv for HO2. Because the signals obey Poisson statistics, the detection limit is less than 105 cm-3 in about 20 minutes.

In addition to measurement of OH and HO2 mixing ratios, we will measure the total OH loss rate. This measurement is made the same way that kinetic rate constant measurements are made using the discharge flow technique. A pump pulls the ambient air down a flow tube, which is connected to an OH/HO2 detection system at one end. A smaller diameter tube, centered in the flow tube, houses an OH and HO2 source. The OH from the source mixes with the ambient air, reacts, and is detected with the detection system. As the inner tube is retracted, the OH reaction time is increased and the OH loss is measured. Hourly injection of zero air and of known amounts of CO calibrate the radial and wall losses as well as test for any system-related OH losses. The flow tube will likely be housed in the trailer. A glass manifold that samples at the top of the 10-meter tower will be attached to the flow tube so that the OH loss rate measurements will be those appropriate for the top of the tower. We tested this instrument during the PROPHET summertime intensive in 1998 and feel that it will be ready for high-quality OH loss measurements by summer 1999.

The third measurement we will make is HO2 flux measurements using the eddy-correlation technique. We have sufficient sensitivity for HO2 that 1-second averages have excellent statistics. We will attach a sonic anemometer near the GTHOS inlet and will use this arrangement to produce 2'> in order to determine the HO2 vertical flux. This measurement will be difficult because of the size of the detection head for GTHOS. However, recent work by Miller and Wyngaard (private communication) suggest that flux measurements with large instruments may be possible.

All of these measurements will be made as continuously as possible, day and night. We hope to have GTHOS operating essentially autonomously by summer 1999. We expect 60-90% data coverage during the entire Houston SOS campaign. The averaging time for the mixing ratio and OH loss rate measurements will be a few minutes; the HO2 flux measurements will require 30 to 60 minutes.