USEPA and others in the vapor intrusion (VI) field have been evaluating a variety of Indicators, Tracers, and Surrogates (ITS) to assess their use in predicting the best time to collect representative indoor air samples for vapor intrusion studies (Schuver, et. al., 2018). The idea is that if a predictive combination of easily obtainable, low cost ITS can be identified, they could be used to improve the collection of actionable analytical data at a lower cost.
For VI assessments, typical indicators include seasons of the year, wind speed, the difference between the indoor and outdoor temperatures (differential temperature), barometric trends, and the difference between sub-slab and indoor air pressure (differential pressure). Last month, I reviewed two ambient factors that are potential contributors to changes in differential pressure – barometric pressure and temperature (indoor and external). As is common in the industry, I obtained the barometric pressure and external temperature data from a remote weather station which was many miles from my site. This month, I will compare the differential pressure data to an onsite weather station to see if we can tease out any details.
Differential pressure is generally measured with a hand-held manometer by measuring sub-slab air pressure and indoor air pressure and taking the difference. Recently, however, sensitive differential pressure sensors have become available as part of the “Internet of Things” (IOT) revolution. These sensors can be connected to permanent sub-slab monitoring points, such as the Vapor Pin®, to collect and transmit differential pressure, temperature, and barometric pressure readings to the web at preset intervals. They can also be used to set alarm point that will notify users of system faults or other unacceptable conditions.
This month (July 2020), I collected differential pressure using a sensor at our warehouse. The sensor was connected to a Vapor Pin® and allowed to run continuously. The warehouse consists of 4,500 square feet, slab-on-grade with a common moisture barrier. The ceiling height is approximately 22 feet, the walls and roof are constructed of insulated steel, and the building interior is heated by two gas fired furnaces hanging from the ceiling. For this experiment, the differential pressure sensor measured and logged differential pressure and internal temperature. External temperature and barometric pressure readings were obtained from both a remote weather station several miles from our warehouse and from an onsite weather station.
Last month’s data indicated that the differential pressure was directly related to the internal temperature of the building and not so much to external temperature nor barometric pressure. The latter observation was puzzling to me because I would have expected barometric pressure to be a primary driver of differential pressure. My first inclination was that the barometric pressure at the site must be significantly different than that reported by the remote weather station. To compensate for this, I rented an onsite weather station to log, among other things, barometric pressure and temperature.
The two graphs below plot a segment of the collected data as differential pressure, exterior temperature, and interior temperature on one graph; and differential pressure and barometric pressure on the other graph.
Upon inspection of the graphs, two conclusions can be made with respect to differential pressure: 1) from July 8 through 14, the differential between the sub-slab environment and the indoor air space was slightly greater than 0, indicating that the flow of soil gas was upward into the warehouse most of the time; and 2) there is a slight diurnal fluctuation in differential pressure during the week.
The graph of temperatures is quite revealing. Based on this week of data (which was consistent with readings collected throughout the month), the differential pressure tends to mirror the changes in internal temperature and external temperature from both the onsite and remote weather stations. The diurnal internal temperature and maximum differential pressure occurs around noon of each day. I suspect that this pattern is due to solar heating of the building. Although not the case last month, external temperature from both the onsite and remote weather stations appeared to be directly relatable to changes in differential pressure.
The graph of barometric pressure is also revealing. I had suspected that differential pressure would be most sensitive to changes in barometric temperature. However, as confirmation to last month’s data, there is no immediately discernable patterns in the differential pressure and barometric pressure readings, either collected onsite, or remotely. It is interesting that the remote and onsite barometric pressure readings track one another very well but are shifted by about one inch of mercury. I believe this may be a calibration issue associated with the onsite instrument.
What these data demonstrate, in our case at least, is that the remotely obtained temperature data, which matches the onsite data during this time period, appears to be useful in predicting the differential pressure changes for this warehouse space.
We must remember, however, that the conditions measured in the spring and summer months of Ohio, may not be representative of those in the fall and winter months. As the weather changes and we move into the heating season, we will revisit this issue.
As we wait for the weather to change, we may set our sights on a surrogate for VI. Stay tuned.
Reference:
Schuver, H, Lutes, C, Kurtz, J, Holton, C, Truesdale, RS. Chlorinated vapor intrusion indicators, tracers, and surrogates (ITS): Supplemental measurements for minimizing the number of chemical indoor air samples—Part 1: Vapor intrusion driving forces and related environmental factors. Remediation. 2018; 28: 7– 31. https://doi.org/10.1002/rem.21557