Recent discussions of Continuous Emission Monitoring Systems have presented some differences in measurements between Metal Oxide Sensors (MOS) and Laser Absorption Spectroscopy (LAS). Although MOS sensors are low-cost solutions for measuring methane concentrations, their absolute accuracy puts their use into question. This may leave one wondering: To MOS, or not to MOS? We argue that MOS sensors are very sensitive and effective sensors to use for the purposes of methane detection and continuous site monitoring because of their high sensitivity to minute changes in methane concentration levels.

Every problem has multiple solutions. A bathroom scale and a high-precision analytical scale, for example, can both find the mass of an object. However, the best tool for a given task depends on the task at hand. At the doctor’s office, for example, a low-cost bathroom scale is adequate for measuring the mass of a human, and a high-cost and high-precision scale that reports mass to the thousandth kilogram is both unnecessary and unhelpful. On the other hand, when highly precise measurements are needed in a laboratory environment, the use of the high-precision scale is justified. Similarly, both MOS and LAS sensors can successfully measure methane emissions. A benefit of the LAS sensors is that they produce a precise measurement of absolute methane concentrations with high accuracy at a price point of $5,000 to $40,000+. Whereas the MOS sensors are not as accurate for measuring absolute methane concentrations, they do measure changes in methane with high accuracy and precision at a price point of $10 to $50.

The goal of Continuous Emissions Monitoring Systems is to accurately detect and locate a leak in a timely manner. Below, we explore why these goals are best achieved using a MOS sensor:

1. Accurate Detection and Location of Leak

For continuous monitoring, the new EPA regulations require the establishment of site-specific baseline emissions. As a result, methane emissions are reported as exceedances above a baseline value (e.g. changes in methane) rather than the absolute concentration of methane. Although we utilize the high accuracy of LAS sensors for accuracy tasks such as initially calibrating the MOS sensors, both the LAS and the calibrated MOS sensors can measure precise changes in methane detection and successfully complete the goal of monitoring methane emissions.

Figure 1. A comparison of methane concentrations as measured by two SOOFIE sensors (with calibrated MOS sensor) and one LDS Analyzer (Aeris Technology Strato)

To prove the capability of the MOS sensors, we conducted a side-by-side field comparison (Figure 1) of the SOOFIE sensors (with MOS sensor) and an Aeris Technologies Strato LDS (with LAS sensor). As shown in Figure 1, the MOS sensors are reading a much higher absolute methane concentration, but the change in methane concentration is very precise and comparable to the LAS sensor.

Figure 2. A comparison of methane concentrations as measured by two SOOFIE sensors (with MOS sensor) and one LDS Analyzer (Aeris Technology Strato) after an applied baseline.

After the removal of a baseline value in Figure 1, the performance of the calibrated MOS sensors is comparable to that of the LAS sensors, as seen in Figure 2.

2. Detection in a Timely Manner

The time to detection depends on several variables, most of which are outside of our control. First, both the ability and speed of detection depends on the wind direction, which is outside of our control, at the time of a leak relative to the location of a given sensor. A change in the wind direction can aid in a larger area of detection for a leak; however the timing of this directional shift and the speed of detection due to a wind shift are also out of our control. Although one can take advantage of a typical wind pattern to optimize the placement of sensors, even with the optimal placement, one sensor cannot quickly predict all potential leaks at a given site across all wind conditions.

In addition to optimized sensor placement, one can increase the number of sensors deployed to assist in fast detection. This number is dependent on numerous factors, including site size and project budget. Given that the MOS and LAS sensors have comparable performance in their ability to detect leaks, as presented in Figure 2, for an equal project budget, the lower price of the MOS sensors allows for the placement of >5x more sensors at optimal detection locations than the LAS sensors. A higher spatial distribution creates a higher density of data across the site, which thus creates a more accurate spatial and quantitative resolution of site emissions.

Conclusion

We acknowledge that MOS sensors are not the ideal sensors to use for applications requiring highly accurate absolute methane detection concentrations. However, for the goal of continuous emission monitoring and quantification, the MOS sensors offer a low-cost and efficient solution for detecting leaks across a site that allows for a higher spatial and temporal resolution for tracking methane emissions.