Grantee Research Project Results
Final Report: Storage Tank Abatement of Methane using Persistent Emissions Detection (STAMPED)
EPA Contract Number: 68HERC22C0013Title: Storage Tank Abatement of Methane using Persistent Emissions Detection (STAMPED)
Investigators: Dobler, Jeremy
Small Business: Spectral Sensor Solutions LLC
EPA Contact: Richards, April
Phase: I
Project Period: December 1, 2021 through May 31, 2022
Project Amount: $100,000
RFA: Small Business Innovation Research (SBIR) Phase I (2022) RFA Text | Recipients Lists
Research Category: Small Business Innovation Research (SBIR) , SBIR - Air
Description:
The primary research objective of this project was to evaluate the ability of a laser-based continuous monitoring technology, known as GreenLITETM, for detecting, localizing, and quantifying methane leaks that may be associated with oil and gas storage facilities. Large uncertainties associated with emissions from such facilities currently exist, primarily due to limited measurements and challenges associated with measuring emissions from these sources using existing technology.
GreenLITETM is designed to monitor either methane (CH4) or carbon dioxide (CO2) over large areas (0.25 - 25 km2 for the current system, up to 50 km2 is possible) in near-real-time and is unique in its ability to provide an estimate of the spatial distribution of the gas over the area being monitored. The system consists of at least two transceivers and a set of retroreflectors, generating a series of typically horizontal, intersecting integrated-path chord measurements between each transceiver and the retroreflectors. The intersecting chords enable a sparsely-sampled tomographic reconstruction of the gas concentration using either a Gaussian plume-based model or an area box model. The hardware and software comprising the measurement portion of the GreenLITETM system were developed by Spectral Sensor Solutions LLC (S3), and the analytics and data dissemination were developed by Atmospheric and Environmental Research (AER).
The key milestones for the program were:
- Plan the required deployment of a GreenLITETM system to the S3 Farm test site needed to demonstrate the accuracy of the leak detection and quantification (LDAQ) in alignment with the established Continuous Monitoring Protocols (CMP) from the Department of Energy funded Advancing Development of Emissions Detection (ADED) program and document the test procedures.
- Work with the Methane Emissions Technology Evaluation Center (METEC) personnel to have the test plan and setup audited prior to start of testing.
- Deploy the system as planned for a series of blind releases with a range of emission rates and area extent to simulate realistic industrial tank leakage. Releases were to be randomized over time of day, location, leak rate, and extent of the source over a period of 2.5-3 months to ensure a wide range of conditions were captured.
- Perform analysis to report detections and quantify/localize detected leaks according to the ADED protocols.
- Compare the detection and emissions results to blind values post-analysis per the ADED protocols using the CMP metrics.
- Prepare final report detailing results versus actual release rates and the metrics outlined in the ADED protocol.
For this research project a test plan for a single-blind study using controlled releases within an area representative of a large oil and gas storage tank farm was designed and executed. The test plan was established to enable the system under test to be evaluated against a set of metrics established by METEC. The metrics were developed for the ADED program led by METEC and are captured in the CMP developed through that program. (see https://energy.colostate.edu/wp-content/uploads/sites/28/2021/03/Continuous-Monitoring-Protocol-R1.0.pdf). The protocol was developed to assess the ability of various technologies to detect, localize, and quantify emissions as end-to-end systems, including all aspects from hardware to analytics, and was designed to be transferrable to other test centers.
The lead scientist from METEC, Dr. Clay Bell, served as an advisor for the test plan development and ensured alignment with the ADED protocols. Dr. Bell also provided an onsite audit of the test setup for the single-blind testing that was conducted at the S3 Farm test site located near Fort Wayne, Indiana (Figure 1).
Figure 1. Layout of GreenLITETM system at the S3 Farm test site. T01 and T02 are the two transceiver locations, R01-R29 are the reflector locations, green are the chords measured by T01, and blue are the chords measured by T02. The red boxes are used as bounding boxes for computing emissions and providing an estimated location of the emission.
Summary/Accomplishments (Outputs/Outcomes):
The results from this single-blind study show that GreenLITETM is capable of detecting emissions as small as 5 standard liters per minute (SLPM) and that, for average wind speeds <5 m/s, it had a 100% detection rate for leaks ≥15 SLPM. Of the 42 releases, 25 were detected as emission events. Releases were localized to the correct equipment unit 9 times, to the correct equipment group 9 times, and elsewhere within the test facility 7 times. Quantification of the 25 detected emissions was within 11.1% on average (average of positive and negative estimate errors) with a standard deviation of 82%. The absolute percent difference for all emissions estimates (average of absolute values of estimate errors) was 58.6%.
A histogram of the releases as a function of release rate is shown in Figure 2. More releases were performed at the lower rates due to a lower probability of detection at those rates, and higher rates were also limited by the quantity of gas available within the program resources and were conducted for shorter release periods (releases in this test series ranged in duration from 12.2-60.4 min with an average of 22.8 min).
Figure 2. Distribution of the 42 releases as a function of release rate.
A total of 42 single-blind controlled releases were performed during a period of 6 weeks. Figure 3 (below) shows bootstrapped logistic-regression probability-of-detection curves based on the detections during any of the 42 releases. The bootstrapping technique envelopes the expected detection performance of the system under test and is performed by randomly selecting N detection samples out of N actual detections (with replacement) prior to performing the logistical regression. This process is repeated 100 times to build the grey curves seen in Figure 3. The right-hand panel provides another way to look at the blind test data detection results, where a logistic regression with bootstrapping was performed, but normalizing the release rate by the average wind speed during the release periods. This normalization results in significant tightening of the spread of the bootstrapped curves, indicating wind speed is one of the primary drivers of variability in the detection results. The plot on the right has only discreet values for the x-axis due to the normalization, which is why there are only probabilities of detection of 0 for no detection and 1 for a detection.
Figure 3. Probability-of-detection curve by release rate (left) and by release rate normalized by wind speed (right).
Conclusions:
The Storage Tank Abatement of Methane using Persistent Emissions Detection (STAMPED) Phase I program accomplished all of the proposed objectives. The GreenLITETM system was evaluated through a series of single-blind tests over a range of environmental conditions. A total of 42 controlled blind releases were performed over 27 test days spanning 6 weeks, while the system operated autonomously over an area of ~0.35 km2. The results of the blind testing were evaluated using the metrics established by the ADED CMP. Probability-of-detection curves were established along with evaluation of the ability of the system to localize and quantify detected emission events. For wind speeds less than 5 m/s, all releases ≥15 SLPM were detected, and emissions sources as small as 5 SLPM were detected with a lower probability. The results of this work will enable comparisons of the performance of GreenLITETM against other continuous monitoring methodologies being tested using the ADED protocol. This research shows that GreenLITETM provides a 90% probability of detection for leaks greater than or equal to 20 SLPM under all conditions tested.
Additional testing is needed that includes the presence of multiple emissions sources, allowed emission events, and larger emission rates in order to fully assess the capabilities of the system for use in quantifying emissions from oil and gas storage tank facilities. However, the results obtained from this research effort demonstrate feasibility of the GreenLITETM system to offer continuous monitoring of relatively large storage tank facilities and to provide localization and quantification of the sources of emissions with high-sensitivity.
The GreenLITETM technology is uniquely positioned to provide continuous monitoring for large area (0.25-50 km2) sources while also providing maps of the spatial distribution of the gas concentration and emissions. The patented technology has application for compliance monitoring in the oil and gas industry upstream and mid-stream processes. Other horizontal markets include: transportation sector, such as large railyards, airports, and shipping ports; industrial sector, where understanding emissions and processes that drive emissions is needed for large industrial facilities like cement factories, steel mills, etc.; mining sector, to track total emissions from large mining operations like coal, oil sands, and precious metals mining; and agricultural sector, for evaluating emissions of large feedlots, biogas facilities, or in field range cattle. Other opportunities exist for long-term monitoring of carbon capture and storage facilities, landfills, wetlands, and reservoirs.
The perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.