Development and validation of a route planning methodology for vehicle-based remote measurements of methane and other emissions from oil and gas wells and facilities

ABSTRACT

Multi-sensor vehicle systems have been implemented in large-scale field programs to detect, attribute, and estimate emissions rates of methane (CH₄) and other compounds from oil and gas wells and facilities. Most vehicle systems use passive sensing; they must be positioned downwind of sources to detect emissions. A major deployment challenge is predicting the best measurement locations and driving routes to sample infrastructure. Here, we present and validate a methodology that incorporates high-resolution weather forecast and geospatial data to predict measurement locations and optimize driving routes. The methodology estimates the downwind road intersection point (DRIP) of theoretical CH₄ plumes emitted from each well or facility. DRIPs serve as waypoints for Dijkstra’s shortest path algorithm to determine the optimal driving route. We present a case study to demonstrate the methodology for planning and executing a vehicle-based concentration mapping survey of 50 oil and gas wells near Pecos, Texas. Validation was performed by comparing DRIPs with 174 CH₄ plumes measured by vehicle surveys of oil and gas wells and facilities in Alberta, Canada. Results indicate median Manhattan distances of 145.8 m between DRIPs and plume midpoints and 160.3 m between DRIPs and peak plume enhancements. A total of 46 (26%) of the plume segments overlapped DRIPs. Locational errors of DRIPs are related to misattributions of emissions sources and discrepancies between modeled and instantaneous wind direction measured when the vehicle intersects plumes. Although the development of the methodology was motivated by CH₄ emissions from oil and gas facilities, it should be applicable to other types of point source air emissions from known facilities.

Implications: This paper presents and validates a method that addresses the challenge of measuring industrial emissions from public roads. The method can increase the effectiveness and efficiency of targeted vehicle-based emissions surveys where the locations of potential sources are known. We believe the method has broad application in addition to the upstream oil and gas context it was designed for.

Acknowledgment

The authors thank Clay Wearmouth, Thomas Fox, Adam Boulding, and Brynn Tarnowsky for operating PoMELO during Summer 2018. The authors thank Marshall Staples for operating PoMELO on 21 February 2020. The authors also acknowledge funding support from the Canada First Research Excellence Fund and the Global Research Initiative in Sustainable Low Carbon Unconventional Resources (GRI).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The authors confirm that the data supporting the findings of this study are available within the article and its supplementary materials. Additional data are available by request.

Code availability

The route planning and analysis were programmed in Python with standard packages. The results can be reproduced by employing the equations, explanation, and parameters provided in the main text. Additional code and data are available by request.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/10962247.2022.2113182

Additional information

Funding

This research was undertaken thanks in part to funding from the Canada First Research Excellence Fund and the Global Research Initiative in Sustainable Low Carbon Unconventional Resources (GRI).

Notes on contributors

Mozhou Gao

Mozhou Gao is a Ph.D. candidate and Research Assistant at the University of Calgary.

Chris H. Hugenholtz

Chris H. Hugenholtz is a Professor and Director of the Centre for Smart Emissions Sensing Technologies, University of Calgary.

Thomas Barchyn

Thomas Barchyn is a Research Project Manager at the University of Calgary.