Review on Airborne quantification of Angolan oil and gas methane emissions

Overview

The authors present an important addition to the literature on offshore oil and gas GHG emissions by expanding observations to Angola for the first time. They conducted an extensive airborne campaign and carefully calculated mass balance observations that they compare with inventories. The implications of their results can be better highlighted with a more careful presentation of their findings and a more integrative comparison with the offshore literature. This includes (A) a more thorough description of the shallow water facility clusters to determine if they are independent facilities or groups of dependent facilities as previously described in other papers, (B) propagating their shallow vs deep water trends into the final figures, including in their discussion on carbon intensity, and how these trends compare to the literature, and (c) raising the profile of CO2 in their story as it appears to contribute to over half of the CO2eq emissions, including in the abstract, body text, figures, and maybe a new figure.

Specific Comments

Comment 1.

This study focuses on methane, but estimates and evaluates carbon dioxide fluxes as well. In addition, CO2 is combined with CH4 into an estimate of carbon intensity. This is relatively rare in the airborne oil and gas literature as this field has historically assumed that O&G production field methane emissions are more important (with GWP scaling) and less well known. This is not necessarily always the case based on the few studies that measure carbon dioxide (see below). In fact, this studies’ CO2 flux of 613 t/h and CH4 flux (with GWP scaling 16.8 X 29.8 = 476.8 t CO2eq/h) shows that CO2 and CH4 both contribute to the carbon intensity of the basin with CO2 contributing a possibly larger fraction. However, this is not shown or explicitly stated, but is an interesting result.

I suggest the authors consider highlighting the CO2 side of their story more in the introduction, show what fraction of the carbon intensity is driven by CH4 and CO2 (for both shallow and deep water), and perhaps, if they think it is within scope, compare with other carbon intensity estimates in other basins. Perhaps a final carbon intensity figure may help communicate this.  

Relevant Literature on O&G CO2 and carbon intensity

1. Liggio J et al 2019 Measured Canadian oil sands CO2 emissions are higher than estimates made using internationally recommended methods Nat. Commun. 10 1863
2. Aircraft and satellite observations reveal historical gap between top–down and bottom–up CO2 emissions from Canadian oil sands https://academic.oup.com/pnasnexus/article/2/5/pgad140/7127723
3. Measurement-based carbon intensity of US offshore oil and gas production https://iopscience.iop.org/article/10.1088/1748-9326/ad489d

Comment 2.

Line 63-66. Can the authors back these statements up with numbers? What is the contribution of Africa or at least Nigeria and Angola to the global O&G industry. Why do we think there are major CH4 emissions arising from the production of these resources? Are there onshore studies that have already demonstrated that?

Following lines. How does the oil and gas production in Angola compare with other fields? How much is there? And how much is reinjected? Re-injection is an energy intensive process. Can this explain the CO2 emissions from the deep water facilities?

Comment 3.

Line 70-72. The text argues that the processing operations for FPSO hydrocarbons occurs at the onshore LNG facility, but the older shallow water facilities do not send hydrocarbons to the LNG facility. Is the correct interpretation that the processing for deep water occurs onshore and the processing for shallow water occurs onsite offshore? This paper finds higher CH4 emissions from shallow water facilities compared to deep water, but is that an artifact of the fact that processing emissions (where the majority of offshore CH4 emissions seem to occur) are just exported to the LNG facility? Are there preliminary processing operations occurring on the FPSOs (like treaters and dehydrators separating the water, gas, and oil) or do even those operations occur at the LNG facility. Can the authors discuss this more?

Was the LNG facility sampled in another flight? If yes, can it be included in this story to round out the processing emissions section of the story?

Comment 4.

Line 83. There are some additional citations for airborne offshore methane studies that the authors should consider adding.

New regions

Zang, Kunpeng, Gen Zhang, and Juying Wang. "Methane emissions from oil and gas platforms in the Bohai Sea, China." Environmental Pollution 263 (2020): 114486.

- They report concentrations and a regional flux

Gorchov Negron, Alan M., Eric A. Kort, Genevieve Plant, Adam R. Brandt, Yuanlei Chen, Catherine Hausman, and Mackenzie L. Smith. "Measurement-based carbon intensity of US offshore oil and gas production." Environmental Research Letters 19, no. 6 (2024): 064027.

-They add offshore Alaska and California in the US

More Gulf of Mexico studies

Gorchov Negron, Alan M., Eric A. Kort, Stephen A. Conley, and Mackenzie L. Smith. "Airborne assessment of methane emissions from offshore platforms in the US Gulf of Mexico." Environmental science & technology 54, no. 8 (2020): 5112-5120.

Ayasse, Alana K., Andrew K. Thorpe, Daniel H. Cusworth, Eric A. Kort, Alan Gorchov Negron, Joseph Heckler, Gregory Asner, and Riley M. Duren. "Methane remote sensing and emission quantification of offshore shallow water oil and gas platforms in the Gulf of Mexico." Environmental Research Letters 17, no. 8 (2022): 084039.

Biener, Kira J., Alan M. Gorchov Negron, Eric A. Kort, Alana K. Ayasse, Yuanlei Chen, Jean-Philippe MacLean, and Jason McKeever. "Temporal variation and persistence of methane emissions from shallow water oil and gas production in the Gulf of Mexico." Environmental Science & Technology 58, no. 11 (2024): 4948-4956.

-They compare with satellite like your study here so may be a good comparison point.

Comments on Measurements and Mass Balance.

Comment 5.

Figure 1. Since no lat or lon is offered in Figure 1a, can the authors include points of the offshore facilities or show a separate map of facilities? The facilities are shown later in the inventory figures and it would make sense to not cover up the aircraft flights with the facilities, but showing them all together might make for an informative figure of what the campaign looked like. In this map, can the authors include a shallow vs deep water dividing line and the LNG facility?

Comment 6.

Line 600 & Figure A2. Can the authors show what CH4 and CO2 look like in the vertical profiles? Since these are your target gases, their level of mixing in the boundary layer is the most important consideration. If these are included these in the criteria would it change your decisions on the height of the PBL or whether the PBL is mixed enough?

Comment 7.

Can the authors include a figure (perhaps a multi-panel plot of multiple examples, perhaps in the appendix), showing what the downwind concentrations used for mass balance looked like (something like distance or longitude for the X-axis, and concentration for the Y-axis). Perhaps the authors can include an example that shows what part of the measurements were used for the background. In at least one example, can the authors show what the upwind concentrations looked like as well? The paper states that upwind transects are rare and not used as the background, but were low and therefore the sides of the downwind plumes were sufficient to be used as a background. It would be good to show this.

Comment 8.

Line 255. What are the satellite products in the IMEO data portal that provided positive detections in your domain in this study? Who estimated the methane emissions (the satellite data product creator/team?, a team at IMEO?). How was uncertainty assigned to these?

Comment 9.

Line 578. How does the method create more flux uncertainty with fewer transect flown?

More Comments

Comment 10.

Figure 3. How many facilities are in each facility group/cluster?

Comment 11.

Line 299. The text states that the operators reported normal operations during high-emission events and therefore, they must be unaware of their high emissions. I’m not sure this logically flows and it implies that the emissions must be unknown fugitives. Operators can emit large volumes under normal operating conditions when especially they intentionally vent gas. This is one of the driving sources of regular intermittent emissions in the Gulf of Mexico (relevant studies cited elsewhere).

Comments on Context around Deep vs Shallow water.

Comment 12.

Line 355+. The authors highlight how facility condition (e.g. age and type) are more predictive of emissions. Can the authors include photos of the facilities and highlight how deep water and shallow water visibly look different? What do the groups of facilities in shallow water look like? If photos were unfortunately not taken during the campaign, can the authors track down operator or satellite imagery of the facility? The importance of this is emphasized in the following comments. There needs to be more work to characterize what the shallow water infrastructure looks and behave like.

Comment 13.

One major conclusion is that older, low producing shallow water facilities had higher emissions than younger deep water facilities. Research in the Gulf of Mexico also found this trend. This is the second oil and gas basin with (a) clearly distinct shallow water (old) and deep water (new) fields that was also (b) sampled with aircraft. It agrees with the Gulf of Mexico shallow-deep water finding. I suggest the authors consider discussing whether this marks a trend.

Comment 14.

Old shallow water facilities across the globe tend to look similar (based on those visible in google earth). They are composed of a large multiplatform central processing facility (composed of vents, flares, compressors, etc) surrounded by smaller satellite production facilities that have almost no infrastructure and basically serve as a well-head. The study on Southeast Asia (Nara), the US Gulf of Mexico studies (Gorchov, Ayasse, Bjeiner), and a satellite Mexican Gulf of Mexico study (see below) all sampled this specific type of infrastructure, and the Southern North Sea paper (Phul), may have sampled at least on of these facilities. These studies agree in that they found relatively high methane emissions with some identifying highly intermittent emissions from these facilities. The Gulf of Mexico studies also found higher shallow water emissions (due to this class of old facility) than deep water.

In this study, the clusters of facilities in shallow water had high methane emissions that were intermittent, so this raises the question of whether this is the same type of facility as in these other studies.

Do your photos show these facility groups to match this description? If you do not have photos and cannot see the facilities from google earth pro/satellite products, what do the pipelines show? If there are pipelines transporting hydrocarbons to one facility from all of the others, this likely matches this facility class. Are you sure these are groups of separate facilities or are they different parts of one multiplatform facility connected by a cat-walk? If it is a multiplatform facility, then it might also match this facility class.

 If yes, I suggest the authors (A) discuss this, contextualize the finding with trends in the literature, and note how previous studies of this facility class also found high intermittent methane emissions. Venting was found to be one of various sources of intermittent high emissions. Do photos show vents on these facilities? This assumes photos were taken during the airborne campaign.

I also suggest the authors (B) consider whether dividing by the count of facilities in the cluster is misleading since they are not independent and treat the cluster as one unit in their denominator for emissions per facility. Please note that the emissions/unit for the Gulf of Mexico calculation treated facility grouping of this specific design as 1 unit (so 1 processing facility with 20 satellite production facilities was considered to be 1 facility). If the cluster of facilities match this type of infrastructure, then make sure the comparison is consistent.

Irakulis-Loitxate, Itziar, Javier Gorroño, Daniel Zavala-Araiza, and Luis Guanter. "Satellites detect a methane ultra-emission event from an offshore platform in the Gulf of Mexico." Environmental Science & Technology Letters 9, no. 6 (2022): 520-525.

Comment 15.

Section 3.4. I suggest you consider if your satellite vs airborne story is similar to satellite vs airborne comparisons in…

Biener, Kira J., Alan M. Gorchov Negron, Eric A. Kort, Alana K. Ayasse, Yuanlei Chen, Jean-Philippe MacLean, and Jason McKeever. "Temporal variation and persistence of methane emissions from shallow water oil and gas production in the Gulf of Mexico." Environmental Science & Technology 58, no. 11 (2024): 4948-4956.

Comments on Presenting Final Results.

Comment 16.

Line 425 & Figure 9. If you are using the emissions/facility reported in Figure 4 from Pühl et al. (2024) for the comparison, please note that the emissions/facility for the Gulf of Mexico are from the 2020 Gulf of Mexico paper and not the 2023 paper.

Gorchov Negron, Alan M., Eric A. Kort, Stephen A. Conley, and Mackenzie L. Smith. "Airborne assessment of methane emissions from offshore platforms in the US Gulf of Mexico." Environmental science & technology 54, no. 8 (2020): 5112-5120.

The 2023 paper is more extensive and can be used to make a complete ratio for shallow water and deep water. The 2020 paper sampled a fraction of GOM facilities so the balance of shallow vs deep water will change that ratio.

Comment 17.

Figure 9. Since the deep water and shallow water regions have such different infrastructure and emissions, I suggest also making a separate comparison with a deep water emissions/facility and a shallow water emissions/facility. Perhaps it may be interesting to compare an emissions/facility for the deep water gulf of Mexico and shallow water Gulf of Mexico to see if they match.

Comment 18.

Line 440. How is error calculated for total emissions with the sum of the mean?

Comment 19.

If the deep water and shallow water systems are different, can you estimate a separate carbon intensity for each in addition to the combined carbon intensity? Are the deep water facilities dominated by CO2 and the shallow water facilities dominated by CH4? How does this compare to the deep water and shallow water carbon intensity and GHG breakdowns in at least the Gulf of Mexico?

Comment 20.

Line 495. Do the authors include reinjected gas in the production denominator for carbon intensity and methane intensity? Be careful with this. I’m of the mind that you should not include it as what policy makers care about is how much emissions come from the marketed oil and gas.

Comment 21.

Line 645-Data accessibility. Will the authors share their flux data too? This data could be useful for meta-analyses. I know there is caution about sharing operator information. Would publishing flux data with just lat and lon get around that?

Technical Comments.

Comment 22.

Line 37-The wording is a little redundant with second most important GHG point in line 34.

Comment 23.

Line 159-vertical or horizontal transect? Is the correct reading… “Summing up the fluxes from all *horizontal* transects, *vertically*,….”? Currently it reads as summing up vertical transects.

Comment 24.

Figure 2. Arrows are hard to see.

Comment 25.

Figure 4. This is a log plot so hard to see if daily variability is in fact generally low or not.