Apex Haus
Published in

Apex Haus

Carbic and the future of Oil Field Management

Founded in 2015, Carbic deploys sensor technology to the oil patch to reduce operating costs and GHG emissions.

By: Chelsea Goddard & Stuart Farris

A nodding donkey rig pumps crude oil up from the ground on an oil field


Carbic brings machine learning (ML) and Internet of Things (IoT) innovation to the oil patch to simultaneously reduce operating costs and GHG emissions.

  • Their sensors allow oil and gas producers to catch wasteful leaks that account for hundreds of millions in lost revenue annually and reduce overall operational costs.
  • Their sensors provide real-time flow volume, pressure, and temperature information that oil and gas companies can use to optimize their operations, reducing production costs, and emitting less GHG in the process.


The rapidly growing share of fossil fuel, electric utility, and automotive companies adopting ambitious climate commitments reveals that many are beginning to accept their role in the climate crisis we all face. Companies are starting to receive pressure from their financiers, like JPMorgan, who are distancing themselves from “companies that do not align their operations with the Paris climate accord” and countries — from Europe to China — are adopting ambitious emission targets⁹. Yet, the path to carbon neutrality or 100% renewable energy can be significantly harder for certain emissions heavy industries. The fossil fuel industry is an obvious example, as the entire process is carbon-intensive. Extraction, transportation, and use of their products all generate GHG emissions. Some O&G companies, like BP and Shell, are trying to reinvent themselves by shifting from vertically integrated hydrocarbon companies to more broadly focused energy companies that invest heavily in wind, solar, and biofuels. With that said, there are more straightforward changes that fossil fuel companies can make that don’t require a complete shift in their business models.

Improving operations and monitoring methane leaks, for example, is a seemingly straightforward change that not only reduces GHG emissions but can increase revenue and reduce production costs. So how are companies currently monitoring leaks? The Environmental Protection Agency says the oil and gas industry emits eight million metric tons of methane a year. Recent studies have suggested that the wasted gas is enough to fuel 10 million homes for a year — that’s worth an estimated $2 billion.


Currently, there is an emerging two-body problem for fossil fuel companies where A) decreased demand for oil means US players need to compete at lower oil prices, and B) societal pressure is placing historic pressure on operations to decrease GHG footprints.

Fossil fuel companies spend billions of dollars extracting oil and gas. As demand peaks and begins to fall, the natural assumption is that the lowest cost producers will survive. Right now the Middle East and Russia have an advantage over the US in that their costs of production are much lower¹⁰. This implies that, in order to stay competitive, US players need to reduce the amount they spend extracting a given barrel of oil and cubic foot of gas.

In addition to the operational overhaul of cutting costs, these companies also need to curtail fugitive leaks. To remain viable, energy companies need options that will lead to a low emissions future, reduce their GHG footprint, and save product. We know in industry, you can not monitor what you do not measure — in the Oil & Gas industry this starts with measuring leaks.


Carbic’s non-intrusive sensors are easily installed on wellheads, pipes, or storage tanks and are powered by a self-contained battery and solar panel. Their technology works by emitting high-frequency sound waves into pipes and tanks, then recording the returning sound waves which have bounced off flowing fluids and particles. A small onboard computer performs some minor processing before sending the data to more powerful, offsite computers via a satellite connection. Once powerful edge nodes receive the sensing data, they pass the information to ML algorithms which extract the flow information. Alongside Carbic’s sensor technology, users are also able to see this data in Carbic’s online interface.

Carbic wireless acoustic sensors will track total production from wellhead to refinery.

So, how can real-time flow information reduce emissions and save companies money? The most simple use case would be performing fluid balance tests, which ensures the volume of hydrocarbons coming out of the wellhead is the same volume going through pipelines and into storage tanks. If there is a fluid balance discrepancy, there is most likely a leak somewhere along the line between Carbic sensors. Patching leaks keeps valuable resources in the pipe and out of the atmosphere. Companies can also benefit from more precise measurements of stimulation fluids injected into the earth and oil and gas flowing from the earth. With more accurate flow information, companies can optimize production efforts by using fewer resources (like water, which is not-so-surprisingly very expensive in the desert of West Texas and the like) to extract more hydrocarbons. And ultimately, this means that if wells can be significantly optimized, fewer wells will need to be drilled.

Market size

The natural gas market in the United States exceeds $100 billion annually and is projected to increase in the coming decades¹. The retirement of coal power plants will increase the demand for natural gas as bridge fuel towards zero net emission targets. Let’s break this down further.

There are over 9,000 independent oil and gas companies as well as major players, like ExxonMobil and Chevron that produce 38 trillion cubic feet (tcf) of gas per year³. The EPA, US Energy Information Administration, and Environmental Defense Fund, estimate ~2.3%, or 0.88 tcf, of produced natural gas is leaked into the atmosphere in the US every year⁴. Assuming moderate market prices of $2.75/thousand cubic foot, that is $2.6 billion of leaked product every year in the US alone. This represents the $2.6 billion of lost revenue that Carbic’s sensors target in the natural gas industry.

The oil industry is even larger than the gas industry. The US produces 12 million barrels of oil per day⁷. In 2016 the production costs per barrel was ~$5 meaning around ~$20 billion is spent on oil production in the US every year⁸. Beyond catching leaks, Carbic’s sensors can optimize the production process and bring the cost/barrel down. Quantifying how much these sensors optimize the production process may be difficult, but even a meager few percent decrease in the cost/barrel would amount to saving the oil industry billions of dollars.

Market strategy post-Covid

Since 2015 more than 200 O&G companies in North America have filed for bankruptcy. This year alone, 18 companies have defaulted on their debts with analysts projecting this number to rise. Once a dominant driver of trade (accounting for more than 15% of the S&P in 2008), the onset of the pandemic coupled with overproduction has slowed demand for oil & gas and collapsed prices. When Covid hit, “global oil demand fell by more than 30 million barrels per day” and prices dropped by more than 40%. This has resulted in an overall shift away from “oil for consumption” to “oil for production” (specifically for single-use plastics; the sector with the highest expected growth in demand for oil).

US crude oil price turned negative for the first time on record in February 2020.

At a high-level, the first pivot Carbic made was away from operators in the United States to a focus on global oil players. This is the right play given that O&G companies that survive will be the nationalized players in Russia and the Middle East. The more interesting strategic questions, however, are beyond this year; when we will see startups move from reactive responses to calculated measures.

For a company like Carbic, there is a strong likelihood that their customer base contracts as a result of larger industry trends. This signals a clear need to heavily invest in R&D initiatives that can account for future use cases outside the O&G industry. Two interesting possibilities include CO2 removal and green hydrogen.


  • Jed Somers, CEO: Wharton; Analyst at Barclays
  • Jeffrey Garoon, COO: BSc, Mechanical Engineering; Oxy; FlowCommand


Ryan Littlefield, a former Facilities Engineering Lead at Occidental, shares his perspectives on the product viability of Carbic, the potential challenges, and his thoughts on their go to market strategy and future.

Can you walk us through the oil extraction process and at what point emissions occur? How does Carbic help alleviate this?

There are two major categories of emissions: carbon dioxide emissions and methane emissions. Methane emissions do far more damage to the environment than carbon dioxide per emitted unit (anywhere from 5X to 30X the effect, depending on the metric). Carbon dioxide emissions primarily occur when gas is burned. This happens at natural gas power plants, when gas is flared, and even on your gas stove burner in your kitchen. Methane emissions occur when gas is released directly into the atmosphere without being burned. Small sources of methane emissions include fugitive leaks at valves and when using a pneumatic automation system. These emissions are relevant and important to consider, but by far the largest source of methane emissions is due to oil tank venting.

When an oil well comes online, it is producing a mix of oil, water, and gas. Surface facilities are used to separate these fluids so they can be sold or sent to disposal. At each stage of separation, the pressure is reduced, which causes more gas to evolve off of the oil. This gas is naturally driven into the gas gathering system when the operating pressure of the separation vessel is greater than the pressure of the gas gathering line. In the final stage of separation, the oil is sent into an atmospheric tank. Because these tanks operate at atmospheric pressure, a compressor must be used to flow the gas into the gas gathering system. When the compressor goes down, operators are faced with a difficult decision. They can either shut in all the production going to the facility or allow the tanks to vent to the atmosphere. Unfortunately, the more economic choice is to vent the gas. Operators are allowed a certain amount of methane emissions according to the permits they file with the state or federal governments, so this is legal practice that is primarily reduced by installing backup compressors. I believe that the best way to reduce these emissions is by eliminating the need for tanks altogether through the use of new technology and clever operational practices.

Carbic’s sensors can provide information on whether or not gas is flowing out of the tanks into the compression system. However, operators generally already know this by tracking when the compressor is on or off. Even if this was not the case, the operator may still choose to continue producing even when they know that gas is venting from the tanks. I do believe Carbic’s sensors provide low-cost, high-value data for optimizing oil production, but I do not see them fundamentally changing an operator’s emissions.

What do you see as the biggest risk to the product viability? How should Carbic navigate this?

Carbic faces a large challenge in data accuracy vs data value. When an operator decides to trial this technology, they are often overly focused on comparing the readings from this meter to older technologies, such as coriolis or turbine meters. This causes the operator to be overly focused on determining the inevitable discrepancy between the two data sources. The challenge is summed up well by Segal’s law, which states “A man with a watch knows what time it is. A man with two watches is never sure.” When too much focus is put into determining which “watch” is giving the correct time, then the value of the data is completely lost.

The benefit of Carbic’s meters is they are easy to install, low cost, and require no infrastructure to transfer the data into a SCADA system. This means an operator can easily monitor their entire lease or asset and determine which wells are producing above average and which ones are below average. They can then focus their efforts on the right wells to maximize impact and respond to issues more immediately.

If Carbic is able to succeed, what do you anticipate as the broader impact on the oil & gas industry at large?

There is a lot of inefficiency in current oil & gas operations. Although operators do collect and analyze a lot of data every day around production efficiency, there are always major gaps that need to be filled in in order to perform more advanced machine learning based optimizations. Carbic provides the means to gather this additional data so the industry can produce more oil at a lower cost basis, which will allow prices to continue to be driven down.

This story was authored independently from Carbic and their investors and represent the thoughts and opinions of Chelsea Goddard and Stuart Farris. Thank you to Jeff Williams, Jed Somers, and Ryan Littlefield for their time.



[2] https://www.eia.gov/todayinenergy/detail.php?id=43115



[5] https://www.eia.gov/todayinenergy/detail.php?id=43115

[6] https://science-sciencemag-org.stanford.idm.oclc.org/content/361/6398/186.full

[7] https://www.eia.gov/dnav/pet/hist/LeafHandler.ashx?n=pet&s=mcrfpus2&f=a

[8] http://graphics.wsj.com/oil-barrel-breakdown/

[9] https://www-ft-com.stanford.idm.oclc.org/content/e1be8a23-1c80-43dd-be7b-18636ff61c46




Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store