14 Jul 2026
Subsurface Detective: 40 Years of Reservoir Engineering Secrets with Jim Clark
A 40-year reservoir engineering veteran breaks down how oil and gas companies actually decide what's underground — and why AI still can't replace the domain knowledge it takes to get it right. Jim Clark, CCS Reservoir Analytics Engineer at JWC3 Resources, joins Local Energy to explain reservoir engineering as detective work, the real economics of carbon capture, and how a $30 billion gas project came down to a single well-log calculation.Jim spent decades working reservoirs from the Gulf of Mexico to deep-water East Africa before moving into CCS (carbon capture and sequestration) analytics. In this conversation with hosts Peter Brecht and Wade Speer, Jim explains why he calls himself a "subsurface detective," walks through the machine learning model he built to predict 20-year gas recovery with 98% accuracy, and unpacks why capturing carbon costs more than storing it. The episode also covers CCS vs. CCUS, well integrity risk for CO2 injection, 45Q tax credits, and the surprising link between data centers and the next wave of carbon storage projects.CHAPTERS00:00 Intro: Meet Jim Clark, 40-Year Reservoir Engineering Veteran01:16 Why a Reservoir Engineer Is a Subsurface Detective03:22 Reading Rock: The Clues Hidden in a Single Core Sample06:13 Before the Bit Drops: How Reservoir Models Get Built Pre-Drill10:21 The Real Signals That Decide When a Well Gets Shut In13:57 The $30 Billion Well: Inside a High-Stakes East Africa Decision19:38 Why AI Still Can't Replace Domain Knowledge21:26 Building an AI Reservoir Advisor to Train Engineers Faster24:40 Prompt Engineering, Hallucinations, and Trusting the Machine27:42 CCS vs CCUS: The Difference Almost Everyone Gets Wrong30:53 The Real Reason Carbon Capture Costs $100 a Ton32:18 Old Wells, New Risk: Why Well Integrity Matters for CO2 Storage36:06 The 45Q Tax Credit Math Behind CO2 Flooding38:20 Data Centers, Green Power, and the Next CO2 Storage Boom41:41 Career Advice: What Jim Would Tell a Future Reservoir EngineerKEY TOPICSReservoir engineering fundamentals and the "subsurface detective" mindsetUsing core samples and pressure data to infer reservoir boundariesBuilding simulation models before and after drillingMachine learning models that predict 20-year well productionWhy domain knowledge is irreplaceable when using AI toolsPrompt engineering and avoiding AI hallucinationsCCS vs CCUS definitions and how they actually differThe true cost breakdown of carbon capture, including the $100-a-ton benchmarkWell integrity risk for CO2 storage in aging wellbores45Q tax credits, CO2 flooding, and enhanced oil recovery economicsHow data center power demand is reshaping CO2 storage opportunitiesCareer advice for aspiring petroleum and reservoir engineersLINKS & RESOURCESConnect with Jim Clark: https://www.linkedin.com/in/jim-clark/JWC3 Resources: mentioned at 00:19 as Jim's current company (CCS/reservoir analytics)BoodleBox (AI platform with academic knowledge bank, mentioned at 26:49): https://boodlebox.aiABOUT THE PODCASTThe Local Energy Podcast explores the people, technology, risk, and stories behind the oil and gas industry. Hosted by Peter Brecht, the show features conversations with operators, engineers, service providers, entrepreneurs, and energy professionals working across upstream oil and gas, insurance, technology, and local energy markets.Local Energy Podcast: https://localenergy.com/Connect with Peter Brecht: https://www.linkedin.com/in/peterbrechtConnect with Wade Spear: https://www.linkedin.com/in/wade-spear-05995b54Subscribe on YouTube: https://www.youtube.com/@localenergyioListen on Spotify: https://open.spotify.com/show/0huYDWqFYOjJScHi26l7Q9Apple Podcasts: https://podcasts.apple.com/us/podcast/local-energy/id1885425019
