Nova: Picture this: every year, humanity pumps about 37 billion tonnes of carbon dioxide into the atmosphere. That is a trillion tonnes since the Industrial Revolution. Now imagine trying to put some of that back where it came from — kilometers underground, locked inside porous rock, forever. That is exactly what the field of geological carbon storage is all about, and it is the subject of a quietly influential book called Geological Storage of CO2: Modeling Approaches for Large-Scale Simulation, by Jan Nordbotten and Michael Celia.

Nova: That is the fascinating part. When this book came out in 2011, it was actually the first textbook ever written on modeling geological CO2 storage. Nobody had pulled together the mathematics, the physics, the computational methods, and the real-world data into one coherent volume before. And today, with 77 commercial carbon capture and storage projects operating worldwide and the planet racing toward gigaton-scale deployment, the questions this book tackles have never been more urgent.

Nova: Exactly. And here is a number that might surprise you: the Intergovernmental Panel on Climate Change estimates that well-designed geological storage sites can trap over 99 percent of injected CO2 for a thousand years or more. That is a remarkable claim. The question is: how do we know that? How do we model something we cannot see, happening deep underground, over timescales that dwarf human experience? That is the intellectual adventure at the heart of this book.

Nova: Before Nordbotten and Celia dive into equations, they start with what they call simply "the carbon problem." Fossil fuel CO2 emissions hit 32 gigatonnes in 2010. That number has only climbed since. The book frames carbon capture and storage, or CCS, as one of the essential "stabilization wedges" — a term from a famous 2004 paper by Pacala and Socolow — needed to keep global warming in check.

Nova: Right. And CCS is a big wedge. The idea is deceptively simple on paper: capture CO2 from power plants, cement factories, or steel mills before it hits the atmosphere, compress it into a supercritical fluid — something between a gas and a liquid — then inject it deep underground into porous rock formations. The book argues that without CCS, the cost of meeting climate targets could more than double.

Nova: Absolutely. And the history supports that. The first large-scale CO2 injection for enhanced oil recovery started in 1972 at the SACROC project in Texas. Over its lifetime, SACROC injected more than 175 million tonnes of CO2. Then in 1996, Norway launched the Sleipner project — the world's first dedicated CO2 storage operation not tied to oil recovery. Since then, Sleipner has stored over 20 million tonnes of CO2 in a deep saline aquifer beneath the North Sea.

Nova: That is the sobering part. Current global CCS capacity sits at about 64 million tonnes per year across 77 operational projects. To put that in perspective, annual global emissions are roughly 37 billion tonnes. So we are capturing less than two-tenths of one percent. The scaling challenge is immense. The book confronts this head-on, arguing that rigorous modeling is essential to deploy storage at the necessary scale safely and efficiently.

Nova: Precisely. And that is where things get really interesting.

Nova: The core of the book is about modeling fluid flow in porous media. Imagine a sandstone formation two kilometers underground. It looks solid, but under a microscope it is riddled with tiny connected pores filled with salty brine. When you inject supercritical CO2, it pushes through these pores, displacing brine. You now have two fluids moving through a complex, three-dimensional labyrinth you cannot see.

Nova: It is a nightmare of complexity. But Nordbotten and Celia walk the reader through it step by step. They start with single-phase flow — just one fluid — introduce the basic equations of pressure and flow. Then they add the second phase and suddenly buoyancy, capillary forces, and relative permeability all come into play.

Nova: Exactly. Supercritical CO2 has a density about 60 to 70 percent that of water at typical reservoir conditions. So it wants to rise. It will migrate upward until it hits an impermeable caprock layer — a geological seal. This is called structural trapping, and it is the primary mechanism that keeps CO2 in place, at least initially.

Nova: Yes. Over time, other mechanisms kick in. Residual trapping happens when CO2 gets disconnected into tiny bubbles trapped in pore spaces by capillary forces — like water droplets on a windowpane that will not budge. Then solubility trapping: CO2 dissolves into the brine, like sugar dissolving in tea. And finally, over thousands of years, mineral trapping: dissolved CO2 reacts with rock minerals to form solid carbonates, locking the carbon away permanently.

Nova: That is one of the most counterintuitive findings in the whole field. The risk of leakage decreases the longer the CO2 sits there, because more and more of it gets immobilized by these trapping mechanisms. The book provides the mathematical frameworks to quantify this evolution over time.

Nova: Great question. Traditional full-physics reservoir simulations are computationally brutal. You are solving equations on a three-dimensional grid with millions of cells, over timescales of centuries. A single run might take days on a supercomputer. The vertical equilibrium approach simplifies things dramatically. Since CO2 and brine segregate rapidly due to gravity — CO2 rising, brine sinking — you can average the equations over the vertical direction, reducing a 3D problem to a 2D problem. It is like taking a tall building and describing it with just its floor plan.

Nova: Exactly. This innovation has been adopted widely. It allows researchers to run hundreds or thousands of simulations quickly, which is critical for uncertainty analysis. And uncertainty is huge in subsurface work because you never have perfect knowledge of the rock properties underground.

Nova: The book grounds its modeling approach with real-world data, and the most famous test case is Sleipner. Located in the Norwegian North Sea, Sleipner has been injecting about one million tonnes of CO2 per year into the Utsira saline formation since 1996. It is the longest-running dedicated CO2 storage project on Earth.

Nova: It has been monitored obsessively. Time-lapse seismic surveys — essentially taking ultrasound images of the subsurface every few years — have tracked the CO2 plume as it spreads. The plume has grown in a predictable, layer-cake pattern: it rises until it hits thin shale layers within the formation, spreads out laterally, then rises again when it finds an opening. Nordbotten and Celia's models reproduce this behavior beautifully, which builds confidence that the physics are right.

Nova: Yes, but not all projects have been smooth. Take Gorgon in Australia, operated by Chevron. It started injection in 2019 and is designed to store up to 4 million tonnes per year — making it one of the largest dedicated storage projects in the world. But it faced significant technical delays and underperformed for years due to pressure management issues in the injection wells. It only started hitting its targets more recently.

Nova: Precisely. And that is a central message of the book: models are essential but must be continuously updated with monitoring data. This is called history matching — you feed observational data back into the model to calibrate it. The book emphasizes that modeling is not a one-and-done exercise. It is iterative.

Nova: Northern Lights is genuinely groundbreaking. It is a joint venture between Equinor, Shell, and TotalEnergies, and it is the world's first open-source CO2 transport and storage service. The idea is elegantly simple: industrial emitters across Europe capture their CO2, ship it to a terminal on Norway's west coast, and from there it gets piped offshore and injected into a saline aquifer 2.6 kilometers below the seabed. Phase one opened in 2024 and can store 1.5 million tonnes per year. Phase two, planned for 2028, will expand that to over 5 million tonnes.

Nova: Exactly. And that business model could be transformative. If the infrastructure exists, emitters do not each need to build their own storage solution. The Nordbotten and Celia framework provides exactly the kind of modeling toolset needed to design, certify, and monitor these large-scale operations.

Nova: No discussion of geological CO2 storage is complete without addressing the elephant in the room: what if it leaks? The book dedicates significant attention to seal integrity — the caprock that keeps CO2 from escaping — and to potential leakage pathways, especially through abandoned wells.

Nova: It is considered the number one leakage risk. There are millions of old oil and gas wells globally, many of them poorly documented or improperly sealed. If a CO2 plume encounters one, the CO2 could migrate up the wellbore. The book includes analytical solutions for leakage rates through wells and faults — simplified mathematical formulas that let you estimate leakage without running a full simulation.

Nova: The evidence is remarkably reassuring, at least for well-characterized sites. A 2025 report from the Global CCS Institute concluded that well-selected storage sites in depleted oil and gas fields or deep saline aquifers have negligible leakage risk. The IPCC estimates retention rates above 99 percent over a thousand years. There is even a real-world stress test: in 2004, the Nagaoka CCS project in Japan experienced a large earthquake during active CO2 injection. They stopped, monitored carefully, and detected zero leakage.

Nova: It is. But the book does not sugarcoat the challenges. Monitoring is expensive and must continue long after injection stops — potentially for decades. You need seismic surveys, well pressure measurements, geochemical sampling, satellite monitoring, and more. Regulators in Europe, the United States, and Australia all require rigorous monitoring plans. And the public needs to trust the process.

Nova: It can, and the book addresses this. The physics is analogous to what happens with wastewater injection from oil and gas operations, which has caused small earthquakes in places like Oklahoma. The key is to manage injection pressures carefully and avoid injecting near critically stressed faults. The book provides the mathematical framework to calculate stress changes around injection wells. It is a manageable risk if you do the modeling upfront.

Nova: That is exactly right. And that is why a book like this matters. It gives engineers, regulators, and policymakers the intellectual foundation to make informed decisions. It is not a cheerleader for CCS — it is a rigorous, sober toolkit.

Nova: So here is where we land. Geological Storage of CO2 by Nordbotten and Celia is not a beach read. It is dense, mathematical, and unapologetically technical. But it arrived at exactly the right moment — just as the world was beginning to grapple seriously with deployment of carbon capture and storage at scale. It gave the field a common language, a shared set of modeling tools, and a rigorous way to answer the most important question: will the CO2 stay put?

Nova: But here is the challenge that remains: we need to go from 64 million tonnes per year to billions of tonnes per year — literally a hundred-fold increase — to make a meaningful dent in global emissions. That means hundreds or thousands of storage sites, each one requiring site characterization, modeling, monitoring, and long-term stewardship. The mathematical and computational foundations laid in this book will be essential to that effort.

Nova: Beautifully put. The Earth has been storing hydrocarbons underground for hundreds of millions of years. We are learning to use those same geological systems to put carbon back. It is one of the most ambitious engineering challenges humanity has ever undertaken, and it starts with understanding the physics at the pore scale.

Nova: This is Aibrary. Congratulations on your growth!