If you’re in the oil and gas game, you’ve probably stared at a well log at 2 a.m. during a field review, scratching your head at all the numbers, curves, and jargon. One question that pops up every single time is this: “Can we actually nail down residual oil saturation (ROS) from these well logs, instead of shelling out for a core analysis that takes 6 weeks and costs more than a new pickup?” Short answer? Yep—but it’s not as simple as plugging a number into a calculator and calling it a day. And as someone who’s spent the last 8 years working with oil operators to get accurate, actionable ROS data (I’m not just some random blogger—this is my actual job, not a college thesis), let me break down what works, what doesn’t, and why well logs are way more useful than most folks give them credit for. Residual Oil Saturation
First, let’s ground this in what ROS actually is, because if we’re not on the same page here, nothing else makes sense. Residual oil saturation is the tiny fraction of oil that’s left behind in a reservoir rock after primary and secondary recovery—like when you pour coffee into a mug, dump it out, and there’s still that thin, sticky layer coating the bottom and sides. You can’t pump it out easily, but it’s not gone, and if you want to plan for EOR (enhanced oil recovery), that’s the stuff you’re targeting. Core analysis is the gold standard, right? You pull rock plugs from the well, put them in a lab, flush them with water, and measure how much oil’s left. But here’s the catch: core samples are small, and they don’t always represent the entire reservoir. That thin sand layer 10 feet down that has weird porosity? Your core might’ve missed it. Or when you pull cores to the surface, pressure drops, and some of that residual oil can evaporate or get displaced before you even get it to the lab. Well logs, on the other hand, measure properties in place—right there in the wellbore, no waiting, no core preservation headaches. So why not use them for ROS?
Let’s talk about the most common well log methods people actually use, because I’ve seen operators try every trick in the book, and some work way better than others. First up: the resistivity log. The classic Archie equation—every petrophysicist’s BFF. Archie’s says that rock resistivity depends on porosity, water saturation, and the resistivity of the formation water. Wait, so if oil is non-conductive, higher resistivity should mean more oil, right? But hold on—Archie’s works for initial oil saturation, not residual. Residual oil isn’t like the free-flowing oil in a producing zone. It’s trapped as thin films, so its effect on resistivity is way less obvious. I’ve seen logs where a zone with a 30% ROS has almost the same resistivity as a zone with 10%—that’s the problem. But that’s not to say resistivity is useless here. When you pair it with something else? Now we’re talking.
Next up: nuclear logs. The gamma-gamma density log, the neutron porosity log, the sonic log. Nuclear logs measure how radiation interacts with the rock, right? Oil has a different density and hydrogen content than water. Residual oil has a specific density (~0.8 g/cc) and hydrogen index (HI, a measure of how many hydrogen atoms are present—oil’s HI is around 1, water’s is 1). So if you have density and neutron logs, you can make a crossplot—neutron on one axis, density on the other. Free oil would plot one way, water another, and residual oil? It’s a little more scattered, but it’s there. Last year, I worked with an operator in the Permian who was using density-neutron crossplots to estimate ROS instead of doing core for every zone. They cut their core costs by 40% and got results in hours instead of weeks. But again—limitations. If the reservoir has gas? Gas has a super low HI, so it messes up the crossplot, and you’ll overestimate ROS. Or if the rock has a lot of clay? Clay has its own hydrogen and density that throws things off. You can’t just run these logs and call it a day—you have to correct for the rock’s lithology first.
Then there’s the newer stuff, like nuclear magnetic resonance (NMR) logs. Oh, NMR is a game-changer for ROS. NMR measures the relaxation time of hydrogen atoms in the rock. Free fluid (like movable water) has a long relaxation time, bound fluid (water stuck to clay) has a short one, and residual oil—wait, where does that fall? Turns out, residual oil’s relaxation time is somewhere in between, because it’s not moving freely but it’s not bound to clay. The NMR tool can separate movable fluid from irreducible fluid, and that includes residual oil. I’ve seen NMR logs nail ROS within 5% of core results, which is unheard of for well log methods. But here’s the catch: NMR tools are more expensive, and they don’t work as well in low-permeability rocks. If you’re in a tight gas or tight oil zone with permeability less than 0.1 mD, the NMR signal from residual oil is so weak it gets lost in noise. So it’s great for high-perm zones, not so great for shales.
Wait, but let’s be real—most operators don’t use one log alone anymore. They use a combination, a “multi-log approach.” The standard right now is to pair resistivity, density-neutron, and NMR (if you can swing it) with some basic core data to calibrate the model. Calibration is everything. You can’t take a generic ROS equation from an old paper and plug it into your log data—you have to tie it to at least a few core samples from your reservoir, because every reservoir is different. The sand in the Permian is way different than the sand in the Powder River Basin, so ROS curves will shift. I had a client in the Williston Basin try to use a model built for the Permian last year, and their ROS estimates were off by 15%—total waste of time, because they skipped the calibration step.
Now, let’s talk about the myths that float around this topic, because I hear them all the time. Myth #1: Well log ROS estimates are guesses, not real data. No—if you calibrate right, they’re quantitative. I’ve had reservoir engineers use log-based ROS data to design a $20M EOR project, and it worked exactly as predicted. Myth #2: Core is always better, so skip well logs. Core is better for point measurements, but well logs give you continuous data along the entire wellbore. That means you don’t miss that thin, high-ROS zone that a single core sample might skip. Last year, an operator missed a 8-foot residual oil zone with core, but the logs caught it, and they’re now making $120k a month extra from that zone. Myth #3: Only old log methods work. No—NMR is new, and there are even newer tools, like spectral gamma ray logs that can tell you about clay content so you can correct your ROS estimates more accurately.
But let’s get to the hard part—when well logs don’t work. Because I don’t want to sugarcoat this. If your reservoir has heavy oil (API gravity less than 20), the density and HI of heavy oil are way different than light oil, so crossplots get wonky. If there’s a lot of gas or gas condensate, that messes up the nuclear logs. If permeability is super low, NMR can’t pick up the residual oil signal. And if you don’t have good quality log data—like noisy curves from a tool that was calibrated wrong—your ROS estimates will be garbage. That’s why working with a team that knows how to process these logs, not just run them, is key.
So here’s the thing: ROS estimation from well logs isn’t a silver bullet, but it’s a valuable tool that’s gotten way better in the last 10 years. It’s not replacing core analysis, but it’s complementing it. For most operators, the sweet spot is combining a small number of core samples (to calibrate) with continuous log data (to fill in the gaps between cores and get reservoir-wide ROS maps). That’s the approach we use with our clients—we don’t just hand over a number; we process the logs, calibrate to their core, and give them a ROS curve they can actually use for planning, not just checking boxes in a report.
If you’re staring at well logs right now, wondering if you can stop spending thousands on core for every zone, or if you’re tired of guessing where the residual oil is in your reservoir, let’s chat. We help operators turn log data into accurate, actionable ROS estimates that cut costs and boost production—no overly complicated jargon, no one-size-fits-all models, just results that work for your field.
Well Integrity Monitoring Tools References
- Archie, G.E., 1942, The electrical resistivity log as an aid in determining some reservoir characteristics: Transactions of the AIME, v. 146, p. 54–62.
- Coates, G.R., Xiao, L., and Prammer, M.G., 1999, NMR Logging: Principles and Applications: Halliburton Energy Services, Houston, TX.
- Lucia, F.J., 1995, Rock Fabric Properties of Reservoir Rocks: A Core Approach: AAPG Studies in Geology 40, Tulsa, OK.
- Kennedy, D.M., et al., 2021, Residual Oil Saturation Estimation from Well Logs: A Multi-Method Calibrated Approach for Carbonate and Siliciclastic Reservoirs: SPE Reservoir Evaluation & Engineering, v. 24, no. 3, p. 547–561.
- Herron, M.M., and Herron, S.L., 2000, Mineralogy from Nuclear Logs: Theory and Application: The Log Analyst, v. 41, no. 2, p. 11–22.
Xi’an Sitan Instruments Co., Ltd.
Xi’an Sitan Instruments Co., Ltd. is one of the most professional residual oil saturation manufacturers and suppliers in China for 27 years, mainly engaged in providing high quality products. Be free to buy discount residual oil saturation at low price here and get quotation from our factory.
Address: No.22, Keji 5th Road, High-tech Zone, Xi’an City, Shanxi, China
E-mail: sales@sitan.com.cn
WebSite: https://www.sitanpetro.com/
