Previous Lecture Complete and continue  

  5. North Sea Case Study

Lecture content locked

Enroll in Course to Unlock
If you're already enrolled, you'll need to login.

Transcript

- Here's another reservoir, this one's a little more complicated. While there's six reservoirs where I have data, there's actually seven in the series, but we think the seventh behaves just like the other six. What we have is a spill-fill sequence involving six reservoirs, so we have oil coming into the deepest reservoir, and then, as this reservoir overfills, it spills into this reservoir. And then this reservoir gets filled up. This one continuously gets overfilled, spilling, spilling, spilling, up, up, up. Okay, so what we would expect? Well, first off, in a normal basin subsidence scenario, which is what the 2-D petroleum system model indicates here. I won't show this, but we have it, Nicole Mass directed that. So, what we have is a low-maturity oil coming in. It's dense, it's low-maturity, so high-density. And then a higher maturity comes in and goes to the top and displaces the dense stuff down below, that's density stacking. And, so you have dense oil spilling out here and, because these reservoirs are cold, they undergo biodegradation at the all-water contact, just like we showed in the Caron Energy Project. So, it's gonna be a biodegraded oil charging into this reservoir, and, just like that, you're gonna fill this reservoir and the biodegraded oil, which will be the heaviest, you're removing the lighter components, concentrating the dense components, asphaltene, is gonna be here. And then that's gonna be the oil that spills, so you expect the following. You expect higher-maturity oil here, lower-maturity oil here, less biodegradation here, more biodegradation here, and you could say, naturally, if you have any water washing, you're gonna have more exposure to water as you move through the reservoir, so you might see more water washing here. That's the kind of expectation. What do you get? Okay, run the DFA data, it's beautiful. We did have an optical light scattering issue that had to be addressed, which we did. It was wavelength-dependent, so it was trippy, but it's all addressed. Sir Ian Bettencourt did that. So what we see is something interesting. We see, roughly, a factor of three between these oils and these oils, so that looks like the Caron Energy. I'm plotting the asphaltene content, or the color, so these oils are low-asphaltene content, high-asphaltene content as we're spilling up into shallower and shallower reservoirs. So, that's a factor of three. It looks the like the Caron Energy case study, but in Caron, we think were going from Peters-Moldowan zero to six, here, this is already biodegraded as indicated here, we'll take a look in a minute, Peters-Moldowan two to six. So we have the same factor of three with less biodegradation in range, so we have to understand why. So we look at the 2-D CG, I'm just showing a section of the 2-D CG. First, we'll look at the deepest reservoir. We can see that the pristane is big, that the norpristane, another isoprenoid, is big, that the farnesane is big, and if you look at the n-alkanes, now I'm gonna move away from the range where we might have mud contaminant, so that's over here, the n-alkane C 17, C 18 is indicated, it's small, C 19, they're small in comparison, for example, to pristane and phytane, so this is biodegraded. If it's unbiodegraded, then the pristane and phytane would be smaller than these associated n-alkanes close by in the GC space. This is about Peters-Moldowan two. If we look at this oil, this is in the intermediate reservoir, there are six so this is an intermediate one, you can see the n-alkanes are largely missing. They're just gone, you don't have 'em. But you still have the isoprenoids present, phytane, pristane, norpristane, et cetera. This is roughly Peters-Moldowan four. If you go to the top one here, then the n-alkanes and the isoprenoids are missing, so the bugs ate 'em, and this would be about Peters-Moldowan six. Now we want to go check do we have 25-norhopanes. So we go check and this is looking for, basically, the hopane- or norhopane-type structures are present and here, this is GC GC MS, so we're gonna go through the first column, the second column, then we're gonna set the mass spec on this fragmentation peak, and that fragmentation peak is missing the normethyl group, the 25-norhopanes and there they are. They're showing up. So we have 25-norhopanes in the sample, so this sample is Peters-Moldowan six as confirmed by GC-GC-MS, a very sophisticated technique. Okay, so that's fine. These assignments are confirmed and I'll just say we have total agreement with what's standard literature in Peters-Moldowan. It's a good thing 'cause Ken Peters's on these papers. Now we're gonna look at water washing. This is the North Sea, and the North Sea doesn't have aquifers like Colombia. You don't have a 80-meter tilt in the cold water contact because of the water flowing past the reservoir. Initially, I expected less water washing in the North Sea, so let's see what we get. Here's how we're gonna measure this, just like we did in Colombia. We're gonna use our scale using the naphthalenes which are two-ring aromatics. I like naphthalenes because they're not that volatile so if you have a flash issue associated with the oil or some improper storage, a little bit, it's not likely that you're gonna lose these compounds. With the single-ring aromatics, it's a little more dicey because of the volatility. So I like these, they're more robust. So we'll see what we can see. There's naphthalene, the two different one-carbon naphthalenes with the methyl group either here at alpha or beta in position on the naphthalene. There's the family of C 2 naphthalenes, C 3, C 4, C 5, and this is in the Peters-Moldowan oil equals two oil. So we don't see any evidence of water washing here. I think Jeremiah's gonna run these oils looking at the benzenes to see if he can see evidence of water washing in the benzenes. Let's take a look at the intermediate oil. That's the same slide I showed before but now I'm emphasizing the naphthalenes and the water washing issues, not the extent of biodegradation. If we look at this Peters-Moldowan four oil, and what do we see? Naphthalene's missing like in Colombia. The C 1 naphthalenes, the two peaks, 1-methyl and 2-methyl naphthalene, they're missing, and the C 3's, well, there's a lot of C 3's missing. Oh, sorry, C 2's and the C 3's are looking more intact, C 4's, C 5's. But this is a rather extensive water washing, so I was pretty surprised. And these compounds are disappearing in order of their water solubility. The water solubility is dropping as you put carbons on by two orders of magnitude over the range that we're looking at, whereas the biodegradation index doesn't seem to be changing much, from the literature that I could find. But the fact that we're having the samples disappear in order, the compounds disappear in order of water solubility, to me, looks like water washing. Then we go up to the severely biodegraded oil and you can see all the naphthalenes, even up to C 5, are missing, they're missing. What's going on here? So the question is why is there so much more water washing in this oil than in Colombia, where Colombia has much more active aquifers? It took us a little while to figure this out. The issue is very simple. In the North Sea, we have active biodegradation in the reservoirs we're looking at. So if any naphthalene migrates to the water, it gets eaten, the bugs eat it. So there's a flux in one direction only. There's no back-flux. In Colombia, you remember we had all the n-alkanes because the reservoir was too hot for biodegradation. It had undergone enough subsidence that the reservoir heat had killed the bugs. So now, in the Colombia case, Llanos Basin, if any naphthalene gets into the water, it loads up the naphthalene and saturates the water with naphthalene and now, you have to wait. Either the naphthalene's gonna have a back-flux, or you'd have to wait for the naphthalene to diffuse away from the oil-water contact or be swept away by those active aquifers, but the velocity of the aquifer that's in our paper, it's about a centimeter a year. It's about the same flow-rate as the movement of continents, actually, for different reasons. What this means is that it's a much slower process to remove the water-soluble components if you're not immediately consuming the naphthalene once it gets in the water because you get a back-flux. So that's why we have such extensive water washing in the North Sea and not in our Colombia sample, even though the aquifers are much more active in Colombia. So I told this to one expert and he said I can't possibly be right and I told this to another expert and he said everybody knows this, but nobody bothers to publish it. So there you have it. So we're gonna put it in our paper and if anybody wants to throw rocks, they're quite welcome. Now the other thing is I had mentioned thermal maturity. So we have not much range of thermal maturity. Remember those compounds, Ts and Tm, I told you the Ts was stable but Tm is metastable? If you make oil at low temperature, you have more of this. And it's dependent on source rock, what the ratio of Ts and Tm starts life to be, but if it's at cold temperatures, you don't do much conversion of metastable to stable, but if it's at high temperature, now you can really convert this over so the limiting range, the equilibrium range, I can't remember, is approaching one, you get more Ts, I can't remember exactly the equilibrium value, but in any event, you get larger ratios as you get more thermally mature. So we look at all the reservoirs, deep to shallow, and indeed, we get a range, you can see this large error bar on these measurements, but you can see that this is more mature than this, and that's exactly what's expected. It was a low-maturity initial charge and that overfilled. The low-maturity is more dense, spills out, goes up, spills out, goes up, spills out, like that, exactly what we would expect if we have, actually, a spill-fill sequence in these reservoirs. I mean, you don't have to have a spill-fill sequence in a series of reservoirs, so this is confirming that, consistent and confirming, these are injectites and it's a little tricky to try to understand spilling out of injectites, but that's what's going on. What we're listing here is the order of magnitude by the font size, so biodegradation going two to six is a big deal for concentrating the asphaltenes. We do have some severe water washing, which I'm claiming is biodegradation-assisted and that contributes to higher concentration of asphaltene in these oils compared to these because we have more water washing with these oils. And we have a subtle thermal maturity variation that adds to that asphaltene gradient, I think not much, but I don't know how much. We're trying to sort that out. And so all of these three factors give rise to this DFA data showing this factor three variation on asphaltenes which corresponds, the viscosity depends very exponentially, essentially, on asphaltene content, so there's a big viscosity range. So, with all these different, spill-fill going on, biodegradation, we can sort everything out linking DFA plus 2-D GC. In conclusion, the studies here support well-developed literature on a variety of issues. Biodegradation can have a big negative impact that's known on oil quality and viscosity. Biodegradation occurs at the oil-water contact. The Caron Energy showed that quite clearly. Something I forgot to mention, the Caron Energy oil at the top of that reservoir at 300 meters was unbiodegraded, which means that you don't get much biodegradation in migration, you get most of it in-reservoir. Why is that? Because the migration process is fast and the reservoirs, well this one was sitting there for 50 million years, so you have some time for the bugs to work. So the biodegradation's primarily in-reservoir. The composition's matching the Peters-Moldowan scale, that's what everybody would expect. We find the standard temperature limits on biodegradation and we get the standard expectation on maturity versus time with the variant. And biodegradation can triple asphaltene content. This literature is not as well developed, but we have a lot of support for this. So we can combine the downhole fluid analysis measurements with thermodynamic modeling with high dependence thrown in on geodynamic processes to understand the viscosity profile in the reservoir. We can see when we have multiple charging when the first charge is biodegraded, the second charge not biodegraded due to reservoir subsidence, and petroleum system modeling is a great help to understand this specifically and, in general, for understanding these different things. Biodegradation-assisted water washing is observed, we're claiming. Using GC GC combined with DFA, we can unravel a variety of different complex reservoir fluid geodynamics including all these different factors listed here, we can unravel. Thank you very much.

- Here's another reservoir, this one's a little more complicated. While there's six reservoirs where I have data, there's actually seven in the series, but we think the seventh behaves just like the other six. What we have is a spill-fill sequence involving six reservoirs, so we have oil coming into the deepest reservoir, and then, as this reservoir overfills, it spills into this reservoir. And then this reservoir gets filled up. This one continuously gets overfilled, spilling, spilling, spilling, up, up, up. Okay, so what we would expect? Well, first off, in a normal basin subsidence scenario, which is what the 2-D petroleum system model indicates here. I won't show this, but we have it, Nicole Mass directed that. So, what we have is a low-maturity oil coming in. It's dense, it's low-maturity, so high-density. And then a higher maturity comes in and goes to the top and displaces the dense stuff down below, that's density stacking. And, so you have dense oil spilling out here and, because these reservoirs are cold, they undergo biodegradation at the all-water contact, just like we showed in the Caron Energy Project. So, it's gonna be a biodegraded oil charging into this reservoir, and, just like that, you're gonna fill this reservoir and the biodegraded oil, which will be the heaviest, you're removing the lighter components, concentrating the dense components, asphaltene, is gonna be here. And then that's gonna be the oil that spills, so you expect the following. You expect higher-maturity oil here, lower-maturity oil here, less biodegradation here, more biodegradation here, and you could say, naturally, if you have any water washing, you're gonna have more exposure to water as you move through the reservoir, so you might see more water washing here. That's the kind of expectation. What do you get? Okay, run the DFA data, it's beautiful. We did have an optical light scattering issue that had to be addressed, which we did. It was wavelength-dependent, so it was trippy, but it's all addressed. Sir Ian Bettencourt did that. So what we see is something interesting. We see, roughly, a factor of three between these oils and these oils, so that looks like the Caron Energy. I'm plotting the asphaltene content, or the color, so these oils are low-asphaltene content, high-asphaltene content as we're spilling up into shallower and shallower reservoirs. So, that's a factor of three. It looks the like the Caron Energy case study, but in Caron, we think were going from Peters-Moldowan zero to six, here, this is already biodegraded as indicated here, we'll take a look in a minute, Peters-Moldowan two to six. So we have the same factor of three with less biodegradation in range, so we have to understand why. So we look at the 2-D CG, I'm just showing a section of the 2-D CG. First, we'll look at the deepest reservoir. We can see that the pristane is big, that the norpristane, another isoprenoid, is big, that the farnesane is big, and if you look at the n-alkanes, now I'm gonna move away from the range where we might have mud contaminant, so that's over here, the n-alkane C 17, C 18 is indicated, it's small, C 19, they're small in comparison, for example, to pristane and phytane, so this is biodegraded. If it's unbiodegraded, then the pristane and phytane would be smaller than these associated n-alkanes close by in the GC space. This is about Peters-Moldowan two. If we look at this oil, this is in the intermediate reservoir, there are six so this is an intermediate one, you can see the n-alkanes are largely missing. They're just gone, you don't have 'em. But you still have the isoprenoids present, phytane, pristane, norpristane, et cetera. This is roughly Peters-Moldowan four. If you go to the top one here, then the n-alkanes and the isoprenoids are missing, so the bugs ate 'em, and this would be about Peters-Moldowan six. Now we want to go check do we have 25-norhopanes. So we go check and this is looking for, basically, the hopane- or norhopane-type structures are present and here, this is GC GC MS, so we're gonna go through the first column, the second column, then we're gonna set the mass spec on this fragmentation peak, and that fragmentation peak is missing the normethyl group, the 25-norhopanes and there they are. They're showing up. So we have 25-norhopanes in the sample, so this sample is Peters-Moldowan six as confirmed by GC-GC-MS, a very sophisticated technique. Okay, so that's fine. These assignments are confirmed and I'll just say we have total agreement with what's standard literature in Peters-Moldowan. It's a good thing 'cause Ken Peters's on these papers. Now we're gonna look at water washing. This is the North Sea, and the North Sea doesn't have aquifers like Colombia. You don't have a 80-meter tilt in the cold water contact because of the water flowing past the reservoir. Initially, I expected less water washing in the North Sea, so let's see what we get. Here's how we're gonna measure this, just like we did in Colombia. We're gonna use our scale using the naphthalenes which are two-ring aromatics. I like naphthalenes because they're not that volatile so if you have a flash issue associated with the oil or some improper storage, a little bit, it's not likely that you're gonna lose these compounds. With the single-ring aromatics, it's a little more dicey because of the volatility. So I like these, they're more robust. So we'll see what we can see. There's naphthalene, the two different one-carbon naphthalenes with the methyl group either here at alpha or beta in position on the naphthalene. There's the family of C 2 naphthalenes, C 3, C 4, C 5, and this is in the Peters-Moldowan oil equals two oil. So we don't see any evidence of water washing here. I think Jeremiah's gonna run these oils looking at the benzenes to see if he can see evidence of water washing in the benzenes. Let's take a look at the intermediate oil. That's the same slide I showed before but now I'm emphasizing the naphthalenes and the water washing issues, not the extent of biodegradation. If we look at this Peters-Moldowan four oil, and what do we see? Naphthalene's missing like in Colombia. The C 1 naphthalenes, the two peaks, 1-methyl and 2-methyl naphthalene, they're missing, and the C 3's, well, there's a lot of C 3's missing. Oh, sorry, C 2's and the C 3's are looking more intact, C 4's, C 5's. But this is a rather extensive water washing, so I was pretty surprised. And these compounds are disappearing in order of their water solubility. The water solubility is dropping as you put carbons on by two orders of magnitude over the range that we're looking at, whereas the biodegradation index doesn't seem to be changing much, from the literature that I could find. But the fact that we're having the samples disappear in order, the compounds disappear in order of water solubility, to me, looks like water washing. Then we go up to the severely biodegraded oil and you can see all the naphthalenes, even up to C 5, are missing, they're missing. What's going on here? So the question is why is there so much more water washing in this oil than in Colombia, where Colombia has much more active aquifers? It took us a little while to figure this out. The issue is very simple. In the North Sea, we have active biodegradation in the reservoirs we're looking at. So if any naphthalene migrates to the water, it gets eaten, the bugs eat it. So there's a flux in one direction only. There's no back-flux. In Colombia, you remember we had all the n-alkanes because the reservoir was too hot for biodegradation. It had undergone enough subsidence that the reservoir heat had killed the bugs. So now, in the Colombia case, Llanos Basin, if any naphthalene gets into the water, it loads up the naphthalene and saturates the water with naphthalene and now, you have to wait. Either the naphthalene's gonna have a back-flux, or you'd have to wait for the naphthalene to diffuse away from the oil-water contact or be swept away by those active aquifers, but the velocity of the aquifer that's in our paper, it's about a centimeter a year. It's about the same flow-rate as the movement of continents, actually, for different reasons. What this means is that it's a much slower process to remove the water-soluble components if you're not immediately consuming the naphthalene once it gets in the water because you get a back-flux. So that's why we have such extensive water washing in the North Sea and not in our Colombia sample, even though the aquifers are much more active in Colombia. So I told this to one expert and he said I can't possibly be right and I told this to another expert and he said everybody knows this, but nobody bothers to publish it. So there you have it. So we're gonna put it in our paper and if anybody wants to throw rocks, they're quite welcome. Now the other thing is I had mentioned thermal maturity. So we have not much range of thermal maturity. Remember those compounds, Ts and Tm, I told you the Ts was stable but Tm is metastable? If you make oil at low temperature, you have more of this. And it's dependent on source rock, what the ratio of Ts and Tm starts life to be, but if it's at cold temperatures, you don't do much conversion of metastable to stable, but if it's at high temperature, now you can really convert this over so the limiting range, the equilibrium range, I can't remember, is approaching one, you get more Ts, I can't remember exactly the equilibrium value, but in any event, you get larger ratios as you get more thermally mature. So we look at all the reservoirs, deep to shallow, and indeed, we get a range, you can see this large error bar on these measurements, but you can see that this is more mature than this, and that's exactly what's expected. It was a low-maturity initial charge and that overfilled. The low-maturity is more dense, spills out, goes up, spills out, goes up, spills out, like that, exactly what we would expect if we have, actually, a spill-fill sequence in these reservoirs. I mean, you don't have to have a spill-fill sequence in a series of reservoirs, so this is confirming that, consistent and confirming, these are injectites and it's a little tricky to try to understand spilling out of injectites, but that's what's going on. What we're listing here is the order of magnitude by the font size, so biodegradation going two to six is a big deal for concentrating the asphaltenes. We do have some severe water washing, which I'm claiming is biodegradation-assisted and that contributes to higher concentration of asphaltene in these oils compared to these because we have more water washing with these oils. And we have a subtle thermal maturity variation that adds to that asphaltene gradient, I think not much, but I don't know how much. We're trying to sort that out. And so all of these three factors give rise to this DFA data showing this factor three variation on asphaltenes which corresponds, the viscosity depends very exponentially, essentially, on asphaltene content, so there's a big viscosity range. So, with all these different, spill-fill going on, biodegradation, we can sort everything out linking DFA plus 2-D GC. In conclusion, the studies here support well-developed literature on a variety of issues. Biodegradation can have a big negative impact that's known on oil quality and viscosity. Biodegradation occurs at the oil-water contact. The Caron Energy showed that quite clearly. Something I forgot to mention, the Caron Energy oil at the top of that reservoir at 300 meters was unbiodegraded, which means that you don't get much biodegradation in migration, you get most of it in-reservoir. Why is that? Because the migration process is fast and the reservoirs, well this one was sitting there for 50 million years, so you have some time for the bugs to work. So the biodegradation's primarily in-reservoir. The composition's matching the Peters-Moldowan scale, that's what everybody would expect. We find the standard temperature limits on biodegradation and we get the standard expectation on maturity versus time with the variant. And biodegradation can triple asphaltene content. This literature is not as well developed, but we have a lot of support for this. So we can combine the downhole fluid analysis measurements with thermodynamic modeling with high dependence thrown in on geodynamic processes to understand the viscosity profile in the reservoir. We can see when we have multiple charging when the first charge is biodegraded, the second charge not biodegraded due to reservoir subsidence, and petroleum system modeling is a great help to understand this specifically and, in general, for understanding these different things. Biodegradation-assisted water washing is observed, we're claiming. Using GC GC combined with DFA, we can unravel a variety of different complex reservoir fluid geodynamics including all these different factors listed here, we can unravel. Thank you very much.