- One of the issues that we have in Uinta Basin because of the nature of the oil and the fact that a lot of it's oil wet is that the resistivities range over four orders of magnitude. And this slide, just to show you an example of some of the sands, and you can see over here sand coming in at just over 8,000 feet. You can see the resistivity comes way over 1,000 in meters. I mean, this is a backup at 1,000 in meters it goes two, three, 4,000 in meters which is sort of the upper limit of that lateral log. So, we ran lateral logs in all these wells because of the high resistivity. Again, if you look at this particular sand down here, the backup is going, again, up to 40,000. This is one to 10,000 out here, so one of the issues that we have to look at is what tools do we want to use in terms of the frequency component of dielectric system to be operable over all this range. So, what I did is calculate what the skin depth is of the various frequencies for various resistivities. And this particular graph shows you the skin depth. Now, the skin depth is when something is defined by electromagnetic propagation for years and years. And there's a reference in the paper that I just stated if anybody's interested in looking at this thing. Reference in the paper that just came out in October, but at any rate, you have skin depths of different frequencies, and skin depth is defined as where you have one over E of the percentage of the current is passing within that region. So, in other words, 100 ohm meters and at one gigahertz, your skin depth is, oh, that'll be about two inches. So, that means that 63% of the current is passing within two inches of the surface of the well board at that frequency. Similarly, at 1,000 ohm meters, you're up around seven inches. One of the things that we want to do is have a system that would measure as much, would have the highest signal to noise ratio and give us the most signal that we possibly could over a very large range. So, it's quite apparent that instead of using some of the lower frequencies, this is 10 megs, this is 30 megs, .1 gigs, 100 megs, 300 megs, and 1 gigahertz. We want to use something that has the most current at the shallowest depth to give us a combination with the shallow resistivity tool which will be about at that same kind of depth because you want to compare the resistivities that we're measuring with a dielectric with the resistivities we're measuring with the shallow resistivity tool which in this case is going to be an MSFL. So, what we did is calculate with a grid model the total amount of water in the system measured. And again, the CRIM references is all in the paper in both the out cover issue and the journal. At any rate, if you plot in a Pickett plot fashion the total water than you measured with the dielectric tool, which is the Y axis, the picket plot form, versus the resistivity that you measure with the MSFL, or shallow resistivity device, plot these in a Pickett plot form, assuming that there is going to be no contribution from resistivity from the matrix. In other words, you don't have any metals in there that are conducting. So, all you're looking at is at water, and you can see that we're talking about all the other clean reservoirs. I'm taking gamma ray less than 60 in this particular case. So, we're looking at the clean formations, and when you plot those data, you can see that you can get a nice fit to an m. So, with just looking at these CRIM data, the dielectric data on the Y axis, the resistivity on the X axis, I can calculate what an m is, and also extrapolate back up to what the resistivity of the water is that is to be measured in this thing. The clay component is almost nil in these reservoirs, and so, it wasn't something I really had to worry about. We'll talk about the clay component a little bit later. But the interesting thing is that I can calculate the slope of this line is one over m, so I can calculate that this slope of this thing has an m of 1.3 and has an intercept up here which is the RW effect of the analysis rate at that point. Now, the question is, I'm sure one of you are going to ask me, what about filtrate? Well, in these rocks, we also discriminated against any kind of invasion effects that we had. We had to throw out a whole bunch of data. And as I pointed out over here, there were well over 300 intervals that were analyzed. There's probably close to 400 intervals. We had to throw out a bunch of them because they were invaded with the filtrate. So we had to eliminate those just to see what the connate water was in the formation. And what we were looking for is this red arrow up here would say that, okay, if we had invasion from the, not invasion, but flooding from the water flood, this would increase in this direction, this RW would increase. And that's what we were gonna try to look at. So, we did some theoretical models and this is, this gives you, the blue line represents the resistivity that you would get at 25,000 ppm which was the connate water salinities, and I picked on this particular case an m of 1.65 which is about the average as we'll see a little bit later on about the average of all these data. So this is the blue line which we would expect to see in normal cases. If it was a flooded water, which has a salinity of about 10,000 ppm, you would be over here on this orangish line. So, what we're looking for is to see when we had the data go from this line to the orange line, but it wasn't going to be just a straight shift, just an alt shift up to that line because in the, as all of us know, when you do a flood experiment in core or in a reservoir, that the water goes into the highest porosity permeability areas first and then it extrapolates back to the lower portions as it starts flooding up. So what we were anticipating is this red line where you would be along the 25,000 parts per million line, and then gradually creep up to somewhere a little bit below the 10,000 ppm line, and probably never really approach it because you're never gonna flush out all the connate water, but you could flush out quite a bit of it. It's always the issue of, the other issue that we have was when we have filtrate invasion. You never, unless you're Darcy rock, you're never gonna flush out all the connate water. So, the resistivity of the resultant is going to be something here in between. So, once we've established an m that way, then you can start looking at how you can calculate n, the Archie saturation exponent. And this particular slide shows you how we actually did this because once you've established an RW, which is the intercept at 100% porosity, but once you've established this RW and you've established m, which we just showed you in this thing, and again, we did this for hundreds of intervals over a number of wells, then you can start looking at n. And the way you look at n is put the total porosity that you calculate from either your petrophysical model or you can use an NMR techniques, which we did a lot, and plot that against the LLD deep resistivity, and if it inclines up in a straight line, and I guess everybody really knows this, if it lines up in a straight line, that line intersects back here at the RL line which is 100% water line, that'll intersect at the bulk volume water irreducible number so you can just take these points, extrapolate that back to this point, and this gives you where the bulk volume water irreducible is. In this particular case, it's about 2.2% porosity, or .022 in decimal, and the slope of this line is the function of m and n, and once we've established m, and we would get the slope of this line, then you can calculate n. In this particular case, we have an n of 1.99. We have an m of 1.63. Again, we did this on a number of wells, or a number of a intervals at a bunch of different wells to get the both m's and n's. This slide's a little bit busy, but there's cases where it's difficult to determine both. The data did not line up in a nice straight line, so one of the things that we did is calculate, is take, over on this right hand figure, you get the bulk volume water from the dielectric tool and take this minimum down here, which we said would probably be the irreducible number, extrapolate that over to the curve, and that would be the BVWI in this particular case. Then you connect all the data points to this BVWI. So, in other words, we already have this data point up here which is RW. We also have this slope of this line, now we got this data point, so now we can connect it to these data, and then calculate an n. In this particular case, you can see the n is 1.8. The slope is negative, rather than positive, so it means that n is less than m. This has an m of 1.93 and a slope of 1.8. So, we're able to, even in data where it was not obvious what the slope of the data from a resistivity total porosity stand point is where they intercept, you could do it this way. So, bottom line is, we did this. We're able to, 265 data points of m. And what I've plotted up here is just the distribution of, in the lower ream, of the m's that we have. You can see the really interesting thing, couple of things that come out of this, you know, from a fracture stand point, the m in a fracture is one. And if you look at all these data points, 265, only one data point ended up down here in what we would think is a fractured system where the m is approaching one. Most of the data lies way out here and way below 2.0 which is the default that people like to use in the saturation equations, but the mode up here is 1.6, 1.65 as we said before. If you then look at the n distribution, we have 108 data points in these various sands for the n distribution. This is even more interesting that in the n, in this particular case, ranged from, you know, one to four, so, the interesting part of this thing is, again, the arrow points down to an n at two which is the default, we have a lot of data that's way out here with very large n's, and as we all know that as n gets way above two, two and a half, you're talking about oil wet or mixed wet system, so I have this large arrow on here saying this is increasingly oil wet clastics, now this is all clastic formation. There is also a course in the Green River, Lower Green River's a lot of carbonates. One of the things we did, I didn't have the access to these data after I left Newfield, so, what I've done since then, done some work in the Wasatch Formation, which is the formation directly below the Green River. Basically, the same environmental stand point. And plotted in the Wasatch Formation, we've got 54 points in the Wasatch Formation, in the clastic component, and you can see that on the X axis, m, and the Y axis, I have n, and you can see this is the unity line, and you can see that there is a huge scatter in m versus n data. In fact, if you look at these data, there's only about two or three points that actually m is equal to n. All these data down here, n is less than m. All these data up in this quadrant n is greater than m. Now, this is for the clastic component. Let me show you what happens when we get to the carbonate component over this same interval. These are 93 data points. Again, this is the unity line, the blue line's the unity line. M is equal to n. And you can see that we have a very few data points on here where n is less than m, but look at the number of data points that we have up in this region where n is much greater than m, and so, of course, what this means is that all this, all these carbonate, in these carbonate sections, all these things are oil wet. Which you'd expect, you know, because of the zeta potential between carbonates and clastics, you expect the carbonates to be more oil wet. This is about we had seen in the Lower Green River section, but as I say, I no longer had those data to calculate into that.

- One of the issues that we have in Uinta Basin because of the nature of the oil and the fact that a lot of it's oil wet is that the resistivities range over four orders of magnitude. And this slide, just to show you an example of some of the sands, and you can see over here sand coming in at just over 8,000 feet. You can see the resistivity comes way over 1,000 in meters. I mean, this is a backup at 1,000 in meters it goes two, three, 4,000 in meters which is sort of the upper limit of that lateral log. So, we ran lateral logs in all these wells because of the high resistivity. Again, if you look at this particular sand down here, the backup is going, again, up to 40,000. This is one to 10,000 out here, so one of the issues that we have to look at is what tools do we want to use in terms of the frequency component of dielectric system to be operable over all this range. So, what I did is calculate what the skin depth is of the various frequencies for various resistivities. And this particular graph shows you the skin depth. Now, the skin depth is when something is defined by electromagnetic propagation for years and years. And there's a reference in the paper that I just stated if anybody's interested in looking at this thing. Reference in the paper that just came out in October, but at any rate, you have skin depths of different frequencies, and skin depth is defined as where you have one over E of the percentage of the current is passing within that region. So, in other words, 100 ohm meters and at one gigahertz, your skin depth is, oh, that'll be about two inches. So, that means that 63% of the current is passing within two inches of the surface of the well board at that frequency. Similarly, at 1,000 ohm meters, you're up around seven inches. One of the things that we want to do is have a system that would measure as much, would have the highest signal to noise ratio and give us the most signal that we possibly could over a very large range. So, it's quite apparent that instead of using some of the lower frequencies, this is 10 megs, this is 30 megs, .1 gigs, 100 megs, 300 megs, and 1 gigahertz. We want to use something that has the most current at the shallowest depth to give us a combination with the shallow resistivity tool which will be about at that same kind of depth because you want to compare the resistivities that we're measuring with a dielectric with the resistivities we're measuring with the shallow resistivity tool which in this case is going to be an MSFL. So, what we did is calculate with a grid model the total amount of water in the system measured. And again, the CRIM references is all in the paper in both the out cover issue and the journal. At any rate, if you plot in a Pickett plot fashion the total water than you measured with the dielectric tool, which is the Y axis, the picket plot form, versus the resistivity that you measure with the MSFL, or shallow resistivity device, plot these in a Pickett plot form, assuming that there is going to be no contribution from resistivity from the matrix. In other words, you don't have any metals in there that are conducting. So, all you're looking at is at water, and you can see that we're talking about all the other clean reservoirs. I'm taking gamma ray less than 60 in this particular case. So, we're looking at the clean formations, and when you plot those data, you can see that you can get a nice fit to an m. So, with just looking at these CRIM data, the dielectric data on the Y axis, the resistivity on the X axis, I can calculate what an m is, and also extrapolate back up to what the resistivity of the water is that is to be measured in this thing. The clay component is almost nil in these reservoirs, and so, it wasn't something I really had to worry about. We'll talk about the clay component a little bit later. But the interesting thing is that I can calculate the slope of this line is one over m, so I can calculate that this slope of this thing has an m of 1.3 and has an intercept up here which is the RW effect of the analysis rate at that point. Now, the question is, I'm sure one of you are going to ask me, what about filtrate? Well, in these rocks, we also discriminated against any kind of invasion effects that we had. We had to throw out a whole bunch of data. And as I pointed out over here, there were well over 300 intervals that were analyzed. There's probably close to 400 intervals. We had to throw out a bunch of them because they were invaded with the filtrate. So we had to eliminate those just to see what the connate water was in the formation. And what we were looking for is this red arrow up here would say that, okay, if we had invasion from the, not invasion, but flooding from the water flood, this would increase in this direction, this RW would increase. And that's what we were gonna try to look at. So, we did some theoretical models and this is, this gives you, the blue line represents the resistivity that you would get at 25,000 ppm which was the connate water salinities, and I picked on this particular case an m of 1.65 which is about the average as we'll see a little bit later on about the average of all these data. So this is the blue line which we would expect to see in normal cases. If it was a flooded water, which has a salinity of about 10,000 ppm, you would be over here on this orangish line. So, what we're looking for is to see when we had the data go from this line to the orange line, but it wasn't going to be just a straight shift, just an alt shift up to that line because in the, as all of us know, when you do a flood experiment in core or in a reservoir, that the water goes into the highest porosity permeability areas first and then it extrapolates back to the lower portions as it starts flooding up. So what we were anticipating is this red line where you would be along the 25,000 parts per million line, and then gradually creep up to somewhere a little bit below the 10,000 ppm line, and probably never really approach it because you're never gonna flush out all the connate water, but you could flush out quite a bit of it. It's always the issue of, the other issue that we have was when we have filtrate invasion. You never, unless you're Darcy rock, you're never gonna flush out all the connate water. So, the resistivity of the resultant is going to be something here in between. So, once we've established an m that way, then you can start looking at how you can calculate n, the Archie saturation exponent. And this particular slide shows you how we actually did this because once you've established an RW, which is the intercept at 100% porosity, but once you've established this RW and you've established m, which we just showed you in this thing, and again, we did this for hundreds of intervals over a number of wells, then you can start looking at n. And the way you look at n is put the total porosity that you calculate from either your petrophysical model or you can use an NMR techniques, which we did a lot, and plot that against the LLD deep resistivity, and if it inclines up in a straight line, and I guess everybody really knows this, if it lines up in a straight line, that line intersects back here at the RL line which is 100% water line, that'll intersect at the bulk volume water irreducible number so you can just take these points, extrapolate that back to this point, and this gives you where the bulk volume water irreducible is. In this particular case, it's about 2.2% porosity, or .022 in decimal, and the slope of this line is the function of m and n, and once we've established m, and we would get the slope of this line, then you can calculate n. In this particular case, we have an n of 1.99. We have an m of 1.63. Again, we did this on a number of wells, or a number of a intervals at a bunch of different wells to get the both m's and n's. This slide's a little bit busy, but there's cases where it's difficult to determine both. The data did not line up in a nice straight line, so one of the things that we did is calculate, is take, over on this right hand figure, you get the bulk volume water from the dielectric tool and take this minimum down here, which we said would probably be the irreducible number, extrapolate that over to the curve, and that would be the BVWI in this particular case. Then you connect all the data points to this BVWI. So, in other words, we already have this data point up here which is RW. We also have this slope of this line, now we got this data point, so now we can connect it to these data, and then calculate an n. In this particular case, you can see the n is 1.8. The slope is negative, rather than positive, so it means that n is less than m. This has an m of 1.93 and a slope of 1.8. So, we're able to, even in data where it was not obvious what the slope of the data from a resistivity total porosity stand point is where they intercept, you could do it this way. So, bottom line is, we did this. We're able to, 265 data points of m. And what I've plotted up here is just the distribution of, in the lower ream, of the m's that we have. You can see the really interesting thing, couple of things that come out of this, you know, from a fracture stand point, the m in a fracture is one. And if you look at all these data points, 265, only one data point ended up down here in what we would think is a fractured system where the m is approaching one. Most of the data lies way out here and way below 2.0 which is the default that people like to use in the saturation equations, but the mode up here is 1.6, 1.65 as we said before. If you then look at the n distribution, we have 108 data points in these various sands for the n distribution. This is even more interesting that in the n, in this particular case, ranged from, you know, one to four, so, the interesting part of this thing is, again, the arrow points down to an n at two which is the default, we have a lot of data that's way out here with very large n's, and as we all know that as n gets way above two, two and a half, you're talking about oil wet or mixed wet system, so I have this large arrow on here saying this is increasingly oil wet clastics, now this is all clastic formation. There is also a course in the Green River, Lower Green River's a lot of carbonates. One of the things we did, I didn't have the access to these data after I left Newfield, so, what I've done since then, done some work in the Wasatch Formation, which is the formation directly below the Green River. Basically, the same environmental stand point. And plotted in the Wasatch Formation, we've got 54 points in the Wasatch Formation, in the clastic component, and you can see that on the X axis, m, and the Y axis, I have n, and you can see this is the unity line, and you can see that there is a huge scatter in m versus n data. In fact, if you look at these data, there's only about two or three points that actually m is equal to n. All these data down here, n is less than m. All these data up in this quadrant n is greater than m. Now, this is for the clastic component. Let me show you what happens when we get to the carbonate component over this same interval. These are 93 data points. Again, this is the unity line, the blue line's the unity line. M is equal to n. And you can see that we have a very few data points on here where n is less than m, but look at the number of data points that we have up in this region where n is much greater than m, and so, of course, what this means is that all this, all these carbonate, in these carbonate sections, all these things are oil wet. Which you'd expect, you know, because of the zeta potential between carbonates and clastics, you expect the carbonates to be more oil wet. This is about we had seen in the Lower Green River section, but as I say, I no longer had those data to calculate into that.