- [Voiceover] The gamma ray is the workhorse of the petrophysical industry to identify and distinguish clean formation from shale. The natural radioactivity is potassium, thorium, and uranium, and so mostly that is coming out of clay minerals, but not entirely. You can get radioactive carbonates charged by uranium water coming into them, and also in the new application of organic-rich shales, TOC, total organic carbon, can have extraordinarily high gamma ray measurements because of uranium. It would be better if you're going to concentrate on clays only, to run a spectral gamma ray that then will isolate potassium, thorium, and uranium, and you can just look at the clay minerals themselves. Here is a schematic that I'm sure all of you are aware of calculating shale volume from gamma ray, identifying gamma ray shale and gamma ray clean, and then either linearizing them to calculate V shale or having a non-linear equation such as Steiber to calculate V shale. I should say that this is not a trivial choice. One should spend quite a bit of time identifying what you believe to be the shale and the clean, because this is starting point of all petrophysics. And you can create almost as you wish V shale values to your desires. The important thing in petrophysics is that it isn't just simply push a button and hear the results and trust me. There's an awful lot of interpretation at every stage. For the SP, which was one of the earliest logs run, the SP log was initially thought to be a permeability log, and it is some measure of permeability. It's naturally occurring counts in the formation due to the juxtaposition of the well bore, a shale zone, and then a clean formation, sand or carbonate where there is mud filtrate invasion. The little current cells are set up as the salinity is trying to equilibrate. The SP is a deflection if and when there is a salinity contrast between the mud and the formation. And so you can then, if you, one of the powerful uses of the SP, is to calculate RW, water resistivity which is essential. But the caveat is that you've got to know mud filtrate resistivity, RMF. And in our experience, the mud filtrate records on rasterized images can be extraordinarily unreliable. So be very careful if you're using the SP as an RW indicator on its own. You should check with other things. Again, you can calculate V shale from the SP if you wish. Okay, so now turning attention to porosity logs a little bit. The so-called porosity logs really aren't measuring porosity. They're measuring other things. But the main petrophysical information coming out of them is porosity. First of all, the sonic or the acoustic log measures reciprocal speed of sound or slowness in microseconds per foot or microseconds per meter. One equation to calculate porosity is given in that you have to know the matrix travel time or speed of sound in matrix. And you also have to know the fluids. So in order to calculate it properly, you've got to know those two as well as the basic data itself. Similarly, the density log. It measures the bulk density of the rock and you need to know matrix and fluid properties in order to calculate correctly a density porosity. The neutron measures concentration of hydrogen and there's a transform to get it into porosity units. The problem is, from the petrophysical viewpoint, that that transform can be done by the service company to either limestone, sandstone, or dolomite. And quite often, if you're getting digital data, it won't tell you. So my suggestion is always, always get the rasterized images because there will be a record, one hopes at least, as to what lithology the neutron log was run on. Many petrophysicists won't use the neutron log because, as compared with the density, it's got a lot of environmental corrections. You may or may not know from the service company which have been applied. And so there's quite a debate in the community as to how one should use neutrons. A lot of people will normalize the neutrons in any main study to try and get consistent readings. The Pe curve measures the photo electric adsorption cross section and is a very powerful tool with particularly the density in the neutron for lithology. Cross plots, and we'll see some in just a moment, of density / neutron and sonic / neutron can yield you porosity values with no requirement for matrix and fluid input. In other words, you don't have to, as those previous equations showed, you don't have to define delta t matrix and delta t fluid and so on. It's done sort of for you. And we'll see how that works. Also, V shale is available from a density / neutron combination as shown. We're finding in some recent work we're doing that that probably is a much better shale indicator in organic rich shales with a lot of TLC because the logs are unaffected by uranium. What we're tending to do these days is to use a V shale in the organic-rich shales using a density / neutron. Calculation is unreliable in gas because the neutron reads too low and the density is too light in gas. All porosity logs are measuring total porosity but we are really interested in effective porosity. Total porosity is defined in the equation in the middle of the graph as effective porosity plus the contribution of shales because clay minerals, they have water in them. And the porosity log is looking at that as porosity. And so, in order to calculate effective porosity, which is what you're really interested in, you have to know V shale accurately and you have to know shale porosity accurately. And you also have to know total porosity accurately. So you can see that we're building on uncertainties, if you want to think of it that way. You have to be, in other words, very careful on these kinds of calculations. A good way of verifying whether you have made correct calculations, and we do this all the time, is to look at all the different porosity calculations from individual porosity logs and cross plot and compare them in the realm of effective porosity, and total for that matter. And in an ideal world, all effective porosities should converge. In one particular instance, they won't converge. But they may not converge if you've got incorrect fluid and matrix input or perhaps a porosity log that is badly calibrated, and that should become obvious. But also, there's one powerful applicaton of this, and that is in carbonates in the realm of moldic porosity. The concept is that in moldic porosity that is kind of isolated in carbonates, the density / neutron, any nuclear device will measure everything because it's a volume measurement, whereas the acoustic / sonic log is a directional measurement and will skip over them. On this example, which is from the brown dolomite of the Texas Panhandle, is shown a mismatch, highlighted in yellow, that is therefore interpreted to be moldic porosity. Let's look at some porosity cross plots, because these are very powerful. Here is a standard density / neutron cross plot. Be careful. You must have the neutron in limestone units in order to solve this equation. You cannot come in, in sandstone units. But also, you see on this cross plot tie lines for sandstone, limestone, and dolomite. Also you can see porosity tie lines: 5, 10, 15, and 25%. That's sort of at 45 degree angles. So from this then, looking at any one datapoint, you can get a first guess at not only its porosity from the cross plot, but also its grain density by extrapolating points down to the zero porosity line. It's not shown on this graph, but another advantage of this is gas effects reduce the neutron and reduce the density. And the vector of that is about at 45 degrees. That means that gas effects will move the datapoints but they don't change porosity that much. Shalely intervals will look like dolomites. They tend to fall to the bottom right of this plot. Here is a density / Pe. You can see now what's really interesting, the location of each of the lithology lines is displaced. There's sandstone, and then comes dolomite, and then limestone on the extreme right. But also, if we look at the sandstone line, we can see that it's almost vertical. In other words, even from a change of porosity of zero to 40%, there's not much change in Pe, which is very helpful, very useful. Here is a density / sonic cross plot. And from the viewpoint of petrophysics and lithology identification, it's not terribly helpful because it's not a good lithology indicator. And also it's a little hard to see, and we don't want to spend time doing it. But the porosity tie lines are also erratic. You'll see a curved correlation on there. That is the Ghana Hunt correlation, an empirical transform because the time series equation that we quoted, which is the straight lines, doesn't work in under-compacted formations. And here again is a sonic / neutron. Again, a little bit similar to a density / neutron. Fairly good lithology differentiation and also fairly good cross plot porosity.