Let's say you have a census unit, or a neighborhood, or whatever it happens to be where you know that the population count is 1,000 people. Then, let's say you have a smaller area. So, this might be within your study area and you want to estimate the number of people that live there. So, we know that the population is 25 percents or so we know that the area is 25 percent of the larger area, what is the population of that smaller area? Actually, we have no idea. Now, you can estimate it using areal interpolation and I'll show you how to do that. Really what we're doing, is we're taking a population and multiplying by the proportion of that area. So, in other words, we know that the larger area is 1,000 people. We know that the smaller area is one quarter or 25 percent of that. So, if we multiply the total count by 0.25 that proportion, that gives us a value of 250. But remember, this is just an estimate. So, all we're doing is saying, we know the proportion of the size of this area in relation to this larger area. So, we're going to use that proportion to estimate the smaller one. That's really the fundamental idea behind areal interpolation. But there's lots of reasons why that might not actually work out. So, for example, it maybe that in this area, it's actually a farmer's field and nobody lives there, or it's an industrial area, or it's a warehouse, or whatever. Then, everyone lives in a condos over here, out of the entire 1,000 people, they all live in this one area. So, the basis of areal interpolation is this assumption, that the thing that you're interpolating often it's people, it doesn't have to be, but often it is. Is you're making the assumption that those people are evenly spread out across the entire study area and that you can then just carve out a proportion of those, and then estimate what that is from there. This can be done, it's done all the time, that's why I'm telling you about it now. Is even though it's a flawed measure and there's some assumptions that are made about it, it still gets used a lot. Often, it's in that category of better than nothing, like at least it gives you some kind of an estimate. You're often not doing it on just one area, you may be doing it for 100's of census units across a city or bigger area. So, then, you may not be as worried about minor assumptions or errors in a data set and it will give you some kind of approximation of that variable. So, this is a long introduction, but the idea I want to get across here is that, this has to do with doing some very simple overlay analysis and field calculations. It's way of tying together some of the concepts we've talked about and demonstrating how that works. It's also a handy tool, in order to make these kinds of estimates. If, for example, we wanted to estimate how many people live within 800 meters of this library. If we zoom in here a bit, we can see that we've got some census tracts. I've got population counts per census tract for this. So, I've done that intentionally as opposed to density, because here I wanted to count or estimate the number of people within a certain area, not the density. So, how can we do that? Well, first we can generate a buffer of 800 meters around that library, around that point and that will produce a new feature class with the new polygon, that we can then start to use for our areal interpolation. So, now, that I've generated my buffer, I can use that to select the census tracts that intersect that. The idea being that I want to isolate those census tracts and put them in a separate feature class. So, here are the population counts for each of those census tracts. I've saved these into a new feature class. I'm not going to go through every dialog box and show you every single one of those, because I mainly want you to understand conceptually the steps that I'm going through. You should be able to, through various other videos or explanations, look at some of the details in terms of how to do this. But I'm just showing you the main steps here as to how this works. So, here are the population counts for each of the census tracts that intersect my 800 meter buffer. Remember, the idea is that I'm trying to estimate the number of people inside that buffer. Okay. So, here's my population for each of the census tracts. Now, I'm going to do a Calculate Geometry to get the areas of each of these census tracts. I'm using NAD 1983 UTM Zone 17, so I know that the areas will be accurate. So, here are my populations I've added in my areas. Now, what I need to do, is estimate the areas that are inside the buffers. I could use an intersection for this, but it's simpler to just use eclipse. So, that's what I'm going to use here. I'm using the buffer to clip the census tracts. So, that's my results. So, that's the parts of the census tracts that are inside the buffer. I've just added in the underlying census tract, so you can compare the part that's inside the buffer versus the part that's outside the buffer. Of course, the idea here, is that I'm going to estimate the proportions of each area that's inside versus outside and exactly the same way I did in the introduction. So, here's the total population for each census tracts. Here's the total area for each census tract. Here's the clipped area, I just did a Calculate Geometry on the clipped part as well. So, now, I can just do a field calculation, to easily calculate what's the proportion of the clipped area in comparison to the total area. So, here, it's well, 62 percent, here's 36 percent, and so on. So, now that I have those proportions, I can and exactly the same way I did up here, apply the same kind of calculation. So, here are the proportions. I multiply that by the total population. So, I'm basically saying, what's 62 percent of this population over here? I'm multiplying that by that, just like I did in my example here and the result is 1,930 people. So, 62 percent of the area is inside the buffer. So, I'm making the assumption that 62 percent of the people are inside the buffer and I do that for each one of these, to get an estimate for each of the parts of the census tract that are inside the buffer. So, all I want to do, is add up all of those proportions, the estimated proportions for all of these. One way to do that, is I could just right-click on the top of that field and create a statistics dialog box. From that, I can see that the sum is 10,131. So, my estimated population inside 800 meters of that library would be 10,131 people. So, that's just a way of showing how areal interpolation works. Like I said, it's not the most accurate thing necessarily, but often it does come in handy.