A Crash Course in GIS

“Green Brothers not included”

This past semester, I had the opportunity to take a course about map-making. It wasn’t all old papers and compasses, though. In fact, it was none of those things; this class was about the maps of today, the maps of the future. It was about the maps made in Geographic Information Systems, or GIS.

Although you may not know it, you use GIS or GIS-assisted programs every day, and you probably see maps made in GIS software pretty regularly. What is GIS, though? Simply put, it’s a way to analyze spatial data. On a deeper level, it’s a way to organize information in space in a way that can be used to analyse data, determine the best solution to problems of distance, find the location and attributes of different objects, and more. It’s fancy maps; anything you can do with a map, you can do with GIS software, and far more efficiently.

Holistically, it’s a framework for placing data in space and working with it to find out more about that data, the location  it’s in, and perhaps even about the world as a whole.  And I absolutely love it; there’s something incredibly satisfying about being able to take a bunch of random numbers and files and put them together in a way that’s not only legible but contains practical, usable information.  And this is the core of GIS; making data accessible.

Still not sure where you might have seen GIS in your every day life?  The easiest example, and perhaps the most relevant of all, is the same thing I use to find the nearest Taco Bell: Google Maps.  The beating heart of Google Maps, the software that lets you figure out where you are in relation to everything else, is GIS.  Need to know how to get home from the city?  Just type in an address, and with the arcane magic of network analysis, Google can tell you exactly where to go.  That’s GIS.

There’s more to it than just Google Maps, though. It’s housing development and urban planning. It’s goods logistics and transport. It’s migratory bird tracking. It’s analysis of the movement of air pollution. It’s a framework for climate change science. It’s search-and-rescue operation assistance. It’s the weather channel. It’s finding out where all the corn is grown. If it has maps, if it has data, if it has something concrete in the real world, it could have applications in GIS.

That’s part what I love about it, too. It’s so incredibly broad and so incredibly versatile that proficiency in GIS software can open the door to exciting career paths that I might not have had the opportunity for before. And that’s another reason why I’m so interested in learning more about it. It looks fantastic on a resume, especially for conservation and ecology positions, both things that I’d consider doing in the future. Federal agencies love people with GIS skills; in fact, the biggest providers of GIS data is the US government.

But enough about what GIS is. I want to show you some of the cool things that I was able to do with it, and how I did these cool things. I want to show you the power over data analysis that GIS can give somebody.

Well, actually, before I move on to that, one last thing of note; there is no one GIS program. My personal favorite, and the most common one around the world, is ArcMap and ArcGIS, made by ESRI, the benign overlords of the GIS world. But there’s dozens of others. There’s several made in part by the US Government, like Grass GIS, or there are others that were just made by a private company, like QGIS. There are plenty of others, too, and some are better at certain types of analysis than others, but ArcGIS is the most common and the only one I’ve used. But, from what I understand, most of the basic principles overlap pretty well across the board.

So, when it comes to making GIS maps, there’s two large categories of data; vector and raster. Vector data are shapes made out of points, lines, and polygons. Think like a city road layout; all those lines are roads and all those polygons are buildings. Raster data, on the other hand, is based on cells, or pixels; aerial photography is a type of raster data. It’s continuous, and every space has a value. You can think of vector vs. raster data as like the default vs. satellite settings on google maps.

And so, using this data and the myriad tools that GIS programs have at their disposal, you can do all sorts of things with data. Let me give you a few examples. Throughout the class I took, we were given various tutorials and introductions to the tools of ArcMap, and saw some real-world applications of data analysis. Most of the data came from either ESRI, the US Census, or our textbook, Getting to Know ArcGIS, 4th ed., by Michael Law and Amy Collins. Here’s a map I made in a tutorial:

This map depicts the relative size of food deserts, or areas with low access to fresh fruits and vegetables, based on Wisconsin county census data. This map isn’t actually particularly good at getting the information across clearly, as it mixes the magnitude and the intensity of data into one graph, and it isn’t exactly clear. But the idea gets across; there are two counties in Wisconsin where 100% of the population is within a food desert. I think. This may have been based on low-income populations only. Like I said, not a great map. Here’s a better one that I made.

I didn’t do much for this map. I just organized the formatting and layout, and chose the colors for the data. But this is the kind of information you can display with GIS maps. Crime data, number of crimes, crime intensity, crime type, and more; it’s a good way to present information, especially to politicians and other people who don’t read very much.

This map, and the one above it, are ones that I was guided through making by our textbook. But in the real world of GIS analysis and data collection, there are no tutorials, and how you display the information can determine the type of story you want to tell. For example, here’s a map about old people in Florida.

I’m pretty proud of this map, personally. I like the layout and the way that I set it up, and the way that I included a locator map in the upper-right corner. See that? See how cool that is? I love maps.

But, anyway, one of the best features of GIS programs is the ability to differentiate shapes and regions based on attribute values within those shapes and regions. For example, changing the color of a census tract based on the percent population of old people. That’s what I did here. I’m not really sure what you would do with this information, but just spitballing here, it might be useful to know where more hospitals would be good, or where retirement centers are likely to be successful, or where not to build new schools, or things like that. It can tell you the demographics of an area at a glance, though in an admittedly limited way. But maps can also have more specific purposes, too.

This map that I made using data from somewhere that I’ve forgotten shows the number of hospitals per Illinois county. It would probably be useful in for the state government to figure out which counties are most lacking in health care infrastructure, or which have a lot of health care infrastructure. You could use this to figure out where to funnel some money for development, for example. Or you can use it to show your friends that, holy shit, there are like sixteen counties in Illinois without any hospitals.

What makes this map interesting is that it doesn’t tell you exactly how many hospitals are in each county, with the exception of counties that have zero hospitals. This map only gives a range for the other counties. It doesn’t tell you that Cook County, the county that Chicago sits in, has like 90 hospitals or something ridiculous. To be fair, Cook county is also the most densely-populated county in the state, but that’s a lot of hospitals. I remember making this map and thinking “huh, that’s a lot of hospitals. Does every county have that many?” No, no they don’t. But this map doesn’t show that.

This gets into some of the intricacies of making maps, some of the things that you have to consider when you put this information together. When you classify data into groups or chunks, you have to figure what kind of scale you want to classify by. Should each group have the same number of features? Should each group have the same level of spread? Should I have made Chicago it’s own fucking group since it’s so different than every other one? Things like that.

Up until now, all the maps that I’ve talked about have been in entirely vector data. But this map up above is a little different; it uses both vector data and raster data. The aerial photograph underneath, taken in 1970 or something, is an example of raster data; each pixel is a value, and if you put them all together on one plane you get something that kind of looks like a place. So this map, for example, is useful for examining the changing landscape of my university’s campus. And you can do the same thing with changing land cover, land usage, and the like. Raster’s good for weather, too, and elevation. Really, it’s good for anything you can find a use for it.

There are so many more maps out there, so many more possibilities for the power of GIS. And like I said earlier, that’s what I love about it. There’s so much you can do with it, so many ways that you can amass and quantify and share data. If you know how to use a GIS program, it effectively turns you into a citizen scientist. Want to see the racial distribution of your town? Download some census data and boom, you’ve got it right there.

I hope one day to be able to use my GIS powers for good, either as a GIS consultant for my local landscape surveyors and planning and zoning commission or as a professional bird-tracker for Cornell. I’d love to work for the National Park Service, too, doing something, perhaps, with mapping forest regeneration after wildfires, or mapping which campgrounds are most heavily used and/or abused. Stuff like that, to me, would be so cool to work on. Or I could work for the CDC, for example, and try to pinpoint the spread of infectious diseases, just like John Snow, the first GIS master. He made maps and everything!

I’ve only scratched the surface of GIS in this article, and I’ve barely dug any deeper myself. I know how to use some of the tools, but the ones I know are only a small portion of the toolbox that ArcMap gives me. I’m quite excited to learn more about the world of digital cartography and spatial analysis, and I hope you are, too. Or at least that this article didn’t sound like some sort of lecture. There’s plenty of places to learn more about this stuff, if you really want. But I at least hope I gave you a sense of context for Google Maps, the Weather Channel, and all those maps you see in the newspaper all the time. Maps are cool, man.

Oh, and one last thing: here’s a website I kind of made as part of my class. I don’t know if you need to sign in or not to see it, but I thought I’d include it anyway. It tracks how happy I felt in different parts of campus. See if you can find the hidden goose.