A Journey of a Thousand Map Units Begins With a Single Dataset–GIS 101

So you heard about this GIS thing and you thought to yourself that it might be kind of cool to be able to make your own maps.  But then you looked at the software and decided that it’s rather more dense and complicated and esoteric than you’d like, so you decided that maybe you’d pass on making your own maps and instead just use whatever you could find on the Internet.

Fair enough, but the truth is that making your own maps really isn’t very hard.  Sure, the software may be a little dense and complicated and esoteric, but so is most every word processor on the market.  But this doesn’t stop any of us from firing up Word and using it to write a letter.  The same goes for email.  How many of you can claim to know more than 2% of the full capabilities of Gmail?  I know I can’t.

GIS software can indeed perform a great many complex and wondrous scientific operations, but a working knowledge of them is not a prerequisite for using the software.  With a little bit of knowledge and the right data, pretty much anyone can make a respectable map.

Which is what we’re going to do.  Relax – I’ll walk you through the process.  It probably won’t hurt at all.

The first order of business is to secure the software we’ll be using.  There are a bunch of good options out there, but for today’s exercise we’ll be using QGIS, because it’s solid and dependable, but also because it’s my personal favorite.  Head on over to their download page and download and install whatever version of the latest release is appropriate for your system (I’ll be using 2.12.3 throughout).  I have chosen the latest release over the LTS (long term support) version because it includes a variety of new features that are well worth having.  Besides – if you decide to stick with this GIS thing you’ll be updating your software on a regular basis.

Once you have installed QGIS, fire it up.  Right off the bat there are a couple things we’ll want to do.  The first is to simply rearrange the UI to suit our purposes.  The UI is completely customizable (and easily so – just click on things and move them around), so it’s a simple matter to arrange things to better suit our own habits.  In time you’ll determine the setup that best suits you – in the meantime I’ll show you the setup that suits me (out of the box, at least).  The next thing we want to do is change the CRS of our project.  When we first fire up QGIS it defaults to EPSG 4326 (also known as WGS 84).  Maps are projected, necessarily so because when mapping we are representing a three-dimensional object (Earth) two-dimensionally.  Therefore we must project a map in order to draw it.  Projecting is inescapably imperfect, therefore every map distorts something in some way.  We use a CRS (co-ordinate reference system) to project a map.  WGS 84 is intended for maps of a global scale, but our map will be on a local scale (in Massachusetts).  Therefore we’ll change the CRS  to a projection that distorts less on our more focused scale.  For this map we’ll use EPSG 26986, NAD 83/Massachusetts Mainland (Meters).  If you want to know more about projections and co-ordinate systems, a great resource can be found here.

 

 

Now we need data.  For this map our task will be easy (but don’t get used to it.  Finding appropriate data is usually the hardest part of GIS), as Massachusetts has a wonderful State GIS agency called (you guessed it) MassGIS.  We’re only going to use six datasets for our map, which really isn’t much as these projects go.  It’s still data, though, and it needs to be managed, so let’s talk about that for a minute.

GIS is driven by data.  As such, practitioners tend to amass enormous amounts of data.  Data that must be organized.  Maps are made out of data, and a failure to properly wrangle said data can be crippling.  Where did I put that data from 6 months ago?  Which one of the 14 folders called “Project X” did I put it into?  Or is it somewhere else?  It’s a road dataset, so did I put it in the huge folder called “Roads”?

I didn’t just make these questions up.  They are questions I have actually had to ask myself at earlier, less organized points in my career.  Trust me.  Properly managing data is the single most important part of modern GIS.  As an added bonus, well organized data is easier to back up.  The value of this cannot be overstated.

This is how I do it.  I start with one master folder called simply “Maps” (in Windows you can make this folder a distinct Library.  I strongly advise doing so).  This folder is where I keep everything – data, projects – everything.  The advantage of this is that I simply have to back up my “Maps” folder and I can sleep soundly at night.  The disadvantage is that my “Maps” folder can get pretty damn large (and I don’t even work with Big Data.  If I did I think I would have to store it elsewhere).  Inside my “Maps” folder are two other folders:  “Data” and “Projects”.  My “Projects” folder contains a myriad of folders (usually one per project), as well as a separate folder called “Saves”.  This is where I keep all the ‘Saves’ of projects, whichever software I happen to be using.  This way all my software defaults to the correct location whenever I ask it to open a project.

The “Data” folder gets considerably more complicated.  And personal.  Its contents are more individualized and dependent upon the work most often performed and the types of data needed to do so.  In my case, I work most often (almost exclusively) in Massachusetts.  Because of this, MassGIS is my go-to source for data.  So it just makes sense for me to organize my data in a manner similar to the method MassGIS uses to organize their data.  Much of their data is organized by town, and I follow suit.  My “Data” folder slowly fills with folders named for towns in Massachusetts.  Inside each town’s folder I further separate the data into folders for vector data (points, lines, polygons) and raster data (images, digital elevation models).  MassGIS organizes some of its data differently, and for this data I include separate folders, appropriately labeled.  For instance, I have a folder called “Statewide” for those datasets that encompass the entirety of Massachusetts.

For this project we will be using two statewide datasets and four town datasets (three from one town, one from another).  Start here to get them.

Again, the choices you make will be your own.  I’m just describing how I do it.

First, I scroll down the MassGIS download page until I reach the section titled ‘Transportation’ under the ‘Infrastructure’ heading in the ‘Vector Data’ section.  There I click on ‘Mass DOT Roads’, then ‘Download these layers’.  I then scroll down to ‘Greenfield’ (in the left-hand column) and click on the filename (eotroads_114.zip) to initiate the download.  I download the file to Maps→Data→Greenfield→Vector→Roads.  I click my way back to the main download page, then in the next section (‘Other Facilities and Structures’) I download ‘Building Structures (2D, from 2011-2014 Ortho Imagery)’  in a similar fashion (structures_poly_114.zip to Maps→Data→Greenfield→Vector→Structures).  Then, under the ‘Physical Resources’ heading, I download ‘Contours (1:5,000)’ from ‘Elevation and Derived Products’ (while I’m there, I also download this file for the next town over: Montague).  Lastly, I move on to ‘Inland Water Features’ and download ‘MassDEP Wetlands (1:12,000)’.  This one is Statewide, and it is divided into two layers, a polygon layer (wetlandsdep_poly.zip) and a line layer (wetlandsdep_arc.zip), both of which I download to Maps→Data→Statewide→Vector→Hydro.

Now that we have our data we can start building a map.  Our first order of business will be to navigate to our ‘Maps’ folder in QGIS’s Browser panel and then add it to our favorites (right-click on it).  This will greatly facilitate our future mapping endeavors.  Then lets add in our two contour datasets (all you have to do is double-click on them in the Browser Panel – QGIS can read a zip file) and symbolize them (by double-clicking on their names in the Layers panel).

 

 

Maps are built in layers, and the order of our layers should roughly mimic the world we are attempting to depict.  So our next order of business should be to add and symbolize our two hydro (water) layers.  For some reason, they come from MassGIS double-zipped, but all that costs us is an extra mouse click.

 

 

Next, lets zoom into a more focused area of interest, then add our road and structures layers, symbolizing them as we go.

 

 

Lastly, lets rename our layers in the table of contents as something more human-readable (just right-click on them individually) and then save our project.

 

 

So now we’ve made a map.  What do you say we put it onto a piece of paper so it can actually be a useful thing?

Click on File→New Print Composer.  Once open, I first change the paper size to the one my printer spits out (ANSI A: Letter).  Then I use the ‘Add New Map’ button to add a map to the composition by clicking and dragging corner to corner.  Then I click on the ‘Item Properties’ tab and change the scale of the map to 1:24,000.  This is because we here in Massachusetts still backwardly use feet and inches, so if I make the scale 1:24,000 on my printed map one inch will equal 2,000 feet.  This makes my paper map a useful and easy-to-use tool in the real world.  Those of you fortunate enough to live in the modern world can do the same thing with the metric system, just with easier math.

Once the scale is set, I change my project from a picture into a map by adding a scalebar, North arrow and title.

 

 

Switch back to your map window and save the project.  Congratulations!  You made a map.  Feel free to print it out and use it in the real world.

Now, in all truth and in the interest of full disclosure, we’ve barely scratched the surface of the cartographic capabilities of QGIS.  And we haven’t touched upon its processing or analytical capabilities at all.

But that’s okay, because we did accomplish what we set out to do – we made a custom map from scratch.  A map that we can print out for use in the world (alternately, we could export it as an image for use in a document or on a web page).  And it wasn’t terribly difficult, which is the point I set out to make.

Baby steps.