Mapping the GIS Adventure – Lab 6: Y’all, That Texas Coastline is Killer

You might notice that I skipped Lab 5. This is due to the unfortunate fact that I accidentally lost my Lab 5 pdf. I still have the map, so I think I’ll be able to recreate it, but I haven’t had a chance to go digging. Since it’s been a little bit since my last post, I decided to skip ahead to Lab 6.

This lab was all about fun of digitizing, which my professor has frequently promised will be a skill I will utilize again and again. When I started the map, I was back in the midst of the crazy semester, getting swamped by all the new information and work coming my way in GIS. I was desperate to not turn in this map late, and also very, very confused.

Lab6

Digitization isn’t too hard per say, but it is time consuming. And potentially painful. I remember the night I sat down to digitize the whole map of Texas. But before that nice recollection, the details: The goal of the map was to provide a clear map of the ecoregions of Texas for state parks of Texas.  The requirements:

  • projection of the Texas data to a Texas centric projection
  • georeferencing of the Texas image
  • correct digitization of the polygons
  • assigning two attributes to each polygon construction
    • attribute 1: integer field containing the ecosystem
      number from the map
    • attribute 2: a text field with enough characters to
      hold each ecosystem name
  • An intelligible legend

So, back to the night I sat down to digitize. I had already successfully georeferenced the Texas image – or so I had thought. One of my friends kindly showed me that my amount of georeferencing points was way too high (28 as opposed to an upper limit of 20) and I hadn’t applied the suggested 2nd polynomial warp. So after georefencing again, I found myself ready to digitize. I began drawing polygons and utilizing a snapping tool, so that way I could avoid the dreaded “polygon slivers.” I can only imagine what kind of ruination would have come to me with slivers.  As I created my polygons, I realized why it was going to be torture: as I descended upon the Texas

I believe this was about an hour after I started making those polygons.
I believe this was about an hour after I started making those polygons.

coastline, I found myself upon these craggy sections. I wanted to make sure I captured most of the detail, but also wanted to avoid having too many nodes. And as I clicked my way through that coast, I realized I was clicking my way to what also felt like carpal tunnel. Oooh, the agony. Worse, the polygons do still have some issues: along that coastline, which I did streamline a bit, you can see that the details aren’t highlighted due to coloring. This doesn’t worry me as much, because of the small number of parks along that coastline and the fact that such details of the coastline were less necessary for the goal of the map.

However, I did persevere and was able to complete the digitization. So I then proceeded to color the polygons and create that intelligible legend. As to how “intelligible” that legend actually was, well. I found I was able to simply pull the polygons into the legend, but I couldn’t figure out how to remove duplicates from the legend. I have since learned that it’s all in the legend details – simply delete the duplicates. But alas, at the time I did not know this, so for every region that exists, there is the color swatch and the number of those pops up. The Texas state parks locations also are included. However, I do wonder if my data had the names of the Texas state parks. If I did, that probably would have been a good addition to the map because I feel like while the state parks probably would have known where they are, thanks to the addition of the county lines. At the same time, there are a lot of state parks on that map and it would have just really added to the overall clarity of the map.  I also could have done well to provide more information about where my data came from. Because don’t we know that I did not make that stuff myself. I also didn’t mention my coordinate system or datum, which does no good.

In general, this is around the time that the labs starting becoming more about the technical aspects behind it. The results were of course extremely important, but the little things done behind the visual started to become very important. We went from simple technical aspects to understanding that while you could hypothetically fake a pretty map on the surface, if you started to leave out all that nitty gritty technical stuff, your map is fairly worthless. Of course, I didn’t figure this out until several maps later.

Mapping the GIS Adventure – Lab 4: Maps Are Where “Y’all” and “Eh?” Can Be Said in the Same Breath

NunavutLab4

I remember this mapping assignment with strong feelings. This was our fourth assignment, and we were into our fifth week of the semester, and I was struggling a little when it came to these maps. I was also feeling inadequate, as I had not yet made friends in the class and was perceiving everyone around me to be plugging along just fine. Of course, actually, everyone was learning this new skill as I was, and we all probably would have benefited if we had started talking to each other a lot sooner. So the map that gave me such terror was actually pretty simple: provide a size comparison between the state of Texas and the Canadian province, Nunavut. Along with this, the skill we were to learn was that of projection, both on-the-fly and permanent.

If there's any question, I reprojected all of these themes into the Canada Lambert Conformal Conic coordinate system.
If there’s any question, I reprojected all of these themes into the Canada Lambert Conformal Conic coordinate system.

The requirements for this map were similar as those of the previous labs, but also required us to get a dose of ArcCatalog and create a table that listed all of the data that we were using for the Canada map, and their original projections, and then the projections we reprojected them to, which needed to be a conformal projection. These were to be permanent reprojections, while the Texas map didn’t require permanent reprojections, but instead on-the-fly projections.  Presently, this doesn’t seem hard at all. At the time, wrapping my mind around it was impossible. The biggest issue was merely that I didn’t understand that on-the-fly projection could trump permanent projections. So even though I reprojected all of my themes correctly, I never changed the coordinate systems of the data frame, so I never saw a difference, until I learned what was going on later.

One big thing: this was where we were supposed to learn the difference between the  “Define Projection” and actual “Project” tools. My professor warned us all, multiple times to be very, very careful that we understood the difference between the two. While I certainly had an unfortunate experience with this map, I do thank the heavens that I understood the difference almost immediately. It would probably have been the straw that broke the camel’s back otherwise.

Otherwise, this map is one of my favorite maps I’ve created so far. I won’t lie – it’s because I find it pretty. The raster I used to show the physiography of Canada utilizes the same color spectrum as most physiography maps, but never before have I wanted to wax poetic on one. That might just be a dedication to Canada’s physical appearance, and I’m not even ashamed. In terms of other map elements, we were also required to show the railroads and major roads of Canada and it took me a while to figure out which colors worked best, but I feel pretty good about my bright purple for major roads and midnight blue for railroads. And I must shout kudos to esri’s ArcMap railroad symbology. Well done.

Otherwise, we were also supposed to clearly highlight Nunavut and clearly show Iqaluit, the capital of Nunavut. You’ll note that someone has definitely learned to use the “select” tool, as Iqaluit is the only city on there. The only one. Hurrah for me, because when I first took a stab at this map, I still had the selection skills of Lab 2. As for highlighting Nunavut, I think I don’t do so as successfully as say, Texas is highlighted in the inset. This is partially due to the fact the physiography needed to be clear, but was also really strong. So even though Nunavut has a different overall coloring, a red outline, and two large names shouting out to the viewer, the eye is not immediately drawn to it. To rectify that, I would have probably made the physiography a lot lighter with transparency, so that way Nunavut would have stood out more. I also would have done something about the outline, such as a black outline to strengthen the border of it.

The inset of the Texas map was pretty easy. Texas was red, outlined in black, and the rest of the visible states were yellow, outlined in gray. Not too hard to make that work, given physiography wasn’t necessary. The only aspect I found giving me issue was the scale, but once I adjusted the data frame for the Texas map, I was able to have them at a matching scale. Based off of the map, I hope you all agree with me that Texas, while my favorite big state, is still definitely smaller than Nunavut. If you don’t, I’ve failed because Nunavut has some 500,000+ miles on Texas. No, not EVERYTHING is bigger in Texas…

As for my technical aspects of the map, such as the legend, scale bar, and North arrow, I’m pretty satisfied with my space usage and placement, though I definitely would have done well to somehow fit that scale bar directly beneath Canada, so that way the ends lined up with the edges of Nunavut. I just did the finger test myself and they totally would have matched up and would have saved a lot of readers the irritation of keeping their fingers the same width apart as they moved them to Nunavut.

What cracks me up now is how this map turned from a pain in my rear to truly, one of my favorite maps. Under the rules of our class, I was allowed one free lab that  could be turned in late with no late penalty. As we already had one lab due a week, I sat on this lab all the way to Spring Break to work on it with no other requirements breathing down my neck. I remember redoing the assigned tutorials and changing the projections and realizing just how not hard this assignment was. I was able to have fun with it, and realize that a colored background wasn’t always the best way to go. I understood things I hadn’t gotten before. Lastly, you can bet getting that map done and turning it in was the best kind of relieving success.

Most of all, I learned the importance of doing your work when you’re assigned it, and crawling before you walk. The struggle to understand a concept may be torture, but once you learn it, you learn it. It helps set you up for the next struggle, with a lot more ease than if you hadn’t survived the prior one.

 

Mapping the GIS Adventure – Lab 3: Prior Planning Prevents Poor Map Presentation

So, I was working on my preparation for my post about Lab 3, only to realize that I had erred: my Lab 2 map that I had posted was actually my Lab 3 map. Apparently, I had since deleted my Lab 2 map because it was poorly done. So I thought to myself, what now?

Since I have already pointed out all of the issues with the map you saw previously, I decided it would be fun to try and redo the map again. I looked up all the requirements my professor had for us, and pulled out the old data, and here we are, the new and improved Texas Transportation Corridor map:Lab3Redo

So, the requirements aren’t much different from what I described for my first Texas Transportation Corridor. I needed the following:

  • Major Texas cities with built up areas, displaying relationship between cities and trans corridors
  • Smaller Texas cities and towns, labeled, showing how non-major metropolitan areas benefit
  • Existing roads and corridors, clearly differentiated
  • Visually pleasing raster image as background
  • Reasonable scale
  • Graphical scale bar and representative scale bar with appropriate units
  • Graticule
  • Legend
  • Title
  • Citation of data source and cartographer
  • Map projection and datum
  • Date created
  • Neatline

In my last map, all that marked the cities, major or not, were circles with varying sizes. I found myself choosing to represent the major cities this time with polygons of their areas, exactly for the proper display of their built up areas. For the major cities, I made sure to also bump up the text size and bold and add a small mask as well, so that way the names would stand out. I also minimized what I considered “major cities.” I originally hadnot only San Antonio, Houston, Dallas-Fort Worth and Austin, but Corpus Christi, Lubbock, and El Paso as major cities too. While these are certainly notable cities in Texas, they differ wildly in size when compared to the original four.

When I decided on the small towns I would include, I relegated Lubbock, El Paso, and Corpus Christi into that group. I chose the following cities for several reasons: they were either well known, near a junction of corridors, major roads, and interstates, or they were in an area where I had yet to include many cities to aid understanding of how areas may be affected. I included these reasons in the following list.city table

For the small cities, I simply used a light blue dot of the same size as representation.  The text size is smaller, and they aren’t bolded, although they do utilize Arial font. They also have a mask, because against that raster it is hard to get anything clear enough for reading.

When looking back on both my first version of this map, and the Department of Transportation’s map, I noticed that the DoT’s map had the corridors separated into two groups by color. This was originally unclear as to why for me, but once I dug through my data a little more, I found that they could be defined by priority, high or low. A google search later, I learned this meant the kind of traffic they handled. I agreed with DoT that this was important for map viewers to know and thus, my high priority roads are defined by a goldenrod color, while my low priority roads are defined by a sort of electric blue. I also bumped up their thickness to ensure they were clearly important. Interstates are a eye-catching but muted magenta and a little thinner than the corridors, with the major roads being a gray that’s even thinner, so not to draw from the most important map elements.

Beyond those major map elements, I am far more comfortable this time around with my usage of technical map requirements. I feel like I utilized my space more and properly balanced it. I also am far more comfortable with my bar scale. My only issue is that I failed to include the raster properties in my legend.

What I probably learned the most from doing this map over? The importance of making a plan before starting a map. I thought I was doing this by listing the requirements and noting why they were necessary, but I found myself wasting time on simple things like deciding what kind of a halo was necessary, what colors to use, and which cities were major cities. If I had taken the time to puzzle these all out before I started on the map, I wouldn’t have wasted so much time in ArcMap. Next map I make, there will definitely be a plan drawn up before I enter ArcMap.

So, what do we think? Improvement?

Mapping the GIS Adventure – Lab 2: The Texas Transportation Corridor Probably Had a Bad Cartographer

Lab 2

My professor enjoyed providing our assignments through these faux-memos he created, as if we worked for a GIS company and he was our boss. We also had the opportunity to be promoted, never, it seemed, to an actual paycheck. At this point, my memo promised me that I was “highly valued, but poorly paid.” Given my professor’s tendency for straight talk, I have a feeling this may be exactly my future.

But it did provide the incentive. Apparently, all I needed to do was make good maps to progress. So, I put my nose to the grind for this map, and promptly found it ground off. Our professor informed us that Governor Perry had proposed a Texas Transportation Corridor, which never made it to fruition. Sad, but worse was that my “boss” required a better map displaying the proposed corridors than what was available. Alas, I worry my map isn’t much of an improvement, but before I offer you an alternative, here’s what’s up with mine.

First, the givens. This was my first time using ArcGIS, and our professor was also working on our cartography skills. We were required to have a legend, scale bar, a north arrow, a scale, graticule, title, and data source. All of which I provided, though looking back, I would have kept my legend where it is, though I certainly would have played with it more so that way its titles were correct. I would have moved the data source information further into the bottom right-hand corner to neaten everything up. I also would have utilized all the space I had. My north arrow could have stayed in the white space, but moved to the top left-hand corner, while I feel my scale would have been better off in the white space in the top right-hand corner. I don’t even want to discuss my scale bar, because these were the days of me completely misunderstanding how to make a scale bar pretty and actually helpful. Alas, this was a problem until probably Lab 10. But that’s another story.

I feel fairly comfortable with my representation of the proposed corridors, noted by the bold, turquoise lines that criss-cross over the map. These are meant to be the most prominent feature of the map, so that’s okay. Everything else I have issue with. I also included any major roads in Texas, which are noted by thin, blue lines, and interstates, which are noted by a golden line, barely visible due to the fact that my data for major roads includes interstates. This oversight is something I think I would have avoided now by extracting the interstates from the major roads data, or I could have just dumped the major roads data altogether. The interstates certainly could have done with some labeling for the sake of clarity.

Unfortunately, the cities are also a mess. I remember thinking that only the big cities of Texas would have been good to have on the map. Unfortunately, I lacked any understanding on how to eliminate all cities but those main cities from my data and thus ended up various other cities included on the map. I also attempted to categorize the sizes of the cities by their dots. So, this would obviously mean that San Antonio, Dallas-Fort Worth, and Houston would have fairly large dots. Definitely, San Antonio would have a bigger dot than Austin. For some reason, my population density circles were messed up, probably due to the attempts to delete cities off of my map.

We were required to included some kind of topographical layer, so I have the elevation in the background at a middling level of opacity, so that way you don’t lose the gradient detail, but it doesn’t overwhelm the other map elements. Lastly, there was a gray color for the county lines, just so that way they could be identified if necessary. If I had taken off the major roads and figured out my city elimination, I probably could have included the county names.

I do remember loving playing with the colors on my map. I’ve always paid attention to what can hurt the eyes and such when it comes to color and design, and what colors work well together. Though the blue is a little bright now, it definitely was fun to get everything to be clear enough to be understood.

Now that you’ve seen my version of this map, what do you think of the Department of Transportation’s version:

trans_texas_corridor_map_1

Did I improve upon it? Or is it no better?

Mapping the GIS Adventure – Lab 1: The Schools of the SEC Are Dense

map1

 

This was the first lab we were assigned in Principles of GIS. Since A&M was in its second official year in the Southeastern Conference, my professor had us make our own accounts with esri ArcGis. He then gave us a simple Excel sheet with the locations of all the universities in the SEC (football, you know). His requirements were that each school had to have distinct symbols, and of course a legend, and along with a base layer that was from esri’s options and then explain later to him why we chose the symbols and base layers selected.

Uploading the shapefile was fairly simple, as was providing each school with a different symbols. The snag resulted when I decided that I could give each school a very different symbol, such as, a picture of their mascot. At the time, this was fairly difficult, because I could not figure out how to provide these distinct symbols to the respective school as long as they were in the same layer. The symbol image change that ArcGIS has changes all of the points in the layer to the same image. So even though I was uploading a distinct image, I still failed to successfully upload and retain a different image for each school. In the end, the only solution I could figure out was to make separate layers for each school, and then upload each individual image for each school/layer. On that complicated note, I later was able to talk with some esri recruiters when they visited A&M. I explained the predicament I had found myself in and watched as a recruiter furrowed her brow and said, “There has to be an easier way.” I assured her I looked through every option and even went to the help and could notfind anything. If anyone has played with that, I’d like to hear what you did.

The base layer I chose was simply population density. I originally wanted to be a bit more out there and put the base layer that shows the levels of binge drinking, but then I chickened out, worrying my professor would find it inappropriate – though it may be worth noting that the levels were actually high in the areas corresponding to the schools. So I just went tame and put population density. In retrospect, my symbols are less helpful because of their shapes actually make it hard to see their county levels at this scale. And I had chosen this scale because I wanted him to be able to see all of the schools on the map. My first lab was so much easier than I had ever expected, and would ever have again though, and I’m thankful.