Data Collection and Manipulation
All coverages were downloaded from the National Park Service (NPS).
The NPS data includes coverages for roads, trails, lakes, streams, topological features, crown cover, the park boundary, and other attributes of the landscape. I used data for the park boundary, lakes, roads, and trails. Their native data sets were designed for ESRI software - either ArcInfo 7.0 or higher, and ArcView 3.0 or higher. I converted them into IDRISI vector format using FME.
The first step was to make a base map, or a template. The raw data was extremely easy to manipulate. I was able to do my analyses using only four layers:
1. CRLABNDYP_POLYGON - This is an outline of the boundaries of Crater Lake National Park. Since its default color came up as black, I created a light green palette to represent the park, and applied that color display to it.
2. LAKES_POLYGON - This is an outline of the lakes contained within the park. The lakes also came up as black. Light blue was the obvious color choice, so I created that palette, and used it with the coverage.
3. ROADS_ARC - All the roads in and around the park, including dirt and gravel roads, as well as maintained and umaintained ones. For purposes of this project, only major paved roads that people would need to get to and from the trails would be sufficient. The attribute label ""ROADTYPE" defines the state of the roads. The only roads in the park that met the criteria had the value "Secondary Highway". Such attributes such as "improved, gravel" or "unimproved, dirt" would likely be unacceptable to the vast majority of tourists. This worked out well, because the limited number of displayed roads covers both the north and south entrances to the park, as well as all analyzed trailheads.
4. TRAILS_ARC - All the trails in and around the park. As one can see by looking at the file, there are simply too many trails to analyze. What then, are the best trails to take? I took a very non-scientific method to answer this question, because of the extreme subjectivity of the answer. I decided that the best trails should be maintained, and offer views of the lake. First, I looked at the metadata for the trails file. The attribute label "MAINTENANC" described the trails as either "maintained", "unmaintained", and "no entry." I made a new vector file of only "maintained" trails, because it is more probable that the majority of tourists will want to use a maintained trail. This was performed with a simple SQL query:
[MAINTENANC] = "maintained"
The next step was to find which of these trails I wanted to do research on. Many websites (for examples, see Uhler, National Park Service, or Yahoo) listed the same eight trails with brief descriptions, including the overall view (ie, "overlook of Phantom Ship", or simply "outstanding views"), length of the trail, and average time of completion. As per those listings, I eliminated the three trails that do not offer a view of the lake, leaving me with five trails out of an original forty-one to analyze: Sun Notch Viewpoint, Watchman Peak, Cleetwood Cove, Garfield Peak, and Mt. Scott. I then added three more trails: the Pacific Crest National Scenic Trail, the Pacific Crest Alternate Trail, and the Wizard Island Trail. I added the Pacific Crest (and its alternate) trail because of its sheer size. It spans from Mexico to Canada, and includes all 33 miles of Crater Lake National Park. I added the Wizard Island trail because Wizard Island is the focal point of Crater Lake. With that in mind, I'm not too sure why other websites don't list that particular hike in with the others, but I felt it to be a hike that is worthy of inclusion. I then made another vector file consisting of these eight trails. In order to do that, I did a boolean enquiry in SQL, using the "NAME" attribute label. The operation was:
[NAME] = "Sun Notch Trail" OR [NAME] = "Mt Scott Lookout Trail" OR "[NAME] = "Pacific Crest National Scenic Trail" OR [NAME] = "Pacific Crest Alternate Trail" OR "[NAME] = "Garfield Peak Trail" OR [NAME] = "Wizard Is. Summit Trail" OR "[NAME] = "Cleetwood Cove Trail" OR [NAME] = "Watchman Trail"
That completed the template. With my base file in place, it was simple to analyze the trails.