Chapter 4
This chapter was all about how you would go about mapping density in GIS software. Obviously, with discrete points, you can observe density to some extent by looking at where a large amount of dots in a small amount of space are, but mapping specifically for density can give you a more specific understanding of which “dense” places are more or less so, rather than just identifying clusters. There are methods of mapping density that use dots randomly placed based on ratio per area by GIS, but this is also visually more accurate to look at than raw data, AND there are ways to make it more visually accurate, like processing this information in small areas.
There are many different ways to map density, including by area and with a density surface. Mapping by area can output a shaded fill map or a dot density map. This type of mapping is good for data that has already been summarized by area or that can be summarized by the GIS. However, it doesn’t pinpoint centers of density and can be very inaccurate for large areas because of this.
A density surface, which is created with the help of the GIS, is good for individual locations, sample points, and lines, so more discrete data. It can output a shaded or contoured map. Compared to mapping by area, it is more precise at pinpointing centers of density, but it requires more data processing.
For already defined areas, density maps can be in the form of dots or shades. As with everything, choosing one of these depends on what kind of information you are working with and what you want your map to convey. It is important to make the information you aim for as easily understood as possible.
Parameters to consider in density mapping are, cell size (if you are using a raster map), search radius, calculation method, and units.
Chapter 5
Chapter 5 was about graphic analysis and how to combine more than one layer of a map to create a map that can show correlation. Rather than focusing on physical features, this chapter talked about how to focus on occurrences in an area and can show viewers where attention is necessary.
There are two main ways to define your analysis, one being in separate layers, and another being a huge combined layer with all information in one.
Finding what’s inside a single area lets you monitor activity or summarize information about an area. This includes:
- Service area around a central facility
- A buffer defining distance around a feature
- Administrative or natural boundaries
- Manually-created area (for some sort of proposal)
For any of these when dealing with several layers of a map intersecting, you can choose to include all features that are even somewhat in your “boundary” layer, include only features that are fully in the boundary, or include parts of features whenever they show up in the boundary of the other layer. Which of these is chosen depends, again, on the map you are creating. It is best to create a map that makes it easiest to see features and important patterns, but from what the book described, this is a pretty intuitive process.
The GIS can place features that occur in both layers in a table, but if there are multiple areas within your data, it cannot recognize them as separate. I would imagine that if every area was a different set of layers, the GIS could recognize them as different that way, but this probably is very tedious and requires lots of processing.
The process GIS goes through to help with this is also described in this chapter. GIS can choose the best overlay method based on your data, and if it chooses a format your map currently isn’t in, it can convert it to that format in order to move forward with the processing.
Choosing a vector map can give you a more precise measure of areal extent, but requires lots of processing to remove silvers (errors in matching layers up) and to calculate the amount of each category.
Choosing a raster map can have varying precision based on the cell size used, and small cell sizes, which are more accurate, require more processing like a vector map would. However, raster maps don’t create the issue of silvers, they are faster to create, and they automatically calculate area extent. It is slower than vector processing though. I think personally that this is the better method as long as you use small cell sizes.
Chapter 6
Chapter 6 addressed how to represent how nearby activity can affect data between two features or using a set distance or cost in the form of a sort of “radius” around a certain area. This radius can have multiple levels if it helps to convey closeness better.
There is also an option of whether or not to include curvature of the earth depending on how large of an area your map is covering. I would imagine that it is also important to account for geographic features or barriers that would change the cost of travel, like if there is a mountain to be climbed or dodged.
There are 3 main ways nearby activity can be represented.
- Straight line distance: specify the source and the radius, and the GIS creates a nice circle using that information. It is good for creating a boundary or selecting features at a set distance, and you only need two very simple layers to do it.
- Distance or cost over a network: specify source location and distances along each linear feature you want included. This one is good for finding what is within a specific travel distance of a location accounting for means of getting there (like roads). You need 3 layers, but one of them can be directly from ArcGIS
- Cost over a surface: specify the source features and a travel cost, and the Gis will show the travel cost as it grows from each feature. This is sort of like the sorted radius, but it is based on cost and not distance. I would assume it accounts for geographic features as mentioned before.
The chapter also goes more in depth about how to actually get these different things to work in the software, like how to get a map to show distance from feature to feature, and how to make the map have several distance ranges. You can also set several different source features, and have the distances set to be “near at least one” of them after defining both.
Overall, so far from reading these chapters, I am getting the gist of how GIS works, but a lot of the specifics on how it works and what to do when I want specific things to happen is losing me a bit without having been on the software. I am hoping that once I begin working with it, I will understand better.