McConkey – Week 2

Chapter 1:

Mitchell opens by explaining how much GIS has grown with developing technologies. Mitchell proceeds to delve into the main theme of Chapter 1 by defining GIS analysis and concisely describing how to conduct such analysis in a recommended series of steps. The steps are framing the question, understanding your data, choosing a method, processing the data, and examining the results. Mitchell then explains the types of geographical features and how they are represented. The three types of features are discrete, continuous, and summarized by area. Discrete features can be easily plotted on a map because they describe whether the feature is present or not. Discrete features may be represented by dots, lines, or other methods. A discrete feature might be a dot representing the location of a well or a curved line representing the path of a river. Continuous phenomena are features such as weather and temperature. Continuous phenomena can be constantly changing. The areas between sample points (ex. weather station)  require interpolation to produce a value. Meanwhile, summarized data represents counts or density of features within an area’s boundary (ex. zip code parcels colored depending on the average number of households). Mitchell goes on to describe vectors and rasters, which are ways of representing features. In regards to vectors, each feature is a row in a table and is defined by x and y values in space. Vectors can include dots, lines, and areas (shapes). Features in a raster are represented as a matrix of cells in continuous space. Mitchell then explains the different types of geographical features in more detail, even though some are self-explanatory.  The geographical features include categories, ranks, counts, amounts, and ratios. The chapter ends with a brief summary of working with data tables. One thing that I admire from Mitchell’s writing is how he includes several examples of potential features or applications. Various maps are included throughout the text, which brings clarity and emphasis to these ideas. The writing has a typical but well-structured pattern in which information is brought forward. Once a new term is introduced, it is defined and an example of its application is typically included. The flow of information is more easily digestible with this structure and the graphics serve as a great visual aid.

Chapter 2:

The second chapter covers the topic of mapping and why it is important. There’s a lot fewer GIS vocabulary terms in this chapter as it pertains mostly to concepts. Ultimately, creating maps helps you see where features are present or absent and to recognize patterns. Sometimes finding these patterns is the goal of the map. Mitchell reminds the reader that it is important to keep the audience in mind when making a map. What data is depicted, and how the data is presented affects the overall clarity of the map. Therefore, mismanaging the presentation of data can make a map unnecessarily complicated. Making sure your map expresses the information in the most efficient and clear way possible is something I’ve put a lot of thought into when working on previous projects, so I’m glad Mitchell touches on this simple but important concept in a lot of depth. I never knew this, but Mitchell states that it is usually best to have no more than seven categories on a map. Apparently the majority of people can distinguish up to seven colors or patterns on a map, but begin to have more difficulty discerning information when there are additional categories. Logically, this makes sense as having too many categories can easily clutter a map. Mitchell makes it clear that the amount of categories or symbols used should be chosen given the purpose of the map. He also expresses that when color coding regions on a map, it is a good practice not to use random colors but to instead assign similar categories to similar colors. In a lot of cases, this can improve clarity. For example, a land use map may use light green for lightly forested areas and a darker green for deeply forested areas. Alternatively, you can combine the use of different colors, shapes, and thicknesses when appropriate. The example Mitchel uses is for a road/transportation map. In this case, it is a good idea to make freeways thicker than highways and highways thicker than local streets. It is also a good idea to use familiar symbols when possible. For instance, most people associate two parallel lines connected with a series of horizontal lines as a railroad track. It would not be good to utilize the common railroad pattern for a highway or vice versa. At the end of the chapter, Mitchell asserts that it is important to use statistical analyses when quantifying the relationship between features or patterns.  

Chapter 3:

Chapter 3 pertains to mapping the most and the least and is the most comprehensive chapter thus far. Mapping the most and least of something is a good way to visualize relationships between places. This includes discrete features, continuous phenomena, and data summarized by area. Mitchell makes note that data is generalized to create patterns. To expand on this, I believe since maps are a reflection of the real world, data is almost always generalized to some degree. Moreover, there is a lot of reiteration in the beginning of the chapter. I appreciate this approach as it helps you remember and retain what was learned in chapter 1. Mitchell reviews counts and amounts, ratios, and ranks specifically, which are important terms to keep in mind as they can all be grouped into classes. I actually went back and discovered that the explanation on ranks is verbatim to that from chapter 1, and the same map is used as an example as well. Mitchell then goes on to explain classes and how to use them both manually or with a standard scheme. The four most common schemes (natural breaks, quantile, equal intervals, and standard deviation) are defined and mapped for an easy comparison. Each type of scheme has its advantages and disadvantages, but none are necessarily better than another because each scheme should be used depending upon the distribution of data itself. For example, if the data is heavily skewed, then an equal interval will provide a misleading result. Next, the ways in which quantities can be expressed are covered. As with the different scheme types, each quantitative type has its purpose and its own advantages and disadvantages. Graduated symbols vary in size, which is good since people naturally associate symbol size with magnitude. Graduated colors accomplish a similar thing, but with areas and continuous phenomena. Colors are not always associated with magnitude, but people tend to assume ‘dark’ means more and ‘light’ means less. I am most unfamiliar with charts, but the text describes the use of charts very well. Charts make it easier to read patterns or feature values, but they may obscure patterns by compromising visuality. Contours, or contour lines, make it easy to see the rate of change over an area, but individual feature values may be harder to determine. Contour lines are great for depicting air pressure gradients or changes in elevation. Additionally, 3D perspective views grant a high visual impact and can depict elevation very well. However, this style of map makes reading individual feature values more difficult. 

Chapter 4: 

Chapter 4 deals with mapping density. While density maps are not very good for examining the location of individual features, they are reliable for looking for patterns and for mapping areas of different sizes. While it is true that you can examine the location of features by plotting the location of all the features on the map, it may be difficult to truly differentiate the different concentrations within the map accurately. Density maps utilize a uniform areal unit, such as square miles or hectares, which allows a clearer more precise image of the distributions. Potential applications for density mapping include census data analysis, crime analysis, or plotting the distribution of businesses. There are two approaches to density maps. The first technique involves basing your density map on features summarized by a defined area(s). The second involves creating a density surface. A density surface is typically created as a raster layer with each cell in the layer being assigned a density value. This provides more detail, but at the cost of more effort. Under defined area(s), a density map can be produced using a dot map or by calculating density values for each area of interest. For a dot map, each dot represents a specific number per feature. For example, a single dot can represent 5 businesses or 100 people. The dots are distributed randomly in their area, but the density can be observed by how close or far dots are relative to each other. The rest of the reading instructs how to create a dot density map, creating a density surface, and explaining the calculations GIS does for these procedures. Mitchell explains how to use the appropriate methods and when, including potential real life examples along with corresponding maps. 

1 thought on “McConkey – Week 2”

  1. “I never knew this, but Mitchell states that it is usually best to have no more than seven categories on a map. Apparently the majority of people can distinguish up to seven colors or patterns on a map, but begin to have more difficulty discerning information when there are additional categories. ” I do cover quite a bit of this stuff in Geog 112 (which is based on the old Geog 353). You are just going backwards through classes! Which is fine. I think our plan now is to get as many ENVS/Geog students in this course first, then elaborate in more detail in upper level courses.

    Overall a great review of an avalanche of concepts, ideas and jargon. You have had enough GIS in other places that hopefully this starts to pull stuff together. As noted, the goal is to have this course first for ENVS / Geog majors.

    Great job. Feel free to toss in more general comments or questions.

Leave a Reply