Chapter 1:Introduction to GIS Analysis
GIS analysis is similar to something that I am sure many of us have been doing, but now we are putting it into a geographic context. Especially when looking for both patterns and relationships, sometimes these maps can be self explanatory, however the more interesting problems are finding ways to explain both the correlation and causation of what occurs and what data is shown with the maps. The GIS analysis process that the first chapter outlines reminds me of the scientific method if we were not to make a hypothesis, finding a question to ask, finding a way to frame the question while also being able to measure said data, understanding the data that is presented and gained, choosing a method of measuring the data and method to present the data for further analysis, processing the data, and looking at the results of the data. One thing that I had not considered before was the many ways of looking at the results that are gained, such as displaying them on a map, looking at a data, or within a chart. Measuring geographic features can be extremely useful, but I was unaware at first of how many different types of features there are, there are things easier to measure such as mountains, city limits, and where rivers are located, those things can be pinpointed and are considered discrete features.
Discrete features seem to answe the question of ‘is this feature here or not?’. Continuous phenomena is something that is a value that can change during the average day, such as precipitation or temperature. Continuous data can be represented in areas enclosed by boundaries assuming that everything within the data is the same type.
Features summarized by data can represent a simple count or density of various features within a certan enclosed area.Within GIS, there are 2 different ways to represent data, through vectors and rasters. With vectors, the system uses a table with shapes & points to represent data at certain sets of coordinates and bounds, this is most similar to a graph or shapes. With a raster model, it is similar to a large excel sheet where you paint the different types of cells to represent different numbers & data points. The book states that both types of representations of the data are good, discrete features are usually represented with vectors, continuous are represented as either vector or raster, and continuous numeric values are represented with raster models.
With attributes, every feature has some attributes that are able to be used to identify whatever we are trying to represent. These attributes are the following: categories, ranks, counts, amounts, and ratios. A category is an overarching topic that contains a group of similar stuff. One example is that if I were to be mapping a city, i would put office buildings, restaurants, and shops in under the category of ‘Businesses’. With ranks, I would put something in order from high to low, or Excellent -> Poor. With ratios, I would put different colors for the amount of features in a map, if I were to try to map something similar to population density, a lighter color would reflect a lower amount of people living there, and a darker color would represent a higher population. If I were to use a count, I woult count something such as the amount of customers going into a business and make a larger circle around the business if there were more customers.
Chapter 2: Mapping where Things Are:
Looking for patterns can be key for helping me understand the are that I am mapping. Something such as a crime map can help me understand what the biggest issues of the area can be in terms of crime, maybe see what parts of the city meet more crimes vs lesser crimes, which could explain where police usually are, or if crimes get reported in general. The main decision is deciding what needs mapped, what to display and how to display them. There are different purposes for different types of maps, such as the example with the police department, a business needing to know its demographics, and other considerations need to be made. How the map gets used is something that becomes a key issue when thinking about how to create the map, while a city zoning map would be useful for bringing up to a committee meeting, it may not be as useful for other purposes such as the case of where crime occurs. The level of detail is needed to be put into consideration for what type of audience will be seeing my map, will it be for a general audience, or some seasoned professionals? Some key considerations for my maps to make sure that I know I have geographic coordinates, and hopefully have both a category & value for every main part of my map. When I assign a map feature with a type, I must have a code within the feature. These types should be stored prior to adding them to the map so I do not have to go back and add them later. Some categories can be hierarchical, and will have a ubtype, such as general zoning vs mixed usage. When I make my nmap, I need to know what features I would like to display and figure out what the symbols I will be wanting to use. With a single type, I will represent all features with the same symbol, like If I wanted to represent sales by delivery, I would represent each sale with a dot. I can also separate data and map only certain types of data, such as amazon deliveries vs uspc deliveries. I am also able to map by category, this is typically done with different colors, but I will only want to do up to 7 different categories at one time as most people wouldn’t be able to distinguish more than those 7. If I were to display more then 7, it would be more wise to group those categories together on a secondary map that is easier to digest such as the one on Page 40. If my map does its job and presents the information pretty clearly, there should hopefully be a pattern that is able to be more easily seen and understood.
Chapter 3: Mapping the Most and Least
The reason people map the most and least of something is to find a pattern or find qualities of features such as those within real estate. Some things I am able to map with Most -> least is any feature associated with discrete, continuous,m or data summarized by an area. The discrete features are typically represented by graduated symbols while areas are often shaded to represent quantities. Continuous phenomena are typically displayed using graduated colors while even a 3d perspective view can be used to represent the continuous surfaces. Data summarized by data is typically displayed with shading the area via the value while using a chart to use it as a representation of the data. While forming the map, I need to keep the indented audience & purpose both in mind. If I were to just use the map for presentation, I will want to explain what the data points mean, however if I were to be trying to explore the data, the map would provide a good baseline of a direction to go in when it comes to patterns and ideas. There are many numerical ideas that I will want to keep in mind, such as amounts, counts, ratios, or rankings of various areas. I will want to find the best way(s) to represent these through the data. This is typically done in the method of using gradients or small -> large shapes. Once I figure out the quantities and type of quantities, I will want to figure out how to classify the data. If they were to be individually presented, I will not need to group them. If I were to group points of data together, I will want to use classes by assigning them the same symbol. I would want to use standard classification schemes if I were to want to group similar values in order to look for patterns. I’ll be able to figure out the best scheme for creating the class break by looking at the distribution of the values of data. One of the best ways I am able to do this is by creating different graphs or charts using the data. But that being said, it is incredibly important to keep things concise and understandable.