Abby Charlton – Week Two

Apologies in advance for the formatting of this blog post. Questions, definitions, etc are all written into the same sections. 

1. Chapter 1
   1. This chapter gives the very basics on making maps with ArcGIS and includes a step-by-step process on how to plan your map. First, it is mentioned that when getting your data,  you should create a specific question to guide your project. The more vague of a question, the more possible ways you could go about your research; therefore. The more specific your guiding question is, the more efficient your research will be. After that you need to understand your data, so you should identify features, attributes, and categories, and then, if needed, calculate different data based on what you already have. Then, based on your first question and guiding questions of what your map will be used for and who will see it, you will choose a method of map that works the best. Finally, you should process the actual data in GIS and analyze the results. 
   2. There are also many definitions that are important as well. Here are what I deemed the most confusing/most important. 
   3. Discrete vs. Continuous data: Discrete are data points of location that do not change. This point is an x,y coordinate, and it either exists or it doesn’t. However, continuous phenomena are measured anywhere and everywhere in an area, and they have no gaps. If there are gaps in the data, they use interpolation–the act of assigning values to these blank spaces in order to keep the map continuous. 
   4. Raster vs vector data: Raster data uses cells to represent locations, while Vector data uses points and lines to represent locations. 
      1. Question though: When is it the best time to use raster data vs vector data? What situations require each one? 
   5. Counts vs amounts: counts are the actual number of features on a map, while amounts are any measure of quantity associated with said features. One example would be how many trees there are in one section of a mapped national park. 
2. Chapter 2
   1. Chapter two is similar to chapter one in which it goes over the basics, except in this case, it goes deeper into each section about making the map. For generating questions, it describes how you should generate your questions based on what information you’re going to need from the final analysis and how you will be using the map. One example of the “how” would be determining if categories would be a good idea for the final project, or if they would just convolute the point. When preparing the data, you should start with assigning coordinates to places (either latitude/longitude or street address)  and giving them categories based on their features.  If available, it’s good to have a category attribute with a value for these.
   2. The next section is all about actually making the maps on ArcGIS. The first step here is to determine what features you want to display and what symbols you want to represent them. When mapping by category, you may have these different categories in a different map layer or subset, but the subset should mostly be used when mapping individual locations. You should make sure to check what you are making a subset out of because it may lead to confusion and incomplete data. An example of this confusion would be mapping certain roads, which then makes it look like infrastructure doesn’t connect. With categories, using multiple can reveal patterns about the data that may have been hidden before–just make sure to have individual maps for each category so that you have the ability to see simple data too. Yet, even with individual categories, you should avoid having more than seven, otherwise it gets confusing. If you end up needing more than seven, grouping the categories may be an option, but note that you may lose good information by doing this. 
   3. Put good thought into the shapes/color that you use to symbolize your data, as they may have underlying meanings, or they may be hard to distinguish from others. 
3. Chapter 3
   1. This chapter focuses primarily on mapping “the most” and “the least” information, ranking information with quantities. Mapping by these qualities introduces an additional level of information other than just the straight locations of each phenomena. In some maps, this information may be more valuable than other mapping goals. For example, if a city wanted to put in a daycare center and wanted to be the most centralized location for all workers, it would be best to map the places of business and by how many people work at each location. Additionally, you can map quantities with most data, meaning that discrete, continuous, and data summarized by area can be mapped with their associated quantities. However, they are mapped mostly in different ways. Discrete data is typically represented by graduated symbols or shaded areas; continuous phenomena  are represented with graduated colors or contours or maybe a 3-d view; finally, data summarized by area is usually displayed by shading each area based on its value. These representations may change with the objective of the map. Again, what is the purpose of the map–when creating a map for presentation, you’ll want to choose representations that make patterns easier to see, which might force you to sacrifice other parts of your data that you would keep if only using the map for pattern recognition. 
   2. Quantities can be counts, amounts, ratios, or ranks, and knowing which one your data is will help you decide which map you should be using. Counts and amounts can be used with both discrete and continuous data, but ratios are best for summarizing by data, as counts and amounts could potentially skew the data towards another conclusion. Typical ratio data are averages, proportions, and densities. Finally, ranks put features in order from highest to lowest.
   3. After determining your quantities, you’ll likely transfer into building classes. Classes are ranges of data that encapsulate several data entries, and they are typically used with counts, amounts, and ratios. You should make classes with group features that have similar values, and how you define each class (how you choose the range) depends on your data set.
      1. How do we choose which class type to do? I understand that if there is a wide range we should just make our own scheme, but when do I choose to use natural breaks vs quantile or equal intervals?
      2.  
4. Chapter four
   1. This chapter is all about map density, which is another useful subset of mapping. Although unlike other kinds, density maps are more useful with pattern recognition than mapping locations, as its way easier to see concentrations of data. You can map both the density of features or the density of feature values. Then, you can map these densities with graphs, dot maps, or simply the values of each area, and these are typically done with raster data. Dot maps are best for individual locations. 
   2. Since density maps are made with raster data, you will need to determine cell size. The bigger the cell size, the rougher the map will look, and the smaller the size the smoother the map will look. Yet, the smaller the cell size, the more storage you’ll need to store it, so there are advantages and disadvantages for each kind of cell. Next, when planning the search radius (the area which surrounds a point), you’ll need to decide if it’s a large or small radius that you need. Larger radii have more generalized patterns and consider more features, while smaller radii will show more variation and intricate patterns. 
   3. Getting the values in the cells can come from multiple ways. 
      1. Simple Calculation – only counting those features found within the search radius of each cell
      2. Weighted calculation – uses mathematical function to give more importance to features closer to the center of the cell. This type of calculation often results in smoother maps with patterns that are generally much easier to distinguish. 
   4. When displaying density, you should use either graduated colors or contours. With graduated colors, you should classify the data and then assign colors to each class. This should let you sense a pattern. With contours, GIS will automatically create the map from the surface without many other steps.