Sunday, March 11, 2018

Module 8: Isarithmic Mapping

This week's module introduced the Isarithmic Map, which is a type of thematic map that displays continuous and smooth phenomena such as temperature or rainfall amounts. The Isarithmic map is the second most widely used thematic map behind the choropleth map.  Every time we watch the local weather forecast, we likely see an Isarithmic map showing temperature values across the viewing area.

The lab assignment had us prepare a map displaying the average annual precipitation for the US state of Washington.  The data file is a raster type and displays average precipitation over a 30 year period.  It was obtained from the USDA Geospatial Gateway and was prepared using the PRISM interpolation model.  PRISM stands for Parameter-elevation Relationships on Independent Slopes Model and was developed in 1991 by Chris Daly, a PhD student at Oregon State University.  The PRISM model arose out of the need for interpolation models to account for the elevation-precipitation relationship. The algorithm PRISM not only takes into account the elevation of each raster cell, but also the direction the slope of a hill or mountain faces.  This accounts for the "rain shadow" effect where less amounts of precipitation are typically found on downsloping areas of the topography. The PRISM model was developed to mimic the thinking of an expert climatologist, and thus is called an "expert model" among the scientific community.

In the first step of the lab, we displayed the rainfall data as a continuous tone by changing the color ramp to "Precipitation" in the layer properties' Symbology tab.  The allowed for precipitation values to be displayed evenly over the plane.  We also used the "hillshade effect" to give the map a sense of relief.  Because the PRISM model incorporates a Digital Elevation Model, we can use this effect. In the next part of the lab, we changed the symbology to Hypsometric Tinting.  This means that we differentiated the contour values with stepped color shading by assigning a certain color to a a range of rainfall values.  We first had to prepare the raster cells by using the Spatial Analyst tool "Int" which converted raster cell values from floating values to integers. With the Raster cell values being integers, it was easier to class the cells into specific ranges. I used the Quantile Classification Method in order to class my data and divided the data values into 10 classes.  Finally we used the "Contour List" spatial analyst tool to add contour lines to the Hypsometric Tint map.  I set the contour values to mimic the color ranges, therefore a separate legend was not needed.  My finished map product can be seen below.

Sunday, March 4, 2018

Module 7: Choropleth and Proportional Symbol Mapping

This weeks Module introduced Choropleth maps and the use of proportional symbols.  Choropleth maps are one of the most commonly used maps because they can easily portray data by shading areas known as enumeration units based on the proportion of statistical data.  Proportional and Graduated Symbols are also another way to visually depict either standardized data or raw count data.  This weeks lab assignment had us compare the population density of European countries with their wine consumption.  I used a Chloropleth map to represent the population density.  This data was standardized to people per kilometer.  Because the population data is considered unipolar, I chose a sequential color scheme that starts at a light tan color to represent countries of lower populations and transitions to a dark brown color to represent countries with higher populations.  I used graduated symbols to depict the amount of wine consumed by countries in Europe.  This data was also standardized to liters per person in order to account for the vast differences in enumeration unit sizes.  My map, along with a description of software used for map production can be found below.

The majority of the map production and the entirety of data analysis was performed using ArcMap 10.4.1.  I first created the Choropleth map by opening the data layer's properties and selecting "Graduated colors" in the Symbology tab.  Also in the Symbology tab, I was able to choose which attribute from the attribute table that I would like to depict on my map.  In this case I chose "POP_DENSIT" for the population density, which is a standardized version of raw population.  Next I had to choose which classification scheme to use in order to best represent the population data.  I chose "Natural Break" because it seemed to do a good job at evenly distributing the shaded enumeration units so that the audience can immediately tell which areas of Europe are most populated.  Also in the classification window, I was able to exclude four countries with very high population densities.  These four countries' values were skewing the data as evident on the histogram and were considered outliers, thus were removed.

To visually depict the wine consumption, I added a new data layer to the map in ArcMap and again opened the Symbology tab in the data Properties.  This time though I clicked on "Graduated Symbols" under Quantities.  By default, this added circles to the map, but I changed the symbol to a wine bottle silhouette in the Symbol Property Editor.  After adding my essential map elements, I imported the map into Adobe Illustrator (AI) to finish it.  I added all country and water body labels to the map using AI.  I then added a drop shadow to a few elements such as the title for emphasis.

There were a few things I wish I could have presented better on my map.  My wine consumption legend is very big and contains distracting white space. I had trouble making this layout any more appealing in arcmap.  I'm hoping to learn later how I could use AI to move legend items around and make it smaller.   Another issue was the fact that I could not treat my wine bottle symbols as separate entities, therefore I could not move them around.  I know it's possible in AI to ungroup this layer, however all my attempts were unsuccessful. Finally when I imported my map from ArcMap to AI, some of my wine symbols were discolored and not completely black.  This was especially an issue in the inset map.  I reordered my layers in AI to ensure my wine bottle symbol layers were all above my country outline layers, but the problem persisted.  Again, I plan to get better at problem solving in Illustrator in the future.

Sunday, February 25, 2018

Module 6: Data Classification

This week's lecture material introduced us to six common methods of data classification, and the lab focused on the four classification methods below.
  1. Natural Break
  2. Quantile
  3. Equal Interval
  4. Mean-Standard Deviation
The assignment tasked us with creating a Thematic Map displaying the distribution of senior citizens (over age 65) in Miami Dade County, Florida.  We created one map displaying this data as a percentage of residents over the age of 65, and a second map which normalized the amount of senior citizens in the county by square mile.  In both cases, we used the four classification methods listed above to compare.  I believe that the map displaying the data as the amount of residents over the age of 65 normalized by square mile most accurately represents the spatial distribution of these residents. My map can be seen below.  

As you can see above, the four classification methods display the spatial distribution of senior citizens in different ways.  Depending on the type of data, each of these methods have their situation where they are the most appropriate method to use.  In the case of the senior citizen data, I believe using the Quantile classification method displays the senior citizen population the best.  The Quantile method is best for displaying ordinal-level data such numbers of people in a population.  This method does not take into account how data is distributed on a number line, but instead is good at easily portraying which areas have a higher amount of data values compared with other areas.  The map reader can quickly identify that the area where most senior citizens live in Miami Dade County is the northeast corner of the county.  This can not be said of other classification methods such as the Equal Area method.  

Saturday, February 17, 2018

Module 5: Spatial Statistics

Our lab assignment this week involved completing an online ESRI training course titled "Exploring Spatial Patterns in Your Data Using ArcGIS."  This training was completed through ESRI's My Virtual Campus.  The training consisted of five exercises which helped us examine our data with the end goal being the ability to select the appropriate interpolation analysis technique.  The first exercise taught us how to display the mean, median, and directional distribution of our Western European weather station data.  Displaying these three values lets us visually access the data and how it is geographically spread out.  Examining out close the mean and median values are to each other helped us determine if the data was normally distributed.  If the mean and median values are plotted close to each other on the map, that is a characteristic of normally distributed data.  Part two of the training taught us how to display our data as a Histogram and as a QQ Plot.  The Histogram helps us determine the frequency that a certain data value occurs and plots these values as bars representing a certain interval of data.  The QQ Plot helps us examine the distribution of the data by plotting the values as points overlay-ed on a line representing a normal distribution of data.   Part three of the training looked at variation in our data.  According to the First Law of Geography, everything is related, but objects closer together are more related than objects farther apart.  ArcMap helps to visualize spatial variation of data points using a Voroni Map.  Part four helps determine spatial patterns in our data by using a Semivariogram Cloud.  This diagram plots the distance between data points vs. the semivariance.  Finally part five helped us identify trends in our data.  A trend is the pattern of spatial variability which can tell us important information regarding the patterns behind the data. 

Following the instructions in part one of the virtual training, using ArcMap I created a map displaying all of the weather stations in certain countries in Western Europe.  On this map I plotted a point for the mean and median by using the Mean tool and the Median tool from ArcToolbox.  The mean is the average of all of the data points, while the mean is the middle point of all of the data points on the map.  These two values were plotted very close together, and this close proximity is a characteristic of normally distributed data.  I then determined the directional distribution by using the Directional Distribution Tool in ArcMap.  The output resulted in a large ellipse stretched along its X axis.  This is means that there is a very wide range of spatially distributed data which spread out more in an east to west direction than north to south. 

Overall is was really cool to see patterns in our data using tools available to us in ArcMap.  I work in the environmental field and deal with a lot of monitoring well data.  Throughout the online training I kept thinking about how I could use some of these tools to help me answer questions about my data at work.  Because of this, I feel that this weeks lab was one of the most useful to me personally.  My map created in part one of the training can be seen below.  I really wanted to orient my map in a landscape direction, but I couldn't get the orientation to change using the print page setup.  I commonly change the orientation of my page at work and during other labs, so I don't know why it wouldn't work this time. 

Sunday, February 11, 2018

Module 4: Cartographic Design

In this week's lab we were instructed to create a map of the Public schools in Ward 7, Washington D.C.  This map had to be created so that it complied with Gestalt's principles of Visual Hierarchy, Contrast, Figure Ground, and Balance.  Gestalt's principles were created in order to describe the way the human eye perceives visual elements and integrates them into a whole image.

I created my map entirely within the ArcMap software.  My first step was to import all my data layers.  These layers included a general outline of the DC area, a layer showing the Ward 7 area, as well as roads, schools, highways, interstates, surface water, and parks.  My goal was to draw the audiences attention immediately to Ward 7, so I zoomed into that area as much as possible within the data frame.  I then decided to contrast the colors and make Ward 7 lighter and the surrounding DC area a darker color.  In order to maintain a Figure-Ground relationship, I screened the Ward 7 roads layer and made them lighter so they are de-emphasized in the background.  I made the roads lighter by choosing the same color as the Ward 7 polygon layer, "lime dust" and changing the Hue, Saturation and Value using the HSV Model color selector in ArcMap.  Next I used the Geoprocessing Clip tool to clip the Schools layer so that only schools located within Ward 7 were displayed.  This left 18 schools on the Map, all located in the Ward 7 area.  I changed the school point symbols to a more appropriate symbol shaped like a school house and changed the color to dark red in order to establish a contrast and enhance the figure ground relationship.  The schools are the main theme in the map, therefore I wanted to make sure they stood out from the background.  I then symbolized the schools by categories of Elementary, Middle, and High schools.  I increased the size of the symbols with High School being the largest and Elementary being the smallest.  This was accomplished by selecting Unique Values in the Symbology tab of Properties.

I established a Visual Hierarchy with the roads by adjusting their line weight in increasing increments starting at 0.7 for local roads, up to 3.4 for interstates.  I used the Spline text tool to label the Anacostia River, then changed the text color to blue, and style to Italic to represent the flow of water. An inset map was used to show the audience where exactly Ward 7 was in relation to the whole Washington DC area.  I tried to create a sense of balance within the map by placing the 3 rectangle features (legend, school list, inset map) in 3 corners, not including the SW corner.  The SW corner of the map was already a busy area of the map, thus these features fit better visually in the upper and right side of the map.  In the past I've made my north arrow too large, so this week I made sure to make it small so it doesn't call too much attention to itself.  My map can be seen below.

Thursday, February 1, 2018

Module 3: Typography

This lab was meant to help us learn the typographic guidelines of labels and features on a map, and to put those to use creating a map of the Marathon, FL area.  The typographic guidelines are rules that have been developed over the years by cartographers and have since become the standard.  The guidelines focus on things such as point and area feature label placement, leader line placement, font, text size, text style, ect.

I created a base map using ArcMap, then imported it into illustrator and completed the map there.  One thing I learned was to create the scale bar and north arrow in ArcMap before importing to Illustrator because it's much easier to add those elements using GIS software.  In Illustrator I created a legend and changed it's border opacity to 62% so it was visible, but didn't draw too much attention away from Marathon map.  I then added a gray drop shadow to the counties layer in order to help it stand out from the background.  I changed the drop shadow color from black to gray to ensure  the adjacent water feature labels were legible.  Finally I added two clouds in the upper left side of the map where there was an abundance of blank space.  I felt the clouds were subtle enough to not become the focus of the map, but instead add a balance to the map.  I added a gray drop shadow to the clouds to make them more realistic.  My finished map can be seen below.

Sunday, January 28, 2018

Module 2: Intro to Adobe Illustrator

This week's lab was meant to introduce us to the graphic design software, Adobe Illustrator (AI).  Before starting this weeks assignments, I had never used Illustrator so it was nice to get my feet wet and learn the basics of what Illustrator can do.  The first step in the lab had us create a basic map of the State of Florida showing major cities, the capital, counties, and surface water.  Next I opened up Illustrator and watched some introductory videos in order to get an overview of what AI is and what it can be used for.  Then following the lab's instructions, I browsed around the artboard space and identified the location of important tools that I would commonly use.  In part three of the lab, I actually used some of these tools and played around with making shapes and changing the fill color.  I learned how to use some of the other major tools as well such as pen, pencil, and text.  Finally I studied the layers panel located on the right side of the window and eventually familiarized myself with the basic concept of using layers.

Finally during part four of the lab exercise, I uploaded my Florida map which I began creating using ArcMap.  Using the skills learned in this lab, I added some features.  First I added the Title, and a state nickname subtitle using the text tool.  I then pasted in pictures of the state animal and state bird from the web and displayed a title and source for each by using the text tool.  In order to move the scale bar and legend around, I grouped all the pieces of these features into their own respective layers.  Once they were in their own layer, I was able to move them and make room for other map elements.  Using a script, I changed the major cities symbol to a different style of circle.  I changed the state capital symbol to a star in order for it to stand out. Finally I noticed I still had too much open space in the middle of the map so I added a sun and palm tree symbol from the "logo elements" section of the symbol library.  I changed to the background to a light green by creating a rectangle and filling in the color, locking the layer, and with that finished my map was completed.  My finished product can be seen below.