Cartographic Skills: Dot Mapping

This week we learned how to create dot distribution maps. Dot maps provide a geographic location of one or more phenomenon. Each dot is geographically placed at its corresponding location and the number of dots changes in proportion to the value for the given geographic location. This provides a view of the overall spatial distribution of data and the density of where the dots exist.

 

We created a dot map for population density in southern Florida. Using Census Bureau data, we portrayed total population statistics for corresponding dots in reference to geographic regions of south Florida. Using Arcmap helped us place dots by using the Symbology tool, we manually created the size of each dot and the density for how many dots we placed on the map. Having dots too large can portray the data inaccurately. Areas might look too dense due to dots being created too large and spanning outside of the dots spatial region. So having smaller dots helps show the actual raw data counts.

When we adjusted the density for the dots, we wanted the dots to just touch each other. If we created more density, we end up with hard to read data as dots start to overlap and cannot be quantified. In Arc we are able to use a slider to manually adjust the density and the dot size, within a preview window, in order to get the right mix of size vs density. 

The great thing about dot maps is we are able to show other layers of data without any loss of visualization for the dots. So we placed a water feature layer on the map. This allowed a view of where water feature are located, which shows why there are no points located within the water features. Without this layer, map readers may question why there are large areas without dots. In Arc we used a masking tool to hide dots that fall within water features. This is any easy option to use but comes at a computer usage cost as masking creates a heavy burden on the speed of the CPU.

With dot mapping, it's useful to show the dots where they are best needed such as urban areas were most of the data exists. Using the masking tool we are able to show dot features that fall within the urban area layer. 

I found this week’s assignment challenging in using the masking tool as it majorly slowed my PC down. It was best to turn off the masking option. So I needed to add and adjust all the map features, before turning on the masking tool. The last option you need to do is turn on the masking tool before you export, making sure everything looks good beforehand was the challenge. I had to go back a few times and redo features and the masking tool slowed me down quite a bit.

Cartographic Skills: Isarithmic Mapping

The week we created an Isarithmic map depicting average annual rainfall for the State of Washington. We used data created from PRISM Climate data. PRISM is a climate model using weather station data over a 30 year period, applied with digital elevation model (DEM). Perception usually increases with elevation and PRISM analyses the regression of the climate data for each grid within a DEM. In other words it compares surrounding weather station data records and how they relate to each other, along with elevation data. The outcome produces a climate weather model referencing the comparison between weather and topographic features. As you can see in my map below, the higher rain fall amounts are found at higher elevation or mountainous regions. The contours on the map are not topographic contours but range differences in the amount of precipitation. The range difference is however directly related to elevation.

We used a file geodatabase downloaded from the USDA Geospatial Gateway. The PRISM data was originally created from the PRISM Climate Group but the USDA built the geodatabase using PRISM data along with other layers related to climate data. We used the average annual rain fall layer which spans yearly averages over a 30-year period.
We used Arcmap to analyze two different methods of displaying Isarithmic maps for displaying smooth continuous data. The two methods are; Continuous Tone and Hypsometric Tinting. We used the PRISM data to visually see the difference in these two methods.
Continuous tone symbology uses color tone proportional to the value of the surface at each point. I used the Precipitation color scheme to show values of warm lighter colors of low rainfall amounts and cool darker colors for higher rainfall amounts.

Hypsometric tinting visually enhances a 3-D surface by using light and dark tints for low and high values. I took the tint a step further and used a range of color schemes, rather than a single color scheme. Using a range of colors allows you to visually understand the order of colors in relation to the values, as each range value corresponds to a unique color. We manually boke our data into 10 classes by manually defining the classes. This allows the data to be smooth and continuous, necessary for hypsometric tinting.

For the final map we used hypsometric tints along with hillside effect and contours, as seen in my map above.  Using contours allows for better visualization for the breaks of ranges in the data. Hillside effect gives a 3D shadow effect with the sun’s relative position taken into account.
Having traveled through the state of Washington, you notice many areas of agriculture in the central and eastern regions. What is interesting about the actual precipitation data, is that the fertile farming regions of the state appear to have the lowest rainfall amounts. With room to grow many types of agriculture, good soil and sun, all you need is water. So it’s interesting to see low rainfall amounts for agricultural regions and the human capacity to bring water to areas that normally would not sustain agriculture.

Cartographic Skills: Data Classification

This week we learned about Data Classification for properly displaying data within a map. There are many methods of displaying data. We looked at four methods of classifications and built a comparison map of each classification. We used 2010 Census Bureau data and preformed classification methods on the percentage and total population of ages 65 and older, within census tracts for Miami Dade County Florida. Below is a map using four methods of classification; Natural Break, Quantile, Equal Interval and Standard Deviation.

The map above displays four comparison methods of classification using the percent of population ages 65 and up for census tracts in Miami Dade County, Florida. The classification methods used are; Natural Breaks, Quantile, Equal Interval and Standard Deviation. Each method of classification displays data differently.

Natural Break classification groups data “naturally” into classes by using an algorithm to find spikes or clusters where the data is found within the range of data distribution. This type of classification works good for the percentage of age 65 and up map as it finds breaks in the data where there might be larger groupings and separates the percentages into appropriate views. However, map readers may not understand the legend very well in determining how the data is broken up into each class.

The Quantile method of classification divides the total distribution of data into equal numbers of observations. Each class contains a range value obtained by determining a midpoint between the highest and lowest value for each class. I believe the Quantile method is the best method of classification for displaying the percentage of population ages 65 and up. The data range shows the highest percent of values placed in the last class. Showing where the highest values exist, map readers can compare against the lower values and easily understand the breakdown of data distribution. Also the legend is easily understood in how the ranges are broken up. 

Equal Interval classification shows the data within equal ranges. Basically the data is equally split up into breaks or classes. Ranges for the data are determined by dividing the total data range by the number of classes used. This type of classification does not display data very well for the percent of population ages 65 and up map. Since the percent values are equally spread across the number line and a class of ranges are equally divided, the data on the map looks very even and is hard to compare neighboring values because the data is equally dispersed.

The Standard Deviation considers how data is distributed along the number line. Classes are formed from a mean deviation and broken up by adding or subtracting values. This classification does not show the data very well for the percent of population map. It’s very confusing to understand and hard to portray the data appropriately as the map reader must have an understanding of statistical analysis.