Sunday, February 24, 2019
GIS 4006 - Module 6 - Data Classification
For this weeks Module 6, we were tasked with displaying data for Dade County, Florida in four different classification methods for symbology: Natural breaks, equal interval, quanitle, and standard deviation. Two maps were created that display the percentage of individuals above 65 for the Dade County area and the number of individuals above 65 normalized by the sq. miles field. I chose a color gradient that reflects increasing values through a "hot spot visualization" with lighter colors reflecting lower values and darker red colors reflecting higher values. Each of the four classification methods separates classes based on a unique calculation method but I feel quantile resulted in the best map for visualizing hot spot areas for senior residency. I would also present this information as a percentage as this allows for easier user comprehension and visualization.
Subscribe to:
Post Comments (Atom)
Spring 2023 semester wrap up
The spring 2023 semester at UWF has been an eventful one in which I finalized the requirements for my bachelors of science in natural scien...
-
The spring 2023 semester at UWF has been an eventful one in which I finalized the requirements for my bachelors of science in natural scien...
-
By constructing three cartographic maps a better understanding of Arcmaps and Arccatalog is developed. The lab presented this week had us ...
-
"Vector - A coordinate-based data model that represents geographic features as points, lines, and polygons." -ESRI For the last ...
No comments:
Post a Comment