By: Brian Truong.
GeoVis Project @RyersonGEO SA8905, Fall 2018
Context
As a student and photographer, I have frequently walked around the streets of Toronto. I would often see homeless individuals in certain neighborhoods in Toronto. While at the time I was aware of some shelters across of Toronto, I never fully understood the Toronto shelter system as I thought organizations in Toronto were one and the same in terms of providing shelters to those who are at-risk. I also noticed that the City of Toronto updates their shelter occupancy data on a more less daily basis, which led me to choose this topic for my GeoViz project. My lack of knowledge of the shelter system and the readily available data, motivated me to choose to make a Time Series map along with incorporating ESRI’s Story Maps into the project. This was to ensure that whoever wanted to see my project could be told the story of Toronto Shelters as well.
Process
Toronto open data provides shelter occupancy data in multiple formations, however, a JSON data format was chosen due to previous experience with working with JSON data in Alteryx. JSON data was provided through a link from Toronto Open Data. Using Alteryx a scrip was created to download the live(ish) data, parse it, put it in a proper format, then filter the data, and along with creating appropriate columns to work with the data.
Above is an example of the JSON data that was used, the data itself is semi-structured as the data is organized in a specific format. The data consisted of multiple of entries for Shelter location, those were filtered out so that only organizational program was present for each shelter location. this usually went down too the program that housed the largest number of people. In order for a proper time series to be created, a date/time column must be present, columns were created through the use of the formula tool where columns such as date/time and occupancy rates were created.
Above is the final Alteryx strip that was used to get the data from a JSON format to a .xlsx format. However, there was one problem with the data. The data itself wasn’t geocoded, so I had to manually geocode each shelter location by running the address of each shelter through Google Maps and copy and pasting the (x and y) values of the shelter locations. These coordinates were then put into the same file as the output of the Alteryx script, except it was in a different sheet. Using VLOOKUP, shelters were assigned their coordinates through matching shelter names.
Time Series Map
The time series portion of this geoviz project was created using Arcmap Pro, the excel file was brought into ArcGIS Pro and points were created using x and y coordinates. A shapefile was created, in order to create a time series map, the time field had to be enabled. Below shows the steps needed to be taken in order to enable time as a field on a shapefile.
In order to actually enable time, a time column must already exist in the format of dd/mm/yyy XX:XX. From that point, the change in shelter occupancy could be viewed through a time-slider going at any interval that the user required. For this project, it went by a daily basis on a 3-minute loop. In order to capture it as a video and export it, the animation function was required. Within the animation tab, the tool append was used.
What the append feature does is that it follows the time series map from the first frame, which is on the first day of the time series map (Jan 1, 2018 00:00) to the last day of the time series map (Nov 11, 2018 00:00). The animation would then be created as per specifications of the settings. The video itself is exported through a 480p video at 15 frames a second. It was then uploaded on YouTube and embeded on the storymaps.
ESRI Story Maps
The decision to use ESRI’s story maps was in part due to what motivated me to work on this. I wanted to tell the story of shelters and who they serve and is affected by them. Especially after two major events in the past year that has led to shelters in Toronto showing up on the news. Both the cold snap in early 2018 and the large influx of migrants has had a huge effect on Toronto’s Shelters.