Canada Line Development

The development of the area around Canada Line is explored by looking at the number of business licences issued at different areas along the track over time.

Software Used

• R 3.2.2 (with the checkpoint library)
• QGIS 2.12.0-Lyon (with the NNJoin and Table Manager plugins installed)
• Python 2.7 (with the virtualenv package)

Other Requirements

• Google Maps API secret key in the environmental variable GOOGLE_MAPS_SECRET

Setup

R Setup

Install the checkpoint package and run src/requires.R.

Python Setup

# start in the src/ directory
virtualenv .proj
. .proj/bin/activate
pip install -r requirements.txt

Reproduce the Graphs

1. Open as an RStudio project and run requires.R (this will take a while)
2. Run s1_business_data_download.R
3. Run s2_transit_selection.R
4. Run

# start in the data directory
touch geocoded_02_08_14.txt
./geocode.py addresses.txt geocoded_02_08_14.txt

(with the virtualenv activated). The output file geocoded_02_08_14.txt is imported in the next step. 5. Run s3_import_geocoded.R 6. Perform instructions from the Prepare the Spatial data section 7. Run s4_combine_geocoded_with_rest_of_data.R

Run s5_graphics.Rmd for the report graphics, s5_table.R for the table statistics and s5_graphics.R for the presentation graphics

Prepare the Spatial data

Additional datasets required:

• Local Area Boundary
• Rapid Transit Lines
• Rapid Transit Stations

Prepare Rapid Transit Lines

1. Import the Rapid Transit Line RTL file into QGIS.
2. Select Canada Line from the RTL layer
3. Reproject the dataset into UTM Z10N (EPSG 32610) using the selected feature and save it disk (I'll call it CanLine)

Prepare Rapid Transit Stations

1. Import the Rapid Transit Stations RTS file into QGIS.
2. Select all Canada Line stations from the RTS layer
3. Reproject the dataset into UTM Z10N (EPSG 32610) using the selected feature and save it disk (I'll call it CanStations)

Prepare Local Area Boundary

1. Import the Local Area Boundary RTS file into QGIS.
2. Reproject the dataset into UTM Z10N (EPSG 32610) using the selected feature and save it disk (I'll call it LocalArea)

Prepare the Geocoded Data

1. Import the data/stage3.csv file into QGIS (should have been produced after running s3_import_geocoded.R)
2. Reproject the dataset into UTM Z10N (EPSG 32610) using the selected feature and save it disk (I'll call it Stage3)

Buffer CanLine

1. Create a 1km buffer around Canada Line using Geoprocessing Tools
2. Save it and call it CanLineBuffer

Select the Geocoded Data

Select business licences from Stage3 that interesect CanLineBuffer and call it Stage3InBuffer

Calculate the Distance to Station and Tracks (Requires the NNJoin and Table Manager plugins)

1. Use NNJoin (input layer: Stage3InBuffer and join layer CanLine). Use Table Manager to delete all columns beginning with "join_", rename the "distance" column to "d_to_track" and call the output Part2Stage3InBuffer
2. Use NNJoin (input layer: Part2Stage3InBuffer and join layer CanStaions). Use Table Manager to delete all columns beginning with "join_", rename the "distance" column to "d_to_stat" and call the output Part3Stage3InBuffer
3. Export Part3Stage3InBuffer as a csv file (call it dist.csv and place it in the data directory for use in step 7 of the reproduce the graphs section)

Create Maps

1. Import data/complete.txt into QGIS and reproject it UTM Z10N (call it BusinessLicences)
2. Add styles to BusinessLicences, CanLineBuffer, CanStations and CanLine to match the final maps shown in the graphics directory.
3. Use QGIS print composer to add the scale bars and projection information export the results filtering the BusinessLicenses layer to the year 2002 and 2014 respectively.