- Landuse - Constraint
- Landuse - Factor
- Slope
- Population Density
- Distance from City Centres
- Distance from Transit Hubs
Creating WATER_con
What is WATER_con?
WATER_con is a Boolean image of the study site and indicates whether locations are considered to be land or water. In this case, land areas have a value of 1 and areas covered in water have a value of 0. Using WATER_con as a constraint in Decision Wizard removes the water covered areas from the Station_Suitability output that will be made.
IDRISI Selva Operations
Here is a Macro Modeler
design that summarizes the operations performed in IDRISI Selva to create WATER_con:
Macro Modeler Legend
Complete Step by Step Process
1) See LU_fuzz for the steps taken to create the LU_RC raster file.
2) Assigned Boolean values to LU_RC, using the ASSIGN module, in which water bodies were given a value of 0 and all other land locations were given a value of 1 (resulting in WATER_con).
Creating LU_fuzz
What is LU_fuzz?
LU_fuzz is a raster image that represents the suitability of a location based on its designated land use, as determined from the land use file acquired from SFU’s SIS Data. For this variable, reasonable judgement was used to assign suitability values to each type of land use; below is a table of the various land uses and their respective values.
Land Use |
Land Use Code (Reclassed Code) |
Suitability % (Value) |
| Industrial - Extractive | M300 (1) |
0 |
| Recreation and Protected Natural Areas | R100 (2) |
0 |
| Lakes and Water Bodes | R200 (3) |
0 |
| Residential - Single Family | S110 (4) |
50 (127.5) |
| Residential - Rural | S120 (5) |
60 (153) |
| Residential - Townhouse and Low-rise Apartments | S130 (6) |
40 (102) |
| Residential - High-rise Apartments | S135 (7) |
20 (51) |
| Commercial | S200 (8) |
70 (178.5) |
| Commercial - Residential/Mixed | S210 (9) |
60 (153) |
| Industrial | S300 (10) |
20 (51) |
| Institutional | S400 (11) |
20 (51) |
| Transportation, Communcation and Utilities | S500 (12) |
20 (51) |
| Open and Undeveloped | U100 (13) |
100 (255) |
Reasoning
- Areas with suitability values of 0 were designated as such because they were deemed to be unsuitable locations for a SkyTrain station.
- Residential areas were assigned suitability percents from 20% to 60%. A general trend of higher population areas having lower suitability values and lower population having higher suitability values was applied. This is because it will cost the developers/government more to complete the Evergreen Line if they are forced to buy citizens’ property and assist them in their relocation.
- Commercial areas were assigned relatively high suitability percents because businesses directly benefit from an increase in traffic around their location. The suitability is not at a maximum because some businesses will have to be bought out in order to make room for a station, which adds resistance to the ease of conversion.
- The other land uses (industrial, institutional, and transportation, etc.) with suitability percents of 20% were assigned as such because they are establishments that are very costly to relocate or buy out.
- Open and undeveloped land is as suitable of a location you can get!
IDRISI Selva Operations
Here is a Macro Modeler design that summarizes the operations performed in IDRISI Selva to create LU_fuzz:
Macro Modeler Legend
Complete Step by Step Process
1) Obtained the original file, gvrd_2001_landuse, from SFU’s SIS Data.
2) Clipped the file to a custom polygon (Station Suitability Area) of the study area (resulting in landuse_clip) in ArcMap.
3) Reclassed the land use codes of each land use type and manually entered the new code for each polygon into attribute table.
4) Converted the file to Raster using the “Polygon to Raster” tool (resulting in LU_RST).
5) Converted the raster to ASCII using the “Raster to ASCII” tool.
6) Imported the ASCII file to IDRISI using the ARCRASTER module.
7) Reclassed all error values (-9999) to be equal to 0 using the RECLASS module (resulting in LU_RC).
8) Assigned the suitability values for each land use using ASSIGN module and a custom .avl file (resulting in LU_fuzz).
| Back |
Creating SLOPE_fuzz
What is SLOPE_fuzz?
SLOPE_fuzz is a raster image that represents the suitability of a location based on its slope. The suitability of particular slopes were determined using the information issued by BuildingAdvisor (2012). They recommended building upon slopes of less than 10% and warned that slopes of 20% or greater will require an extensive amount of work to properly prepare the site for construction. Thus, locations less with less than 10% slope were considered to be suitable and sites with slopes greater than 20% were considered unsuitable.
IDRISI Selva Operations
Here is a Macro Modeler design that summarizes the operations performed in IDRISI Selva to create SLOPE_fuzz:
Macro Modeler Legend
Complete Step by Step Process
1) Downloaded the original digital elevation model files, 092g02_0102_demw.dem and 092g07_0100_demw.dem (Zones 092G02 and 092G07), from GeoBase.
2) Re-projected them to NAD_1983_UTM_Zone_10N using the “Project Raster” tool.
3) Joined the files using “Mosaic to New Raster” tool (resulting in DEM_suit_proj).
4) Resampled the raster to 20m by 20m resolution using the “Resample” tool.
5) Clipped the file to a custom polygon (Station Suitability Area) of the study area (resulting in dem_resa_clip).
6) Converted the raster to ASCII using the “Raster to ASCII” tool.
7) Imported the ASCII file to IDRISI using the ARCRASTER module.
8) Used RESAMPLE module to align cells and extent of file (resulting in DEM_RES).
9) Calculated the slope (percent) of each cell using the SURFACE module (resulting in SLOPE).
10) Added SLOPE to the Decision Wizard and defined its fuzzy parameters.
11) The function was defined as sigmoidal and monotonically decreasing with maximum suitability assigned to slopes of 10% or less and minimum suitability assigned to slopes of greater than 20%.
12) After finished the Decision Wizard step, SLOPE_fuzz is created!
| Back |
Creating POP_fuzz
What is POP_fuzz?
POP_fuzz is a raster image that represents the suitability of a location based on the population in a 1020m by 1020m area around a location. I considered the maximum population calculated (9403) to be the most suitable (a value of 255) and a population of 0 as not suitable (a value of 0).
The size of the square around each location was 1020m by 1020m because of the maximum walking time it would take to reach the centre (if a station was located there). The maximum time it would take for an average person to walk to the centre of such an area would be 8.7 minutes (if they lived at one of the corners of the square; the distance would be 721m). A person living along the middle of one of the edges of the square would have an average walking time of 6.2 minutes (510m in distance) to the centre. These walking times were deemed to be reasonable for potential SkyTrain passengers to endure in order to use the service of the nearest SkyTrain station.
The distances and walking times were calculated using pedestrian walking data collected by TranSafety, Inc. (1997). They found that the average walking speed of a pedestrian was 4.43ft/s, which equates to 4.97km/hr or 82.8m/min.
The square was also 1020m by 1020m because a filter was used in the creation of POP_fuzz and filters require a filter size to be in an odd number of cells. The cell size used in this analysis was 20m by 20m, therfore a neat size such as 1000m by 1000m could not be used.
IDRISI Selva Operations
Here is a Macro Modeler design that summarizes the operations performed in IDRISI Selva to create POP_fuzz:
Macro Modeler Legend
Complete Step by Step Process
1) Downloaded the original dissemination area boundary file, gda_059b06a_e.shp, from Statistics Canada.
2) Reprojected it to NAD_1983_UTM_Zone_10N using the “Project Raster” tool.
3) Clipped the file to a custom polygon (Station Suitability Area) of the study area (resulting in dissem_2006_clip).
4) Calculated the Area of each polygon (dissemination zone) using Calculate Geometry in the attribute table.
5) Gathered and manually entered population data (From Census Canada) for each polygon (dissemination zone) in the clipped file.
6) Calculated the population density (per sq. m) for each dissemination zone in a new field in the attribute table (POP_DENS).
7) Corrected the population density values for clipped dissemination zones using the area provided from Census Canada.
8) Converted the population density values to raster using the “to Raster” tool.
9) Converted the raster (pop_dens) to ASCII using the “Raster to ASCII” tool.
10) Imported the ASCII file to IDRISI using the ARCRASTER module.
11) Calculated the mean population density (POP_DENS_1k) in 1020mx1020m squares around each cell by using a custom filter (51x51 cells with each cell being equally weighted).
12) Reclassed all error values (-9999) to be equal to 0 using the RECLASS module (resulting in POP_DENS_1k_RC).
13) Created a base raster (squared_1k), using the INITIAL module, in which all cells were equal to 1040400 sq. m (which is the area of each 1020mx1020m square).
14) Overlaid (multiply) the population density with the base raster to calculate the total population around each cell (resulting in POP_6mins).
15) Added POP_6mins to the Decision Wizard and defined its fuzzy parameters.
16) The function was defined as linear and monotonically increasing with maximum suitability at a population of 9403 and a minimum suitability at population of 0.
17) After finishing the Decision Wizard steps, POP_fuzz is created!
| Back |
Creating CITY_fuzz
What is CITY_fuzz?
CITY_fuzz is a raster image that represents the suitability of a location based on its distance away from a city centre. I considered a location to be at the maximum suitability (a value of 255) if it was within 5 minutes of walking distance (414m). Locations that were further than 30 minutes of walking distance (2485m) away from a city centre were considered unsuitable (a value of 0).
These classifications are based off of reasonable judgement. I determined that a 5 minute walk to a city centre from a SkyTrain station would be quite convenient (suitable) for a passenger, and that if it takes over 30 minutes to get there, it would be unreasonable and inconvenient (unsuitable) for a passenger. The distances and their walking times were calculated using pedestrian walking data collected by TranSafety, Inc. (1997). They found that the average walking speed of a pedestrian was 4.43ft/s, which equates to 4.97km/hr or 82.8m/min.
City centres were defined as high traffic areas, within the study area, as recognized by the author’s judgement. Locations that were considered as city centres were: Town Centre Park, Coquitlam Centre, Rocky Point and Lougheed Town Centre.
IDRISI Selva Operations
Here is a Macro Modeler design that summarizes the operations performed in IDRISI Selva to create CITY_fuzz:
Macro Modeler Legend
Complete Step by Step Process
1) Created a shapefile (City_Centres) in ArcMap using personal knowledge of high traffic Coquitlam/Port Moody areas.
2) Imported the shapefile (City_Centres) to IDRISI using SHAPEIDR.
3) Converted City_Centres to Raster using the PolyRas module (resulting in City_Cent_Ras).
4) Calculated distance away from City Centres using the Distance module (resulting in DIST_City_Cent).
5) Added DIST_City_Cent to the Decision Wizard and defined its fuzzy parameters.
6) The function was defined as linear and monotonically decreasing with an upper bound of 414m (5mins of walking) and a lower bound of 2485m (30mins of walking).
7) After finishing the Decision Wizard steps, CITY_fuzz is created!
| Back |
Creating TRANSIT_fuzz
What is TRANSIT_fuzz?
TRANSIT_fuzz is a raster image that represents the suitability of a location based on its distance away from a transit hub. This raster was made using the same logic as CITY_fuzz, so locations were considered most suitable if they were within 414m and unsuitable if they were greater than 2485m away from a transit hub.
Transit hubs were classified as high traffic TransLink transit exchanges, which included: Coquitlam Central Station, Port Moody Station and Lougheed Town Centre Station.
IDRISI Selva Operations
Here is a Macro Modeler design that summarizes the operations performed in IDRISI Selva to create TRANSIT_fuzz:
Macro Modeler Legend
Complete Step by Step Process
See CITY_fuzz, the same steps were followed to create TRANSIT_fuzz except that the original shapefile was created using the locations of the 3 transit hubs and was named Transit_Centres. The above Macro Modeler design shows the file names used in the creation of TRANSIT_fuzz.
| Back |