SMT201 AY2019-20G1 EX2 Eugene Choy Wen Jia

From Geospatial Analytics for Urban Planning
Revision as of 21:32, 7 November 2019 by Eugene.choy.2016 (talk | contribs)
Jump to navigation Jump to search

Objective

To identify a suitable location for building a National Communicable Disease Quarantine Centre.
Quarantine-1313617-1279x263.jpg


Part One - Map Views w Roads, Buildings, Natural Areas and Elevation

Eugene Part1 Maps.png

Map View w Roads

The road network in Gombak seems to be somewhat sparse as there are large areas which are not covered by roads. This will be a consideration as we are looking for suitable location that are accessible. However, the lack of road network could also suggest that there are large plot of area that might be available for our selection process. For the map view, I used black to represent the roads that can be used for transportation of materials to a suitable site. The grey roads represents paths/walkways or roads that leads to residential area which will not be used for this study. The roads indicate in grey will be excluded for the next part of the analysis such as the proximity map.

Map View w Buildings

The Buildings in Gombak is also somewhat sparse which means that are large area of land that are available for us to pick from as suggested previously by the sparse road network in Gombak. Most of the buildings in Gombak seem to be concentrated near the perimeter of the Gombak area.

Map View w Natural Areas

There are also some Natural Area features found in the Gombak Area. There are some forest and park area represented in dark green as well as water bodies represented in blue in this area. They are located near to the perimeter of the Gombak Area. However most of the natural areas seem to be near already existing buildings so it probably will not play a very huge part in the selection of area.

Raster w Elevation

The raster is represented with green being the low-lying land, yellowish to be medium elevated land and red being land that are elevated at the highest. From the raster map, it seems like the lower middle of Gombak is the most elevated part, which probably is the reason why that area seem to be the least populated with road networks or even buildings.

Part Two Proximity Raster w Roads, Buildings, Natural Areas and Slope Raster layer

Eugene Part2 Proximity.png

Proximity Raster w Roads

The Gombak road area is first processed to export a layer with non Residential roads first before a Road raster is generated on the non Residential Road followed by the Proximity Raster for the layer. This is because roads that links to residential area should not be considered since we will already be using buildings to represent the population. It is also unlikely for resources to be transported on those road networks.

Filter Residential Roads Eugene.jpg

As Accessibility is one of the important consideration for the selection of the ideal area, The Proximity Raster map is used to determine the distance of a road from a particular (5*5) pixel in the area. Green is used to represent pixels that are closest to a particular road, yellow being moderately close and red representing that a particular pixel is a big distance away from the road. The bins for the colour gives preference to pixels less than 150m away from road with the colour green and anything else more than that will be either yellow or red

Min = 0 meters Max = 768.8464965820312 meters

Proximity Raster w Buildings

Since we are trying to avoid building the Quarantine center near to the population (HDB, Schools, Public Area...), the Proximity Raster map gives preference to pixels which are at least 250 meters away from the building which is indicated with the color green as ideal. Anything lower than 250 will be labelled either yellow or red with red being alarmingly close to the buildings in Gombak.

Min = 0 meters Max = 844.0379638671875 meters

Proximity Raster w Natural Areas

It will also be ideal if the selected site can be a distance away from any natural areas in the Gombak Area. Since we are trying to avoid the area. The Proximity Raster is labelled similarly to the Proximity Raster for buildings. The raster map gives preference to pixels which are at least 200 meters away from any natural areas in the Gombak area. Any raster pixel which is at least 200meters away from a natural area is in green, anything nearer than that will either be yellow or red.

Min = 0 meters Max = 887.144287109375 meters

Slope Raster Layer

The Slope Raster Layer is created to show the degree of slopes in the Gombak Area. Green represent slopes which are less than 15 degrees while anything more than 15 will either be yellow or red. Judging from the mass of green in the slope raster, it seems like most of the slopes in the Gombak Area is somewhat of a low degree. Most of the slopes are concentrated around the center of the area.

Min = 0 Degrees Max = 34.21861267089844 Degrees


Part Three - Criterion Scores for various Layers

The Min-Max method is used to standardise the various raster layer to derive the various criterion score from the Proximity Rasters. Most of the reasoning for the color scheme is also similar as the one in the proximity maps.

Eugene Part3 Criterion Score.png


Criterion Score for Roads

The colour scheme for Criterion score is similar to the Proximity raster except that the score has been standardised w Min-Max method on the values of the Proximity Map for Roads. To get the Min-Max i ran the Raster Layer statistic on each proximity layer to get the statistics.

Formula: (X - 0 meters ) / 768.8464965820312 meters

Criterion Score for Buildings

Criterion Score for Natural Areas

Criterion Score for Slope

References

The Colors on Map - https://www.thoughtco.com/colors-on-maps-1435690ra
Photo by Thad Zajdowic from https://freeimages.com/