Using LEGO blocks to visualize the landscape elevation throughout the province of Ontario was an the objective of this project and the steps I took to execute this project will be outlined below.
I first sourced the elevation data from Scholar’s GeoPortal and used the north and south PDEM files for Ontario as the foundation for the elevation model. Using ArcGIS I added the north and south PDEM layers and merged the two files using Mosaic To New Raster tool. This produced a merged PDEM.
Next, the merged PDEM needed to be resampled, to increase the pixels size so that it would align with the size of a 1×1 LEGO block. Using the Resample tool, I resampled the pixel size from 30x30m to 30,000×30,000m resolution. This resolution was influenced by a number of factors:
- maintaining the integrity of the elevation levels (699m was the highest peak at 30x30m, but it reduced to 596m when resampled to 30kx30k)
- scale of physical model as it relates to size and cost of the LEGO blocks
Below is the resampled layer to 30k resolution and clipped to a raster tiff of Ontario (also at same resolution)
In the Properties dialiogue box I converted the Stretched symbology to Classificled symbology which would allow me to isolate specific elevation interval classes. I seleccted seven classes based on the following criteria:
- Wanted to isolate the high and low values
- Using intervals of 75m depicted the more visually appealling variation in elevation and did so most effectively. It allowed for a <75m and a >450m class
- No more than seven classes because of LEGO colour options and available stock
- Equal interval of 75m increments
Colour selection at this stage was preliminary and a divergent scheme from green to dark burgundy seemed to be most aesthetically pleasing.
To isolate each elevation layer to determine the number of pixels (i.e. LEGO blocks) each layer requires the raster layer had to be converted to a vector layer.
Using Raster Calculator and the Int Tool, I converted the current raster from a float to an integer raster layer which is needed to be done to convert raster to polygon. This converted each cell value of the raster to an integer.
This new raster file was then converted to a polygon layer using the Raster to Polygon tool, creating this output.
Activating the raster layer from a previous step, I was able to then manually select each pixel for each respective layer to determine the number of pixels (ie LEGO pieces) that comprised the layer.
Each pixel was selected using the Selection tool and then onces all pixels for the appropriate layer were selected, the Create Layer from Selected Features was used to create an individual layer for each elevation level.
This process was repeated 7 times, producing 7 layers of elevation. Each layer’s Attribute Table was then used to identify the total number of pixels present in the layer and then was used to determine the number of LEGO pieces needed for that layer, where 1 pixel = 1 single-block LEGO piece.
These individual layers will also be used during the build, as a guideline for the distribution and placement of each LEGO piece.
Each colour class is an individual layer. Colours are still preliminary and the number of LEGO pieces per layer is as follows:
- <75m: 1089 pcs
- 75-150m: 987 pcs
- 150-225m: 809 pcs
- 225-300m: 657 pcs
- 300-375m: 455 pcs
- 375-450m: 221 pcs
- >450m: 51 pcs
Using BrickLink, I was able to purchase 1×1 LEGO bricks for each layer. Factors that influenced the colour selection for each layer are as follows:
- Quantity of colour available
- Price of individual bricks
- Location of supplier (North American)
The resulting colour scheme selected is a divergent scheme, as follows:
- <75m: dark green
- 75-150m: medium grey
- 150-225m: light green
- 225-300m: tan
- 300-375m: light lavender
- 375-450m: medium lavender
- >450m: dark purple
Here is the final product.
Here is a time lapse video of the LEGO build:
https://www.youtube.com/watch?v=RP6PxkPlK1w&feature=youtu.be