Week 05 Lecture: Intro to GIS / Spatial Data

WK05

Intro to GIS / Spatial Data

GIS, or the social construction of spatial data — a map is an argument about the world, not the world

Lecture Slides

GIS — or the Social Construction of Spatial Data

LAAR 61400 · Week 05

Part I

A map is an argument

Part II

From globe to grid

Part III

How space becomes data

Part IV

Analysis

Cholera deaths plotted over time as a bar chart

“Science manipulates things and gives up living in them. Constructive scientific activities see themselves and represent themselves to be autonomous, and their thinking deliberately reduces itself to a set of data-collecting techniques which it has invented. To think is thus to test out, to operate, to transform—the only restriction being that this activity is regulated by an experimental control that admits only the most "worked-up" phenomena, more likely produced by the apparatus than recorded by it.”

Maurice Merleau-Ponty · Eye and Mind · 1961

Part I

A map is an argument

Part II

From globe to grid

Part III

How space becomes data

Part IV

Analysis

The geoid — exaggerated gravity-field surface of the earth

Latitude and longitude graticule on a globe

Projecting the spheroid onto a flat plane

Conic, cylindrical and planar developable surfaces

A face projected onto a globe and back — distortion illustration

A grid of different world map projections

Part I

A map is an argument

Part II

From globe to grid

Part III

How space becomes data

Part IV

Analysis

RASTERvsVECTOR

Vector stores objects with edges — points, lines, polygons. Raster stores a field sampled on a grid of cells. The model you pick pre-decides which questions are easy.

GIS layers stacked: elevation, land use, parcels, streets

A digital elevation model rendered in grayscale

Raster terrain zoomed to reveal the grid of cell values

Map algebra / neighborhood operation on a raster grid

Part I

A map is an argument

Part II

From globe to grid

Part III

How space becomes data

Part IV

Analysis

Colored digital elevation model — Elevation (DEM)

Elevation surface to flow direction, with the 8-direction coding scheme — Elevation → flow direction

Flow direction raster derived from the DEM — Flow direction

Flow direction grid to flow accumulation, with the 8-direction coding scheme — Flow direction → accumulation

Extracted stream network — Extracted stream

Terrain grid driving a parametric model in Rhino/Grasshopper

Terrain re-represented as mesh and contours in a Grasshopper definition

Worked example: elevation surface to flow direction, with the 8-direction coding scheme

# flow direction: for every cell, where does water leave?
DIR = {E:1, SE:2, S:4, SW:8, W:16, NW:32, N:64, NE:128}

for each cell (r, c) in DEM:
    center  = DEM[r][c]
    maxDrop = 0
    flow    = 0
    for each neighbor (nr, nc) around (r, c):   # the 3×3 box
        drop = (center - DEM[nr][nc]) / dist(nr, nc)
        if drop > maxDrop:
            maxDrop = drop
            flow    = DIR[ direction(nr, nc) ]
    FLOWDIR[r][c] = flow      # one of 8 codes — or 0 if a sink

“…to think is to test out, to operate, to transform.”

Maurice Merleau-Ponty · Eye and Mind · 1961

/ 26