GEOG 328 Final Unit / Exam Prep Flashcards

1
Q

What is a Binary model? What is it most commonly used for?

A

A Binary model is a model which presents explicit data, generally does not include a suitability scale but rather denotes suitable/unsuitable.

Uses a Boolean output 1 (yes, suitable) or 0 (no, unsuitable)

Binary modelling is generally used for when criteria must absolutely be met.
Commonly used for siting analysis
Example: siting a new school

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What does Map calibration consist of for multi-criteria modelling

A

Binary - apply thresholds: exclude (0), include (1)

Multiple categories - Apply thresholds: high, medium, low (figure out levels of acceptance)
Range from 0-1 (or 100) - Linear / function

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What does layer combination mean with regards to Multi-criteria modelling?

A

Combining the quantitative variables into one layer

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What does Weighted Criteria Mean with regards to Multi-criteria Modelling?

A

The process whereby each input can be weighted according to its importance or its percent influence.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What does “constrain” refer to with regards to multi-criteria modelling?

A

It removes areas that should be excluded - keeps the output from looking strange.

e.g. Beetle infestations don’t occur in water/human illness does not occur where population density is 0.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is a model?

A

A model is a representation of the real world.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is a Spatial Model?

A

A spatial model is a representation of some phenomena and incorporates spatial data and relationships

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What steps do you need to do before modelling?

A

Before modelling, data must be calibrated, layers of data integrated and logical/spatial constraints applied.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Before performing multi-criteria spatial modelling what must be done with the data?

A

For multi-criteria spatial modelling the data must be calibrated, layered, and constrained to aid in guiding the problem solving and analysis.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is terrain?

A

Terrain - Also know as “relief”, this is the third dimension (elevation) of the land surface. The “lie of the land”.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

How is terrain Expressed?

A

Expressed as elevation, slope, and aspect.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What does Terrain influence?

A

Influences climate, water, and water flow.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What is Bathymetry?

A

Underwater terrain.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

How is Digital terrain surface modelling achieved?

A

Elevation points are used to create models of terrain.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

T?F - Both DEM and TIN (triangulated irregular network) are considered to be digital elevation models?

A

TRUE

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Of DEM and Tin, Which is a Raster and which a Vector?

A

DEM is used to refer to raster models.

Tin is used to refer to vector models.

17
Q

What are some methods for obtaining elevation data to create DEM’s

A

Lidar-light detecting and ranging
Stereo photogrammetry from aerial surveys
Multi-view stereo applied to aerial photography.
Interferometry from radar data.
Real time kinematic gps
Topographic maps
Theodolite or total station
Doppler radar
Surveying and mapping drones range imaging.

18
Q

What are some problems with Ground survey and GPS?

A

It is - expensive, time consuming, not viable over large areas

19
Q

Problems with Stereo photogrammetry?

A

Problems due to tree cover / land cover and it is very time-consuming.

20
Q

What are some pros and cons of LiDAR?

A

Pros: Post-processing results in raster surface / Gives you millions of points of data / It can map bare earth even with vegetation.

Cons: Very Very expensive and time-consuming in post processing.

21
Q

Pro’s and Con’s of Modern Photogrammetry?

A

Pros: Advances in computing power, software and drones / Less expensive than LiDar / accuracy ~5cm possible / Results in both DEM and ocular imagery, so good for visualization

Cons: Does not do will with vegetation (>60%)

22
Q

What are some benefits of using InSAR to map DEMs (satellite)

A

Generate high quality DEM models and measurements of displacement (by comparing two images). Not affected by weather. effective for mapping large area’s

23
Q

How do you get collected data points to become surfaces?

A

Input points, whether collected by survey, photographs, LiDAR, ect, must be processed to create terrain models. Some require extensive post-structuring such as LiDAR.

The points are interpolated to create a surface with raster models (DEM), and triangulated to create vector models (TIN).

24
Q

Of Tin and DEM, which represents Gradual changes and which Abrupt?

A

Tin = Abrupt (creates a surface of triangles between collected data points and thus has the ability to represent sharp changes in values like cliffs)

DEM = Gradual (interpolated surface using all of the collected points, creates a gradual surface between values)

25
Q

What are the two types of DEM?

A

DTM (digital terrain model) a bare earth model.

DSM (digital surface model) = DTM + human made elements (buildings, roads) or natural features that are on the bare earth (Such as vegetation).

26
Q

What are Breaklines? Why are they important?

A

features on the landscape that act as discontinuities: they interrupt the surface.

important to try to help increase the accuracy of your in your interpolation. If you have an understanding of terrain and features you can include coordinates of lines or features that discounts data outside of those boundaries

27
Q

In what situation would you choose to use a TIN model? When would you use a DEM model?

A

TIN’s are generally used in smaller areas and are great with a good point cloud, they can also better represent complex terrain features. Processing time is longer + data storage is greater.

DEM (because of generalization) complex features are often smoothed out as a result. they can also include redundancy data. More visually appealing and widely used in GIS.

28
Q

How are Slope / Aspect / Ruggedness determined?

A

All three are derived using a neighbourhood analysis (looking at the input cell and the differences in the cells surrounding it)

29
Q

What is a Watershed?

A

How water is forced to move at the surface through an area as a result of the elevation.

30
Q

What are some applications of TSA applied within hydrological Modelling?

A
Sample applications: 
Flood management (climate change)
Watershed delineation
Stream delineation 
Stream networks
31
Q

What are the steps to delineate a Watershed?

A
  1. Fill DEM (raster processing / pixel resolution / rounding (interpolating) can result in depressions (sinks, which are low points with no output.) Removing sinks is called “hydrological conditioning”
  2. Flow direction Raster. Shows the direction that water will flow out of each cell. Uses a neighbourhood analysis to code and shows direction based on elevation/slope/aspect (like a backlink except not based on cost surface)
  3. Specify pour points (essentially start point in cost distance) cells with high accumulated flow become points - everything upstream will become a watershed.
  4. Delineate watersheds (use flow direction raster and pour raster as imputs)
32
Q

What was the goal of Alex Lausanne’s project?

A

To use suitability analysis to find potential archeological dig sites.

here she could use GIS tools to answer Archaeological problems.

End goal was to create a step by step guide to locate sites within a given time frame.

33
Q

What types of models did Alex Lausanne’s use to try and validate her study?

A

She first used a weighted variable model to find the highest areas of potential archaeological discoveries based on all of the sites variables. The model was also deductive (as no training data was available).

second was a decision tree to try modelling the predictive modelling. (No good)

Final model was binary decision tree (each cell went through the process and had to answer yes to each of the variables, then they would be considered high archeological potential)

34
Q

What was the result of Alex Lausanne’s study?

A

The result was a study area 23% of its original size with a roughly 80% discovery rate of archaeological finds as a result of two surveys undertaken.

35
Q

Alex Lausanne describes what a model is in her presentation. what was that?

A

A model is a representation of reality that must be interpreted by an informed researcher.