Lecture 13 Error and uncertainty

1
Lecture 13 Error and uncertainty

• Outline
• terminology, types and sources
• why is it important?
• handling error and uncertainty

2
Introduction

• GIS, great tool but what about error?
• data quality, error and uncertainty?
• error propagation?
• confidence in GIS outputs?
• NCGIA Initiative I-1
• major research initiative?
• dropped because too hard?
• Be careful, be aware, be upfront...

3
Terminology

• Various (often confused terms) in use
• error
• uncertainty
• accuracy
• precision
• data quality

4
Error and uncertainty

• Error
• wrong or mistaken
• degree of inaccuracy in a calculation
• e.g. 2 error
• Uncertainty
• lack of knowledge about level of error
• unreliable

5
Accuracy vs. Precision
Inaccurate
Accurate
1
2
Imprecise
4
3
Precise
6
Question

• What does accuracy and precision mean for GIS
  co-ordinate systems?

7
Quality

• Data quality
• degree of excellence
• general term for how good the data is
• takes all other definitions into account
• error
• uncertainty
• precision
• accuracy

8
Types and sources of error

• Group 1 - obvious sources
• age of data and areal coverage
• map scale and density of observations
• Group 2 - variation and measurement
• positional error
• attribute uncertainty
• generalisation
• Group 3 - processing errors
• numerical computing errors
• faulty topological analyses
• interpolation errors

9
Age of data
Northallerton circa 1999
Northallerton circa 1867
10
Global DEM
National DEM
European DEM
Scale of data
Local DEM
11
Digitiser error

• Manual digitising
• significant source of positional error
• Source map error
• scale related generalisation
• line thickness
• Operator error
• under/overshoot
• time related boredom factor

12
Regular shift
original
digitised
13
Distortion and edge-effects
original
digitised
14
Systematic and random errors
original
digitised
15
Obvious and hidden errors
original
digitised
16
Vector to raster conversion error

• coding errors
• cell size
• majority class
• central point
• grid orientation
• topological mismatch errors
• cell size
• grid orientation

17
Effects of raster size
Fine raster
Coarse raster
18
Effects of grid orientation
Original
Original raster
Tilted
Shifted
19
Attribute uncertainty

• Uncertainty regarding characteristics
  (descriptors, attributes, etc.) of geographical
  entities
• Types
• imprecise (numeric) or vague (descriptive)
• mixed up
• plain wrong!
• Sources
• source document
• misinterpretation (human error)
• database error

20
Imprecise and vague
505.9
500
500-510
238.4
240
230-240
21
Mixed up
505.9
238.4
238.4
505.9
22
Just plain wrong...!
505.9
100.3
238.4
982.3
23
Generalisation

• Scale-related cartographic generalisation
• simplification of reality by cartographer to meet
  restrictions of
• map scale and physical size
• effective communication and message
• can result in
• reduction, alteration, omission and
  simplification of map elements
• passed on to GIS through digitising

24
Cartographic generalisation
13M
110,000
1500,000
125,000
City of Sapporo, Japan
25
Question

• An appreciation of error and uncertainty is
  important because

26
Handling error and uncertainty

• Must learn to cope with error and uncertainty in
  GIS applications
• minimise risk of erroneous results
• minimise risk to life/property/environment
• More research needed
• mathematical models
• procedures for handling data error and
  propagation
• empirical investigation of data error and effects
• procedures for using output data uncertainty
  estimates
• incorporation as standard GIS tools

27
Question

• What error handling facilities are their in
  proprietary GIS packages like ArcGIS?

28
Basic error handling

• Awareness
• knowledge of types, sources and effects
• Minimisation
• use of best available data
• correct choices of data model/method
• Communication
• to end user!

29
Question

• How can error be communicated to end users?

30
Quantifying error

• Sensitivity analyses
• Jacknifing
• leave-one-out analysis
• repeat analysis leaving out one data layer
• test for the significance of each data layer
• Bootstrapping
• Monte Carlo simulation
• adds random noise to data layers
• Simulates the effect error/uncertainty

31
Conclusions

• Many types and sources of error that we need to
  be aware of
• Environmental data is particularly prone because
  of high spatio-temporal variability
• Few GIS tools for handling error and uncertainty
  and fewer still in proprietary packages
• Need to communicate potential error and
  uncertainty to end users

32
Practical

• Error in off-the-shelf datasets
• Task Assess the error in land cover data
• Data The following datasets are provided for the
  Leeds area
• Streets and buildings (110,000 OS LandLine data)
• 25m resolution land cover data (ITE LCM90)

33
Practical

• Steps
• Display OS LandLine data over ITE LCM90 data
  using ArcMap. You can also add the OS 150,000
  colour raster image and set transparency 70.
• From your knowledge of the area identify areas of
  erroneous classification
• What might these errors be due to?

34
Learning outcomes

• Familiarity with error in classified satellite
  imagery
• Familiarity with ITE land cover map 1990 (LCM90)
  data
• Experience with new GRID functions

35
Useful web links

• The Geographers Craft lecture on error
• http//www.colorado.edu/geography/gcraft/notes/err
  or/error_f.html
• GIGO
• http//www.geoplace.com/gw/2000/1000/1000gar.asp
• Disaster waiting to happen
• http//www.osmose.com/utilities/articles_press_rel
  eases/data_quality/

36
Next week

• Interpolating environmental datasets
• creating surfaces from points
• interpolation basics
• interpolation methods
• common problems
• Practical Interpolating surfaces from point data