Assessing and Managing Error in Spatial Data - Exam 2 Notes | CSS 4200, Exams of Agricultural engineering

Download Assessing and Managing Error in Spatial Data - Exam 2 Notes | CSS 4200 and more Exams Agricultural engineering in PDF only on Docsity!

CSS 4200

Geographic Information Systems

Lecture 17: -^

Course Update

-^

Assessing and Managing Error in Spatial Data

(Bolstad, Chap. 14)

Announcements

•^

ArcView 9.3 and Extensions (15)

-^

Laboratory Quizzes: 10 November, 01 December weeks

-^

Prelim Exam #2 Review: Thursday, 20 November

-^

Prelim Exam #2:

Tuesday, 25 November

•^

Textbook Chapters: #5, #11-

-^

Practice problem set: 10 November week

-^

Final Examination (“N”):Wednesday, 17 December, 9-11:30a, G14 Fernow

Errors vs. Mistakes

•^

Errors– Systematic

• Conform to physical parameters
  • Random
    • Conform to laws of probability

•^

Mistakes (Blunders)– Don’t conform to anything– Can only be avoided, not modeled

GIS Data Quality

•^

Accuracy and precision– Spatial and attribute

•^

Logical consistency

•^

Completeness

•^

Timeliness

•^

Spatial variability

•^

Appropriateness– scale or resolution– classification (intent)– format

Timeliness

•^

Age and lineage of geospatial data– Some data age well, some poorly

•^

Using data from the wrong time can causeproblems– Particularly if different layers are different ages

•^

Assessing change may be a goal–

One theme, multiple dates

Scale & Resolution (“grain”)

•^

Generalization mayimproperly representsize and shape

-^

Cartographic aesthetics

-^

Entire regions may beeliminated (islands,peninsulas, etc.)

-^

Scale vs. resolution

Classification

•^

What is intended use of source data?– Source data produced for other purposes– Scale limitations

• Minimum mapping unit
  • Themes
    • Land use vs. land cover, for example

•^

Secondary data issues

Processing Errors

•^

Different precision of numerical data

•^

Topological problems– Assumption of sharp boundaries– Digitizing procedures

• 1:100,000 scale map: a line 1 mm in width on maps

represents

?

m on ground surface

Error Propagation

•^

Additive v. multiplicative errors

•^

Universal soil loss equation:A = R * K * L * S * C * P

• A = annual soil loss• R - rainfall-erosivity factor (297 +/- 72)• K - soil erodibility (0.1 +/- 0.05)• L - slope length (2.13 +/- 0.045• S - slope steepness, gradient (1.169 +/- 0.122)• C - cover-management factor (0.5 +/-0 .15)• P - conservation practices 0.5 +/-0.

Thematic Accuracy

•^

Accuracy of attribute data

•^

Associated with spatial, temporal, topologicalaccuracy– Is an attribute wrong because of improper analysis, or

because everything is shifted, or input data was out ofdate, etc.?

• Important if you want to improve quality, accuracy,

and usefulness of project results.

Error Matrices