GEOMORPHIC CHANGE DETECTION

Transcription

1 GEOMORPHIC CHANGE DETECTION I. THRESHOLDING ALTERNATIVES IN GCD April 30 th to May 1 st, 2014 Sponsored by:

2 WORKSHOP PLAN DAY 1 (AFTERNOON) E. Essential Best Practices to Support GCD F. Traditional Approaches to Change Detection G. Change Detection in Raster Calculator vs. GCD H. Getting to know GCD Software I. Thresholding Alternatives4:00 to 4:45 J. Recap & Preview of Day 2 A Plan.. Not a Contract 1:00 to 1:45 1:45 to 2:15 2:15 to 3:00 3:00 to 3:15 BREAK 3:15 to 4:00 4:45 to 5:00

3 DETAIL PLAN H. H. THRESHOLDING ALTERNATIVES FOR GCD 1. Three Types in GCD 6 2. Recall minlod 3. Error Propagation 4. Probabilistic Thresholding 5. Tutorial 6. Thresholding by Subtraction vs. Exclusion vs. Weighting

4 GCD 6 THRESHOLDING Simple defined min LoD Propagated Errors Probabilistic Confidence Interval

5 DETAIL PLAN H. H. THRESHOLDING ALTERNATIVES FOR GCD 1. Three Types in GCD 5 2. Recall minlod 3. Error Propagation 4. Probabilistic Thresholding 5. Tutorial 6. Thresholding by Subtraction vs. Exclusion vs. Weighting

6 APPLICATION OF A MIN LoD You take original DoD, and remove all changes <= min LoD For example +/- 20 cm

7 VARYING min LoD THRESHOLDS z, zf z DEM old DE new 2 zz 2 z DEM old DEM new Wheaton 2008

8 DETAIL PLAN H. H. THRESHOLDING ALTERNATIVES FOR GCD 1. Three Types in GCD 5 2. Recall minlod 3. Error Propagation 4. Probabilistic Thresholding 5. Tutorial 6. Thresholding by Subtraction vs. Exclusion vs. Weighting

9 MINLoD USING ERROR PROPAGATION Distinguish those changes that are real from noise Use standard Error Propagation DEM Errors can vary temporally and spatially Elevation (Time 1) DEM old z Elevation (Time 2) DEM new z 10cm 20cm z z DEM old 2 z 2 DEM new e.g. z cm 8.8 in See Brasington et al (2000): ESPL Lane et al (2003): ESPL Brasington et al (2003): Geomorphology

10 ERROR PROPAGATION GETS APPLIED SAME WAY AS MIN LoD Does not matter whether the min LoD is specified, or calculated from error propagation Just on a cell-by-cell basis! In background a perror grid is produced

11 WHAT ARE TYPICAL ERRORS? Remotely Sensed or Aerial Surveys LiDaR : +/- 12 to 25 cm Aerial Photogrammetry : +/- 10 to 15 cm Ground-Based Surveys Total Station Surveys : +/- 2 to 10 cm GPS: : +/- 3 to 12 cm Terrestrial Laser Scanning: +/- 0.5 to 4 cm

12 SO WHAT WOULD PROPAGATED ERRORS BE? Remotely Sensed or Aerial Surveys LiDaR : +/- 12 to 25 cm (17 to 36 cm minlod) Aerial Photogrammetry : +/- 10 to 15 cm(14 to 22 cm min LoD) Ground-Based Surveys Total Station Surveys : +/- 2 to 10 cm (3 to 14 cm min LoD) GPS: : +/- 3 to 12 cm (4 to 17 cm min LoD) Terrestrial Laser Scanning: +/- 0.5 to 4 cm (0.7 to 6 cm min LoD)

13 DETAIL PLAN H. H. THRESHOLDING ALTERNATIVES FOR GCD 1. Three Types in GCD 5 2. Recall minlod 3. Error Propagation 4. Probabilistic Thresholding 5. Tutorial 6. Thresholding by Subtraction vs. Exclusion vs. Weighting

14 HOW COULD I REPRESENT AS PROBABILITY? Using inferential statistics, we ll calculate a t-score σ DoD is the characteristic uncertainty In this case σ DoD = minlod t z DEM new z DEM old DoD Just the ratio of actual change to min LoD change Assuming two-tailed test, t is significant at: 68% confidence limit when t= 1 95% confidence limit when t=1.96

15 PROBABILITY THAT CHANGE IS REAL Even when min LoD is spatially constant, probability varies in space why? Wheaton (2008)

16 SENSITVITY OF THRESHOLD?

17 DETAIL PLAN H. H. THRESHOLDING ALTERNATIVES FOR GCD 1. Three Types in GCD 5 2. Recall minlod 3. Error Propagation 4. Probabilistic Thresholding 5. Tutorial 6. Thresholding by Subtraction vs. Exclusion vs. Weighting

18 WHERE DOES THIS FIT? Choosing the Threshold Method

19 GCD 6 THRESHOLDING - TUTORIAL Simple defined min LoD Propagated Errors Probabilistic Confidence Interval

20 DETAIL PLAN H. H. THRESHOLDING ALTERNATIVES FOR GCD 1. Three Types in GCD 5 2. Recall minlod 3. Error Propagation 4. Probabilistic Thresholding 5. Tutorial 6. Thresholding by Subtraction vs. Exclusion vs. Weighting

21 CONCEPT: EXCLUSION OR SUBTRACTION? LoD raster must be applied independently to cut raster and fill raster to preserve signage. Is it better to exclude points whose LoD > DoD or subtract LoD from DoD? Slide from Greg Pasternack (2012) USBoR Lecture

22 COMPARISON OF EXCLUSION VS. SUBTRACTION Exclusion raster Subtraction raster Raw DoD Net Change 33,456 yds 3 Net Change 17,555 yds 3 Blue is deposition; grey is within 1 ; others are erosion Subtraction is always a more conservative estimate Slide from Greg Pasternack (2012) USBoR Lecture

23 WEIGHTING BELOW THE THESHOLD Instead of subtraction or exclusion, values below the threshold can be weighted based on their probabilities For example, if the probability that change is real is 80%, for a vertical change of + 10 cm, the value used could be 8 cm. See Lane SN, Westaway RM and Hicks DM (2003). DOI: /esp.483.

24 DETAIL PLAN H. H. THRESHOLDING ALTERNATIVES FOR GCD 1. Three Types in GCD 5 2. Recall minlod 3. Error Propagation 4. Probabilistic Thresholding 5. Tutorial 6. Thresholding by Subtraction vs. Exclusion vs. Weighting