1 Field- And GIS-Based Measurements of Coastal Change for the Southeast Chukchi Sea, Alaska William F. Manley INSTAAR, University of Colorado Diane M.

1

Field- And GIS-Based Measurements of Coastal Change for the Southeast Chukchi Sea, Alaska

William F. Manley INSTAAR, University of ColoradoDiane M. Sanzone Arctic I&M Program, National Park ServiceJames W. Jordan Dept. of Environmental Studies, Antioch University New EnglandOwen K. Mason GeoArch AlaskaEric G. Parrish INSTAAR, University of ColoradoLeanne R. Lestak INSTAAR, University of Colorado

AGU Presentation -- December 14, 2006

See also: instaar.colorado.edu/QGISL/ARCN/

2

Coastal Erosion

Rapid, observable change to the environment

Multiple impacts on a variety of habitats

Fragile coast is a sensitive indicator of “stressors”:

direct human disturbance climate change:

• longer ice-free season• increased permafrost melting• change in frequency and intensity of storms• sea level rise

3

Goals

Field measurements as test of GIS approach

Preliminary GIS results

4

Study Area

C h u k c h

i S

e a

5

Coastal Monitoring Stations

• 27 sites• first established 1987-1994• revisited in 2006• measured on “bluff top”

6

Field Methods

7

Remote Sensing & GIS Approach

High-resolution base imagery• 2003 orthophoto mosaic

Imagery and data useful for other concerns

Historic aerial photographs• orthorectified photos for ca. 1980• orthorectified photos for ca. 1950

Comparison of different “time slices” allows us to detect and measure change

2003

ca. 1980

ca. 1950

8

• from NOAA & NPS 1:24,000 natural color photos

• mosaic created by Aero-Metric

• 0.6 m resolution

• accuracy: 1.2 m (RMSE)

• 112 tiles, 98 GB: lots of imagery!

• highest res. in Alaska for this large of an area

• available to the public early 2007

• valuable for other types of research

9

60 framesColor IR1.0 m res.

• 57 frames• Color IR• 1:64,000• 1.0 m resolution• 1.5 m accuracy (RMSE)

10

130 framesBlack & White1.0 m res.

• 108 frames• Black and White• 1:43,000• 1.0 m resolution• 2.0 m accuracy (RMSE)

11

Mainland bluff Beach ridge complex

Barrier island or spit

Shoreline Reference Feature (SRF): “bluff top” (wave-cut scarp)

12

13

14

15

16

17

18

19

20

21

22

2003

23

1949

24

1985

25

2003

26

1949

27

1985

28

2003

29

30

DSAS

Thieler et al. (2005)

31

baseline

Baseline

32

Transects

33

Transects

51.4 m÷ 36 yr =1.4 m/yr

1949

1985

34

“Early” Period ca. 1950 – ca. 1980

m/yr

1949

1985

35

“Late” Period ca. 1980 – 2003

m/yr

2003

1985

36

accr

etio

n

37

GIS Errors

ca. 1950 ca. 1980 2003

Orthorectification 2.0 1.5 NA

Digitizing 3 4 3

Total (SRSS) 3.6 4.3 3.0

Shoreline Position (m)

“Early” “Late” “Long Term”

ca. 1950 – ca. 1980

ca. 1980 - 2003

ca. 1950 - 2003

Position (m) 5.6 5.2 4.7

Rate (m/yr) ± 0.20 ± 0.23 ± 0.09

Coastal Change

38

1:1

r2 = 0.80n = 21

39

1:1

meandifference:0.12 m/yr

r2 = 0.80n = 21

40

GIS Errors

ca. 1950 ca. 1980 2003

Orthorectification 2 1.5 NA

Digitizing 3 4 3

Total (SRSS) 3.6 4.3 3.0

Shoreline Position (m)

“Early” “Late” “Long Term”

ca. 1950 – ca. 1980

ca. 1980 - 2003

ca. 1950 - 2003

Position (m) 5.6 5.2 4.7

Rate (m/yr) ± 0.20 ± 0.23 ± 0.09

Coastal Change

Rate (m/yr) ± 0.12

Field Test (mean difference)

41

• spatial variability• “early” erosion

42

• accretion

43

• station eroded• “late” erosion

44

xy plot

“Early” Period ca. 1950 – ca. 1980

Avg ± SD

ca. 1950 – ca. 1980 0.0 ± 1.1

ca. 1980 – 2003 -0.5 ± 0.6

Avg ± SD

ca. 1950 – ca. 1980 -0.6 ± 0.4

ca. 1980 – 2003 -0.2 ± 0.3

CAKR (n = 1628)

BELA (n = 1146)

45

xy plot

“Late” Period ca. 1980 – 2003

Avg ± SD

ca. 1950 – ca. 1980 0.0 ± 1.1

ca. 1980 – 2003 -0.5 ± 0.6

Avg ± SD

ca. 1950 – ca. 1980 -0.6 ± 0.4

ca. 1980 – 2003 -0.2 ± 0.3

CAKR (n = 1628)

BELA (n = 1146)

46

xy plot

Is coastal erosion increasing with Arctic warming?

47

• acceptably low errors• comprehensive spatial analysis

Conclusions

Field:• more precise• measurements more often

GIS:

Is global warming responsible?:

• storm climatology important• high spatial and temporal resolution needed