Jon D. Sellars1,2 Claudia L. Jolls1

Success of seabeach amaranth (Amaranthus pumilus Raf.) using habitat selection based on light detection and ranging (LIDAR) data

• Jon D. Sellars1,2

• Claudia L. Jolls1

• Cass A. Wigent1

Discriminant Function Analysis

• Variables extracted from 2000 LIDAR were used to model suitable habitat in 2001 for Shackleford Banks and Cape Lookout Spit, Cape Lookout National Seashore, NC

• Elevation
• Elevation in the North American Vertical Datum of 1988 (m)
• Distance from shore
• Distance from the mean high water line (m)
• Surface complexity
• Standard deviation of the surface normal vectors in a 9 x 9 m2 neighborhood
• Slope
• Slope of the surface in degrees
• Grey scale reflectance
• Passive LIDAR data able to distinguish sand/water/vegetation

Discriminant Function Analysis

• Plant locations were captured with a Differential Global Positioning System (DGPS)

• Estimated accuracy ± 1.2 m

• Plants (n = 168 represented by 126 unique DGPS locations) were compared to randomly generated points (n = 426) with in the study area

• 3 x the number of random points were used to capture the greater background variation

• Plant occurrences (n = 26) originally withheld from the model were used as a validation set

• Every fifth point that represented a single plant location

• All statistical analyses were performed in SPSS 11.5.0

• Variables were square root transformed to keep the DFA robust despite outliers

Discriminant Function Analysis

Discriminant Function Analysis

Discriminant Function Analysis

• Distance and reflectance were best able to distinguish suitable habitat based on their ability to differentiate between Random and Occurrence Points

• Variable coefficients from SPSS were used to model habitat in ArcView (3 m2 cells)

• Using Map Calculator function

• Model correctly identified 24/26 (92 %) of validation points as occurring in suitable habitat

• Overlay analysis in ArcView

Conclusions

• Habitat variables can be extracted from remote sensing data

• LIDAR data can be used to delineate and model suitable habitat for Amaranthus pumilus

• Distance from shore and passive data efficiently model suitable habitat

• Potentially distance from shore is a surrogate measure of disturbance
• Passive data delineate areas free of vegetation

ACKNOWLEDGMENTS

• ACKNOWLEDGMENTS

• North Carolina Plant Conservation Program

• National Park Service

• East Carolina University

• ALACE (NOAA, NASA, USGS)

• Mike Aslaksen, NGS-NOAA

• Marj Boyer, NCDA-PCP

• Jeff Colby, ECU

• Karl Faser, ECU

• Cecil Frost, NCDA-PCP

• Michael Hearne, NOAA

• Mark Jansen, NOAA

• Marcia Lyons, NPS

• Karen Trueblood, ECU

• Keith Watson, USFWS

• Jason Woolard, NGS-NOAA