Using LIDAR Data to Examine Habitat Complexity and Ecology of a Coral Reef
2 likes1,047 views
1. The study evaluated the relationship between LIDAR-derived rugosity and coral reef fish assemblages in Hanauma Bay, Hawaii. 2. LIDAR rugosity was highly correlated with in-situ rugosity measurements at a 4m grid scale. 3. Fish abundance, richness, and biomass at hard bottom sites were significantly correlated with LIDAR rugosity at various grid scales, but not at sand sites. Higher rugosity was associated with greater fish numbers and biomass.
Using LIDAR Data to Examine Habitat Complexity and Ecology of a Coral Reef
- 1. Using LIDAR Data to Examine Habitat Complexity & Ecology of a Coral Reef Lisa Wedding a,b, Alan Friedlander b,c a University of Hawaii at Manoa, Department of Geography b NOAA/NCCOS/CCMA/NOS Biogeography Branch c The Oceanic Institute
- 2. Presentation outline • Research objectives • Background – habitat complexity • Data & methods – Fish & habitat surveys – LIDAR data & GIS rugosity analysis • Results – in-situ/LIDAR-derived rugosity – Relationship between fish community structure • Discussion & conclusions – Implications for conservation & MPA design – Future research directions
- 3. Research objectives 1. Evaluate the utility of LIDAR technology for deriving rugosity (a measure of habitat structural complexity) on a coral reef in Hawaii 2. Examine the relationship between coral reef fish assemblage characteristics & LIDAR-derived rugosity
- 4. Importance of habitat structural complexity • Habitat complexity plays a major role in the distribution & structure of fish assemblages • Provide niches, refuge from predation –harbor high species diversity, richness & biomass • Significant management implications - high complexity areas offer greater natural protection - ID these locations can help prioritize areas for conservation - inform MPA placement & design
- 5. Study site – No-take MPA, Est. 1967, 41 ha
- 6. Sampling design Random stratified design • Fish Censuses • 25m x 5m transects • Habitat metrics • biotic cover (coral, algae, inverts) Habitat Complexity • abiotic in-situ (chain method) (depth, habitat complexity) Rugosity : R = dc/dl dc = distance of chain across surface contour dl = linear distance of the transect line 5m 25m
- 8. Shoals LIDAR data at Hanauma Bay USACE Horizontal Accuracy + 1.5 m Vertical Accuracy + 20 cm Min. Depth Range 0-1 m Max Depth Range 40 m Sounding Density 4x4m N (Hanauma Bay) 38,743 •USACE Shoals LIDAR surveys 1999-2000 •Irregularly spaced data, need to interpolate into DEM
- 10. Work flow: LIDAR-derived rugosity LIDAR data acquisition LIDAR collects x,y,z data Data processing (QA/QC, project, clip to AOI) DEMs created in GIS (4, 10, 15, 25 m) LIDAR-derived rugosity product Rugosity grid created from DEM
- 11. Benthic terrain analysis • ArcGIS Benthic terrain modeler extension (Lundblad et al. 2004) – www.csc.noaa.gov/products/ btm/ • Developed by NOAA Coastal Services Center & OSU – to classify habitats & derive slope and rugosity measures from multibeam data
- 12. Calculating rugosity from a bathymetric grid • Obtains the surface area for the central cell (165) based on the elevation values of the eight surrounding cells • Index of Rugosity = surface area planimetric area •Calculated by dividing the surface area of the cell with the planimetric area of the cell to get a measure of habitat complexity In-situ Rugosity = distance of chain linear distance of transect Jenness (2004)
- 13. Research objectives 1. Evaluate the utility of LIDAR technology for deriving rugosity on a coral reef 2. Examine the relationship between coral reef fish assemblage characteristics & LIDAR-derived rugosity
- 14. Correlation between in-situ chain rugosity & LIDAR-derived rugosity Spearman rank correlation coefficient (P-value) Grid Size (m) 4 10 15 25 Chain rugosity 0.61 -0.01 -0.12 -0.09 (<0.01) (-0.98) (-0.60) (-0.70) • LIDAR-derived rugosity was highly correlated w/ in-situ rugosity (4 m grid)
- 15. Research objectives 1. Evaluate the utility of LIDAR technology for deriving rugosity on a coral reef in Hawaii 2. Examine the relationship between coral reef fish assemblage characteristics & LIDAR-derived rugosity
- 16. Relationship between LIDAR-derived rugosity & fish assemblage characteristics (hard bottom) Fish assemblage metrics LIDAR-derived rugosity 25 m 15 m 10 m 4m Numerical abundance 0.73 0.67 0.58 0.68 (<0.01) (<0.01) (<0.05) (<0.01) Species richness 0.66 0.51 0.65 0.64 (<0.01) (0.06) (<0.01) (<0.05) Biomass ( t ha-1) 0.65 0.61 0.50 0.52 (<0.05) (<0.05) (0.07) (0.06) Species diversity (H’) 0.41 0.21 0.51 0.41 (0.14) (0.45) (0.06) (0.14) Values are Spearman Rank Correlation (P-value) Wedding et al. (in press)
- 17. Relationship between LIDAR-derived rugosity & fish assemblage characteristics (hard bottom) Fish assemblage metrics LIDAR-derived rugosity 25 m 15 m 10 m 4m Numerical abundance 0.73 0.67 0.58 0.68 (<0.01) (<0.01) (<0.05) (<0.01) Species richness 0.66 0.51 0.65 0.64 (<0.01) (0.06) (<0.01) (<0.05) Biomass ( t ha-1) 0.65 0.61 0.50 0.52 (<0.05) (<0.05) (0.07) (0.06) Species diversity (H’) 0.41 0.21 0.51 0.41 (0.14) (0.45) (0.06) (0.14) Values are Spearman Rank Correlation (P-value) Wedding et al. (in press) •Hard bottom sites had sig. correlations w/ LIDAR rugosity & numerical abundance, richness & biomass
- 18. Relationship between LIDAR-derived rugosity & fish assemblage characteristics (hard bottom) Fish assemblage metrics LIDAR-derived rugosity 25 m 15 m 10 m 4m Numerical abundance 0.73 0.67 0.58 0.68 (<0.01) (<0.01) (<0.05) (<0.01) Species richness 0.66 0.51 0.65 0.64 (<0.01) (0.06) (<0.01) (<0.05) Biomass ( t ha-1) 0.65 0.61 0.50 0.52 (<0.05) (<0.05) (0.07) (0.06) Species diversity (H’) 0.41 0.21 0.51 0.41 (0.14) (0.45) (0.06) (0.14) Values are Spearman Rank Correlation (P-value) Wedding et al. (in press) •Hard bottom sites had sig. correlations w/ LIDAR rugosity & numerical abundance, richness & biomass
- 19. Relationship between LIDAR-derived rugosity & fish assemblage characteristics (hard bottom) Fish assemblage metrics LIDAR-derived rugosity 25 m 15 m 10 m 4m Numerical abundance 0.73 0.67 0.58 0.68 (<0.01) (<0.01) (<0.05) (<0.01) Species richness 0.66 0.51 0.65 0.64 (<0.01) (0.06) (<0.01) (<0.05) Biomass ( t ha-1) 0.65 0.61 0.50 0.52 (<0.05) (<0.05) (0.07) (0.06) Species diversity (H’) 0.41 0.21 0.51 0.41 (0.14) (0.45) (0.06) (0.14) Values are Spearman Rank Correlation (P-value) Wedding et al. (in press) •Hard bottom sites had sig. correlations w/ LIDAR rugosity & numerical abundance, richness & biomass •Sand sites were not correlated with fish assemblage characteristics
- 20. Relationship between fish biomass (t/ha) and LIDAR-derived rugosity Fish biomass (t/ha) observed on transects Least-squares Simple Linear Regression Grid Size (m) 4 10 15 25 R2 0.643 0.462 0.397 0.386 P-value <0.001 <0.001 <0.01 <0.01 • LIDAR-derived rugosity was a statistically significant predictor of fish biomass in Hanauma Bay at all spatial scales
- 21. Summary • Lidar-derived rugosity (4 m) was highly correlated w/ in-situ rugosity & is a viable method for measuring habitat complexity • Lidar-derived rugosity was a good predictor of fish biomass and demonstrated a strong relationship with several fish assemblage metrics in hard bottom habitat • Relating LIDAR-derived rugosity to various fish assemblage characteristics is an important step is applying remote sensing for resource management applications
- 22. Implications for MPA design & function • LIDAR data provides rugosity measures in a min. amount of time at broad geographic scales (~100km2/day) relevant to regional-level management actions • LIDAR id specific areas that offer greater natural protection to fish through habitat complexity – Predict fisheries potential of an area – support optimal location & design of MPAs
- 23. Future work • Continue to examine the associations between habitat complexity & fish assemblages at a broader geographic scale – Expand pilot work to Hawaiian Archipelago • Explore various measures of complexity (e.g. texture measures, fractals) • Predictive mapping of fish communities to inform MPA design and management actions
- 24. Predictive mapping GIS data layers Modeled Distribution Future MPA design Geomorphic structure Species richness Biological cover Species diversity Fish assemblage data Depth Biomass Slope Current MPAs Rugosity
- 25. Acknowledgements • Eric Brown, Alan Hong, Brian Hawk, Ariel Rivera- Vicente • Hawaii Geographic Information Coordinating Council • NOAA NOS NCCOS CCMA Biogeography Branch • NOAA Coral Reef Conservation Program • State of Hawaii, Division of Aquatic Resources • UH, Department of Geography & Ecology, Evolution & Conservation Biology