1 February 2017
For this assignment I reviewed a study conducted by Jeffrey Barrell and Jon Grant titled Detecting hot and cold spots in a seagrass landscape using local indicators of spatial association.
In their study the authors use hydroacoustic technology and local spatial statistics to detect and quantify the patch structure of a select bed of seagrass in the Richibucto estuary in New Brunswick, Canada.
Below are brief summaries of their findings.
- Using ArcGIS v10 and an analysis of isotropic and anisotropic variograms, slight anisotropy in the north-south direction was found to be present within the bed of seagrass.
- A contoured seagrass cover map was produced through kriging using ArcGIS Geostatistical Analyst to model global spatial patterns (such as anisotropy) and patterns of clustering.
- The Getis-Ord Gi* statistic was used to determine hot spots and cold spots of seagrass cover.
- Hot and cold spot seagrass patch polygons were created using ArcGIS v10 with 30, 40, and 50 m buffer rings drawn around each statistically significant cluster. Overlapping zones were identified to be boundaries of rapid change between high and low seagrass cover.
The largest number statistically significant clusters of seagrass were found at a 30 m search radius and was thus was determined to be the optimal scale at which to model within patch spatial patterns.
I really enjoyed this study because it was well written and I felt that the authors effectively achieved the goal of their study—to identify local spatial distribution of seagrass. Because Barrell and Grant were interested in analysing local (as opposed to global) spatial patterns, they made use of hydroacoustic technology to collect data at the patch scale (as opposed to using conventional large scale aerial data) and then used local spatial statistics (as opposed to using global spatial statistics) to conduct their geostatistical analyses. Given this and Barrell and Grant’s well thought-out method and analysis, I gave their study a rating of 10 out of 10.
References
Anstee, J., Brando, V., Dekker, A., Fyfe, S., Karpouzli, E., & Malthus, T. (2006). Remote sensing of seagrass ecosystems: use of spaceborne and airborne sensors. In A.W.D. Larkum, C. Duarte, & R.J. Orth (Eds.), Seagrasses: biology, ecology and conservation (pp. 347-359). Dordrecht: Springer.
Barrell, J. & Grant, J. (2013). Detecting hot and cold spots in a seagrass landscape using local indicators of spatial association. Landscape Ecology. 28, 2005–2018. doi: 10.1007/s10980-013-9937-2
Finkbeiner, M., Kirkman, H., & McKenzie, L. (2001). Methods for mapping seagrass distribution. In F. Short & R. Coles (Eds.), Global seagrass research methods (pp. 101-121). Amsterdam: Elsevier.