Kelp-urchin dynamics: stability and thresholds for phase shifts in Newfoundland and the Gulf of St. Lawrence
In eastern Canada, studies of kelp-urchin systems have been generally restricted to small spatial (few 100s m²) and temporal (<5 years) extents by the traditional scuba-based monitoring techniques employed. Investigation of the drivers of kelp distribution over multiple spatiotemporal scales (inc...
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
Memorial University of Newfoundland
2020
|
| Subjects: | |
| Online Access: | https://research.library.mun.ca/15091/ https://research.library.mun.ca/15091/1/thesis.pdf |
| Summary: | In eastern Canada, studies of kelp-urchin systems have been generally restricted to small spatial (few 100s m²) and temporal (<5 years) extents by the traditional scuba-based monitoring techniques employed. Investigation of the drivers of kelp distribution over multiple spatiotemporal scales (including broad spatial [<km²] and temporal [years] extents) and in regions poorly studied is key to assessing the stability of these systems and understanding regional specificities of kelp dynamics across eastern Canada. This thesis investigates the factors controlling kelp distribution and the stability of kelp-urchin systems in southeastern Newfoundland (SEN) and the northern Gulf of St. Lawrence (nGSL) over multiple spatiotemporal scales by applying traditional and novel techniques. In a scuba-based manipulative field experiment in SEN, no significant effect of urchin density was observed on the rate of kelp bed destruction from urchin grazing, suggesting that the minimal urchin density required to maintain destructive feeding on kelp beds may be equal to the lowest density tested (88 urchins·m⁻²) or lower. The suitability of remote sensing and geographic information system (GIS) approaches for mapping kelp in the nGSL was assessed by comparing three image classification methods applied to aerial and satellite imagery. Supervised classification of satellite imagery (89% accuracy) and visual classification of aerial imagery (90% accuracy) were the best methods. Visually classified imagery from the nGSL was used to compute spatial pattern metrics quantifying kelp distribution patterns. These metrics showed that kelp distribution is not uniform, as kelp patches exhibited considerable variation in size and geometric complexity. Kelp presence was negatively correlated with depth, urchin density, and exposure to waves. Investigation of kelp distribution patterns from imagery acquired in six years between 1983 and 2016 in the nGSL revealed an increase in kelp cover since 1999. Harsh oceanographic conditions in late ... |
|---|