An Ocean Divided

A New Approach to Classifying Marine Ecosystems

As part of the ongoing effort to conserve key marine mammal populations in the North Pacific Ocean, scientists are seeking ways to effectively classify marine ecosystems. Fisheries managers are increasingly required to consider ecosystems when assessing commercially exploited stocks, while conservation efforts focus more and more on identifying and protecting critical habitat for species at risk, in all stages of life.

Fig. 1. Ecosystem boundaries according to two well-known marine classification systems applied to the North Pacific by (a) Sherman (1986) and (b) Longhurst (1998).

Even phytoplankton, the simplest of marine life, are patchy, ephemeral, and often quickly consumed by predators. However, the patterns in phytoplankton distribution are largely a biological response to physical forcing – the movement of ocean currents, for example. Thus an ecosystem represents a complex combination of physical oceanography, species behavior, and population dynamics.

The challenge of ecosystem mapping – characterizing a physical environment and its associated plant and animal life – in the ocean has encouraged scientists to develop an innovative new system to identify biologically meaningful regions in the North Pacific. The system is proposed in a new study authored by Edward J. Gregr and Karin M. Bodtker (both of the Marine Mammal Research Unit at the University of British Columbia), and was recently published in the journal Deep Sea Research I.

Searching for Similarities
“In this study, our objective was to apply image classification techniques, proven in the realm of terrestrial ecology, to the marine environment as a method for classifying this environmental structure,” the authors write. Because of the dynamic nature of ocean, and the mobility of marine mammals and commercial fishes, any method for identifying ecological boundaries in the ocean must be adaptable to variability in space and time.

Gregr and Bodtker analyzed comprehensive physical data describing the surface of the North Pacific, using image classification (a method for identifying classes in remotely sensed images) to find regions of similarity within the seascape. By comparing the images with established oceanographic information such as surface currents, the scientists identified 15 distinct, biologically meaningful marine regions in the North Pacific Ocean based on physical oceanography.

“Our results illustrated that relatively contiguous ocean regions can be identified using decadal averages of physical oceanographic variables and a seasonal temporal resolution,” the authors write. “Also, these regions can be related, by size and location, to previously well-known water masses (e.g., the Alaskan Gyre, the subarctic current) before and after the 1976-1977 regime shift, and differences between partitions can be related to seasonal variations in these water masses.”

Hypothesizing that regions of similar physical parameters would have both physical and biological significance, the scientists then looked for seasonal and long-term changes in the pattern of regions and their boundaries over two 10-year time periods, one on either side of the shift in ocean climate that took place in the North Pacific Ocean in 1976-77. It is believed that this natural shift in ocean climate began a series of changes that led to a sharp decline in Steller sea lion populations in Western Alaska. It is also likely that the same regime shift led to an increase of Steller sea lion populations in southeast Alaska.

The authors found that seasonal patterns were more similar between regimes (i.e., before and after the 1976-1977 shift in ocean climate) than from one season to the next within a regime. Also, the magnitude of seasonal transitions appeared to differ before and after the regime shift.

Finally, the scientists assessed the biological significance of the proposed regions by testing for seasonal differences in primary production (chlorophyll-a), derived from remote sensing data, among the regions following the regime shift.

Robust Approach
“We found that the number of [proposed] regions with distinct chlorophyll signatures, and the associations between different regions, varied by season,” the authors write. “The overall pattern of association between the regions was suggestive of observed, broad-scale patterns in the seasonal development and distribution of primary production in the North Pacific. This demonstrated that regions with different biological properties can be delineated using only physical variables.”

Fig. 2. Upper zone domains of Dodimead et al. (1963) overlaid with regions identified by our classification method for summer pre-1976.

The current study suggests a method of marine ecosystem classification that is robust, quantitative, and can be adapted to different spatial and temporal scales and a range of input variables. The authors suggest that this study may be the first report in which biological relevance is quantified in relation to physical patterns – traditionally, one or more species of interest is included in the analysis.

This represents the first step towards a quantitative description of marine ecosystem boundaries that could apply across a range of biota, rather than a single species or taxonomic group. Future studies may integrate data on species distribution, which may help to visualize realistic ecosystem boundaries.

“Understanding the dynamics of ecosystem boundaries, which may differ according to the species of interest or the management objectives, is a fundamental challenge of ecosystem-based management,” the authors write. “We present an adaptive ecosystem classification that can accommodate these different needs.”

October 1, 2007

Publication:
Adaptive classification of marine ecosystems: identifying biologically meaningful regions in the marine environment. Gregr, E.J. and K. Bodtker. 2007. Deep-Sea Research Part 1 54:385-402. abstract The move to ecosystem-based management of marine fisheries and endangered species would be greatly facilitated by a quantitative method for identifying marine ecosystems that capture temporal dynamics at meso-scale (10?s or 100?s of kilometers) resolutions. Understanding the dynamics of ecosystem boundaries, which may differ according to the species of interest or the management objectives, is a fundamental challenge of ecosystem-based management. We present an adaptive ecosystem classification that can accommodate these different needs. To demonstrate the approach, we quantitatively bounded distinct, biologically meaningful marine regions in the North Pacific Ocean based on physical oceanography. We identified the regions by applying image classification algorithms to a comprehensive description of the ocean?s surface, derived from an oceanographic circulation model. Our resulting maps illustrate 15 distinct marine regions. We investigated seasonal and long-term c! hanges in the pattern of regions and their boundaries by dividing the oceanographic data into four seasons and two 10-year time periods, one on either side of the 1976 ? 1977 North Pacific Ocean climate regime shift. The size and location of our mapped regions related well to previously described water masses in the North Pacific. We compared our results for each season across the regime shift and for sequential seasons within regimes using the Kappa Index of Agreement and the index of Average Mutual Information. Seasonal patterns were more similar between regimes than from one season to the next within a regime. The magnitude of seasonal transitions also appeared to differ before and after the regime shift. We assessed the biological relevance of the identified regions using seasonal maps derived from remotely sensed chlorophyll-a concentrations ([chl-a]). We used Kruskal-Wallis and Wilcoxon rank sum tests to evaluate the correspondence between the [chl-a] maps and our pos! t-regime shift regions. There was a significant difference in ! [chl-a] among the regions in all seasons. We found that the number of regions with distinct chlorophyll signatures, and the associations between different regions, varied by season. The overall pattern of association between the regions was suggestive of observed, broad-scale patterns in the seasonal development and distribution of primary production in the North Pacific. This demonstrated that regions with different biological properties can be delineated using only physical variables. The flexibility of our approach will enable researchers to visualize the geographic extents of regions with similar physical conditions, providing insight into ocean dynamics and changes in marine ecosystems. It will also provide resource managers with a powerful tool for broad application in ecosystem-based management and conservation of marine resources.