Processes at the interface between biological and physical realms are fascinating.
All life depends on its environment, and we live in a time when the environment is changing.
I am undertaking a study that will reveal one such interface at a time when it is rapidly changing to reach a new equilibrium, the study of the effects of a changing sea ice regime on bio-optical indicators in Hudson Bay, Canada.
This study involves the use of both in situ and orbital data collection. The former is undertaken using a combination of above- and below water light sensors and direct measurements to determine the optical characteristics of the water and the corresponding bio-mass composition. This will allow me to choose an algorithm that best relates the actual biomass content of water to its spectral characteristics. Hudson Bay is an optimal area for this due to seasonal ice cover and spatially-stable polynyas—areas of continually open water. From the geographer’s perspective this provides for understanding of an area of interest, the polynya, and a seasonally ice-covered area as a control.
This direct research is the key to a larger-area analysis of the waters in northwest Hudson Bay. These measurements will be analyzed for primary productivity, biomass, non-organic matter, and suspended particles so that satellite measurements can be calibrated to the water’s character. This is complicated by the notoriously difficult ‘Case 2’ waters of Hudson Bay where the dominant optical properties are influenced by sediments and coloured dissolved organic matter rather than biomass.
The second stage of this research will use the two major types of satellite data—microwave and optical—to relate the spatial patterns of biomass and primary production in Hudson Bay to the character of the sea ice in an attempt to predict how changing sea ice will influence the basis of ecology. The microwave portion will use a new technique for understanding spatial and temporal patterns called hypertemporal remote sensing (Piwowar and LeDrew 1995). This is accomplished by combining dozens or hundreds of spatially registered images and temporally unmixing their spatial features into endmembers—a sort of synoptic climate of spatial patterns. This will produce highly useful, current spatial characteristics of sea ice in Hudson Bay.
The parts fall into place in the final analysis. A data fusion will be performed that combines the in situ ocean colour optical data with remotely sensed satellite imagery to discern patterns in primary production or biomass. This is then fused with the endmembers from the microwave data and statistically related to find whether ice presence or absence is a strong contributing factor for amount of biomass and the magnitude of this difference.
This research is important because the Canadian Arctic is not well understood as a region, while climate change presents a great risk to humans and the environment alike. The bottom of the food chain—the primary producers—can serve as an indicator for the health of the marine ecosystem and as such we need to know how they will be affected by a changing ice regime. I have no doubt there will be many challenges ahead, but I am confident that it is something that can be accomplished through hard work and the support network I have found here at the University of Manitoba.
Reference:
Piwowar, J., and LeDrew, E. (1995) Hypertemporal analysis of remotely sensed sea-ice data for climate change studies. Progress in Physical Geography 19 pp. 216-242
