Trophic ecology is a key component in describing patterns of variation between and within populations, particularly in Arctic marine systems wherein climate change is impacting food webs. This thesis investigates the trophic ecology of Arctic char (Salvelinus alpinus) in the Cumberland Sound region using a multi-indicator approach. My data show trophic niche differences between resident and anadromous ecotypes and evidence for estuarine feeding by residents. I document a shift in the marine diet of Arctic char from zooplankton to capelin (Mallotus villosus) – a novel prey species in this region – that has occurred in less than a decade. Changes in Arctic char growth imply population-level effects of this shift; however, more research is required. Finally, I find lipid effects on δ13C and lipid-extraction effects on δ15N and δ34S for Arctic char muscle tissue. Lipid-correction models did not provide adequate δ13C estimates; thus, chemical extraction or ecotype-specific validation of models is recommended.

Mercury concentrations in freshwater and marine biota are an ongoing concern, even in areas remote from local point sources, such as in the Canadian Arctic and sub-Arctic. Anadromous Arctic charr, which feed in the marine environment, have lower mercury concentrations than non-anadromous Arctic charr, which feed strictly in freshwater, but the two life-history forms have rarely been studied together, and the mechanisms driving the difference are unclear. Here, data from nine pairs of closely-located anadromous and non-anadromous Arctic charr populations were used to explore the impact of biological and life-history factors on individual total mercury concentration ([THg]) across a range of latitudes (49 – 81° N) in eastern Canada. From six of these sampling locations, additional samples of lower trophic level biota (i.e., algae, invertebrates, and forage fishes) were obtained in order to investigate patterns of total mercury (THg) and methylmercury (MeHg) biomagnification in the marine and lacustrine foodwebs supporting Arctic charr. Arctic charr mean [THg] ranged from 20 to 114 ng/g wet weight (ww) in anadromous populations, and was significantly higher in non-anadromous populations (all p < 0.01), ranging from 111 to 227 ng/g ww. Within-population variations in Arctic charr [THg] were best explained by fish age, and were also positively related to fork-length and δ15N-inferred trophic level. Across all sampling sites, the relationship between Arctic charr [THg] and fish age was significant and statistically similar in both life-history types, but only the non-anadromous fish demonstrated a significant relationship with trophic level. Fork-length and site latitude did not explain significant additional variation in Arctic charr [THg] across sampling locations. Trophic magnification factors were 1.98 – 5.19 for THg and 3.02 – 6.69 for MeHg in lacustrine foodwebs, and 1.59 – 2.82 for THg and 2.72 – 5.70 for MeHg in marine foodwebs, and did not differ significantly between the two feeding habitats for either THg or MeHg. The biomagnification rate of MeHg exceeded that of THg in both habitats. Mercury concentrations at the base of the foodweb were higher in the lacustrine environment (estimated at 17 – 139 ng/g dw for THg and 5 – 42 ng/g dw for MeHg) than in the marine environment (8 – 39 ng/g dw for THg and 1 – 11 ng/g dw for MeHg). The proportion of mercury in the methylated form was related to trophic level, and the relationship was statistically similar in the lacustrine and marine habitats. There was no effect of site latitude on mercury concentrations in marine or lacustrine biota, thus the difference between feeding habitats was consistent across a range of latitudes (56 – 72°N) in eastern Canada. We conclude that a difference in prey mercury concentration, driven by differential mercury concentrations at of the base of the lacustrine and marine foodwebs, is important for explaining the difference in mercury concentration between anadromous and non-anadromous in Arctic charr.

The Arctic region has not been a topic of discussion or a reason for a conflict in the past, but with the development of environmental issues which has resulted in melting of the ice, so has the interest for the region developed. This has resulted in a conflict over the area, with five countries as main actors. This paper investigates the five main actors’ goals with increased influence in the area, using the thoughts of Rational Actor Model, with states as utility-maximizing actors and foreign policy as the expression of national interest. Furthermore the paper investigates the means of getting to the goal, what kind of tools they use.The paper shows that all of the countries are interested in the economically benefits that are buried under the melting ice and the short cuts in sea routes that will be a possibility to use ones the ice melt more. It is also revealed that there is a security concern when the region is more accessible. The main tools that the countries use to get the influence that they wish are military and diplomacy means.

Boundaries between air masses, called frontal zones, have been associated with vegetation boundaries (Bryson, 1966; Hare and Ritchie, 1972). Using gridded climate reanalysis data, we analyze the air masses and frontal zones of North America. The position of the July Arctic front varies significantly through the period 1948-2007, with a mean position similar to that found by Bryson (1966). The variability of the front position can be associated with changes in the general circulation; when the AO and SOI are positive (negative), the position of the July Arctic front is further north (south). There is also more variability in the July Arctic front position in Eastern versus Western Canada. When the July Arctic front is north (south) of the mean position, the boreal forest and tundra vegetation is more (less) productive. There is some evidence that climate warming is only starting to shift the July Arctic front to the north. / This study was funded by the Natural Sciences and Engineering Research Council (NSERC) and the Polar Climate Stability Network (PCSN) project funded by the Canadian Foundation for Climate and Atmospheric Sciences (CFCAS).

Mercury (Hg) is a known neurotoxin that is often found in concentrations exceeding safe consumption guidelines in aquatic biota. This is evident in northern Canada, where northerners consume significant amounts of animals such as beluga, seals and burbot. In the Mackenzie River Basin, recent increases in Hg concentration in many of these animals over the past 25 years have been observed. The warming climate, and with it, the changing carbon cycle, are hypothesised in this thesis to play a role in the increases. Within the context of the two major zones (mountainous and peatland), with distinct geomorphology, hydrology and geology, traditional fossil fuel exploration methods (Rock-Eval pyrolysis, organic petrography) have been employed in a novel manner on recent sediments to qualify and quantify the OM and several geochemical analyses have been used to determine the geochemical sources of Hg. The mountainous zone is composed mostly of refractory OM, from forest fire char and heavily reworked OM. It also contains, and fluxes, most of the Hg, which derives from oxidative weathering and erosion of widespread sulfide minerals. However, Hg from this zone is in chemical forms of limited bioavailability. The peatland zone has a greater proportion of labile OM, with higher concentrations of DOC and algal-derived OM. Lake-fed tributaries in this zone contain even higher proportions of labile OM. At one of these sites, the sediment core record shows that Hg has been increasingly associated with labile OM over time, due to increasing primary productivity accelerated by climate change, and is resulting in an increase in scavenged Hg. The temporal trend in algal-bound Hg in the sediment record matches very well with the temporal trend of Hg in burbot sampled from the area, providing one of the first and strongest lines of evidence for the climatic impact on Hg bioaccumulation in Arctic ecosystems.

An understanding of the adaptability of belugas (Delphinapterus leucas) to changing ice-conditions is required to interpret and predict possible changes in habitat selection in response to projected loss of sea ice throughout the circumpolar Arctic. Beluga spring distribution in the eastern Beaufort Sea was described by analyzing observations from aerial surveys conducted from 1975 to 1979. Repeated surveys along the Franklin Bay fast-ice edge in June 2008 were used to study the distribution and behaviour of belugas and bowheads. Despite inter-annual variability in ice extent, belugas consistently selected areas with water depths of 200-500 m, heavy ice concentrations (8/10 to 10/10) and seafloor slope ≥0.5 degrees in spring 1975 to 1979. While predator avoidance may partially explain the observed distribution, foraging success likely has more influence on beluga habitat selection in the spring. In ice-covered offshore regions, belugas may be engaged in under-ice and deep water foraging on Arctic cod (Boreogadus saida). In lighter ice years, belugas may expand their distribution and shift shoreward to take advantage of high prey densities along fast-ice edges. Both belugas and bowheads appeared to be feeding along the Franklin Bay ice edge in June 2008. More research is required to examine and compare possible changes in distribution since the late 1970s and to investigate the factors driving the patterns described.

The veracity of a Radiometric Microwave Profiling Radiometer (MWRP) while mounted onboard a ship in the Arctic marine environment was assessed. The MWRP was validated against radiosonde data by calculating the RMS and bias for simultaneous measurements taken for temperature and absolute humidity profiles. The vertical resolution of the MWRP was calculated using the inter-level covariance method. Based on the comparisons, the MWRP provided reliable measurements of both temperature and absolute humidity while mounted on the CCGS Amundsen. Satellites CloudSat and Calipso were assessed over the Arctic marine cryosphere. Temperature and absolute humidity from the ECMWF-aux data product was compared with profiles from the ship based MWRP. The cloud base heights measured by the ceilometer and MWRP were compared to CloudSat and Calipso's GeoProf-lidar. Due to a large number of possible false detections, the constraints used by the GeoProf-lidar data product for cloud detection may need to be further refined.

This dissertation presents new research into the study of simulation and measurement techniques for microwave remote sensing of sea ice. We have embarked on a major study of the microwave propagation and scattering properties of sea ice in an attempt to link the physics of the sea ice medium to experimentally obtained concomitant scatterometer measurements. During our fieldwork, we studied the polarimetric backscattering response of sea ice, focusing on newly-formed sea ice under a large assortment of surface coverage. Polarimetric backscattering results and physical data for 40 stations during the fall freeze-up of 2003, 2006, and 2007 are presented. Analysis of the co-polarization correlation coefficient showed its sensitivity to sea ice thickness and surface coverage and resulted in a statistically significant separation of ice thickness into two regimes: ice less than 6 cm thick and ice greater than 8 cm thick. A case study quantified the backscatter of snow-infiltrated frost fl owers on new sea ice, showing that the presence of the frost flowers enhanced the backscatter by more than 6 dB. In our simulation work, an efficient method for simulating scattering from objects in multi-layered media was incorporated into a scattered-field formulation of the FVTD method. A total-field 1D-FDTD solution to the plane-wave propagation through multi-layered meda was used as a source. The method was validated for a TE-polarized incident-field through comparisons with other numerical techniques involving examples of scattering from canonically-shaped objects. Methods for homogenization of inhomogeneous media were developed and validated using well-known dielectric mixture models. A Monte Carlo Method for simulating scattering from statistically rough surfaces was developed and was validated through favorable comparison with the SPM method for rough surface scattering. Finally, we presented a new Monte Carlo Method for simulating sea ice remote sensing that utilized the framework of the FVTD method for scattering simulations. The modeling process was driven by actual physical measurements of sea ice, wherein dielectric and physics-based modeling techniques were employed. The method was demonstrated through a series of case studies where the scattering from newly-formed sea ice was simulated using a TE-polarized incident- eld. Good agreement between experimental scatterometer measurements and simulated results was obtained for co-polarized returns, whereas cross-polarized results indicated that more depolarizing features must be taken into account.

Melt ponds greatly increase the transmission of solar radiation through sea ice relative to snow covered or bare ice. This rise in transmittance has the potential to enhance water column heating and primary production. I examine how spatially variable sea ice surfaces control the under-ice salinity, temperature and photosynthetically active radiation (PAR) and provide estimates of solar heating and primary production during melt. Conductivity, temperature and PAR profiles were measured in the Canadian Arctic under snow covered ice, leads, bare ice and melt ponds. The under-ice light field to a depth of 10 to 13 m was highly variable, controlled by increased transmission under melt ponds and shading by bare ice. Below, the light field became relatively homogeneous showing the depth the surface heterogeneity had an effect on transmitted PAR. Furthermore, one water column profile is not representative of the PAR, salinity or temperature under a spatially heterogeneous surface.

During the fourth International Polar Year, an interdisciplinary study was conducted to examine the couplings between sea ice, ocean, atmosphere, and ecosystem in the southeastern Beaufort Sea. This thesis examines components of the system that control the air–sea exchange of carbon dioxide. Using eddy covariance measurements, we found enhanced CO2 exchange associated with new ice formation in winter flaw leads. This exchange was typically directed towards the surface, although we also measured one instance of outgassing. Sea surface dissolved CO2 measurements (pCO2sw) in Amundsen Gulf showed significant undersaturation with respect to the atmosphere at freeze–up, followed by a slow increase over the winter until spring phytoplankton blooms caused strong undersaturation at break–up. Over the summer, pCO2sw increased until becoming slightly supersaturated due to surface warming. Along the southern margins of Amundsen Gulf and on the Mackenzie Shelf we found pCO2sw supersaturations in the fall due to wind–driven coastal upwelling. In the spring, this upwelling occurred along the landfast ice edges of Amundsen Gulf. By combining observations of enhanced winter gas exchange with observations of pCO2sw in Amundsen Gulf, we derived an annual budget of air–sea CO2 exchange for the region. This exercise showed that uptake through the winter season was as important as the open water season, making the overall annual uptake of CO2 about double what had previously been calculated. Prior to this work, the prevailing paradigm of air–sea CO2 cycling in Arctic polynya regions posited that strong CO2 absorption occurs in the open water seasons, and that a potential outgassing during the winter is inhibited by the sea ice cover. As a new paradigm, we propose that the spatial and temporal variability of many processes – including phytoplankton blooms, sea surface temperature and salinity changes, upwelling, river input, continental shelf processes, and the potential for high rates of winter gas exchange – need to be considered in order to understand the carbon source/sink status of a given Arctic polynya region. A paradigm that considers such varied processes is useful in understanding how climate change in the Arctic can impact air–sea CO2 exchange.