<p> Periodic temperature measurements in the DOI/GTN-P Deep Borehole Array on the western Arctic Slope of Alaska have shown a strong near-surface permafrost warming over the last 40 years, particularly since &sim; 1990. Due to the manner in which these deep wells were drilled, the portion of the observed permafrost warming caused by climate change has remained unclear. Other factors that have strongly influenced temperatures near the wellbores include the heat deposited into permafrost during drilling and local-landscape changes associated with drilling operations (creation of reserve pits and drill pads). Multidimensional heat-transfer models capable of assessing the magnitude of the drilling and local-landscape disturbances near the wellbores have not been available. For the western Arctic Slope, such models must be capable of simulating heat-transfer processes in layered fine-grained mudrocks whose thermal properties are highly nonlinear due to the occurrence of unfrozen water at temperatures well below 0&deg;C. An assessment of the drilling and landscape-change effects also requires knowledge of the specific thermophysical properties occurring at the well sites. Little information has been available about these properties on the western Arctic Slope.</p><p> To establish the portion of the observed permafrost warming related to drilling and landscape-change effects, multidimensional (2-D cylindrical, 3-D cartesian) numerical heat-transfer models were created that simulate heat flow in layered heterogenous materials surrounding a wellbore, phase changes, and the unfrozen water properties of a wide range of fine-grained sediments. Using these models in conjunction with the borehole temperature measurements, the mean thermophysical properties of permafrost rock units on the western Arctic Slope were determined using an optimization process. Incorporation of local meteorological information into the optimization allows a more refined estimate of the thermal properties to be determined at a well site. Applying this methodology to the East Simpson #1 well on the Beaufort Sea coast (70&deg;55.046'N, 154&deg;37.286'W), the freezing point of permafrost is found to be &ndash;1.05&deg;C at this site and thermal diffusivities range 0.22&ndash;0.40 &times; 10^&ndash;6 m² s^&ndash;1. Accounting for the drilling and landscape-change effects, tundra adjacent to East Simpson is found to have warmed 5.1 K since the mid-1880s. Of this, 3.1 K (60%) of the warming has occurred since 1970.</p>

Two lakes were studied in detail for rock magnetic properties: Lake El'gygytgyn, a crater lake formed 3.6Ma in the Far Eastern Russian Arctic, and Heimerdalsvatnet, a Holocene coastal lake located in the Lofotens off the coast of northern Norway. These two lakes have vastly different environmental histories, the former a terrestrial lake formed from a meteor impact and never covered by continental ice sheets whereas the latter went from a coastal marine setting to a completely lacustrine environment due to isostatic rebound and sea level fluctuations. Their differences are considerable, however they provide the opportunity to compare Arctic lake systems to discern similarities and differences in their magnetic properties for application to future climatic investigations. Paleomagnetic measurements and down-core magnetic susceptibility were performed at the GFZ German Research Centre for Geosciences in Potsdam for Lake El'gygytgyn and at the Laboratoire de paleomagnétisme sédimentaire at ISMER for Heimerdalsvatnet. Rock magnetic properties were measured at the University of Massachusetts Amherst, Institute of Rock Magnetism, and/or Trinity College. These measurements included: magnetic susceptibility, hysteresis parameters, Curie temperatures, and low-temperature magnetic behavior. Imaging of magnetite grains was also performed. Magnetic susceptibility measurements in Lake El'gygytgyn suggested a correlation between glacials (interglacials) and low (high) susceptibility. The large range in susceptibility indicated there could be magnetite dissolution. The first study supported this hypothesis with evidence at low temperatures (10-35K) of minerals such as siderite, rhodochrosite, and/or vivianite which could form from iron released during dissolution. Marine Isotope Stage 31 was investigated for rock magnetic properties that could continue to support or oppose findings from the first study. It was determined the presence of siderite only occurred in interglacial periods whereas its absence (and probably presence of vivianite) related to glacial periods, indicating more reduced environments during glacials versus interglacials. Heimerdalsvatnet paleomagnetic data from the marine environment (lower part of the core) revealed scattered directions whereas data from the upper part of the core (lacustrine environment) showed better consistency. Rock magnetic measurements showed some variation downcore, however the measurements are not dependable since the amount of paramagnetic material was overwhelming compared to any ferromagnetic mineral present.