Ancient permafrost and a future, warmer Arctic
An understanding of the history of permafrost, and in particular the impacts of past climate on terrestrial permafrost, provides an important analogue for future changes. Numerical models predict its degradation over extensive areas of the Arctic and Subarctic in the near future, with potential for concomitant release of large volumes of stored carbon. In a general sense, little is known of the past record of permafrost from which to evaluate these models. Here we present recent research on the history of permafrost in unglaciated Yukon and Alaska highlighting the role cryostratigraphy. This region provides a unique archive of paleoenvironmental change by virtue of the numerous distal tephra beds interbedded with long, rich sedimentary records and their association with relict ground ice. The earliest appearance of permafrost is marked by well-developed frost cracks in the early Pliocene of northern Yukon, and well-developed ice wedge casts by the middle Pliocene in central Yukon and Alaska. Permafrost was ephemeral through the middle Pliocene and early Pleistocene, and likely absent during interglacials prior to 780,000 years ago. The oldest relict ground ice is associated with ice wedges overlain by the ca. 740,000 yr old Gold Run tephra in the discontinuous permafrost zone of central Yukon. Surprisingly these ice wedges are within several metres of the surface, demonstrating the survival of permafrost through several interglaciations. Additional sites associated with the last interglacial (ca. 120,000 years ago) indicate that widespread ground thaw and thermokarst development occurred with past warming. However, thaw was limited to the uppermost several metres of permafrost, and relict ice wedges that pre-date the last interglaciation are present at several sites separated by over 700 km. The relict ice wedges indicate that the antiquity and resilience of discontinuous permafrost is regional in nature. However, the ubiquity and magnitude of last interglacial thermokarst suggests that terrain effects associated with current permafrost degradation foreshadow more widespread and severe thaw under even modest future warming scenarios.
Detecting changes in permafrost and attributing them to the changes in physical and biological parameters
Recent observations indicate a warming of permafrost in many northern and mountain regions with a resulting degradation of ice-rich and carbon-rich permafrost. Permafrost temperature has increased by 1 to 2°C in northern Russia during the last 30 to 35 years. This observed increase is very similar to what has been observed in Alaska where the detailed characteristic of the warming varies between locations, but is typically from 0.5 to 2°C. In the last 30-years, warming in permafrost temperatures observed in the Russian North and Alaska has resulted in the thawing of natural, undisturbed permafrost in areas close to the southern boundary of the permafrost zone. Most of these changes can be attributed to the recent changes in climate observed in high northern latitudes, and mainly to the increase in air temperature and changes in snow thickness and seasonality. The largest response in permafrost temperatures was observed during the time periods when both of these climatic parameters experienced positive trends. It is much more difficult to correlate changes in permafrost with changes in hydrology or with changes in biological parameters. The major problem in doing this relates to generally slow rates of changes in vegetation and the very limited duration of available permafrost temperature records. In this presentation we will use "space-for-time" substitute approach to show how different types of ecosystems affect the stability of permafrost. Another possibility is to investigate the impact of forest fires on permafrost stability. Disturbances related to forest fires significantly increase the probability of permafrost degradation at the present and in the near future. Severe forest fires usually destroy the surface organic layer, exposing the underlying mineral soil to the surface. This can reset the mean annual temperatures at the bottom of the active layer to a level above 0°C. In this case, permafrost will start to thaw and if permafrost is ice-rich, the thermokarst processes will take off, significantly affecting ecosystems. Both observations and modeling results will be used in this presentation to demonstrate the impact of fire disturbances on the stability of permafrost.
Assessing the Influence of Melting Permafrost on Streamflow in Discontinuous Permafrost Catchments
There is a growing body of literature exploring the role permafrost plays in hydrological and biogeochemical cycling, and how recent changes in permafrost has affected these cycles. For example, over the last 40 years, the Yukon River has seen changes in both the timing, magnitude and quality of streamflow, which are attributed to permafrost melt. As near-surface permafrost degrades, deeper subsurface flow pathways are activated, and the nature of biogeochemical reactions change. However, large-scale studies do not explicitly link streamflow response to direct observations of catchment processes. This talk will explore the role of frozen ground on water and biogeochemical cycling in discontinuous permafrost alpine headwater catchments of the Yukon River. Results from the Wolf Creek Research Basin near Whitehorse, Yukon, will be utilized to show that where permafrost is present, it enhances runoff due to restricted drainage, which also limits dissolution and increases DOC flux due to the ubiquitous nature of surface organic soils. However, in discontinuous permafrost environments, there is considerable intra- and sub-permafrost groundwater that supplies ion-rich waters to streams throughout the year. By using current differences in permafrost disposition as an analogy for future changes, it is possible to hypothesize what hydrometric and/or biogeochemical indicators we expect to respond first. Furthermore, a decade of research suggests that large inter-annual variability and water and chemical fluxes will make change detection from infrequent synoptic samplings schemes challenging.
Transformation of Upland Water and Carbon Dynamics by Thawing Permafrost in the Alaskan Interior
Large arctic rivers can provide an integrated signal of regional permafrost thaw and associated carbon dynamics. A long-term (30-y) decrease in dissolved organic carbon (DOC) and increase in dissolved inorganic carbon in the Yukon River Basin (YRB) suggest increased flow through mineral soils as a result of permafrost thaw. We used U series isotopes to test for the influence of thaw on soil and surface waters in small upland catchments at two sites within the YRB. In natural waters, 234U/238U activity ratios exceed 1.00 (secular equilibrium) as a function of water-rock contact time. Previous work has shown that in major YRB rivers, seasonally and spatially variable 234U/238U ratios could indicate both groundwater inputs and permafrost thaw, with ratios ranging from 1.1 to 2.6. We show that 234U/238U ratios in soil and surface water from these small catchments span the range of values observed in the major rivers, and indicate greater influence of older water where the mineral soil and underlying sediment facilitate drainage and permafrost degradation. Analysis of deep, ice-rich loess permafrost cores (2-10 m) reveals that thaw of Pleistocene ice can release high concentrations of DOC (>1000 ppm) and ammonium in thaw waters. The age and chemical composition of these waters allows for improved prediction of downstream carbon dynamics upon thaw. Field observation of hillslope soil sequences indicates that both topography and mineral substrate influence the effects of thaw on water and carbon dynamics in small catchments.
Climate Drivers and Ecological Triggers of Permafrost Formation and Thaw in Boreal Permafrost Peatlands
Discontinuous permafrost in boreal peatlands is strongly controlled by interactions among climate, hydrology, plant community succession, fire, and Picea mariana (black spruce) establishment. I synthesized data from several experimental, descriptive, and paleoecological studies from northern Manitoba, Canada, to determine how climate change and ecological processes control permafrost formation and thaw and the mass balance of permafrost in this region. In the modern environment, permafrost formation in climatically suitable zones (MAAT < 0°C) is triggered by P. mariana recruitment and subsequent impacts of the forest canopy on snow density. Experimental analyses of seed rain, seed germination, and seedling survival in thawed peatlands indicate that P. mariana recruitment is extremely poor, suggesting that spruce establishment is a major ecological constraint on the rate of new permafrost formation. Meanwhile, thaw rates in modern environments are accelerating as a result of significant temperature increases over the past half century. Over longer time scales, climate has been important for controlling peatland succession, fire, carbon accumulation, and permafrost. During the Holocene Thermal Maximum (~~6000-4000 BP), warmer conditions drove plant communities from wetter fens to drier forested bogs, resulting in greater fire severity and slower rates of carbon accumulation. Decreasing temperatures and increasing moisture from 2000-1000 BP increased the occurrence of wetter poor fen communities but with no apparent onset of permafrost. The Medieval Warm Period (~~1150-650 BP) appeared to drive succession back to drier, forested bog communities, setting the stage for modern permafrost development with the onset of cooler temperatures during the Little Ice Age (~~650-150 BP). Although succession to forested bogs increased fire severity over much of the Holocene, permafrost over the past 700 years has shown a remarkable resistance to thaw following severe fires, possibly as a result of cooler LIA temperatures stabilizing frozen soils. These results indicate the combined importance of climate change, peatland succession, fire, and spruce recruitment on permafrost dynamics. In the modern environment, rapid thaw combined with poor spruce recruitment suggest that the mass balance of permafrost in boreal peatlands is currently negative. Warmer conditions in the future may decrease further the stability of frozen soils and push peatland landscapes to drier bog communities with a greater likelihood of fire and thaw, thereby amplifying the loss of permafrost in these regions.
The influence of seasonal thaw and water table dynamics on soil carbon and trace gas flux in an ecosystem gradient from Interior Alaska
Variations in depth to permafrost, thickness of organic soil layers, vegetation structure, and water table dynamics form the basis for an ecosystem gradient within the Bonanza Creek Long-term Experimental Research station in Alaska. Vegetation at the five stations along the gradient is dominated by black spruce, tall shrub, tussock grass, Equisetum, and brown moss species, respectively. We hypothesize that seasonal variations in carbon dioxide and methane efflux are governed to a large extent by exposure of thawing soil to aerobic or submerged conditions over the growing season. Based on field measures, these systems varied widely from predominantly aerobic conditions (black spruce) to mixed aerobic and submerged conditions (tussock grass) to predominantly submerged conditions (moss). As a result, trace gas efflux patterns were also variable and were dominated by carbon dioxide in the black spruce community, a mixture of trace gases in the tussock grass community, and methane efflux in the brown moss community.
Influences of Glacial and Permafrost Thaw on Dissolved Organic Carbon-14 in the Yukon River Basin
The response of frozen soils to climate warming is of particular significance for understanding long term climate effects on global carbon cycling and carbon export by high latitude rivers. A critical question in carbon cycling is how climate change could alter the fate and chemical nature of dissolved organic carbon (DOC) released from high latitude watersheds, particularly those influenced by permafrost and rapidly melting glaciers. It is generally assumed that with warming conditions, the age of DOC in high latitude rivers will increase due to increased metabolism of older carbon stocks associated with soils and peats, while the role of melting glaciers on DOC age is currently not known. Here we present data on DOC concentration, composition and natural 14C abundance from the Yukon River, 16 tributaries of the Yukon River, glacial meltwaters, groundwater and soil water draining to the Yukon River. We use these data to determine the effects of older carbon sources on DOC in this system. Our results indicate that the transport of DOC in the river and its major tributaries is seasonally dependent with large variations in chemical nature. The oldest DOC ages were measured in headwater streams and tributaries of the Yukon River strongly influenced by glacial meltwater and groundwater discharge. In contrast, DOC from watersheds dominated by peat soils and underlain by permafrost was enriched in 14C relative to modern atmospheric values, suggesting that the contribution of old carbon from these soils to the DOC exported from these watersheds is minor compared the release of modern carbon. DOC exported by the Yukon River at Pilot Station was youngest during the spring flush period (May 1 to June 30), which accounted for approximately 52% of the annual DOC load over a 5 year period (2002-2005). With decreased discharge, older DOC is transported during the summer-fall period (July 1 to Oct 31) followed by the oldest DOC in winter (Nov 1 to Apr 30), presumably transported by groundwater. The export of older C, especially during the summer-fall period may be the result of increased metabolism of soil or peat derived organic matter and/or the influence of greater glacial contributions in summer when glacial melting is greatest. Our data indicate that the mobilization of ancient DOC from rapidly melting glaciers could be an important source of pre-anthropogenic reduced C in the Yukon River in coming decades.