H53D-1450
Geographic Analysis of Alaska Lake Districts
The state of Alaska has over 400,000 lakes greater than 0.01 km2 in surface area covering approximately 3.3% of the landscape. As in most lake-rich regions, these lakes are unevenly distributed on the landscape. So in order to better understand how lakes are organized on the landscape and relate this geographic organization to other climatologic, geologic, and biogeographic characteristics, we analyzed the spatial distribution of Alaska lakes. Using a combination of numerical abundance and surface-area extent of lakes, we selected lake density thresholds to identify and delineate 22 lake districts in Alaska. The total area of these 22 lakes districts occupy 16% of Alaska, yet encompass 64% of lakes and 76% of lake surface-area. The three largest lake districts are associated with the Yukon-Kuskokwin Delta, the Northern Arctic Coastal Plain, and the mountain front of the Alaskan Range on the Alaska Peninsula. Interestingly, these largest lake districts are covered by >17% lakes, while most of the smaller lake districts we identified have <10% lake cover. Of the remaining smaller lake districts, 9 are associated with mountain fronts or intermountain basins, 4 are associated with coastal plains, 3 are associated with floodplains and deltas, and 3 occur in high-elevation or mountain terrain. The highest numerical lake densities occur at deltas, while relatively lower densities occur in mountainous areas where individual lakes are often larger in surface area and likely volume. Comparison of these lake districts were made to permafrost distribution, glacial history, lithology, watershed position, and regional hydrologic budgets and regimes to better understand where lake-rich regions occur, why, and how they might change in the future. Ten of the 22 lake districts occur in areas dominated by continuous permafrost, 6 occur in areas of discontinuous or sporadic permafrost, and the other 6 occur in regions without perennially frozen soils. The majority of lake districts, 63% occur in regions that were glaciated during the last glacial maximum, yet several of the largest lake districts occur in unglaciated terrain. We also made comparisons of lake districts to other natural units used for landscape analysis including ecoregions and watersheds, and human- delineated units including National Parks (NP) and National Wildlife Refuges (NWR). Not surprisingly, 12 of these lake districts occur partly or wholly within NWRs, and all but two are associated with other state or federally managed parks or wildlife refuges. Analysis of lake districts in Alaska, or other regions on Earth, may prove to be useful for better understanding lake change, aquatic and terrestrial habitats, and regional hydrologic budgets and regimes. Such fresh views of the landscape may become increasingly important for improving natural resource science and management in Alaska, where future climate change is predicted to be very rapid.
H53D-1451
Potentially Non-Reversible Changes in Biogeochemical Cycling of an Alpine Wetland in Responses to Changes in Climate, Green Lakes Valley, Colorado
Alpine wetlands have been shown to be among the most sensitive types of wetlands to changes in climate. Yet, little is known about the hydrology of alpine wetlands and how the biogeochemical cycling of these wetlands may respond to changes in climate. Here we report on the results of surface and subsurface water samples collected weekly from May to October from 2003-2007, from a 2-ha wetland located at an elevation of 3593m in Green Lakes Valley (GLV). These results are compared to historical samples collected from 1986-1990. GLV is within the city of Boulder Watershed and part of the Niwot Ridge LTER. Mean annual air temperatures were about +1C higher in the 2000's compared to the late 1980's. Samples were analyzed for all major solutes, dissolved organic carbon (DOC), and stable water isotopes. Geochemical weathering products during baseflow (e.g. Ca++, Mg++, Na+, SO4-)are ten times higher since the late-1980's, with little change during June and July. Nitrate concentrations during baseflow have also doubled over the same time period. For 2003-2007, nitrate was retained by the wetland with retention as much as 99% during baseflow. In contrast, DOC is produced within the wetland, and concentrations in the outflow ( ~ 1.0 mg/L) generally higher than the inflow ( ~ 0.5 mg/L). DOC concentrations in subsurface wells were an order of magnitude greater than in surface waters, suggesting that biogeochemical cycling played an important role in the retention and transformation of the chemistry of inflowing waters before export to down gradient ecosystems. We evaluated potential hydrologic controls by determining the relationships between water chemistry, outlet discharge, residence time, source waters, and flow paths of the wetland. Outflow discharge peaked at 8.0 L/s on June 17 consistent with a snowmelt-dominated source waters. However, results from a constant injection LiBr tracer yielded a residence time of ¬ ~ 35 hours, suggesting a significant amount of hydrologic storage within the wetland. Seasonal Δ18O values range from -20 ‰ to -9 ‰, suggesting changing source waters and flow paths. We will use end member mixing analysis (EMMA) and mixing models parameterized with stable isotopes and biogeochemical tracers to evaluate changing source waters and flow paths during years 2003 to 2007. Changes in climate may in turn have caused changes in the hydrology that have resulted in potential irreversible effects on the biogeochemical cycling of this alpine wetland.
H53D-1452
Surface Elevation, Carbon Sequestration Potential and Rising sea Levels in Estuarine Wetlands
Estuarine wetlands are among the most productive ecosystems on Earth, providing habitat for commercially important fish species and migratory shorebirds, serving as nurseries for many other marine organisms and supporting the productivity of adjacent coastal waters. Typically, these wetlands are driven by tidal hydrodynamics and are net sinks for sediment and soil carbon. Their distribution in the tidal frame depends on a delicate balance between topographic gradient, the rate of vertical soil development, and the rate of sea level change. The complex interactions between hydrodynamics, ecology and soil processes that govern this balance produce positive feedbacks and system self-organization. As complex systems, these wetlands demonstrate resilience under a wide range of conditions but they have been observed to collapse or move to another equilibrium state above certain thresholds. Research at a wetland in the Hunter estuary, southeast Australia has tracked changes in relative sea level and surface elevation in mangrove and saltmarsh wetlands over a five year period (2002- 2006) and soil carbon over a two year period (2005-2006). Mangrove surface elevation was strongly correlated with relative sea level (R2=0.715, p=0.004) but there was no correlation between relative sea level and saltmarsh surface elevation (R2=0.093, p=0.424). Soil carbon levels were high in both vegetation types (% loss on ignition of 16.2% and 18.8% for mangrove and saltmarsh soils, respectively) and not significantly different (ANOVA F=1.36, p=0.270). A 16% increase in soil carbon was recorded in each vegetation type over the period 2005-2006. Mean annual sea level rose by 55 mm and net annual precipitation (rainfall minus evaporation) fell by 189 mm over the same period. The ability of mangrove to respond rapidly to changes in relative sea level and the indicative positive trend between soil carbon and relative sea level suggest that this wetland type is both resilient to future sea level rise and has the potential to sequester carbon. Saltmarsh exhibited a similar potential for carbon sequestration, but low resilience to rising sea level, particularly in areas with steep or urbanised landward topography. Incorporation of these findings into general models of wetland hydrodynamics will inform strategies for adaptive management of estuarine wetlands in response to future climate change.
H53D-1453
Sensitivity of Carbon Inventories to Natural Climate Cycles
Terrestrial carbon inventories and fluxes carry memories of climate variations and other perturbations for time scales of seasons to centuries. We have developed a simple multi-box model of the terrestrial carbon cycle forced by cyclic climate perturbations on net primary production (NPP) and residence times for each carbon pool. The model is used to explore the variability of carbon fluxes under "natural" conditions which serve as background variability to any changes induced by global warming. Preliminary model results suggest that a lower biospheric carbon inventory for an oscillatory climate (as in a drought/pluvial cycle) than for a constant climate. No secular trend in productivity is necessary to cause this loss, and the effect (increased atmospheric CO2) could create a positive feedback by leading to further drought/pluvial cycles (e.g. increased hydrologic cycle intensity associated with global warming). We will present results from sensitivity calculations that explore the spectrum of carbon inventory variations for different climatic periodicities.