Surface Water Dynamics Using MODIS Data
Accurate characterization of the location and areal extent of inland water bodies represents an important parameter for a variety of scientific studies, ranging from landscape-scale studies of surface hydrology to continental and global-scale studies of land surface-atmosphere interactions. A highly distinctive feature of the sub-arctic and arctic regions in High Northern Latitudes (HNL) is the presence of large numbers of small lakes and ponds especially in peatlands. There is evidence from a number of case studies showing that ponds and lakes in the Arctic are undergoing substantial changes. Additionally arid and semi-arid areas can experience rapid, dramatic changes in the condition of surface water. For example the Aral Sea has lost approximately 40% of its surface area in just the last 8 years. Such dramatic changes in surface area of water can have significant impacts for the ecology of the region, for the carbon cycle and locally for energy balance and local climates. We demonstrate use of the MODIS data record, which is approaching 10 years in length, to generate a 250 meter resolution annual time series of lake occurrence and area for the whole of the global HNL region and for arid and semi-arid regions of the world. The increased temporal frequency and broad spatial coverage of MODIS provide a unique opportunity to generate a global, synoptic product to depict changes in lake cover for the past decade. Preliminary results show dramatic changes in lake area and abundance both locally and regionally. The observed changes in lakes are significant ecologically, for the carbon cycle in this area, for local energy balances as well as the hydrological cycle.
Recent drought-induced mortality of aspen forests along a water-balance tipping point for ecosystems in western Canada
In western Canada, the boundary between boreal forest and prairie grasslands marks a dramatic change in nearly all aspects of ecosystem functioning. These include a steep spatial gradient in hydrological characteristics of the landscape (lake level variability, water runoff and stream flow patterns) that coincides with the southern range limit of peatlands and several species of boreal conifers. Previous studies indicate that the forest-grassland boundary in this region represents a critical "tipping point" (Lenton et al. 2008) where long-term water input by precipitation is barely sufficient to satisfy the water use demands of productive, closed-canopy forests. This concept is consistent with the observed, regional gradient in the character of forests dominated by aspen (Populus tremuloides), the most abundant and widespread deciduous tree in North America. Aspen-dominated forests are productive and continuous in the boreal zone, but are stunted and patchy in the boreal-grassland transition zone, often referred to as the aspen parkland. Based on the "tipping point" concept, there are concerns that aspen forests in this region are especially sensitive to the projected trend toward warmer and drier conditions under human-induced climate change. In response to these concerns, a large-scale study was established across west-central Canada in 2000, entitled "Climate Impacts on Productivity and Health of Aspen" (CIPHA). The study has hierarchical sampling design that is aimed at "scaling up" forest-climate responses from individual trees to the region. During 2001-2002, the region was affected by an exceptionally severe drought that subsequently led to massive dieback and mortality of aspen forests within the boreal-grassland transition zone. Drought severity and extent was quantified using a simple climate moisture index (CMI), and drought impacts were quantified using tree-ring analysis, in combination with plot-based and remotely-sensed measures. Results showed that stand-level productivity, dieback and mortality were governed primarily by moisture variation. Furthermore, during and following this drought there was increasing damage by wood-boring insects and elevated, regional-scale mortality of aspen over at least 6 years (2002-2008). Although it is premature to attribute these impacts to anthropogenic climate change, they provide an excellent analog for what may be expected in future, even under a modest trend toward drying over the next few decades. Furthermore, the recent aspen mortality in western Canada shares many features common to other recent episodes of drought-induced forest mortality that have been documented on all of the earth's forested continents. This suggests the need for an integrated, global research and monitoring system that would enable early detection and attribution of large-scale ecosystem changes, especially in climatically-sensitive regions along forest-grassland boundaries around the world.
Drought as a Disturbance: Implications for Peatland Carbon Budgets in the Hudson Bay Lowland
Climatic and Vegetation Controls on Peatland CO2 Fluxes in Alaska: Early Response to Ecosystem-Scale Drought and Soil Warming Manipulations
Peatlands store 30% of the world's terrestrial soil carbon (C) and are located primarily at northern latitudes, where they are expected to experience severe climate warming. We monitored growing season carbon dioxide (CO2) fluxes across a factorial design of in situ water table drawdown (i.e., drought) and soil warming treatments for 2 years in a rich fen located just outside the Bonanza Creek Experimental Forest in interior Alaska. We hypothesized that the sensitivity of microbial activity and decomposition rates to soil environmental change would lead to large changes in ecosystem respiration (ER) and net ecosystem exchange (NEE) across our experimental treatments. Our results showed that the lowered water table treatment did not alter ER of CO2, but did lower gross primary production (GPP) and increase NEE relative to the control treatment, making this plot more of an atmospheric source relative to the control. Surface soil warming increased ER and GPP by 16% compared to un-warmed plots, with no evidence of water table x soil warming interactions. Nonlinear modeling showed that both hydroclimate (water table position, soil temperature) and vegetation (seasonality of LAI) were important controls on of CO2 fluxes. Overall, our initial results suggest that drought will impact CO2 fluxes in northern rich fens by reducing ecosystem C storage due to plant stress.
Climate Change Impacts on Soil Organic Matter: New Insights from Molecular-Level Studies
Natural organic matter is ubiquitously found in the environment and plays a critical role in several biogeochemical processes such as the regulation of atmospheric CO2, agricultural sustainability, and the fate and transport of problematic organic chemicals in the environment. Organic matter preserved within the sedimentary record also holds key information about early life on earth, insights into past climates, and the presence of specific organic matter structures is often used in the search for life on other planets. Despite the importance of natural organic matter in several disciplines, the vast majority of organic matter remains "molecularly uncharacterized" (Hedges et al. 2000, Org. Geochem. 31:945-958). The lack of organic matter structural information is mostly due to the complex nature and uniqueness of organic matter but also due to the lack of sophisticated analytical strategies designed specifically for the study of organic matter structure and environmental reactivity. Organic matter is a collection of compounds from various plant, microbial, and anthropogenic sources, all at various stages of oxidation (decomposition) and represents the most naturally occurring complex mixture on earth. Organic geochemists have long used biomarker methods to study the sources, structures, and stage of organic matter oxidation however biomarker methods only extract and measure a small fraction of the total organic matter composition. Nuclear magnetic resonance (NMR), namely solid-state 13C NMR spectroscopy, has been used extensively to study organic matter structure but suffers from poor spectral resolution due to organic matter heterogeneity and strong dipolar coupling in solids. This presentation will highlight the development of molecular-level analytical methods for organic matter and demonstrate their utility in studying soil organic matter biogeochemistry with global warming. The use of biomarker methods with conventional and innovative NMR methods in tandem provides an unprecedented insight into the dynamics of organic matter in the environment. Examples from molecular-level changes in soil organic matter composition after in situ soil warming experiments will be shown that enable an accurate, long- term projection to be made about the response of soil organic matter to global warming.
The Holocene Thermal Maximum as a Time of Rapid Peat Accumulation and Peatland Expansion in Alaska
High latitudes are particularly sensitive to climate warming resulting from a number of important positive feedbacks, including increasing albedo from changing sea ice extent, snow and vegetation cover, and feedbacks to the carbon cycle. The fate of high latitude ecosystems and associated climate feedbacks in response to warming remains uncertain, particularly in boreal peatlands, which store roughly one-third of the global carbon pool. In order to understand how peatlands respond to climate warming, we examined Holocene carbon accumulation rates from four peatlands on the Kenai Peninsula, Alaska, focusing on the early Holocene (~11,000-9000 cal yr BP), a time when the climate was warmer than today. Basal dates from over 200 peat cores across Alaska were compiled to examine the timing and spatial distribution of peatland initiation across Alaska, and available pollen data from the North American Pollen Database (NAPD) and the Paleoenvironmental Arctic Sciences (PARCS) databases were used to examine associated vegetation distribution patterns. Our study reveals that the highest rates of carbon accumulation on the Kenai Peninsula occurred during the early Holocene Thermal Maximum (HTM), which also corresponds to the highest number of peat basal dates both on the Kenai and across Alaska, indicating that not only vertical peat growth but also lateral peatland expansion was high. We suggest that the warm summers and longer growing season during the early Holocene in Alaska resulted in high net primary productivity (NPP), rapid peat burial, and the greatest carbon accumulation rates. Rapid rates of accumulation and burial may have minimized the effects of aerobic decomposition. In addition, a change in the seasonal timing of precipitation and moisture availability and an increase in summer precipitation may have decreased drought stress, promoting peatland initiation and peat growth. We also speculate that the dominance of broad-leafed deciduous forests and abundant ferns at that time resulted in localized vegetation-climate feedbacks that would have increased relative humidity, minimizing the potential effects of peatland drying as a result of warm summer temperatures. These findings have important implications for carbon storage and climate feedbacks in Arctic ecosystems under expected high- latitude warming, as warmer temperatures can potentially lead to net carbon storage and a negative climate feedback as long as summer growing season moisture availability remains high.