New techniques and old mud: Reconstructing past warmth to predict Arctic change
The Arctic is warming faster than any other region of the planet, and the effects are already apparent. Landscapes that have been continuously ice-covered since the start of the last glacial cycle are now being revealed as ice recedes. What will these changes mean for the Inuit living there? How will vegetation change? Those are some of the questions that Giff Miller, Julio Sepúlveda, grad students Sarah Crump and Jon Raberg, and postdoc Greg deWet, along with colleagues at the University at Buffalo, University of Alaska Fairbanks, and Curtin University, are tackling under a new NSF-funded project, PACEMAP.
Capitalizing on lake sediment deposited in past warm times across the eastern Canadian Arctic, including the Early Holocene, the Last Interglacial (MIS 5e), and the penultimate interglacial (MIS 7), our field teams recover pristine, continuous sedimentary archives that capture these warm intervals. We are using molecular approaches to reconstruct changes in climate, hydrology, and vegetation through past warm times. Julio, Greg, and Jon are using bacterial membrane lipids, called branched glycerol dialkyl tetraethers (brGDGTs), to reconstruct climate back through time. The distribution of these compounds in modern environments has been empirically shown to track environmental temperature, and one of the goals of the project is to create a calibration for Arctic Canada. Sarah, working with Mike Bunce at Curtin University, is using ancient sedimentary DNA extracted from interglacial sediment to provide a more authentic reconstruction of local vegetation communities, where long-distance pollen dispersal compromises pollen records. This approach, relatively untested on such long timescales, also involves modern validation work, where ecologists working with Skip Walker (CU PhD) at UAF, map the modern vegetation around our study sites to compare with DNA extracted from surface sediments.
Our goal is to derive algorithms that will allow us to predict the evolution of Arctic vegetation by 2100 CE using summer temperature estimates from climate models under various future emissions scenarios. In 2019 we will deploy five field teams collecting sediment cores and mapping modern vegetation along a transect from the forest-tundra ecotone to polar desert sites in Canada's far north.
The Australian megafaunal extinction
The timing and cause of the extinction that killed all animals heavier than 60 kg in Australia has been elusive. Using ratites' (flightless birds) eggshells, we have been able to narrow the time window of the extinction to 50,000 +/- 5,000 years ago. Other research has since supported our age estimate for this extinction (Roberts et al., 2001). We have also used the eggshells of these birds to understand dietary changes through time and from region to region. Stable isotopes of carbon, nitrogen and oxygen are preserved in the crystal matirx of the eggshells and present us with a proxy of diet. Based on the evidence to date, our hypothesis is that after human colonization of Australia (60,000 years ago), the ecosystem changed due to over-burning of the landscape. This would have reduced evapotranspiration over the interior of the continent, thereby increasing aridity and eliminating food supplies for the large animals.
The racemization of amino acids in charophyte oogonia
Charophytes, a type of green algae, are prevalent both in the modern and fossil lacustrine environments of Australia and elsewhere. They generally tolerate a wide range of water conditions, which might explain their ubiquity. Some species (e.g., Lamprothamnium papulosum) create a calcareous shell (oogonium) around the female gamete (oospore). These shells are about the size of coarse sand, but are fairly robust for their size and manage to preseve well in the fossil record. This project aimed to determine if oogonia are a viable source of sample material for amino acid geochronology. The racemization rate for each amino acid in a given taxon must be calculated in order to understand the age or thermal history of a sample. So, this was a primary thrust of the project. Then, site-specific chronologies and thermal histories were hoped to be calculated, which would give us more insight into the lake level histories in central Australia. In short, oogonia do not appear to be extremely useful for amino acid geochronology due to wide ranges of D/L ratios and amino acid concentrations within a single stratum. Furthermore, D/L ratio and concentration do not appear to be related. This translates to enormous error bars associated with age or paleotemperature estimates.
The Late Quaternary glacial history of Baffin Island
Ice sheets are an integral part of the earth’s climate system, but the chronology and dynamics of Pleistocene ice sheets is in most places only loosely constrained. In the eastern Canadian Arctic, for example, the history of the northeastern Laurentide Ice Sheet (LIS) during the Last Glacial Maximum (LGM) is only known in selected regions, but is extrapolated to broader areas. More than 160 cosmogenic exposure ages from the Clyde Region, northeastern Baffin Island, reveal that the northeastern LIS was more extensive and dynamic than previously depicted. Tors on weathered uplands surrounding Clyde Inlet are covered with perched erratics that have cosmogenic exposure ages of 20-10 ka, indicating that the uplands were glaciated by cold-based, non-erosive ice during the LGM. In contrast, warm-based, erosive ice probably occupied Clyde Inlet throughout the LGM, indicating strong gradients in basal thermal regimes and the operation of an ice stream in Clyde Inlet. In the most distal sectors of the Clyde Foreland, where cold-based ice hardly modified the landscape, erratics yield a multi-modal exposure age distribution that may indicate numerous advances and retreats of cold-based ice across the foreland throughout Marine Isotope Stages 3 and 2. Ice retreated from the Clyde Foreland ~13 ka, from the mouth of Clyde Inlet ~12-10 ka, and had reached the fiord head by ~8.3 ka. Upland tors surrounding outer Clyde Inlet have single-nuclide apparent exposure ages >60 ka. However, paired 26Al and 10Be concentration data reveal that they are at least several hundred thousand years old, indicating that while the uplands have been covered by ice off and on throughout the Quaternary, they have only been slightly modified.
These new data depict an extensive LIS in the Clyde Region during the LGM, possibly terminating at the continental shelf break beyond Clyde Inlet. Strong gradients in basal thermal regimes suggest highly variable patterns in glacier thickness, velocity, and erosion, an overall pattern indicative of ice stream activity. A northeastern LIS this dynamic and extensive would have been closely linked with fluctuating ocean circulation and sea level.