Special Sessions
Sponsored by the Geomorphology Specialty Group
Association of American Geographers 102nd Annual Meeting
Chicago, Illinois, March 7-11.

Soil Geography and Geomorphology III: Climate and Eolian Influences

Paper Session 3616
Thursday, 3/9/06, from 5:00 PM - 6:40 PM

Sponsorship(s):

Geomorphology Specialty Group

Organizer(s):

Randy Schaetzl - Michigan State University
Peter Scull

Chair(s):

Peter Scull

Abstract(s):

• 5:00 PM Author(s):
  *Randy Schaetzl, PhD - Michigan State University
  Walter Loope, PhD - US Geological Survey
  

  Abstract Title: Silty upland soils and glaciolacustrine environments in northern Michigan: Using soil “spatial signatures” to understand landscape evolution

  
  

  Sandy soils are widespread across the glaciated uplands of the UP (Upper Peninsula) of Michigan. Extremely clayey or sandy soils tend to dominate the extensive lacustrine plains, e.g., Glacial Lake Algonquin, nearby. Silty parent materials are, however, uncommon. Nonetheless, some uplands in the eastern UP have sandy soils with a silty upper profile, i.e., a silt and very fine sand “cap” that ranges up to a meter in thickness. The geologic origin of these silts has long been unresolved. In our research we sampled many upland soils with this silty cap, as well as small areas of soils formed in interstratified silts and sands from the nearby floor of Glacial Lake Minong. Samples were analyzed for their geochemical and mineralogical signatures, and the data examined spatially to elucidate patterns. Our results suggest that the silts in the upland soil caps are loess and that the source of the loess was the recently subaerially exposed Lake Minong plain. Parabolic dunes on the lake plain support the notion that paleoenvironmental conditions subsequent to the demise of Lake Minong were extremely windy and that the surface was deflating. Silt caps on soils slightly farther downwind, and on slightly higher locations, also support the same conclusion. Saltation and eolian transport of sand on the lake floor could have been the mechanism that initiated entrainment and transport of the silts, as has been shown by previous work on the Great Plains. This research highlights the importance of spatial sampling techniques in soil geomorphic research.

  Keywords: loess, geochemistry, Glacial Lake Minong, soils, geomorphology

  
  
• 5:20 PM Author(s):
  *Mark C Nelson - University of Wisconsin-Eau Claire
  Ashley Wong - University of Wisconsin-Eau Claire
  Karen G Havholm, PhD - University of Wisconsin-Eau Claire
  Garry L Running, PhD - University of Wisconsin-Eau Claire
  

  Abstract Title: Dunes, Forests, and People during the Late-Holocene: Evidence from Buried Podzolic Soils in the Crepeele Dune Field, Southwestern Manitoba

  
  

  The Crepeele Dune Field (CDF) is one of 18 late-Holocene dune fields in the Glacial Lake Hind Basin. Previous research has shown that such dune fields are characterized by comparatively greater geomorphic and ecological complexity and were, therefore, important loci of pre-contact human activity. Buried soil profiles with weak podzolic morphology were also observed within the CDF. The purpose of the research reported in this paper is to determine the distribution of these buried podzolic soils and to characterize the spatial relationship between them and archaeological material. Soil profiles were described and sampled from the walls of 12 archaeological excavation units. These locations were mapped. A total station was used to create a topographic map of the area around the units. Soil, archaeological, and topographic data were then combined in a GIS for analysis (ArcMap and ArcView). Based on preliminary analysis, buried podzolic soils are widely observed in intermediate-elevation positions on dunes. Drainage on dunes permits podzolization to occur but dune crests are subject to frequent erosion so podzolic morphology is not preserved in high landscape positions. Evidence for past human activity follows the same spatial pattern. The presence of buried podzolic soil profiles strongly suggests the forest communities that dominate the CDF today were present throughout the late-Holocene and that forest-related resources were an important factor in attracting humans to the CDF. However, more soil data collected from a wider range of slope positions and slope aspects across the CDF is needed before our model can be accepted.

  Keywords: soil geomorphology, Great Plains-Canadian Prairies,geoarchaeology

  
  
• 5:40 PM Author(s):
  *Greg Okin - University of Virginia
  Junran Li - University of Virginia
  Lorelei Li - University of Virginia
  Howard Epstein - University of Virginia
  

  Abstract Title: Impact of Aeolian Processes on Soil Surface Resource Distribution

  
  

  Since the publication by Schlesinger et al. of a general model, arid and semiarid land degradation has been thought of as the reduction or redistribution of net primary production, accompanied by a similar reduction or redistribution of soil resources into resource islands. Both water and wind have been implicated as possible mechanisms for the removal and reorganization of soil resources but recent studies suggest that water erosion cannot, by itself, account for the depletion of soil fertility associated with land degradation. An experiment was established to study the impact of wind erosion on vegetation and soil nutrient status at the Jornada Experimental Range in New Mexico. In this experiment grasses were removed and the soil nutrients and vegetation cover have been monitored. Sediment flux and nutrients concentration in windborne sediments have been measured and compared with meteorological measurements at the sites. To date, results from experiment indicate that: 1) biotic processes continually add nutrients to surface soil that are removed by the wind, 2) detectable soil nutrient depletion happens within a single wind erosion season, 3) wind erosion has led to a dramatic increase in the scale of spatial heterogeneity of nutrients only two years, 4) wind erosion is capable of creating islands of fertility without water erosion, and 5) wind erosion and dust flux are best predicted by the size of gaps between vegetation, instead of other parameters often used in wind erosion models, like lateral cover.

  Keywords: soils, aeolian processes, nutrients, deserts, vegetation

  
  
• 6:00 PM Author(s):
  *Stephani Michelsen-Correa - Colgate University
  Peter R Scull - Colgate University
  

  Abstract Title: The Impact of Reforestation on Soil Temperature

  
  

  As the world’s largest source of organic carbon, soils play an important role in the cycling of carbon in our atmosphere. The amount of carbon stored in the soil is dependant on the rate of photosynthetic uptake by plants and the rate of plant and microbial respiration, which are in part a function of soil temperature. Areas of Central NY are succeeding from fields back to forests. How this change in land cover will impact the underlying soil temperature is an important question to consider as our global climate continues to change as a result of increasing carbon dioxide concentration in the atmosphere. Using a hillside in Hamilton, NY as a case study, our research attempts to determine the differences in soil temperature between forests and fields. Results demonstrate that during the fall (Nov-Dec 25) and winter (Dec 26- Mar 31) forested soils were warmer than field soils. The spring (Apr 1- May 31) showed an opposite trend with field soils being warmer than forested soils. The field soils also show greater temperature variations than the forest soils. These results suggest that during the fall and winter the soil in the forests have the potential to release more carbon than the field soils. As succession occurs on abandoned fields in Central New York soil temperatures will change, which will affect their capacity to cycle organic carbon.

  Keywords: soil temperature, succession, microbial decomposition, carbon cycle,

  
  
• 6:20 PM Author(s):
  *Peter R Scull, Assistant Professor - Colgate University
  Jennifer A. Miller - West Virginia University
  

  Abstract Title: Using multivariate adaptive regression splines to assess the extent to which climate controls landscape-scale soil carbon variability

  
  

  The purpose of the research is to develop models that describe the spatial variation in soil carbon properties on the landscape. Our hope is that such models will provide insight into how climatic variables drive soil carbon formation at very large spatial scales. A firm understanding of landscape scale soil carbon variation provides the foundation from which we can begin to further understand the role soils play in global change research. While it is generally assumed that climate controls soil formation at large spatial scales (e.g. continents), it has not been demonstrated quantitatively using actual point soil samples. Neither is it clear which aspects of climate are most important from a soil carbon perspective. For example, are there significant differences between summer and winter precipitation? Is precipitation or temperature more important at different spatial scales or at different latitudes? New tools (primarily GIS and more flexible statistical methods) and extensive soil and climate databases (e.g. PRISM data) have made it possible to begin to answer such questions. We used a relatively new machine learning technique, multivariate adaptive regression splines (MARS), to begin to explore the relationship between climate and soil carbon in the eastern United States. MARS allows for more flexible relationships than commonly used statistical techniques and, preliminary results suggest that it can be a very useful tool in unraveling the role climate plays in landscape-scale soil carbon variability.

  Keywords: MARS, soil carbon, climate