Introduction

The Americas contain rich late Quaternary eolian records in the form of loess (Fig. 1), eolian sand sheets, and dunes. Eolian deposits are important in paleocli-matic studies because they (1) put limits on the moisture balance of a region at the time of formation, (2) can be sensitive indicators of the amount of vegetation cover, and (3) are records of wind direction. The record of

FIGURE 1 Map showing the distribution of loess (shaded areas) in North and South America and locations of detailed figures: loess distribution in North America from Thorp and Smith (1952) and Pewe (1975), and loess distribution in South America from Sayago (1995).

loess and eolian sand is summarized for South and North America with emphasis on three time periods: the last glacial maximum (LGM), the mid-Holocene, and the late Holocene. Throughout this chapter, all ages given, unless otherwise noted, are in uncalibrated radiocarbon years before present (14C years B.P.).

The evolution of concepts about the genesis of loess make for a fascinating history of scientific thought (see Follmer, 1996). A now classical and generally accepted model of loess formation near glaciated terrain (Fig. 2) is that silt-sized particles are formed by glacial grinding and are delivered as outwash to valley trains (Ruhe, 1983). Wind entrains the particles from floodplains of the valley trains and deposits them in downwind-thinning blankets that also show systematic downwind decreases in particle size and carbonate content. The classical concept of loess stratigraphy, therefore, is that thick, unaltered loess deposits record glacial periods and intercalated paleosols record interglacial or interstadial periods (Fig. 2). In this chapter, certain aspects of the classical loess model are supported by loess studies in the Americas, and many aspects are not.

Although eolian sand deposits of the Americas have not received the attention that loess deposits have, there is also a traditional interpretation that such deposits are related to glacial periods. Several investigators (e.g., Sarnthein, 1978; Kutzbach and Wright, 1985) have considered large dune fields in the midlatitudes to be dominantly relict features that were last active during the LGM, when conditions are thought to have been cold, dry, and windy. In addition, eolian sand and loess were thought to be genetically related: i.e., essentially one deposit with a coarse, proximal facies (sand) and fine, distal facies (loess). In some regions of the Americas, this model seems to apply, and in other regions, it does not. New studies from both South and North America are now demonstrating that many dune fields have a rich record of Holocene activity and that dunes and loess may have distinctly different sources.

0 0

Post a comment