Eolian Records In The Northwestern United States

Loess is extensive over parts of the northwestern United States, including a small part of western Wyoming, a large part of eastern Idaho, a small part of northern Oregon, and much of eastern Washington (Fig. 17). Loess in this region has not been studied for as long as loess in the midcontinent of North America. However, recent work indicates that a long strati-graphic record is present and that loess origins in parts of the region differ from those found elsewhere.

In Idaho and adjacent parts of westernmost Wyoming, loess covers large areas to the north and south of the Snake River Plain, which is probably one of its major sources (Lewis et al., 1975; Scott, 1982; Pierce et al., 1982; Glenn et al., 1983). Loess in this region has a thick ness of up to 12 m in places, but most is 2 m or less (Lewis et al., 1975). Thickness and particle size data (Lewis et al., 1975; Lewis and Fosberg, 1982; Pierce et al., 1982; Glenn et al., 1983) suggest that loess deposition took place under northwesterly or westerly winds (Fig. 17).

Stratigraphic studies of loess in Idaho have been conducted by Scott (1982), Pierce et al. (1982), and Forman et al. (1993). Pierce et al. (1982) studied the loess stratigraphy at numerous localities over an ~400-km long transect from western Wyoming to central Idaho and found two major units, informally designated loess A and loess B. The two loesses are separated by a well-developed paleosol. Based on the degree of soil development and stratigraphic relations with K/Ar-dated lava flows and well-dated Lake Bonneville flood deposits, they suggested that loess A was deposited during the last glacial period and loess B was deposited during the penultimate glacial period. Forman et al. (1993) studied loess at two closely spaced sites in eastern Idaho and reported TL data that support correlation of loess A with the last glacial period (late Wisconsin time). However, they suggested, on the basis of TL data, that loess B was also deposited during the last glacial period (early Wisconsin time). At present, probably too few data exist to determine which age estimate is correct.

Loess Sand Dune

FIGURE 14 Map showing the distribution of eolian sand (stippled) in the central and southern Great Plains of the United States, modern potential drift directions, and inferred late Holocene paleo-winds based on dune orientations. Sand distribution data from Muhs and Holliday (1995); drift directions and paleowinds data from Madole (1995), Muhs et al. (1996, 1997a), Arbogast (1996), and unpublished data of D.R. Muhs and V.T. Holliday.

FIGURE 14 Map showing the distribution of eolian sand (stippled) in the central and southern Great Plains of the United States, modern potential drift directions, and inferred late Holocene paleo-winds based on dune orientations. Sand distribution data from Muhs and Holliday (1995); drift directions and paleowinds data from Madole (1995), Muhs et al. (1996, 1997a), Arbogast (1996), and unpublished data of D.R. Muhs and V.T. Holliday.

Busacca and McDonald (1994) designated the two youngest loess units in eastern Washington as L1 (upper) and L2 (lower). Well-dated Holocene and late glacial tephras from the volcanically active Cascade Range to the west are found within eastern Washington loess and provide valuable time lines for dating and correlation (Busacca et al., 1992; Richardson et al., 1997). In addition, TL ages by Berger and Busacca (1995) and TL and infrared stimulated luminescence (IRSL) ages by Richardson et al. (1997) confirm correlations inferred from tephra data that L1 is of late Wisconsin age and L2 is of early to middle Wisconsin age. Agreement between TL and IRSL ages is good back to ~ 70,000 years ago, which strengthens the age estimates made by these investigators. However, there are differences between the two studies; TL ages reported by Richardson et al. (1997) for the same sections are generally younger than those reported by Berger and Busacca (1995).

Stratigraphic and geochronologic data show when in a glacial cycle most loess deposition takes place in eastern Washington. Because the loess is thought to be derived primarily from proglacial Lake Missoula flood sediments, most loess deposition probably takes place late in each glacial cycle, and some deposition has continued into the Holocene. Eastern Washington loess becomes thinner and finer grained from southwest to northeast, indicating that late glacial winds were from the southwest (Busacca, 1991; Busacca and McDonald, 1994).

0 0

Post a comment