ABSTRACT Lake Coyote, California, which formed in one of five basins along the Mojave River, acted both as a part of the Lake Manix basin and, after the formation of Afton Canyon and draining of Lake Manix ca. 24.5 calibrated (cal) ka, a side basin that was filled episodically for the next 10,000 yr. As such, its record of lake level is an important counterpart to the record of the other terminal basin, Lake Mojave, following the draining of Lake Manix. We studied lake and fluvial deposits and their geomorphology and identified five principal periods of recurring lakes in the Coyote basin by dating mollusks. Several of these periods in detail consist of multiple lake-rise pulses, for which we identified specific fluvial deposits that represent the Mojave River entering the basin. The pulsed record of rapid lake rise and decline is interpreted as switching of the Mojave River between Lake Coyote and Lake Mojave. A composite lake record for both basins shows nearly continuous lake maintenance by the Mojave River from 24.5 cal ka to ca. 14 cal ka. One potential gap in the lake record, ca. 22.7–21.8 cal ka, may indicate either temporary river routing to yet another basin or a dry climatic period. The Mojave River discharge was sufficient to maintain at least one terminal lake throughout most of the Last Glacial Maximum and deglacial periods, indicating that paleoclimate was moist and/or cool well into the Bølling-Allerød and that the lake records may not be sensitive to variations from moderate to high discharge. Nuances of lake-level changes in both the Coyote and Mojave basins are difficult to interpret as paleoclimatic events because the current chronologic control on lake levels from nearshore deposits does not provide the necessary precision. Mojave River avulsion leading to flow to Coyote basin may have been influenced by rupture on a dextral-oblique fault. Earliest post–Lake Manix stream deposits of the Mojave River leading to the Coyote basin are faulted, and most subsequent streams were confined to the downthrown fault block. This fault rupture and possible enhanced river routes to Lake Coyote, rather than Lake Mojave, are bracketed by dated beach deposits to the period ca. 20–19 cal ka. Later, headward erosion through the fluvial plain by the Mojave River eliminated flow to Coyote basin after ca. 14 cal ka and completed incision of the plain after ca. 12 cal ka.

Radiocarbon analyses of pollen, ostracodes, and total organic carbon (TOC) provide a reliable chronology for the sediments deposited in Bear Lake over the past 30,000 years. The differences in apparent age between TOC, pollen, and carbonate fractions are consistent and in accord with the origins of these fractions. Comparisons among different fractions indicate that pollen sample ages are the most reliable, at least for the past 15,000 years. The post-glacial radiocarbon data also agree with ages independently estimated from aspartic acid racemization in ostracodes. Ages in the red, siliclastic unit, inferred to be of last glacial age, appear to be several thousand years too old, probably because of a high proportion of reworked, refractory organic carbon in the pollen samples. Age-depth models for five piston cores and the Bear Lake drill core (BL00-1) were constructed by using two methods: quadratic equations and smooth cubic-spline fits. The two types of age models differ only in detail for individual cores, and each approach has its own advantages. Specific lithological horizons were dated in several cores and correlated among them, producing robust average ages for these horizons. The age of the correlated horizons in the red, siliclastic unit can be estimated from the age model for BL00-1, which is controlled by ages above and below the red, siliclastic unit. These ages were then transferred to the correlative horizons in the shorter piston cores, providing control for the sections of the age models in those cores in the red, siliclastic unit. These age models are the backbone for reconstructions of past environmental conditions in Bear Lake. In general, sedimentation rates in Bear Lake have been quite uniform, mostly between 0.3 and 0.8 mm yr ‒1 in the Holocene, and close to 0.5 mm yr ‒1 for the longer sedimentary record in the drill core from the deepest part of the lake.

Study of slope movements triggered by the storm of November 3–5, 1985, in the central Appalachian Mountains, U.S.A., has helped to define the meteorologic conditions leading to slope movements and the relative importance of land cover, bedrock, surficial geology, and geomorphology in slope movement location. This long-duration rainfall at moderate intensities triggered more than 1,000 slope movements in a 1,040-km 2 study area. Most were shallow slips and slip-flows in thin colluvium and residuum on shale slopes. Locations of these failures were sensitive to land cover and slope aspect but were relatively insensitive to topographic setting. A few shallow slope movements were triggered by the same rainfall on interbedded limestone, shale, and sandstone. Several large debris slide-avalanches were triggered in sandstone regolith high on ridges in areas of the highest measured rainfall. Most of these sites were on slopes that dip 30 to 35° and lie parallel to bedding planes, presumably the sites of least stability.