The sea-level rise due to ice-sheet melting since the last glacial maximum was not uniform everywhere because of the deformation of the Earth's surface and its geoid by changing ice and water loads. A numerical model is employed to calculate global changes in relative sea level on a spherical viscoelastic Earth as northern hemisphere ice sheets melt and fill the ocean basins with meltwater. Predictions for the past 16,000 years explain a large proportion of the global variance in the sea-level record, particularly during the Holocene. Results indicate that the oceans can be divided into six zones, each of which is characterized by a specific form of the relative sea-level curve. In four of these zones emerged beaches are predicted, and these may form even at considerable distance from the ice sheets themselves. In the remaining zones submergence is dominant, and no emerged beaches are expected. The close agreement of these predictions with the data suggests that, contrary to the beliefs of many, no net change in ocean volume has occurred during the past 5000 years. Predictions for localities close to the ice sheets are the most in error, suggesting that slight modifications of the assumed melting history and/or the rheological model of the Earth's interior are necessary.

A slight cooling can induce the formation of ice sheets in the Scandinavian mountains and in the American Arctic. The increasing albedo and the appearance of cold air masses above the glaciers cause glaciation to spread over a vast area. As a result, the sea level lowers and a large part of the Barents and Kara seabeds dries up. Ice sheets are formed there, which spread over the northeastern part of the Kola Peninsula, the Pechora River basin, and over northwestern Siberia. The glacier barrier extending nearly from the North Pole to central Europe hinders latitudinal atmospheric circulation. Precipitation decreases sharply in the areas east and southeast of the glaciers. As a consequence, glaciers in the mid-latitudes retreat and sea level rises. Increased iceberg formation is induced in the periphery of the Barents Ice Sheet, causing it to disappear. An interglacial sets in.

Samples of algal tufa, gastropods and calcite-cemented sand were collected from the Walker and Pyramid Lake areas of the Lahontan Basin, Nevada. X-ray diffraction petrographic and radiocarbon analyses show that massive forms of tufa such as the dendritic variety contain secondary carbon-bearing material and therefore yield unreliable radiocarbon dates. Dense coating of tufa (lithoid), however, gave radiocarbon ages in agreement with dates on coexisting aragonite gastropods. Radiocarbon data from the study were combined with previously dated noncarbonate materials [Born, S. M. (1972). “Lake Quaternary History, Deltaic Sedimentation, and Mudlump Formation at Pyramid Lake, Nevada”, Center for Water Resources, Desert Research Inst., Reno, Nevada] to give an internally consistent record of lake level fluctuations for the past 40,000 years. The main features of the Lahontan chronology are (1) extreme high stands (1330 m above sea level) 13,500 to 11,000 and 25,000 to 22,000 B.P., (2) a moderate high stand (1260 m above sea level) 20,000 to 15,000 B.P., (3) a low stand of unknown elevation 40,000 to 25,000 B.P., (4) an extremely low stand 9000 to 5000 B.P., and (5) an overall increase in the size of Walker and Pyramid Lakes during the past 5000 years, until the late 19th century. Pore fluid data indicate that Walker Lake desiccated sometime during the period 9050 to 6400 B.P. Salts deposited as a result of this dessication are still undergoing dissolution causing a flux of chloride, carbon, and other solute species from the sediments to the overlying lake water. Pore fluid data obtained from Pyramid Lake sediments do not indicate the presence of a concentrated brine at depth. This suggests that Pyramid Lake did not dry completely during this period although it may have been severely reduced in size. There has been considerable disagreement regarding the occurrence of extreme arid conditions (altithermal period) since 10,000 B.P. [Mehringer, P. J. (1977). “Models and Great Basin Prehistory”. Desert Research Inst. Pub, Reno, Nevada]. The data of this study suggest that such a climatic regime did occur in the western Great Basin during the period 9000 to 5000 B.P.

Thorium-230 dating on saline of the Lower Salt unit in pluvial Searles Lake, California, shows that this unit was formed between 24,500 and 32,000 years ago. The initial apparent 14C age of the lake is estimated to be about 900 years. After correcting for nonradiogenic 230Th and for the initial 14C age, excellent agreement between 230Th and 14C ages is obtained. The reliability of 230Th dating on salt deposit opens a new way for continuation of absolute chronology below the Lower Salt in Searles Lake.

Alternative kinetic models for amino acid racemization in Pleistocene molluscan samples are compared by examination of results for samples from marine Pleistocene deposits in California, Washington, and Florida. Linear kinetic models previously have been applied to the Florida samples [ Mitterer, R.M. (1975). Earth and Planetary Science Letters 28 , 275–282.] because these kinetics were observed in laboratory pyrolysis experiments with the particular genus involved (Mercenaria). Nonlinear kinetic models, extrapolated from deep-sea foraminifera racemization kinetics, are applied to samples of Protothaca and Saxidomus from California and Washington and seem more consistent with their local chronologic and stratigraphic control. Average or effective diagenetic temperatures can be estimated by each of these models if reliably dated samples are available. Linear models applied to such samples from California and Florida suggest average diagenetic temperatures that are cooler (by as much as 10°C) than would be inferred from available paleoclimatic records. Nonlinear kinetic models yield estimates of average diagenetic temperatures that are more consistent with these records: full-glacial (i.e., approximately 18,000 yr BP) temperature reductions of between 2 and 6°C are inferred for coastal California and southern Florida. The nonlinear kinetic model is used to expand (by a factor of 2.5 to 3.0) the time scale proposed by Mitterer (1975) for five marine Pleistocene units of Florida.

Changes in fossil pollen assemblages from a 2-m core from Linsley Pond, North Branford, Connecticut, are compared with historically documented land use changes in the lake watershed. Dating with 210Pb and 14C reveals two sedimentation rate changes in the core which are associated with the arrival of European farmers; the building of cabins and suburban housing subdivisions on the lake shore. At European settlement in 1700 ad Ambrosia and Rumex pollen first appear, Gramineae-type pollen increases, and Tsuga decreases. Just before the beginning of agricultural disturbance Fagus pollen declines. The chestnut blight of 1913 causes a reduction of Castanea pollen and a subsequent vegetational succession through Betula to Quercus. The sedimentation rate determined by the chestnut blight horizon is consistent with the rate deduced from 210Pb analysis.