Denton and Hughes (1983, Quaternary Research 20, 125–144) postulated that sea level linked a global ice-sheet system with both terrestrial and grounded marine components during late Quaternary ice ages. Summer temperature changes near Northern Hemisphere melting margins initiated sea-level fluctuations that controlled marine components in both polar hemispheres. It was further proposed that variations of this ice-sheet system amplified and transmitted Milankovitch summer half-year insolation changes between 45 and 75°N into global climatic changes. New tests of this hypothesis implicate sea level as a major control of the areal extent of grounded portions of the Antarctic Ice Sheet, thus fitting the concept of a globally interlocked ice-sheet system. But recent atmospheric modeling results (Manabe and Broccoli, 1985, Journal of Geophysical Research 90, 2167–2190) suggest that factors other than areal changes of the grounded Antarctic Ice Sheet strongly influenced Southern Hemisphere climate and terminated the last ice age simultaneously in both polar hemispheres. Atmospheric carbon dioxide linked to high-latitude oceans is the most likely candidate (Shackleton and Pisias, 1985, Atmospheric carbon dioxide, orbital forcing, and climate. In “The Carbon Cycle and Atmospheric CO2: Natural Variations Archean to Present” (E. T. Sundquest and W. S. Broecker, Eds.), pp. 303–318. Geophysical Monograph 32, American Geophysical Union, Washington, D.C.), but another potential influence was high-frequency climatic oscillations (2500 yr). It is postulated that variations in atmospheric carbon dioxide acted through an Antarctic ice shelf linked to the grounded ice sheet to produce and terminate Southern Hemisphere ice-age climate. It is further postulated that Milankovitch summer insolation combined with a warm high-frequency oscillation caused marked recession of Northern Hemisphere ice-sheet melting margins and the North Atlantic polar front about 14,000 14C yr B.P. This permitted renewed formation of North Atlantic Deep Water, which could well have controlled atmospheric carbon dioxide (W. S. Broecker, D. M. Peteet, and D. Rind, 1985, Nature (London) 315, 21–26). Combined melting and consequent sea-level rise from the three warming factors initiated irreversible collapse of the interlocked global ice-sheet system, which was at its largest but most vulnerable configuration.

Time series depicting mountain glacier fluctuations in the Alps display generally similar patterns over the last two centuries, as do chronologies of glacier variations for the same interval from elsewhere in the Northern Hemisphere. Episodes of glacier advance consistently are associated with intervals of high average volcanic aerosol production, as inferred from acidity variations in a Greenland ice core. Advances occur whenever acidity levels rise sharply from background values to reach concentrations ≥1.2 μequiv H+/kg above background. A phase lag of about 10–15 yr, equivalent to reported response lags of Alpine glacier termini, separates the beginning of acidity increases from the beginning of subsequent ice advances. A similar relationship, but based on limited and less-reliable historical data and on lichenometric ages, is found for the preceding 2 centuries. Calibrated radiocarbon dates related to advances of non-calving and non-surging glaciers during the earlier part of the Little Ice Age display a comparable consistent pattern. An interval of reduced acidity values between about 1090 and 1230 A.D. correlates with a time of inferred glacier contraction during the Medieval Optimum. The observed close relation between Noothern Hemisphere glacier fluctuations and variations in Greenland ice-core acidity suggests that sulfur-rich aerosols generated by volcanic eruptions are a primary forcing mechanism of glacier fluctuations, and therefore of climate, on a decadal scale. The amount of surface cooling attributable to individual large eruptions or to episodes of eruptions is simlar to the probable average temperature reduction during culminations of Little Ice Age alacier advances (ca. 0.5°–1.2°C), as inferred from depression of equilibrium-line altitudes.

The Ross Ice Shelf δ18O profile at station J-9 covers at least the last 30,000 yr. It identifies the depth in the core of ice from (i) the last glacial-interglacial transition (266 to 286 m) and (ii) the 1000-m surface elevation (about 140 m). Various processes contribute to the δ18O change observed in the core: (i) climatic warming, mainly caused by a decrease in winter sea ice extent around Antarctica of about 6° latitude early in the glacial-interglacial transition, (ii) decreasing ice sheet thickness later in the glacial-interglacial transition and during the Holocene, and (iii) decreases in elevation and effective distance from the open ocean as the source of the ice in the core shifts along the flow line toward J-9. Average δ18O values of the last 3000 yr imply a fairly stable climate. Yet shorter (102 to 103 yr) δ18O climatic oscillations up to 6‰ are seen in both the Holocene and the glacial portion of the record.

Ibyuk Pingo, which is 49 m high and 300 m in basal diameter, is one of the largest pingos in the world. Precise surveys carried out for the 1973 to 1983 period indicate that the lower half of the pingo, below a height of 25 m, shows no perceptible growth. From 25 m to the top, the growth steadily increases to a maximum of 2.3 cm/yr at the summit. The source of growth is believed to lie at a depth of about 65 m below the highest peak. Evidence based upon the depth of permafrost near the pingo, radiocarbon dating of wood in the pingo overburden, heat conduction theory, and the measured growth rate suggests an age of about 1300 ± 200 yr. Ibyuk Pingo is in an unstable state and collapse could be initiated by further slumping, a thermal disturbance to the crater, or erosion along a radial dilation crack. A study of the potential collapse pattern of Ibyuk Pingo suggests a number of criteria that might be employed in the environmental reconstruction of large circular features that might be identified as collapsed pingo remnants.

A working hypothesis for the development and maintenance of sorted circles, which are finegrained soil domains delineated by borders of coarser material, involves clast motion relative to soil and intermittent circulatory motion of soil in the active layer. The hypothesis is compatible with the observed size, regularity, sorting, microtopography, surface soil motion, vegetation cover, and subsurface organic carbon content of soil patterns studied in Spitsbergen. Theoretical considerations of buoyancy-driven convection suggest that intermittent convection of thawed soils is plausible, and that the requisite bulk density decrease with depth arises naturally from moisture gradients that develop during the thaw consolidation of ice-rich soils. These considerations also help in understanding the dependence of sorted circles on soil characteristics and on moisture. Moreover, they point to the possible use of sorted patterns as paleoclimatic indicators based on the expected relation between pattern size and active layer depth.

Pedogenic zonation in the soils of the polar regions is a result of gradients in environmental factors and attendant chemical processes. Along a latitudinal transect, it is best manifested at well-drained sites and by soils developed on predominantly silicate rocks. Selected sites in arctic Alaska, in the Canadian arctic, Greenland, and Svalbard adequately fulfill these prerequisites. The processes of podzolization, decarbonization-carbonization, pervection, and salinization as models of arctic pedogenesis demonstrate that processes occurring in the temperate region also operate in the Far North. Brunification, melanization, and oxidation are recognized for the first time as current geochemical and pedogenic mechanisms of the Arctic region. Traditional genetic soil names have been retained because they represent a closer relationship to pedogenic processes than the more modern nomenclature. The identification, the chemical behavior, the strength, abundance, and mobility of the proton donors and conjugate bases are keys to the genesis of soils and the distinction of contrasting soil processes. This new approach to the understanding of arctic pedogenesis can be better fulfilled by collecting, analyzing, and interpreting soil solution obtained in situ.

Published oxygen isotope results from benthic and planktonic foraminifera are used to constrain the extent to which surface ocean temperatures in the tropics changed between glacial and interglacial time. The possible impacts of bioturbation and of the redistribution of 18O/16O ratios within the sea, putting aside the oxygen isotope results, are consistent with the CLIMAP project conclusion that tropical ocean temperatures remained within ±2°C of their present value.

Biostatistical analysis of modern pollen assemblages in 152 Japanese surface samples shows that Cryptomeria japonica can normally grow in areas with a mean January temperature of approximately −7° to 7°C, a mean August temperature of 19° to 27°C, and an effective precipitation (total precipitation during the growing season) of over 1000 mm. The full-glacial distribution of the species on both the Sea of Japan and the Pacific coasts (35–36°N lat) indicates that in these areas the maximum possible reduction of temperature was 8.7–10.2°C in January and 6.0–7.3°C in August, and that the effective precipitation was low, being 1000–1350 mm, or 40–55% below the modern level, provided that the species has not changed its physiological characteristics. Populations of C. japonica expanded northward and upslope from their full-glacial areas of distribution immediately after late-glacial climatic amelioration. This expansion appears to have been regulated mainly by the availability of effective precipitation which became high in northeastern Honshu about 4000 yr ago. After about 2500 yr B.P., C. japonica was planted extensively by humans in moist, temperate climatic regions (excluding Hokkaido), and now has its widest distribution since at least the last full-glacial interval.

The relationship between carnassial length and latitude south is analyzed for 17 African carnivore species to determine if individuals tend to be larger in cooler climates, as predicted by Bergmann's Rule. With modern data in support, middle and late Quaternary temperatures might then be inferred from mean carnassial length in fossil samples, such as those from Equus Cave, Elandsfontein, Sea Harvest. Duinefontein, and Swartklip in the Cape Province of South Africa. One problematic aspect of the study is the use of carnassial length and latitude as necessary but imperfect substitutes for body size and temperature, respectively. For some species, another difficulty is the relatively small number of available modern specimens, combined with their uneven latitudinal spread. Still, in 14 of the species, carnassial length does tend to increase with latitude south, while mean carnassial length in the same species tends to be greater in those fossil samples which accumulated under relatively cool conditions, as inferred from sedimentologic, palynological, or geochemical data. Given larger modern samples from a wide variety of latitudes, refinement of the mathematical relationship between carnassial length and latitude in various species may even permit quantitative estimates of past temperatures in southern Africa.