Some of the best-preserved traces of the boundary are found in southern Scandinavia, where the transition from the latest glacial stage of the Pleistocene to the Holocene was accompanied by a marine transgression. These beds, south of Gothenburg, have been uplifted and are exposed at the surface. The boundary is dated around 10,300 ± 200 years bp (in radiocarbon years). This boundary marks the very beginning of warmer climates that occurred after the latest minor glacial advance in Scandinavia. This advance built the last Salpausselkä moraine, which corresponds in part to the Valders substage in North America. The subsequent warming trend was marked by the Finiglacial retreat in northern Scandinavia, the Ostendian (early Flandrian) marine transgression in northwestern Europe.

Arguments can be presented for the selection of the lower boundary of the Holocene at several different times in the past. Some Russian investigators have proposed a boundary at the beginning of the Allerød, a warm interstadial age that began about 12,000 bp. Others, in Alaska, proposed a Holocene section beginning at 6,000 bp. Marine geologists have recognized a worldwide change in the character of deep-sea sedimentation about 10,000–11,000 bp. In warm tropical waters the clays show a sharp change at this time from chlorite-rich particles often associated with fresh feldspar grains (cold, dry climate indicators) to kaolinite and gibbsite (warm, wet climate indicators).

The Holocene Epoch resisted formal subdivision until June 2018, when the International Union of Geological Sciences (IUGS) and the International Commission on Stratigraphy divided the epoch into three stages. The start of the Greenlandian stage (11,700 to 8,300 years ago), known from Greenland ice cores, coincides with the lower boundary of the Holocene. The onset of the Northgrippian stage (8,300 to 4,200 years ago), also determined using ice cores from Greenland, coincided with a period of cooling that occurred in the North Atlantic about 8,300 years ago. In contrast, the Meghalayan stage (4,200 years ago to the present) was determined using a speleothem, or cave deposit (in this case, a stalagmite from Mawmluh Cave in Meghalaya, India). The stalagmite captured a 200-year period of worldwide drought and cooling dating to about 4,200 years ago. The climatic shift produced severe disruptions in natural resources that were felt by civilizations in the low and middle latitudes around the world, including those of ancient Egypt, Mesopotamia, and the Yangtze River Valley.

Deep oceanic deposits

The marine realm, apart from covering about 70 percent of the Earth’s surface, offers far better opportunities than coastal environments for undisturbed preservation of sediments. In deep-sea cores, the boundary usually can be seen at a depth of about 10–30 centimetres, where the Holocene sediments pass downward into material belonging to the late glacial stage of the Pleistocene. The boundary often is marked by a slight change in colour. For example, globigerina ooze, common in the ocean at intermediate depths, is frequently slightly pinkish when it is of Holocene age because of a trace of iron oxides that are characteristic of tropical soils. At greater depth in the section, the globigerina ooze may be grayish because of greater quantities of clay, chlorite, and feldspar that have been introduced from the erosion of semiarid hinterlands during glacial time.

During each of the glacial epochs the cooling of the ocean waters led to reduced evaporation and thus fewer clouds, then to lower rainfall, then to reduction of vegetation, and so eventually to the production of relatively more clastic sediments (owing to reduced chemical weathering). Furthermore, the worldwide eustatic (glacially related) lowering of sea level caused an acceleration of erosion along the lower courses of all rivers and on exposed continental shelves, so that clastic sedimentation rates in the oceans were higher during glacial stages than during the Holocene. Turbidity currents, generated on a large scale during the low sea-level periods, became much less frequent following the rise of sea level in the Holocene.

Studies of the fossils in the globigerina oozes show that at a depth in the cores that has been radiocarbon-dated at about 10,000–11,000 bp the relative number of warm-water planktonic foraminiferans increases markedly. In addition, certain foraminiferal species tend to change their coiling direction from a left-handed spiral to a right-handed spiral at this time. This is attributed to the change from cool water to warm water, an extraordinary (and still not understood) physiological reaction to environmental stress. Many of the foraminiferans, however, responded to the warming water of the Holocene by migrating poleward by distances of as much as 1,000 to 3,000 kilometres in order to remain within their optimal temperature habitats.

In addition to foraminiferans in the globigerina oozes, there are nannoplankton, minute fauna and flora consisting mainly of coccolithophores. Research on the present coccolith distribution shows that there is maximum productivity in zones of oceanic upwelling, notably at the subpolar convergence and the equatorial divergence. During the latest glacial stage the subpolar zone was displaced toward the equator, but with the subsequent warming of waters it shifted back to the borders of the polar regions.

The distribution of the carbonate plankton bears on the problem of rates of oceanic circulation. Is the Holocene rate higher or lower than during the last glacial stage? It has been argued that, because of the higher mean temperature gradient in the lower atmosphere from equator to poles during the last glacial period, there would have been higher wind velocities and, because of the atmosphere–ocean coupling, higher oceanic current velocities. There were, however, two retarding factors for glacial-age currents. First, the eustatic withdrawal of oceanic waters from the continental shelves reduced the effective area of the oceans by 8 percent. Second, the greater extent of floating sea ice would have further reduced the available air–ocean coupling surface, especially in the critical zone of the westerly circulation. According to climatic studies by the British meteorologist Hubert H. Lamb, the presence of large continental ice sheets in North America and Eurasia would have introduced a strong blocking action to the normal zonal circulation of the atmosphere, which then would be replaced by more meridional circulation. This in turn would have been appreciably less effective in driving major oceanic current gyres.