interest in the community, as shown by multiple paleo-data
compilations and model-data comparison exercises (OttoBliesner et al. 2017).
The Holocene
The Holocene period started about 11 ka ago. The last major
ice sheets had not completely disappeared, but major changes had occurred since the early deglaciation, both in terms
of sea level and continental and oceanic temperatures. At
first look, the climate over these past 11 ka seems stable, but
this apparent stability hides very pronounced regional variations in the hydrological cycle, in the circulation of surface
waters (especially during the final stage of melting of the
residual ice caps), and in the general circulation of the
Mediterranean Sea, a basin surrounded by land and with
limited connections to the open ocean and thus very strongly
affected by changes in rainfall intensity over its watershed.
The Holocene is a period of major movement and
development of populations. However, for the most part of
this period, human activities still had a negligible impact on
the global environment, so the study of climate changes over
recent millennia provide a benchmark against which disturbances caused by industrial and agricultural activities can
be detected. The reconstruction of Holocene climate changes
is facilitated by the precise chronology offered by carbon-14
analysis.
The forcing of summer insolation at 65°N at the beginning of the Holocene reached more than 390 W/m
2 and
caused a global warming that would last until about 6 ka.
The temperature optimum affected the high latitudes of the
North Atlantic basin, including Iceland, the Norwegian Sea
and the Scandinavian coast (Koç et al. 1993). In the Barents
Sea, the temperature maximum was limited to the period
from 7.9 to 6.9 ka due to the dissipation of the heat brought
by the North Atlantic Drift by the melting of the surrounding
ice. At lower latitudes, the temperature increase was
accompanied by a northward shift of the Intertropical Convergence Zone (ITCZ) and a major change in monsoon
dynamics, and therefore in the atmospheric water cycle. The
increase in the thermal contrast between ocean and continent, for example, accentuated the African monsoon as far as
the center of the continent.
The study of sediment cores from both the Mediterranean
Sea and African lakes indicates the existence of major climate reorganizations. For example, before 6 ka, the Sahara
was not the wide-ranging desert that it is today, but grassland
dotted with lakes conducive to farming settlement. This
period is called the African Humid Period (AHP). Around
6 ka, this wet period ended and conditions degraded at a rate
that is still debated (Collins et al. 2017; Shanahan et al.
2016; Tierney and deMenocal (2013). The tropical vegetation of canopy forests along the rivers declined, and the
Sahelian vegetation in turn disappeared about 2.7 ka ago to
make way for the desert conditions present today. This major
change could be related to the gradual decrease in insolation
over the past 10 ka aided by the albedo feedback induced by
the gradual disappearance of vegetation. Alternatively, a
rapid termination of the AHP could have been triggered by
northern-latitude cooling combined with biogeophysical
feedbacks (Collins et al. 2017).
During this wet period, the Mediterranean Sea received
more fresh water, especially in the eastern basin (Kallel et al.
1997). The sinking of well-ventilated, shallow water masses in
winter became impossible in the Levantine basin, and bottom
waters there became completely anoxic, leading to the disappearance of benthic fauna below 800 m depth. A layer of black
sediment rich in organic matter, called a sapropel, marks this
event (Rossignol-Strick et al. 1982; Rohling et al. 2015).
Although ventilation of the eastern waters of the Mediterranean resumed at 6 ka, the deep fauna of this basin, whose
colonization rate is slow, is still very poor.
In addition to these long-term reorganizations, the
Holocene also recorded an abrupt event of short duration
8.2 ka ago. Without reaching the amplitude of the rapid and
sudden climate changes of the last ice age, this event still left
a significant imprint on northern hemisphere temperatures.
Like its glacial counterparts, the ‘8.2 ka event’ is associated
with a freshwater discharge, in this case due to the rupture of
a proglacial reservoir, Lake Agassiz, formed by the retreat of
the Laurentide ice sheet (Barber et al. 1999; Wiersma and
Renssen 2006; Hoffman et al. 2012). The sudden release of
tens of thousands of km
3 of water (estimates vary from
50,000 to 120,000 km
3 ) over just 1–5 yrs had strong consequences, such as a reduction in the SST (about 1 °C) and
salinity of the North Atlantic, a reduction of 2–6 °C in the
atmospheric temperature above Greenland, a decrease in the
temperature of air and water in the lakes of western Europe,
and a decrease in the intensity of ocean circulation for a
period of about 100 yrs after the freshwater discharge.
The study of the 8.2 event has shown that interglacial
ocean circulation, such as the one of the early Holocene,
may also be sensitive to an intense, although brief, freshwater discharge. Recent studies have pointed out that this
may also have been the case during earlier interglacial
periods (Galaasen et al. 2014).
The climate of the last two millennia has also been the
subject of much focus, since it provides a relatively
long-term perspective for recent observations from the
World Meteorological Organization (WMO) network (restricted to the last 150 years) and from satellites dedicated to
the observation of the Earth (limited to a few decades). The
reconstructions of air temperature in the northern hemisphere, used as projections for the whole planet, have primarily been based on continental data (Mann et al. 1998).
248
T. Caley et al.
compilations and model-data comparison exercises (OttoBliesner et al. 2017).
The Holocene
The Holocene period started about 11 ka ago. The last major
ice sheets had not completely disappeared, but major changes had occurred since the early deglaciation, both in terms
of sea level and continental and oceanic temperatures. At
first look, the climate over these past 11 ka seems stable, but
this apparent stability hides very pronounced regional variations in the hydrological cycle, in the circulation of surface
waters (especially during the final stage of melting of the
residual ice caps), and in the general circulation of the
Mediterranean Sea, a basin surrounded by land and with
limited connections to the open ocean and thus very strongly
affected by changes in rainfall intensity over its watershed.
The Holocene is a period of major movement and
development of populations. However, for the most part of
this period, human activities still had a negligible impact on
the global environment, so the study of climate changes over
recent millennia provide a benchmark against which disturbances caused by industrial and agricultural activities can
be detected. The reconstruction of Holocene climate changes
is facilitated by the precise chronology offered by carbon-14
analysis.
The forcing of summer insolation at 65°N at the beginning of the Holocene reached more than 390 W/m
2 and
caused a global warming that would last until about 6 ka.
The temperature optimum affected the high latitudes of the
North Atlantic basin, including Iceland, the Norwegian Sea
and the Scandinavian coast (Koç et al. 1993). In the Barents
Sea, the temperature maximum was limited to the period
from 7.9 to 6.9 ka due to the dissipation of the heat brought
by the North Atlantic Drift by the melting of the surrounding
ice. At lower latitudes, the temperature increase was
accompanied by a northward shift of the Intertropical Convergence Zone (ITCZ) and a major change in monsoon
dynamics, and therefore in the atmospheric water cycle. The
increase in the thermal contrast between ocean and continent, for example, accentuated the African monsoon as far as
the center of the continent.
The study of sediment cores from both the Mediterranean
Sea and African lakes indicates the existence of major climate reorganizations. For example, before 6 ka, the Sahara
was not the wide-ranging desert that it is today, but grassland
dotted with lakes conducive to farming settlement. This
period is called the African Humid Period (AHP). Around
6 ka, this wet period ended and conditions degraded at a rate
that is still debated (Collins et al. 2017; Shanahan et al.
2016; Tierney and deMenocal (2013). The tropical vegetation of canopy forests along the rivers declined, and the
Sahelian vegetation in turn disappeared about 2.7 ka ago to
make way for the desert conditions present today. This major
change could be related to the gradual decrease in insolation
over the past 10 ka aided by the albedo feedback induced by
the gradual disappearance of vegetation. Alternatively, a
rapid termination of the AHP could have been triggered by
northern-latitude cooling combined with biogeophysical
feedbacks (Collins et al. 2017).
During this wet period, the Mediterranean Sea received
more fresh water, especially in the eastern basin (Kallel et al.
1997). The sinking of well-ventilated, shallow water masses in
winter became impossible in the Levantine basin, and bottom
waters there became completely anoxic, leading to the disappearance of benthic fauna below 800 m depth. A layer of black
sediment rich in organic matter, called a sapropel, marks this
event (Rossignol-Strick et al. 1982; Rohling et al. 2015).
Although ventilation of the eastern waters of the Mediterranean resumed at 6 ka, the deep fauna of this basin, whose
colonization rate is slow, is still very poor.
In addition to these long-term reorganizations, the
Holocene also recorded an abrupt event of short duration
8.2 ka ago. Without reaching the amplitude of the rapid and
sudden climate changes of the last ice age, this event still left
a significant imprint on northern hemisphere temperatures.
Like its glacial counterparts, the ‘8.2 ka event’ is associated
with a freshwater discharge, in this case due to the rupture of
a proglacial reservoir, Lake Agassiz, formed by the retreat of
the Laurentide ice sheet (Barber et al. 1999; Wiersma and
Renssen 2006; Hoffman et al. 2012). The sudden release of
tens of thousands of km
3 of water (estimates vary from
50,000 to 120,000 km
3 ) over just 1–5 yrs had strong consequences, such as a reduction in the SST (about 1 °C) and
salinity of the North Atlantic, a reduction of 2–6 °C in the
atmospheric temperature above Greenland, a decrease in the
temperature of air and water in the lakes of western Europe,
and a decrease in the intensity of ocean circulation for a
period of about 100 yrs after the freshwater discharge.
The study of the 8.2 event has shown that interglacial
ocean circulation, such as the one of the early Holocene,
may also be sensitive to an intense, although brief, freshwater discharge. Recent studies have pointed out that this
may also have been the case during earlier interglacial
periods (Galaasen et al. 2014).
The climate of the last two millennia has also been the
subject of much focus, since it provides a relatively
long-term perspective for recent observations from the
World Meteorological Organization (WMO) network (restricted to the last 150 years) and from satellites dedicated to
the observation of the Earth (limited to a few decades). The
reconstructions of air temperature in the northern hemisphere, used as projections for the whole planet, have primarily been based on continental data (Mann et al. 1998).
248
T. Caley et al.