date: Fri Jul 13 15:54:48 2001
from: Keith Briffa <k.briffaatXYZxyz.ac.uk>
subject: HIHOL report
Following on from our breakfast chat ,
Below is a part draft of the report that was put together after the HIHOL
meeting . The point of sending it to you now is so you can judge whether it will constitute
the basis of a PAGES newsletter report, and if so , what you think should be done to it.
You can see that it is a document in progress so to speak, but before working further on it
we would appreciate your thoughts. The more I think about it perhaps a HIHOL newsletter
would work if we can distill some specific contributions from various participants (perhaps
the organisers of the various topics being written up for the Holocene issue? - see the
outline in the report we did for the PAGES/CLIVAR meeting ) . Anyway over to you. I am away
from the 17th to 31st July.
Keith and Dominique
Today, the paleo-community maintains its fascination with the large magnitude changes that
have occurred in the Earth's climate system as it shifts periodically from glacial to
interglacial modes. This research has led to a wealth of information concerning the roles
and associated changes in insolation, atmospheric composition, oceanic circulation, and ice
cover, and more recently to the discovery of rapid and dramatic climatic changes during the
last glacial period and the last deglaciation.
Somewhat in contrast, interest in the Holocene, the present warm period following the last
deglaciation, is gaining added momentum. There are at least two main reasons for this :
the increasing numbers of high resolution environmental records (of both climate and
potential climate forcings) from continental, ice core and oceanic sources ; and
the fact that the Holocene is recognised as a period of significance interest for
understanding the natural variability of climatic and wider environmental systems that
provide background context for the study of anthropogenic changes.
Although the polar record show the Holocene as a period of minor changes (relative to the
glacial periods) from the viewpoint of low latitudes, the Holocene has been a time of large
changes especially in hydrological conditions in the inter-tropical zone. Such changes are
critically important to society today, given the high population in these regions.
During the 24-26 October 2000, 42 scientists from 13 countries gathered to attend the
meeting of HIHOL (High Resolution Variability in the Holocene) held in L'Isle sur la
Sorgue, near Avignon, France, with the support of PAGES, SCAR and NSF.
The meeting was structured to include presentations and discussions on a number of general
themes, mainly focussing on different geographic regions associated with specific climate
climate forcing and global signals (insolation, solar variability, greenhouse gases,
volcanoes, thermohaline circulation, sea level changes)
low latitudes (tropical warm pool, Hadley cell, ENSO, monsoons)
middle latitudes (westerly zones, NAO)
high latitude, polar regions
The last afternoon was devoted to a general discussion and the organization of a special
HIHOL issue of the journal HOLOCENE, in which the latest information on the general themes
will be summarized in 7 review articles, each addressing one or more of the following
What is the "best resolved" picture of Holocene climate variability that can be synthesized
in the different regions on millennium, century and annual-to-decadal timescales?
What were the "significant" changes at these various timescales? Were they in-phase or
out-of-phase in the different regions and why?
Were there major synchronous abrupt events and what were their magnitudes - e.g., at about
8.2K calendar years ago; at 4K years ago; at 2K years ago; at 540 A.D.?
Was there a Holocene "Climate Optimum", that we prefer to call here "Thermal Maximum"?
What is the role of changing seasonal insolation and irradiance on climate changes, e.g.,
low latitude effects on monsoon variability?
How the carbon cycle changed and why?
The following is a brief summary of some of the main conclusions of the meeting concerning
potential climate forcings, the ice core, terrestrial and oceanic records, and recent
results of climate modeling of Holocene variability.
Changes of orbital parameters are precisely calculated. Short description of the main
charateristics during the Holocene (including 6 kyr) ; comparison with stage 5 and 11.
The solar activity affects the solar irradiance and thus contribute to the radiative
balance of the Earth-Atmosphere system. Direct observations show changes of the solar
irradiance of 0.15% over an 11-year solar cycle. Monitoring of solar type stars indicates
that fluctuations of up to 1% are potentially possible on longer time scales. The changes
take mainly place in the UV and infrared wavelengths of the solar spectrum. Long term
reconstructions of solar activity are based on proxies such as sunspots, aurorae and the
geomagnetic aa-index, but are limited to the past centuries. These proxies show a generally
increasing trend since 1600 AD which is interrupted from time to time by short periods
(50-100 y) of low solar activity. The most prominent one is the Maunder minimum (1645-1715
AD) a period which is characterised by an almost complete absence of sunspots. Records of
cosmogenic nuclides in ice cores (10Be, 36Cl) and tree rings (14C) reveal longer cycles of
solar variability (e.g. 90 and 205 years) and several grand minima, periods when the sun
was quiet as during the Maunder minimum.
Since the physics of the relation between solar activity and solar irradiance is not yet
fully understood it is premature to derive a solar forcing function in W m-2. However,
there is a growing number of striking coincidences between climate changes and the
occurrence of abrupt changes in solar activity such as grand minima.
The ice core record of greenhouse trace gases is now documented for CO2, CH4 and N2O. These
species experienced small long term Holocene changes prior the industrial revolution. CO2
concentrations varied by about 25 ppmv ,with a minimum of about 260 ppmv around 8.2 kyr BP
and the highest concentrations during the last 2000 years. The CH4 variability is of the
order of 150 ppbv with the lowest concentrations close to 5,000 BP and the highest during
the early and late Holocene. The N2O concentrations changed by about 10 ppbv (between 257
and 268 ppbv)) and seem to parallel the CO2 trend. Concerning the corresponding radiative
forcing: Jacqueline calculated the radiative forcing of the CO2, CH4 and N2O according to
the IPCC Draft 3 (22.10.2000), chapter 6, p.77. As input I used splines through the CO2 and
N2O record measured along the Dome C ice core. For methane I used a spline through mean
concentrations of the GRIP and the Dome C record. The maximal reduction of the radiative
forcing during the Holocene compared to pre-industrial radiative forcing was 0.38 W/m2 for
CO2, 0.08 W/m2 for CH4 and 0.04 W/m2 for N2O, respectively. This calculation don't take
into account the indirect forcing due to methane changes. DR should compare these radiative
values with those given in NATO book (1992).
Most of the information arises from the ice record, especially for the mid- to
early-Holocene. Explosive sulfur-rich volcanic eruptions not only force annual climate, but
they can also play a significant role in decadal-scale forcing throughout the Holocene.
These eruptions seem to be especially effective in enhancing and possibly lengthening cold
periods, as seems to occur during the Little Ice Age. However the climatic impact of the
abundant volcanism recorded in the earliest Holocene appears to be overhelmed by high
insolation level at this time. Periods when volcanism decreases often are characterized by
warmer temperatures as this cooling component of the climate system is removed.
The ice record
Ice cores provide information not only about climate forcings (see above), but also on the
climatic variability itself. Several records from Antarctica, Greenland, different smaller
arctic ice caps, and low latitude glaciers in China and South America document the changes
of the climate through the Holocene at different latitudes. Furthermore the borehole
temperature profiles keep a memory of the temperature changes occurring at the surface.
During the HIHOL meeting we essentially discussed the Greenland and Antarctic records. The
main source of information concerning the Holocene temperature changes over Greenland
arises from the temperature profile measured in the deep boreholes, GRIP and GISP2, drilled
in the most central part of the ice sheet (Dorthe, is it only GRIP ?). A clear thermal
maximum appears during the 8-5 kyr time interval and we also recognise well the medieval
climatic anomaly (corresponding to a warming in Greenland around xxxx BP) and the little
ice age in the xx-yy BP time interval. Thank you Dorthe for completing the dates. In
Antarctica, most of the evidence arise from the isotopic profiles measured along the ice
cores (Francoise and Tas, can you provide a list). The different profiles all show an early
thermal maximum around 7 to 9 kyr BP but no other evidence for coherent climatic changes
including around the medieval and little ice age time intervals.
The study of the changes in the mass balance of the glaciers under different latitudes (but
mainly at middle latitudes) also provides useful insights in the climate variability
through the Holocene. A striking feature is the record of " retreats " of the glaciers in
Scandinavia , the Alps and in South America than today during long periods of the Holocene.
Furthermore the mass balance history of certain glaciers (in Scandinavia and the Alps) can
provide a record of the North Atlantic Oscillation. Attle, please check this part. How long
can be the NAO record from the glaciers?
The ice record finally provides a unique information about global changes having occurred
in the biosphere and in the oceans. The main observed pre-industrial Holocene change in CO2
concentrations (25 ppmv increase from 7 to 1 kyr BP) is interpreted as a cumulative
continental release of about 195 GtC in connection with a change from warmer and wetter
mid-Holocene climate to colder and drier conditions. The CH4 record confirms, as revealed
by the terrestrial record, that the latitudinal distribution of continental ecosystems and
the hydrological cycle experienced very significant Holocene changes. Jacqueline, should we
say something here about N2O, or wait until the work is submitted?
The oceanic record
Thank you to our HIHOL "oceaners " to provide here the main conclusions (in the spirit of
this document). In my mind and according to my notes important conclusions are about the
evidence of abrupt changes (Bond events,, synchronous or not ? duration ? ; the marine
event at 8.2kyr last longer than the companion recorded in ice ) and the fact that we have
no robust evidences for changes in thermohaline circulation (except maybe around 10.3 kyr
?). Is this absence of evidences due to a lack of available appropriate proxies ?
The macro-scale relationship between climate and sea level is well established. The glacial
and interglacial cycles of the Quaternary period were characterised by sea-level
fluctuations of c. 120m. During the early Holocene, after termination of the last glacial
epoch, sea levels rose rapidly as terrestrial ice masses wasted releasing huge volumes of
water into the oceans. It is generally believed that most of this ice melt had ceased by c.
5000 to 6000 cal BP and that smaller sea level variations after this time primarily reflect
redistribution of water masses. These more subtle, natural variations in climate and sea
level that typify the late Holocene period, though close in time to the present, is a
period for which we have remarkably limited information regarding local, regional and
global sea-level change.
To rectify the above, recent studies have concentrated on the development and application
of AMS radiocarbon dating, various palaeo-environmental indices and transfer function and,
the integration of century and decadal records. The challenges of the next decade may well
be best met through cooperation of the palaeoclimate, climate modelling and late Holocene
sea-level communities. The focus in this work will be closely linked to the need to better
understand hemispheric ands regional variability in past sea-level change and climate.
(Text provided by Lloyd Keigwin which integrets the contribution written by Sicre; we are
waiting for the comments of black, Bond, deMonacal and Eystein on this text)
Several authors presented new results from marine sediment cores spanning a latitudinal
range from the tropics to the Arctic. These cores were chosen for study because of their
unusually high accumulate rates, and all reveal evidence of significant Holocene
variability in the ocean climate. At the lowest latitude, in the Cariaco Basin off
Venezuela, Black et al reported on the upwelling signal preserved in the laminated (varved)
sediments. Seasonal variability in the intertropical convergence zone leads to terrigenous
and diatom layers which are preserved on the anoxic basin floor. Changes in the abundance
of planktonic foraminifera species indicate that over the course of the past 1000 years
there has been substantial variability in the trade wind induced upwelling at decade to
In the subtropical Sargasso Sea, where sediment doesn't accumulate as rapidly as Cariaco
Basin, only centennial to millennial scale changes can be resolved. For this region,
Keigwin showed that the Little Ice Age (LIA) interval was 1-2 degrees colder than today as
well as the interval around 1000 years ago. These oscillations continue back into the
Holocene, but rates of sedimentation decrease so the marine events cannot be matched
precisely with ice core and subpolar results. However, Keigwin asserted that there is no
convincing evidence for abrupt changes in thermohaline circulation (THC) associated with
the late Holocene climate events. Likewise, he was unable to establish THC changes in deep
sea cores north of the Gulf Stream, where the LIA and other Holocene events have been
Farther to the west, off Africa, the subtropical ocean is influenced by the cool waters
advected by the Canary Current. At that location, deMenocal et al showed SST cooling during
the LIA of 4 degrees, approximately twice that observed in the Sargasso Sea. Earlier events
were recognized as well, but the LIA event is most dramatic because the core shows
increasing climate variance since the middle Holocene.
In subpolar waters off Newfoundland, Bond presented the best record yet of Holocene ice
rafted debris (IRD) events and associated changes in SST. These results show without
question the episodic nature of these features and that the LIA appears to be the most
recent example. Bond's results differ significantly from the Greenland ice core paradigm.
Although he finds the "8.2ka event" to be the most significant in the Holocene, he also
reported that the event lasted longer in the ocean record than on Greenland.
Finally, Sicre et al. reported on a latitudinal transect of cores in the North Atlantic.
Using SST estimates based on planktonic foraminiferal faunal counts and d18-O, as well as
alkenone unsaturation measurements, they found that at low latitude SSTs regularly increase
from the beginning of the deglaciation to present time. In contrast, the SST estimates at
latitudes higher than 40�N exhibit a maximum around 9-7 Kyrs and then slowly decrease
towards the top of the core. None of these records indicates millennial scales changes.
Although the benthic records do not provide any evidence for drastic changes in deep water
circulation, the isotope record of a Barents Sea sediment core shows that the temperature
optimum was associated with a maximum flux of Atlantic waters to the Barents Sea and the
Arctic Ocean. Several mid-and high- latitude cores show a short duration cooling which may
be correlated with the 8.2 kyrs event described in the GRIP/GISP record. These records may
reflect the combined effects of mean annual insolation changes and progressive melting of
continental ice sheets.
The terrestrial record
Thank you to our HIHOL "terrestrial friends" to provide here the main conclusions keeping
the "style" of the present document. According to my notes we should highlight the
existence of large hydrological changes (green Sahara,.). Although heterogeneous, they are
coherent at certain times (5.2, 4.2 kyrBP,). What should we write about 8.2 BP? How to
summarize the temporal and spatial distribution of the thermal maxima? The Little Ice Age
appears as a major Holocene event and could be an appropriate case study? We have
indications of increasing variability (temperature? hydrological cycle?) during the late
Please contribute here with your conclusive view about thermal maxima during the Holocene.
Is it fair to say that the thermal maximum occurred during early Holocene(9-7 kyr BP?) at
high latitudes than at low latitudes (6kyr BP)?
Greg Zielinski mentions that many of the ice cores from smaller ice caps in Arctic clearly
show the earliest Holocene to be the warmest and that it should not be so definitively
stated that 8-5 kyr is thermal maximum.
Modelling the Holocene climate
Waiting the contribution of Michel Crucifix. Should include time slice and transient
experiences. Modeling includes now vegetation and oceanic feedbacks.
A common point between the records of data presented during the meeting could be that they
all stress the non-linear character of the climatic system. Indeed, the proxies do not
simply reproduce the time variations of external forcings like insolation and solar
activity. On the contrary, they reveal abrupt events, changes in variability and complex
tele-connections. Hence, no wonder that all the model studies discussed during the meeting
agree on the importance of feedbacks because these are huge potential sources of
In order to understand the interactions between the different components of the climatic
system, complementary modelling approaches are necessary. On the one hand, models with
coarse resolution and including semi-empirical parameterisations for some of the physical
processes (often called models of intermediate complexity) are good tools to simulate the
transient evolution of climate over several millenia, because their relatively simple
formalism is well suited to couple synchronously the main components of the climatic system
with advantageous computational cost, such that several sensitivity studies can be
performed over long time spans. On the other hand, only coupled General Circulation Models
(GCMs) can address specific questions such as the climate variability at decadal time scale
or the impacts of sea-surface and continental temperatures as well as vegetation cover on
The meeting focused on modelling results with one model of intermediate complexity
(MoBidiC/LLN) and two GCMs (LMD-LODYC and IENA?). Furthermore, model inter-comparison
performed in the framework of the PMIP project were also discussed. The main results can be
summarised as follows.
Transient simulations suggest that mean annual temperature over Greenland passed through a
maximum between 5 and 10 kyr. The occurrence of a thermal maximum in the boreal latitudes
(between 50-75�N) can be understood as the combination of the slow retreat of Northern
Hemisphere ice sheets and the extension of boreal forest that is enhanced by high summer
insolation in 9-6 kyr BP interval. In mid and southern latitudes, inter-comparison between
GCM's show that the variations in summer sea surface temperature induced both by changes in
incoming solar radiation (mainly in Spring) and ocean circulation can either play a
positive feedback on precipitation, as in Africa and America, or a negative one, as in
Indonesia, South Africa and Australasia. The vegetation-precipitation feedback is also
thought to be a necessary mechanism to explain the Green Sahara that characterised the 9 -
5 kyr BP period. The conclusion is that vegetation and ocean feedbacks, that turn out to be
so important in past climate change, should definitely be considered in the context of
future climatic change assessments.
Finally, the model session of the meeting concludes over quite promising perspectives.
First, it becomes possible to compare the internal variability generated in GCMs with
high-resolution palaeo-climate indicators. This exercise should help both model and data
communities to understand more in depth the mechanisms involved in climatic variations at
the decadal scale or less. Besides, the recent strides in carbon cycle modelling enable us
to consider the variations in CO2 concentration as a response of the climate system,
associated to specific feedbacks, as opposed to an external forcing.
The main " holes ", which have not been enough discussed during the meeting or which are
Ice core records from the arctic ice caps and tropical glaciers.
Chemistry and pollen ice-core records.
Global sea level record: general trends, uncertainties
The Southern Hemisphere
Professor Keith Briffa,
Climatic Research Unit
University of East Anglia
Norwich, NR4 7TJ, U.K.