Preview

The opening two chapters by John Birks and Frank Oldfield, respectively, provide a comprehensive introductory context for the chapters that follow. John Birks traces Holocene research back to its roots in the early 19th century and describes early debates, principally in Scandinavia, about the interpretation of plant remains preserved in peat bogs and their relevance to climate change. His account takes in the development of pollen analysis and radiocarbon dating, the use of transfer functions in an attempt to quantify past climate reconstruction from proxy records and the pioneering work of COHMAP (Cooperative Holocene Mapping Project) in paleoclimate modeling. He highlights the principal debates and developments in Holocene climate change research that have taken place in recent years and points specifically to the importance of understanding the spatial as well as temporal component of natural climate variability.

Frank Oldfield's chapter is concerned with the role of people in the Holocene. He stresses the need to take into account a much longer history of interactions between human activity and climate change than simply the very recent past. Using data from many different regions he shows that people, especially in the Old World, have had a major impact on land-use and land-cover over many millennia. He argues that the extent of land-cover change may have been sufficient to modify

Holocene climate variability and global warming | 3

local and regional climate and that these changes in turn were responsible for causing alterations in the hydrologic cycle and in soil erosion. The evidence for land-cover change presented is not inconsistent with Ruddiman's claim (see above) that early agriculture may have been the cause of increased atmospheric greenhouse gas concentrations over the past 8000 years and Oldfield stresses the need for further paleoecological research to test this hypothesis. Finally he reviews evidence for the interaction between climate change and human society in the past and calls for a more balanced dialogue between the physical and social science communities in debating this issue, calling on the need to develop models that couple biophysical and social systems and that acknowledge the adaptive nature of human society.

Chapter 4 by Michel Crucifix is divided into two main sections. The first describes the principles of climate modeling. He stresses the difficulty of modeling an inherently complex and chaotic system and the need for models of different kinds: conceptual, comprehensive and intermediate. He points out the importance of specifying initial conditions and boundary conditions, describes some of the problems of parameterization and the different ways in which equilibrium or transient experiments are conducted. He also indicates how paleodata are used by modelers, not only for direct comparison of output, but also, using data assimilation techniques, for providing improved model parameterization. The second section of the chapter is concerned with the results of two model applications. The first asks the question "how long will the Holocene last?", a question relevant to the debate opened by Ruddiman (see above) about the role of human activity in the early Holocene in increasing atmospheric concentrations of greenhouse gases. Output from models of intermediate complexity do not rule out the Ruddiman hypothesis in scenarios where early Holocene CO2 concentrations are allowed to fall below 240 ppmv (parts per million by volume). On the other hand projections forward from the present-day suggest that glacial conditions are not now likely to return for approximately 50 000 years.

The final section of Crucifix's chapter focuses on ocean stability through the Holocene. He argues that regional ocean instabilities, such as sudden coolings related to the reduction in deep-ocean convection, could have occurred throughout the Holocene in the North Atlantic by convective feedback related to interactions with sea-ice and atmosphere dynamics.

Eystein Jansen and colleagues review data from the North Atlantic region that indicate the Holocene "climate optimum" is recorded in many but not all marine sediment cores and that different proxies from the same core have different patterns. The differences are attributed to the seasonality of the insolation forcing and the relationship of the proxy to surface ocean stratification. The results indicate that the thermal maximum is mainly caused by orbital forcing, enhanced by sea-ice albedo feedbacks. The data also show that on shorter century to millennial time-scales variability is an important aspect of the marine climate in the high-latitude Atlantic Ocean, possibly increasing after the end of the thermal maximum. There is little evidence for stationary cyclicity in this variability and the authors conclude that the variability may be a response to long time-scale dynamics of the climate system and not necessarily to a specific external forcing factor.

Jürg Beer and Bas van Geel give an overview of the mechanisms causing natural climate change on decadal to millennial time-scales focusing especially on solar forcing. They argue that although the change in total solar irradiance over the course of an 11-year Schwabe cycle is quite small, the variability of the solar radiation is strongly wavelength dependent and that large changes in the spectral solar irradiance strongly influence photochemistry in the upper atmosphere, and in particular the ozone concentration, which may cause shifts in the tropospheric circulation systems and therefore climate. As direct measurements of solar irradiance are lacking prior to the advent of satellite technologies, evidence for centennial-scale change needs to be derived from proxy records, especially from 10Be from ice cores and 14C from tree rings.

In the second half of their chapter Beer and van Geel argue that there are a rapidly growing number of examples of Holocene climate change that point to the Sun as a major forcing factor. They present the well-known 850 bc event, equivalent to the Sub-boreal-Sub-atlantic transition in the original Blytt and Sernander scheme for the Holocene as a good example. This event is associated with increased peat bog growth and lake-level increase in north-west Europe and with changing husbandry and agricultural practices in south-central Siberia and central Africa. The beginning of this event is coincident with a significant increase in the atmospheric production of 14C. By extension they argue that as the amplification mechanisms for changing solar activity are not well understood, and therefore cannot yet be sufficiently quantified in climate models, solar forcing of climate change may be more important than has been suggested to date. They argue that if the Little Ice Age and the subsequent warming were mainly driven by changes in solar activity this component of natural forcing may well play an important role in estimating future trends in climate.

In Chapter 7, Hugues Goosse, Michael Mann, and Hans Renssen present a strong defence of the "hockey-stick" curve of Northern Hemisphere temperature trends for the past 1000 years, pointing out that since the first curve was presented (Mann et al. 1998) there are now several additional independent analyses covering the same period. All are essentially in agreement in showing anomalously high temperatures over the past few decades. Goosse et al. also show from data-model comparisons how natural (especially volcanic) forcing could explain many features of pre-19th climate variability, including the regional patterns of change associated with the North Atlantic Oscillation (NAO) and El Niño.

Goosse et al. present simulations of the past 1000 years that explore the separate and combined role of internal and forced variability. They present model results that show how temperature differences between regions could be due to spatial responses to particular forcings and/or to internal variability. They also show how progress could be made in data-model comparisons by using paleoproxy data to select the best realization in an ensemble. In this way a climate reconstruction could be derived that was consistent with the paleorecord, model physics, and the forcings.

Dirk Verschuren and Dan Charman stress the difficulty of relating past hydrologic variability on decadal to century time-scales to external forcing. Using

Holocene climate variability and global warming | 5

proxy evidence from Europe and Africa, however, they argue that a number of periods of cooler and wetter conditions inferred from peatland and lake-level changes in Europe correspond to periods of reduced solar activity. In Africa there is evidence for substantial spatial variation across the continent, but some evidence, especially in eastern Equatorial Africa, for an inverse relationship between solar activity and moisture.

Martin Claussen presents evidence that rapid climate change, capable of affecting early civilizations, occurred in North Africa during the Holocene, and that the climate at 5500 years BP was especially unstable. Earth system models are now capable of simulating rapid swings between arid and wet phases in the past but may not yet be able to reliably predict future transitions.

Finally, Ray Bradley provides a perspective on Holocene climate change and presents an array of evidence to demonstrate the relevance of understanding past climate in order to provide insights for the future. He stresses the importance of reconstructing the history of climate forcing and the need to understand the causes and consequences of rapid changes especially AUPs (abrupt, unprecedented and persistent climate anomalies), for which there are many examples, but mainly droughts, in the paleorecord.

0 0

Post a comment