Climate Dynamics (2006) 27: 677–694 DOI 10.1007/s00382-006-0153-5 B. G. Hunt The Medieval Warm Period, the Little Ice Age and simulated climatic variability Received: 19 April 2005 / Accepted: 24 March 2006 / Published online: 11 May 2006 Ó Springer-Verlag 2006 Abstract The CSIRO Mark 2 coupled global climatic caused by natural processes within the climatic system, model has been used to generate a 10,000-year simula- the inability of the model to reproduce the observed tion for ‘present’ climatic conditions. The model output hemispheric mean temperature anomalies associated has been analysed to identify sustained climatic fluctu- with these events indicates that external forcing must ations, such as those attributed to the Medieval Warm have been involved. Essentially the unforced climatic Period (MWP) and the Little Ice Age (LIA). Since no system is unable to sustain the generation of long-term external forcing was permitted during the model run all climatic anomalies. such fluctuations are attributed to naturally occurring climatic variability associated with the nonlinear pro- cesses inherent in the climatic system. Comparison of simulated climatic time series for different geographical locations highlighted the lack of synchronicity between 1 Introduction these series. The model was found to be able to simulate climatic extremes for selected observations for century The Medieval Warm Period (MWP) is generally con- timescales, as well as identifying the associated spatial sidered to be an episode of above average temperatures characteristics. Other examples of time series simulated for the years 950–1200 AD, while the Little Ice Age (LIA) by the model for the USA and eastern Russia had sim- was an episode of below average temperatures from ilar characteristics to those attributed to the MWP and 1550 to 1850 AD (Brazdil et al. 2005). However, there is the LIA, but smaller amplitudes, and clearly defined no universal agreement as regards the precise dates of spatial patterns. A search for the frequency of occur- these events, and they were not episodes of continuous rence of specified surface temperature anomalies, defined above or below average temperatures. In fact, contro- via duration and mean value, revealed that these were versy continues to exist over the reality of both the primarily confined to polar regions and northern lati- MWP and LIA. For example, Hughes and Diaz (1994) tudes of Europe, Asia and North America. Over the question whether the MWP ever really occurred, while, majority of the oceans and southern hemisphere such on the other hand, Grove (1998) has documented the climatic fluctuations could not be sustained, for reasons LIA in considerable detail. Bradley et al. (2003) have explained in the paper. Similarly, sustained sea–ice also discussed this issue. Recently, Jones and Mann anomalies were mainly confined to the northern hemi- (2004) seriously question the basis for the existence of sphere. An examination of mechanisms associated with both the MWP and the LIA. the sustained climatic fluctuations failed to identify a In IPCC (2001, Chapter 2) there is a lengthy discus- role for the North Atlantic Oscillation, the El Nin˜ o- sion of the climatic characteristics associated with the Southern Oscillation or the Pacific Decadal Oscillation. MWP and LIA, which again queries the basis for these It was therefore concluded that these fluctuations were events. The principal concerns listed are the lack of generated by stochastic processes intrinsic to the non- temporal correspondence between climatic variability in linear climatic system. While a number of characteristics different regions, regional variability itself, the modest of the MWP and the LIA could have been partially amplitude of the hemispheric mean temperature anom- alies, and the sparseness and derivative nature of the observations. Jones and Mann (2004) are even more B. G. Hunt CSIRO Marine and Atmospheric Research, PMB1, critical, and question the relevance of much of the col- Aspendale, VIC 3195, Australia loquial evidence for these events, such as the existence in E-mail: [email protected] England of vineyards during the MWP and frost fairs 678 B.G. Hunt: The Medieval Warm Period, the Little Ice Age and simulated climatic variability during the LIA. These issues primarily arise because of 40% of climatic variation was attributable to solar uncertainties in the climatic reconstructions. influences, with 50% to greenhouse gases and 10–20% On the other hand, there are many observational to internal variability. Using an energy balance model studies documenting climatic perturbations, primarily Crowley and Kim (1999) concluded that about 1/3 of temperature, that advocate the occurrence of these low frequency climatic variability was attributable to events (see, for example, Petersen 1994; Villalba 1994; solar influences, with the remainder due to greenhouse Keigwin 1996; Broecker 2001; Cronin et al. 2003 and gases and internal variability. Numerous climatic model Cook et al. 2004). A particularly comprehensive review simulations have been made, see below, that include of historical climatic fluctuations over Europe has been solar perturbations over past centuries, and these typi- presented recently by Brazdil et al. (2005). cally indicate a major contribution of such perturbations This issue is not just an academic disagreement, as it to climatic variability. has entered into the discussion of the credibility of the However, substantial questions have been raised attribution of the current warming to the greenhouse concerning the reality of the impact of solar fluctuations effect. Mann et al. (1999), based largely on proxy data, on climate. For example, Foukal et al. (2004) have have derived a northern hemispheric temperature time queried the observational evidence used to claim that series from 1000 AD to the present. This time series has a solar forcing has varied over the past few centuries. ‘hockey stick’ shape with the curve of the stick repre- While not dismissing such a possibility they demonstrate senting the global warming over the past few decades, that supporting evidence to date is inadequate because while the shaft of the stick represents the fairly station- of the derivative basis of past estimates of solar vari- ary time series of hemispheric temperature anomalies ability. More importantly, based on an analysis of the back to 1000 AD. The crucial issue is that no MWP is physical processes occurring in the sun, Lean et al. identified by the time series, thus justifying the proba- (2002) have questioned whether the proxies used to bility that 1998 was the warmest year, and the 1990s the estimate radiative forcing actually imply fluctuations in warmest decade, of the past millennium in the northern solar forcing. In addition, Damon and Laut (2004) have hemisphere (Mann et al. 1999). The significance of this strongly challenged some specific claims that solar statement is that this implies natural climatic variability forcing is the cause of the present observed global cannot account for the present observed global warm- warming. ing, as concluded, for example, by Davies and Hunt Perhaps most critically, is the recent observational (1994), from model simulations. analysis of Turney et al. (2005). They used trees in Ulster Natural, or internal, climatic variability is defined to obtain a common-dated radiocarbon calibration here to be that which occurs solely due to internal curve (to derive solar activity) and reconstruct temper- interactions of the climatic system, without any external ature values for the past 9,000 years. Such common- forcing whatsoever. This internal forcing is a conse- dating has not been available previously. They con- quence of the complex physical processes which occur, cluded that at centennial to millennial timescales North together with nonlinear interactions, that can generate Atlantic climatic variability was not forced by solar stochastic outcomes. fluctuations. This outcome raises major questions con- A major, currently unresolved, issue regarding past cerning the numerous claims that have been made con- climatic fluctuations is the individual role played by the cerning the substantial role of solar forcing of climate. various possible forcing agents: natural variability, vol- The final contributor to past climatic perturbations, canic eruptions, solar perturbations and varying green- namely internal climatic variability, is generally consid- house gas concentrations. ered to have an influence, albeit of a subsidiary nature In the case of greenhouse gases, specifically CO2, its (Rind and Overpeck 1993; Collins et al. 2002) A more concentration has increased from about 280 ppm prior substantive role has been advocated by Hunt (1998), in to about 1850 to over 370 ppm currently (IPCC 2001). disagreement with the conclusions of Bell et al. (2003). The present global warming is widely accepted as being Given the spatial and temporal limitations of obser- attributable to this recent increase in CO2, implying that vations, and the problems of deriving quantitative cli- the lower CO2 concentrations over the previous millen- matic values from proxy data, a natural recourse is to nium were associated with a generally cooler climate use global climatic models to investigate past climatic (see, for example, Rind et al. (2004)). There is also fluctuations. Jones and Mann (2004) provide a brief general agreement that large volcanic eruptions have a discussion of this issue. transient impact on climate for 2–3 years following an Jones et al. (1998), in a study primarily concerned eruption. Differing estimates of the radiative impact of with the quality of proxy data, compared such data with eruptions and their occurrences currently exist (see two model outputs. They used principal component and Crowley and Kim 1999; Zorita et al. 2005). cross spectra analyses and noted substantial differences The role of solar perturbations is, however, much between the ability of the two models to represent spatial more controversial. There have been many attempts to structures. Collins et al. (2002) compared two estimates estimate the impact of solar perturbations on climate of proxy observations with a 1,200-year simulation with over the past millennium.