Simulating the Maunder Minimum Print
Outreach - Showcase
Written by Tom Rees   
Tuesday, 20 December 2005
ImageOur second Community Showcase article involves a simulation of the Maunder Minimum. Author Tom Rees recreates the work done in a recent paper by Shindell et al., offering his own insights on the subject of the Maunder Minimum and climate change.

Introduction

The climate of Europe in the late 17th century was characterised by generally cooler weather, punctuated by some exceptionally severe winters1. Temperatures in Central England were 1.5oC cooler than the early 20th century, and the winter of 1685 was 5.2oC cooler. Sea ice in the Baltic Sea increased, while Lisbon experienced six winters with snowfall in the period 1680-1704 (compared with only one snowfall in the second half of the 20th Century). As a result of these cooler temperatures, this time period later became known as the "Little Ice Age". However, although there is clear evidence for cooling in Europe during this time, evidence from the rest of the globe is more patchy. Estimates of average Northern Hemisphere temperatures from this time vary by as much as 0.6oC.2

A leading contender for the cause of this cooler period is a reduction in solar output known as the "Maunder Minimum".3 During this time, sunspots were generally absent and solar output fell by around 0.2%. However, other factors, such as volcanic eruptions and small changes in the greenhouse gases CO2 and methane, could have contributed to climate change. Modelling this time period is therefore interesting for two reasons. Firstly, it provides a test of the climate model's ability to recreate past climate change in a global environment that was similar to today's. Second, results from climate models can help elucidate the causes of observed climate phenomena.

The model scenario

A simple approach to simulating the Maunder Minimum was employed by Shindell et al. (2001), who ran a climate model at two levels of solar irradiance, the estimate for 1680 (the late Maunder Minimum), and the estimate for 1780 (post Maunder Minimum) until it reached equilibrium.4 In both cases, they kept greenhouse gases at a constant, pre-industrial level (their objective being to investigate the climate effects of a purely solar change). They found that, although there was generalised cooling (annual global mean surface air temperatures decreased by 0.34oC), the degree of cooling was greater in winter and greater over continents than over the sea. The model used by Shindell et al was similar to the model used in EdGCM, with the same sensitivity to increases in greenhouse gases and differing primarily in resolution (23 atmospheric layers rather than 9, and a horizontal resolution of 5ox4o). It also allowed for stratospheric ozone to respond to temperature changes (with warming in the stratosphere causing ozone degradation).

To recreate this experiment using EdGCM, two experiments were run; one with solar irradiance set at the 1780 level, the other with irradiance set at the 1680 level. Greenhouse gases were set to levels applicable to 1780 in both cases. At equilibrium the 30-year mean global surface temperatures were lower in the 1680 run by 0.44oC. This difference may be a result of the lack of ozone feedback, but the values used for the solar parameter may also have differed slightly.

Unlike the Shindell et al. result, however, the cooling is approximately uniform, with the greatest cooling occurring in the Southern Ocean as a result of increased ice cover (Figure 1). In the northern hemisphere, there is no evidence of higher continental cooling, and no evidence of amplification of the cooling during winter.
Figure 1. Annual and winter time temperatures: difference between 1780 and 1680 using solar changes only.
Figure 1. Annual and winter time temperatures: difference between 1780 and 1680 using solar changes only.

The Arctic Oscillation

The Arctic Oscillation (AO), and the closely related North Atlantic Oscillation, is a climate phenomenon that manifests as a fluctuation in sea level pressure over arctic regions compared with temperate regions.5 In its negative phase, high pressure over the Arctic and a decrease in zonal wind speeds allows cold air to penetrate into temperate latitudes, reducing temperatures (especially in the wintertime). In recent decades, global warming has been associated with a increased tendency towards a positive phase AO, and furthermore a negative phase AO would explain several of the climate features of the Maunder Minimum.

The link between the AO and global climate change is, however, not fully understood. Shindell et al. found that there was a shift to the negative phase in their simulation.4 However, the response of the AO to climate forcing depends upon the model used. The 9-layer GISS model used in EdGCM exhibits a realistic AO but, unlike the 23-layer model, it does not shift to a positive phase in conditions of increasing GHG.6 The probable explanation is that that a highly resolved stratosphere is required to successfully simulate the refraction of planetary waves.7 However, troposphere-only models can show a more modest AO response,8 and the 9-layer GISS model has been found to show an AO phase shift in response to strong changes in latitudinal SST gradients.9 Furthermore, a transient simulation of the Late Maunder Minimum using a model with low stratospheric resolution found that the AO shifted to a negative phase as climate cooled, followed by a positive phase as it warmed.10

With this in mind, the results of a second simulation using EdGCM (Run 2) are of interest. As well as adjusting solar irradiance to 1680 and 1780 values, levels of CO2 and methane were set appropriately, resulting in additional cooling (global mean difference 0.66oC). This time, however, cooling was especially evident in the northern hemisphere winter, and was accompanied by a worsening of minimum temperatures (Figure 2). Other features characteristic of a negative phase AO were evident: sea level pressure over the polar region was significantly increased (Figure 3), and zonal winds were reduced between latitudes 60 and 80oN (Figure 4). Notably, these features were only present as the simulation approached equilibrium, and disappeared at full equilibrium. Although increased sea level pressure is an expected consequence of cooler polar temperatures,4 the fact that the increase was a transient suggests that these observations are related to a phase shift in the modelled AO.
Figure 2. Winter time average and minimum air temperatures in the second run using solar and GHG changes between 1780 and 1680.
Figure 2. Winter time average and minimum air temperatures in the second run using solar and GHG changes between 1780 and 1680.
Figure 3. Wintertime sea level pressure in Run 2: difference between 1780 and 1680.
Figure 3. Wintertime sea level pressure in Run 2: difference between 1780 and 1680.
Figure 4. Zonal wind speed in Run 2: difference between 1780 and 1680.
Figure 4. Zonal wind speed in Run 2: difference between 1780 and 1680.

Conclusion

The results indicate that, in certain circumstances the AO in EdGCM can respond to plausible changes in climate forcing. However, whether such effects occurred during the Maunder Minimum is uncertain. Of the three available reconstructions of the NAO during this period, one indicates a generally negative phase while two indicate a generally positive phase.11. Investigation of the effects of transient climate forcing change based on the Maunder Minimum is a logical next step.

Bibliography
  1. Zinke J, et al. Evidence for the climate during the late Maunder Minimum from proxy data and model simulations available within KIHZ. In The KIHZ project: towards a synthesis of Holocene proxy data and climate models (eds. Fischer et al.), Springer, Heidelberg, Berlin, 2004. Available at: http://w3g.gkss.de/G/Mitarbeiter/storch/pdf/lmm.kihz.summary.pdf
  2. For a summary and a plot of the various millennial reconstructions, see Temperature record of the past 1000 years.
  3. Beckman JE and Mahoney TJ. The Maunder Minimum and Climate Change: Have Historical Records Aided Current Research? Library and Information Services in Astronomy III. ASP Conference Series, Vol. 153, 1998. U. Grothkopf, H. Andernach, S. Stevens-Rayburn, and M. Gomez (eds.). Available at: http://www.stsci.edu/stsci/meetings/lisa3/beckmanj.html.
  4. Shindell, D.T., G.A. Schmidt, M.E. Mann, D. Rind, and A. Waple 2001. Solar forcing of regional climate change during the Maunder Minimum. Science 294, 2149-2152. Available at: http://pubs.giss.nasa.gov/abstracts/2001/ShindellSchmidtM1.html.
  5. For a summary of the AO, see http://nsidc.org/arcticmet/patterns/arctic_oscillation.html. For more in-depth information, see http://horizon.atmos.colostate.edu/ao/.
  6. Shindell, D.T., R.L. Miller, G.A. Schmidt, and L. Pandolfo 1999. Simulation of recent northern winter climate trends by greenhouse-gas forcing. Nature 399, 452-455. Available at: http://pubs.giss.nasa.gov/abstracts/1999/ShindellMillerS.html.
  7. Rind, D., Ju. Perlwitz, and P. Lonergan 2005. AO/NAO response to climate change: 1. Respective influences of stratospheric and tropospheric climate changes. J. Geophys. Res. 110, D1210. Available at: http://pubs.giss.nasa.gov/abstracts/2005/RindPerlwitzL1.html.
  8. Gillett, N.P., M.R. Allen, R.E. McDonald, C.A. Senior, D.T. Shindell, and G.A. Schmidt 2002. How linear is the Arctic Oscillation response to greenhouse gases?. J. Geophys. Res. 107, no. D3. Available at: http://pubs.giss.nasa.gov/abstracts/2002/GillettAllen.html.
  9. Rind, D., Ju. Perlwitz, P. Lonergan, and J. Lerner 2005. AO/NAO response to climate change: 2. Relative importance of low and high latitude temperature changes. J. Geophys. Res. 110, D12108. Available at: http://pubs.giss.nasa.gov/abstracts/2005/RindPerlwitzL2.html.
  10. Fischer-Bruns, Cubasch U, von Storch H, Zorita HE, Gonzales-Rouco F, and Luterbacher J. Modelling the Late Maunder Minimum with a 3-dimensional OAGCM. CLIVAR Exchanges, No 25, September 2002. Available at: http://www.ifm.uni-kiel.de/other/clivar/publications/exchanges/ex25/spaper/s2515.pdf.
  11. Jones, P.D., Mann, M.E., Climate Over Past Millennia, Reviews of Geophysics, 42, RG2002, doi: 10.1029/2003RG000143, 2004. Available at: http://holocene.meteo.psu.edu/shared/articles/JonesMannROG04.pdf.
 
< Prev   Next >
EdGCM in Antarctica

EdGCM Forum Posts

Terms Of Use | EdGCM: Climate Modeling for Research and Education