Nature: Reconstruction of global surface temperature since the last glacial maximum.

Rebuilding the earth's past climate, especially the earth's climate before instrument observation, is the key to understand the natural climate variability. Geological records are important evidence of the earth's past climate change, but the results are often difficult to be continuous in space. The continuous development of climate models provides an important means for paleoclimate simulation, and its simulation results can directly give the continuous climate distribution in time and space. However, due to the uncertainty of geological records and climate models in different degrees, the comparison between them has been lacking, and even the conclusions are inconsistent. John E. Kutzbach, a pioneer of paleoclimate simulation, once pointed out that climate research must closely combine simulation with records in order to better understand the past climate and predict the future climate change. How to combine geological records with simulation results to restore the earth's historical climate more comprehensively is a hot spot in the field of paleoclimate research. In recent research, scholars from the Department of Atmospheric Sciences of the University of Washington, USA, based on the data assimilation technology in modern numerical weather forecasting, constrained and fused the simulation results of geological records (tree rings, ice cores, coral isotopes, etc.). ) to obtain the global gridded paleotemperature series in the past Millennium (Hakim et al., 20 16, Tardif et al., 2019); The application of this method has opened up a new world in the field of paleoclimate reconstruction (Tierney et al., 2020).

Recently, Dr. Mattew B. Osman and Dr. Jessica E.Tierney from the Department of Earth Sciences of the University of Arizona, Dr. Zhu Jiang from the National Center for Atmospheric Research, and Robert Tardif and Gregory Hakim from the Department of Atmospheric Sciences of the University of Washington, USA, used the data assimilation method again. Combining more than 500 sets of SST reconstruction sequences since the last 24,000 years (24 ka) with the test results of 17 oxygen isotope coupling model, the world's first set of temperature reanalysis field (LGMR) with high time resolution (200 years) is obtained, which provides an important basis for understanding the response of climate change to external forces since the last ice age.

The research results of Osman and others show that the earth has been in the cold ice period from 24 ka to 17 ka; From 16.9 ka, the global melting period was suddenly established and the global temperature rose rapidly. After the Millennium-scale New Fairy Wood Cold Event (12.8ka-11.7ka), the earth's climate entered the final transition stage to the modern interglacial period. From the early and middle Holocene to the industrial revolution, global warming slowed down, but it still maintained a significant weak warming of 0.5 (figure 1). Osman and others believe that the climate change since 24 ka can be mainly attributed to two aspects: 1) radiation forcing caused by ice sheet and greenhouse gas changes; 2) The seasonal variation of solar radiation is superimposed with the variation of Atlantic meridional circulation. Osman et al. further pointed out that compared with the historical climate in the past 24 ka, the speed and degree of modern climate warming are extraordinary, which echoes the sixth assessment report of IPCC.

Compared with the reconstruction of existing geological records, the main difference of reconstruction results of Osman et al. lies in the period from the early and middle Holocene to the pre-industrial revolution (Figure 2). The temperature series reconstructed from marine and terrestrial records show that the global surface temperature showed a downward trend from the early and middle Holocene (7 ka) to before the industrial revolution (Figure 2: red curve; Marcotte et al, 2013; Kaufman et al., 2020); The reconstruction of Osman et al. is consistent with the results of transient simulation trace (Figure 2: blue curve; Liu et al., 2009), showing a slow warming state. The explanation given by the author is that the spatial distribution of geological recording points is uneven, which leads to the deviation of global average calculation. It is worth noting that there are some challenges in the reliability of the reconstruction results of Osman et al. First, the assimilation results are all based on a single pattern of ICESM1.21.3. Although the author uses different verification methods to emphasize the reliability of the results, it really needs more model work to verify. In addition, the work of Osman and others only integrates SST reconstruction data without adding land reconstruction data, which is bound to have a certain impact on the existing results, which is also the direction that can be further explored in future research work.

In a word, the work of Osman and others is a successful example of combining geological records with simulation, which not only provides an important basis for people to study the climate change since 24 ka, but also provides a new idea for paleoclimatologists to rebuild the earth's climate in a deeper time.

Main Reference (Swipe Up and Down to View)

Zhang Zhiyong, et al. The latest progress of global climate reanalysis [J]. Beijing: Science and Technology Press, 200 1. Journal of Geophysical Research: Atmosphere, 20 16,121(12): 6744.

[10] Liu Jianmin, et al. Multi-method reconstruction of Holocene global average surface temperature [J]. Scientific data, 2020,7 (1):1-13.

Herry Liu, Liu Zhiping, He Feng, et al. Transient simulation of the last deglaciation period and its new mechanism for Bolling-Allerod climate warming [J]. Science, 2009,325 (5938): 310-314.

Marcotte, Shaken, Pu, et al. Reconstruction of regional and global temperatures in the past1,300 years [J]. Science, 2013,339 (6124):1/kloc.

Zhang Zhijun, Zhu Jun, et al. lasg, 200 1. Nature, 202 1, 599: 239-244. (original link)

Zhang Zhiyong, et al. Simulation of global climate change in 2000 [J]. Beijing: Science Press, 200 1. Past climate, 20 19,15 (4):1251-

Zhu Jun, Wang, et al. Further discussion on glacier cooling and climate sensitivity [J]. Nature, 2020,584 (7822): 569-573.