Abstract

The effect of ice sheet topography on the East Asian summer monsoon (EASM) during the Last Glacial Maximum is studied using CCSM3 in a hierarchy of model configurations. It is found that receding ice sheets result in a weakened EASM, with the reduced ice sheet thickness playing a major role. The lower ice sheet topography weakens the EASM through shifting the position of the midlatitude jet, and through altering Northern Hemisphere stationary waves. In the jet shifting mechanism, the lowering of ice sheets shifts the westerly jet northward and decreases the westerly jet over the subtropics in summer, which reduces the advection of dry enthalpy and in turn precipitation over the EASM region. In the stationary wave mechanism, the lowering of ice sheets induces an anomalous stationary wave train along the westerly waveguide that propagates into the EASM region, generating an equivalent-barotropic low response; this leads to reduced lower-tropospheric southerlies, which in turn reduces the dry enthalpy advection into East Asia, and hence the EASM precipitation.

1. Introduction

Continental ice sheets have experienced large glacial–interglacial cycles in the last 2 million years during the Pleistocene. At their maximum, such as the Last Glacial Maximum (LGM) (21 000 years ago), the Laurentide Ice Sheet (LIS) covered much of North America and extended southward to approximately 40°N, while the Eurasian Ice Sheet (EIS) covered much of northern Europe. These ice sheets can reach a height of 3–4 km with a total volume equivalent to ~130 m in global sea level. The waxing and waning of the ice sheets have long been hypothesized as a major forcing factor on global climate (e.g., Clark et al. 1999). This hypothesis has been corroborated by climate model simulations, which showed a significant impact of ice sheets on the climate over North America (e.g., Manabe and Broccoli 1985; Anderson et al. 1988), the North Atlantic (e.g., Li and Battisti 2008; Pausata et al. 2011; Hofer et al. 2012; Zhu et al. 2014), and the tropical ocean–atmosphere system (e.g., Chiang et al. 2003; Chiang and Bitz 2005; Lee et al. 2015; Lu et al. 2016).

Observational and modeling studies also suggest a potential role of ice sheets on global monsoon systems, in spite of the recognition that the primary driving forcing factor for monsoon evolution is the orbital forcing and the associated seasonal insolation contrast during the Pleistocene (Kutzbach 1981; Kutzbach and Street-Perrott 1985; Ruddiman 2006; Cheng et al. 2009). In observational studies, traditionally, the impact of ice volume on monsoons has often been thought of conceptually from a simple thermodynamic view: an ice sheet retreat in the Northern Hemisphere (NH) reduces surface albedo and therefore leads to a warming over the NH, which will enhance the land–sea thermal contrast and, in turn, the summer monsoon. Based on this simple thermodynamic view, ice sheets have been speculated to contribute to the phase difference between some Asian monsoon records and the precessional forcing in the Pleistocene (e.g., Clemens and Prell 1990, 2003), including the Holocene (e.g., Overpeck et al. 1996). This simple thermodynamic view, however, is challenged by recent modeling studies, especially of the East Asian summer monsoon (EASM). In an Earth system model of intermediate complexity (LOVECLIM; Yin et al. 2008, 2009), the retreat of ice sheet weakens the EASM, just opposite to what expected from the simple thermodynamic view. In a study of EASM in MIS-13, Yin et al. (2008) showed that the presence of Eurasian and North American ice sheets together enhances the summer precipitation over East China through a stationary wave train propagating eastward from the ice sheets. This result is further confirmed in an AGCM (ARPEGE) experiment by Sundaram et al. (2012). This result appears to be also consistent with the evolution of the EASM forced by the ice sheet retreat in the last 19 000 years in a coupled GCM CCSM3. Wen et al. (2016) showed that the retreating ice sheet in CCSM3 weakens the EASM monsoon wind over North China. These results pose the question: why is the ice sheet impact on the EASM different from what is expected from the simple thermodynamic view?

Conceptually, there are potentially several problems with the simple thermodynamic view. First, many modeling studies suggest that the dynamic effects of the ice sheet topography, rather than the simple thermodynamic effects of the ice sheet albedo, are the dominant impacting factor of the ice sheet on the atmospheric and oceanic circulation and, in turn, regional climate in the NH (e.g., Li and Battisti 2008; Pausata et al. 2011; Lee et al. 2015; Lu et al. 2016). The large and tall LIS splits the westerly jet over North America, leading to a southward displacement of the jet stream and storm track locally over North America and the nearby downstream region over the North Atlantic (Manabe and Broccoli 1985; Anderson et al. 1988; Hofer et al. 2012; Ullman et al. 2014). The ice sheet topography can also exert a significant impact remotely on the climate and climate variability over the tropical Pacific (Lee et al. 2015; Lu et al. 2016) and the EASM (Yin et al. 2008, 2009; Sundaram et al. 2012). Second, the direct thermodynamic effect (e.g., warming associated with the reduced albedo and ice sheet thickness) may be reduced or even overwhelmed by cooling caused by the removal of the ice sheet. Indeed, in recent fully coupled ocean–atmosphere model simulations, the lowering of NH ice sheets leads to a broad cooling over the North Atlantic, the North Pacific, and northeastern Asia, in spite of a local warming over the North America continent and the midlatitude North Atlantic and North Pacific; this broad cooling appears to be caused by the changed planetary wave and sea ice expansion (Zhu et al. 2014; Ullman et al. 2014; Lu et al. 2016). Third, recent fully coupled ocean–atmosphere model simulations showed that a retreat of ice sheet can also induce a secondary dynamic impact associated with the change of the Atlantic meridional overturning circulation (AMOC). Indeed, in some CGCMs, a lowering or removal of LIS tends to weaken the AMOC (Zhu et al. 2014; Ullman et al. 2014; Klockmann et al. 2016; Sherriff-Tadano et al. 2018; Galbraith and de Lavergne 2019), which will then induce a bipolar seesaw response in SST and associated climate changes. Finally, it should be pointed out that, in the real world reconstructions, it is difficult to separate the ice sheet effect from the CO2 effect. Indeed, during the major glacial cycles of the last 800 000 years, ice volume roughly covaries with the atmospheric CO2 concentration level, with an interglacial (glacial) period corresponding to higher CO2/less ice sheet (lower CO2/more ice sheet) (e.g., Petit et al. 1999; Shackleton 2000). This covariation makes it difficult to isolate the climate impact of the ice sheet from that of CO2 in the reconstructions. The isolation is made even more difficult by the potentially opposite effects of the ice sheet and CO2 on some regional climate systems. For example, a retreat of ice sheet weakens the AMOC, opposite to the enhancement effect of AMOC by the slow deglacial increase of CO2 (Zhu et al. 2014; Ullman et al. 2014; Klockmann et al. 2016; Sherriff-Tadano et al. 2018; Galbraith and de Lavergne 2019); retreating glacial ice sheets tend to intensify El Niño–Southern Oscillation (ENSO), whereas increasing deglacial atmospheric CO2 concentrations weaken ENSO (Liu et al. 2014; Lu et al. 2019); and a retreat of ice sheet weakens the EASM, opposite to the strengthened EASM by the deglacial rise of CO2 level (Wen et al. 2016).

Here, we will focus on the impact of ice sheet topography on the EASM during the LGM. This is a follow-up study of Wen et al. (2016), which has shown a decreasing EASM in response to the ice sheet retreat in the transient TRACE-ICE simulation in CCSM3. There is, however, no analysis of the mechanism of the response of the EASM. The objective of this paper is a thorough understanding of the mechanism of the ice sheet impact on the EASM in CCSM3. This is done with further analyses and sensitivity experiments in a hierarchy of models. Consistent with Yin et al. (2008, 2009) and Sundaram et al. (2012), our experiments show that the ice sheet retreat decreases the EASM; furthermore, this impact is caused primarily dynamically by the orographic effect of the NH ice sheets via jet migration and stationary Rossby waves. The paper is arranged as follows. After describing our model and experiments in section 2, we present the general climate response over the NH in section 3. In section 4, we show that the removal of NH ice sheet topography in summer decreases the zonal-mean zonal wind over the subtropics and generates an equivalent-barotropic low geopotential height anomaly accompanied by reduced lower-tropospheric southerlies over East Asia. In section 5, we show how the atmospheric circulation changes discussed in section 4 contribute to the reduced EASM precipitation from the perspective of moist static energy. Separate influences of the Laurentide and Eurasian ice sheets on the EASM and role of the ocean are presented in sections 6 and 7, respectively. A summary is given in section 8.

2. Model and experiments

We study the impact of ice sheet on the EASM using CCSM3 (Collins et al. 2006). CCSM3 is composed of the Community Atmospheric Model version 3 (CAM3) at T31 resolution (96 × 48 horizontal grids cells, approximately 3.75° resolution) with 26 hybrid coordinate vertical levels, the Community Land Model version 3 with dynamic vegetation also at T31 resolution, the Parallel Ocean Program (POP) at a nominal 3° horizontal resolution with 25 vertical levels, and the Community Sea Ice Model version 5, a dynamic–thermodynamic sea ice model on the same grid as the ocean model. The model simulates the current climatology well in both the summer and winter. CCSM3 at T31 resolution simulates the EASM reasonably well in terms of the atmospheric wind field, with the EASM characterized by a southeasterly penetrating deep into northern China [see supplementary Figs. 1a,c in Wen et al. (2016)]. Figure 1 shows the present-day summer stationary waves in the slab ocean version of CCSM3, which will be used for sensitivity experiments, and the NCEP–DOE Reanalysis II data (Kanamitsu et al. 2002). It is seen that both the spatial patterns and amplitudes of the stationary waves are reproduced by the model. Notably, the summer stationary waves feature a baroclinic structure, with surface ridges and upper-level troughs over the Pacific and Atlantic and a surface trough and an upper-level ridge over Asia, which is likely induced mainly by diabatic heating [as in Ting (1994)].

Fig. 1.

Geopotential height at lower and upper levels in JJA for present-day simulations. Eddy geopotential height at 200 hPa (contour; gpm) and 850 hPa (shading; gpm) for (a) NCEP–DOE Reanalysis II and (b) CAM3 + slab ocean. Contour intervals are 40 gpm, and dashed contours indicate negative values.

Fig. 1.

Geopotential height at lower and upper levels in JJA for present-day simulations. Eddy geopotential height at 200 hPa (contour; gpm) and 850 hPa (shading; gpm) for (a) NCEP–DOE Reanalysis II and (b) CAM3 + slab ocean. Contour intervals are 40 gpm, and dashed contours indicate negative values.

First, we analyze a transient simulation of the global climate of the last 19 000 years forced only by the deglacial evolution of the ice sheets (TRACE-ICE; He et al. 2013) in CCSM3. In TRACE-ICE, the ice sheet configuration is altered every 1000 years from 19 to 17 ka BP and every 500 years after 17 ka BP based on the ICE-5G reconstruction (Peltier 2004), with all other forcing factors and boundary conditions fixed at their values at 19 ka. Second, we assess the climatic effect of ice sheet topography in CCSM3 with six sensitivity experiments, in which ice sheet thickness in the Northern Hemisphere is set at 100%, 80%, 60%, 40%, 20%, and 0% of the LGM ice sheet thickness value (i.e., elevation differences between LGM and preindustrial levels). In all simulations, orbital parameters, greenhouse gas concentrations, coastlines, and sea level remain the same as those at 22 ka. All simulations are initialized with the LGM vegetation. Each simulation is integrated for 200 years, and the last 50 years are used as model climatology for study.

To further isolate the role of oceanic dynamic response, we conducted two sensitivity experiments in the atmosphere model CAM3 coupled with a slab ocean model (SLAB). The slab ocean contains an ocean mixed layer coupled with a thermodynamic sea ice, with the mixed layer depth and global climatology heat flux (Q flux) prescribed as the seasonal cycle of their values in the 100% fully coupled simulation. Therefore, SLAB considers the effect of local oceanic thermodynamic response by fixing the oceanic heat transport as the climatology. The difference between the fully coupled CCSM3 and SLAB is the effect of ocean dynamics. Two sets of sensitivity experiments are performed in SLAB with different combinations of 100% and 0% heights of the two major ice sheets in the form 100L + 100E, 0L + 0E (Table 1), where letters L and E indicate the LIS and EIS, respectively. For example, 100L + 100E represents the experiment with 100% LIS and 100% EIS, while 0L + 0E represents the experiment with 0% LIS and 0% EIS. We can obtain the combined effects of LIS and EIS on the EASM as the difference between 0L + 0E and 100L + 100E experiments (“noLIS&noEIS”) as shown in Table 1. For the SLAB model, each simulation is integrated for 100 years, and the last 50 years are analyzed as model climatology. Note that the extent of ice sheet in all of the sensitivity experiments in CCSM3 and SLAB is the same as in the LGM climate by keeping land surface properties the same as in 100%, as such the albedo effect of the ice sheet remains the same in all experiments and thus is removed between different experiments.

Table 1.

Effects in sensitive experiments in the SLAB model.

Effects in sensitive experiments in the SLAB model.
Effects in sensitive experiments in the SLAB model.

3. Climate response to ice sheet forcing

Our experiments show that a deglacial ice sheet retreat weakens the EASM, with the reduced ice sheet thickness playing a major role. Figure 2a shows the evolution of ice sheet volume in TRACE-ICE experiment. The volume of ice sheets in the Northern Hemisphere decreases by an equivalent sea level of about 117 m from 19 to 7 ka BP. In contrast, the changes in the ice sheet volume in the Southern Hemisphere are much smaller, only of 20-m sea level rise. We choose the EASM index as the summer (June–August) southerly anomalies at 850 hPa averaged in East China (110°–120°E, 20°–40°N; 18 grid points) (Wen et al. 2016) and calculate this index in the TRACE-ICE simulation (red line in Fig. 2b) as well as the sensitivity experiments in the fully coupled model (black circles in Fig. 2b) and SLAB model (100L + 100E and 0L + 0E; blue squares in Fig. 2b). The red line indicates that the EASM weakens in response to retreating ice sheets (reduced albedo and reduced ice sheet thickness). As the ice sheet thickness decreases from 100% to 0% of the LGM level in the sensitivity experiments (black circles, associated with the effect of reduced ice sheet thickness only), the strength of EASM shows a weakening trend similar to that in the TRACE-ICE simulation (the red line), suggesting the dominant role of ice sheet topography on the EASM.

Fig. 2.

Monsoon index in TRACE-ICE and various simulations. (a) The evolution of ice sheet volume (in equivalent sea level height) in the Northern Hemisphere (solid) and Southern Hemisphere (dashed). (b) EASM wind index in TRACE-ICE (red) and various simulations (black circles, from left to right representing simulation 100%, 80%, 60%, 40%, 20%, 0%; blue squares, representing simulation 100L + 100E and 0L + 0E, respectively). The x coordinates of the black circles and blue squares are determined according to (a).

Fig. 2.

Monsoon index in TRACE-ICE and various simulations. (a) The evolution of ice sheet volume (in equivalent sea level height) in the Northern Hemisphere (solid) and Southern Hemisphere (dashed). (b) EASM wind index in TRACE-ICE (red) and various simulations (black circles, from left to right representing simulation 100%, 80%, 60%, 40%, 20%, 0%; blue squares, representing simulation 100L + 100E and 0L + 0E, respectively). The x coordinates of the black circles and blue squares are determined according to (a).

To understand the response of the EASM to ice sheet topography, we first briefly discuss the hemispheric climate response in the NH to ice sheet orographic forcing, as shown in Figs. 3a,b and 4a (experiment 0%–100% in the fully coupled model). As expected, as the ice sheet thickness decreases from 100% to 0%, northern North America features a notable warming associated with the lapse rate (Fig. 3b). Interestingly, however, the rest of the NH, such as Alaska, Beringia, and northeastern Asia, is dominated by a cooling response, consistent with annual mean surface temperature anomalies in Lee et al. (2015) and Ullman et al. (2014). We also find a cooling anomaly southeast of the Eurasian Ice Sheet, also consistent with previous modeling studies, such as Manabe and Broccoli (1985), Abe-Ouchi et al. (2007), and Löfverström et al. (2014). In the absence of NH ice sheet topography, the Northern Hemisphere upper-tropospheric circulation is less meandering (Löfverström et al. 2014), as shown by the opposite sign of the climatology and anomaly of eddy geopotential height at 200 hPa (Fig. 3a). The reduction of the ice sheet topography generates a negative geopotential height anomaly over western Canada and positive anomalies over Greenland, the United States, and southeastern Canada at 200 hPa, suggesting a weakened climatological stationary wave [Fig. 3a, consistent with Fig. 7c in Hoskins and Karoly (1981)]. The geopotential height anomalies show an anomalous wave train with negative anomalies over the North Atlantic and the eastern European plain and positive anomalies over Europe, Lake Baikal, and the North Pacific. The East Asian region (around 110°–130°E, 30°–60°N) exhibits a relatively lower height in the 0% simulation relative to the 100% simulation, associated with the reduced precipitation (Fig. 3b) and the reduced southerlies at 850 hPa (Fig. 4a). Therefore, the EASM weakens in response to the reduced NH ice-sheet thickness in both the precipitation and wind anomalies.

Fig. 3.

Anomalies induced by reduced NH ice sheet thickness in summer in (a),(b) fully coupled experiments (0% minus 100%), (c),(d) SLAB experiments [(0L + 0E) minus (100L + 100E)], and (e),(f) AGCM experiments [(0L + 0E_AGCM) minus (100L + 100E_AGCM)] for the following atmospheric variables in JJA: (left) eddy geopotential height at 200 hPa (Z200; shading; gpm) and (right) precipitation (shading; mm day−1) and surface air temperature (SAT; contours; K). Contours in (a), (c), and (e) represent Z200 climatology at LGM in corresponding models; contour intervals are 30 gpm; dashed contours indicate negative values. In (b), (d), and (f), red solid contours are for SAT anomalies of 1, 2, 3, 6, 10, and 15 K, indicating warming anomalies; blue dashed contours are for SAT anomalies −1, −2, −3, −6, −10, and −15 K, indicating cooling anomalies. The model domains for calculating the EASM meridional wind index at 850 hPa (110°–120°E, 20°–40°N) are represented in the blue boxes.

Fig. 3.

Anomalies induced by reduced NH ice sheet thickness in summer in (a),(b) fully coupled experiments (0% minus 100%), (c),(d) SLAB experiments [(0L + 0E) minus (100L + 100E)], and (e),(f) AGCM experiments [(0L + 0E_AGCM) minus (100L + 100E_AGCM)] for the following atmospheric variables in JJA: (left) eddy geopotential height at 200 hPa (Z200; shading; gpm) and (right) precipitation (shading; mm day−1) and surface air temperature (SAT; contours; K). Contours in (a), (c), and (e) represent Z200 climatology at LGM in corresponding models; contour intervals are 30 gpm; dashed contours indicate negative values. In (b), (d), and (f), red solid contours are for SAT anomalies of 1, 2, 3, 6, 10, and 15 K, indicating warming anomalies; blue dashed contours are for SAT anomalies −1, −2, −3, −6, −10, and −15 K, indicating cooling anomalies. The model domains for calculating the EASM meridional wind index at 850 hPa (110°–120°E, 20°–40°N) are represented in the blue boxes.

Fig. 4.

Wind anomalies at 850 hPa (vectors; m s−1) induced by reduced NH ice sheet topography in summer in (a),(b) fully coupled experiments (0% minus 100%), (c),(d) SLAB experiments [(0L + 0E) minus (100L + 100E)], and (e),(f) AGCM experiments [(0L + 0E_AGCM) minus (100L + 100E_AGCM)].

Fig. 4.

Wind anomalies at 850 hPa (vectors; m s−1) induced by reduced NH ice sheet topography in summer in (a),(b) fully coupled experiments (0% minus 100%), (c),(d) SLAB experiments [(0L + 0E) minus (100L + 100E)], and (e),(f) AGCM experiments [(0L + 0E_AGCM) minus (100L + 100E_AGCM)].

Ocean dynamics plays a relatively minor role in the ice sheet impact in the coupled model. Indeed, when the ice sheet thickness is reduced from 100% to 0%, EASM weakens in SLAB by a similar magnitude (blue squares in Fig. 2b) as in the fully coupled CCSM3 (black circles in Fig. 2b). Furthermore, the overall climate response over the NH land to ice sheet orographic forcing is similar in SLAB (Figs. 3c,d and 4b) and CCSM3 (Figs. 3a,b and 4a). Therefore, we will focus on the direct orographic influence of the ice sheets by analyzing SLAB simulations in sections 4, 5, and 6, leaving the discussion on the additional indirect influence associated with ocean dynamics to section 7.

4. Atmospheric circulation changes in response to the direct orographic influence of the ice sheets

To understand the direct orographic influence of the ice sheets on the EASM, we show the three-dimensional structure of the global atmospheric anomalies after the removal of both LIS and EIS, that is, the “noLIS_noEIS” effect calculated as the difference of experiments (0L + 0E) − (100L + 100E) (Table 1). The diminished topography of NH ice sheets leads to a weaker and more northward Asian-Pacific jet and North Atlantic jet [Figs. 5a,c,e; consistent with Löfverström et al. (2014)]. The difference of zonal winds in the troposphere shows a dipole mode anomaly: zonal winds are strengthened in high latitudes and weakened in the subtropics (the latitude of EASM), with the wind anomaly centers located over the North Pacific and North America (Figs. 5e,f). The summer North Atlantic jet is featured with a more zonal orientation in 0L + 0E experiment compared to a southwest–northeast tilt in the 100L + 100E experiment at 200 and 850 hPa (Figs. 5a–d; consistent with Löfverström et al. 2014).

Fig. 5.

JJA eddy geopotential height (shading; gpm) and jet (contours; m s−1) at (left) 200 and (right) 850 hPa: (a),(b) 100L + 100E simulation, (c),(d) 0L + 0E simulation, and (e),(f) their differences [(0L + 0E) minus (100L + 100E)]. Contour intervals are 10 m s−1 in (a) and (c), 4 m s−1 in (b), (d), and (e), and 2 m s−1 in (f); dashed contours indicate negative values.

Fig. 5.

JJA eddy geopotential height (shading; gpm) and jet (contours; m s−1) at (left) 200 and (right) 850 hPa: (a),(b) 100L + 100E simulation, (c),(d) 0L + 0E simulation, and (e),(f) their differences [(0L + 0E) minus (100L + 100E)]. Contour intervals are 10 m s−1 in (a) and (c), 4 m s−1 in (b), (d), and (e), and 2 m s−1 in (f); dashed contours indicate negative values.

When the ice sheet thickness is removed, the NH upper-tropospheric circulation is less meandering and an anomalous stationary wave train is generated in the midlatitudes at 200 hPa (Fig. 5e) similar to that in the fully coupled simulations (Fig. 3a), confirming the dominant role of the direct orographic influence of ice sheet. The geopotential height anomaly shows clearly a midlatitude stationary wave train with four anticyclonic anomalies lying over the United States and southeastern Canada, Europe, Lake Baikal, and the midlatitude North Pacific and four cyclonic anomalies over western Canada, the North Atlantic, the east European plain, and East Asia. These anomalies exhibit an equivalent barotropic structure, as seen in the zonal–vertical section of eddy geopotential height anomalies (Fig. 6a). Figure 6a shows a clear wave train of zonal wavenumber 4. Forced stationary Rossby waves excited by ice sheet topography can radiate wave energy to the east of the ice sheet and into the far field. Following the theory of Hoskins and Ambrizzi (1993), Rossby waves can be trapped in westerly jets as a waveguide. The perturbations over/near the LIS and EIS are transmitted to East Asia along the waveguide. The energy of the stationary wave propagates eastward into East Asia with an equivalent-barotropic low response (Figs. 5e,f; Figs. 6a,b), accompanied by reduced lower-tropospheric southerlies and a weakened EASM (Fig. 6b). At 850 hPa, a low eddy geopotential height anomaly is generated over the East China Sea (Fig. 7a), accompanied by a weakened EASM characterized by reduced lower-tropospheric southerlies and, in turn, reduced moisture import (Fig. 7c), descending air (Fig. 7b), and anomalous dryness (Fig. 7d) in East Asia.

Fig. 6.

(a) Zonal–vertical section (averaged between 35° and 50°N) of eddy geopotential height anomaly [shading, (0L + 0E) minus (100L + 100E); gpm]. (b) Meridional–vertical section (averaged between 110° and 120°E) of anomalies [(0L + 0E) minus (100L + 100E)] of meridional–vertical wind (arrow) and eddy geopotential height (shading; gpm). In (a) and (b) the eddy geopotential height climatology (100L + 100E) is superimposed with contours (every 20 gpm). Scaling of meridional–vertical wind is given at the top-right corner of (b). (Units for the horizontal component are m s−1 and for the vertical component 10−2 Pa s−1.) The grayed out regions show the surface topography.

Fig. 6.

(a) Zonal–vertical section (averaged between 35° and 50°N) of eddy geopotential height anomaly [shading, (0L + 0E) minus (100L + 100E); gpm]. (b) Meridional–vertical section (averaged between 110° and 120°E) of anomalies [(0L + 0E) minus (100L + 100E)] of meridional–vertical wind (arrow) and eddy geopotential height (shading; gpm). In (a) and (b) the eddy geopotential height climatology (100L + 100E) is superimposed with contours (every 20 gpm). Scaling of meridional–vertical wind is given at the top-right corner of (b). (Units for the horizontal component are m s−1 and for the vertical component 10−2 Pa s−1.) The grayed out regions show the surface topography.

Fig. 7.

Atmospheric anomalies induced by reduced NH ice sheet thickness [experiment (0L + 0E) minus (100L + 100E)] over East Asia in summer for (a) eddy geopotential height (gpm) at 850 hPa; (b) vertical velocity −ω (Pa s−1) at 500 hPa; (c) wind (vectors; m s−1) at 850 hPa; (d) precipitation (shading; mm day−1). The model domain for calculating the EASM meridional wind index at 850 hPa (110°–120°E, 20°–40°N) is represented by the blue box in (c). Pink shadings and black dots represent anomalies that are statistically significant at a 95% confidence level.

Fig. 7.

Atmospheric anomalies induced by reduced NH ice sheet thickness [experiment (0L + 0E) minus (100L + 100E)] over East Asia in summer for (a) eddy geopotential height (gpm) at 850 hPa; (b) vertical velocity −ω (Pa s−1) at 500 hPa; (c) wind (vectors; m s−1) at 850 hPa; (d) precipitation (shading; mm day−1). The model domain for calculating the EASM meridional wind index at 850 hPa (110°–120°E, 20°–40°N) is represented by the blue box in (c). Pink shadings and black dots represent anomalies that are statistically significant at a 95% confidence level.

5. Atmospheric dynamics of the reduced EASM precipitation

We now focus on the mechanism of the weakening EASM from the moist static energy (MSE) perspective and show that the reduced horizontal advection of dry enthalpy contributes predominantly to the decreased precipitation in the EASM region. Sampe and Xie (2010) used thermodynamic energy equation at 500 hPa to illustrate the dominant role of midtropospheric temperature advection in inducing ascending motion along the mei-yu–baiu rainband. Previous work has emphasized the advantages of the MSE budget over the dry thermodynamic equation in the understanding of tropical–subtropical circulations, because the MSE perspective takes into account the effects of temperature, humidity, and diabatic processes as well as the interaction between moist convection and the circulation comprehensively (e.g., Chou and Neelin 2003; Neelin 2007). Chen and Bordoni (2014) indicated that positive horizontal moist enthalpy advection, primarily dry enthalpy advection, plays an essential role in sustaining the mei-yu rainfall band in a region of otherwise negative net energy input into the atmosphere. This role of dry enthalpy advection is also found in all our simulations (not shown). By comparing vertical integral of horizontal moist enthalpy advection vE¯ anomaly (Fig. 8a) with the separate contributions from the dry enthalpy (Fig. 8b) and latent energy (Fig. 8c), it is clear that dry enthalpy advection anomaly dominates the MSE advection anomaly and in turn the ascending motion and precipitation changes in the EASM region (Figs. 7b,d and 8b). This effect of column-integrated horizontal dry enthalpy advection anomaly (Fig. 8b) is actually very similar to the midtropospheric temperature advection anomaly suggested by Sampe and Xie (2010), confirming the value of the MSE budget analysis.

Fig. 8.

Anomalies induced by reduced NH ice sheet thickness [experiment (0L + 0E) − (100L + 100E)] in JJA for vertical integral of (a) horizontal moist enthalpy advection vE¯ (W m−2), (b) horizontal dry enthalpy advection cpvT¯ (W m−2), and (c) latent energy advection Lυvq¯ (W m−2). Black dots represent anomalies which are statistically significant at a 95% confidence level.

Fig. 8.

Anomalies induced by reduced NH ice sheet thickness [experiment (0L + 0E) − (100L + 100E)] in JJA for vertical integral of (a) horizontal moist enthalpy advection vE¯ (W m−2), (b) horizontal dry enthalpy advection cpvT¯ (W m−2), and (c) latent energy advection Lυvq¯ (W m−2). Black dots represent anomalies which are statistically significant at a 95% confidence level.

The advection of dry enthalpy can be further shown to be caused by the stationary eddy–mean flow interaction. We decompose the horizontal dry enthalpy advection anomaly into mean, transient, and stationary eddy fluxes following Chen and Bordoni (2014). Therefore, the response of the dry enthalpy advection can be decomposed as

 
δcpvT¯=δcp[v¯][T¯]+δcp[v¯]T*¯+δcpv*¯[T¯]+δcpv*¯T*¯+δcpvT¯.
(1)

where δ(·) is defined as the difference of (0L + 0E) − (100L + 100E) and ⟨·⟩ is defined as a vertical mass integral [i.e., ∫(·)dp/g]. The prime denotes the deviation from the monthly time mean for each individual year (e.g., u=uu¯), and the star denotes the deviation from the global zonal mean (e.g., u* = u − [u], where the square brackets indicate the global zonal mean). The five terms on the right side represent, in order, the zonal-mean dry enthalpy advection by the zonal-mean flow, the stationary eddy dry enthalpy by the zonal-mean flow, the zonal-mean dry enthalpy by the stationary eddy velocity, the advection of the stationary eddy dry enthalpy by the stationary eddy velocity, and the advection of the transient eddy dry enthalpy by the transient eddies.

In the region associated with EASM precipitation (110°–120°E), we calculate the zonal-mean anomaly of each term in Eq. (1) (Fig. 9). When the ice sheet is lowered, total dry enthalpy advection δcpvT¯ reduction between 27° and 37°N (black line) results mainly from the reduction in the advection of the stationary eddy dry enthalpy by the zonal-mean flow δcp[v¯]T*¯ (red line) and the advection of the zonal-mean dry enthalpy by the stationary eddy velocity δcpv*¯[T¯] (blue line). These two terms represent stationary eddy–mean zonal flow interactions.

Fig. 9.

Eddy decomposition of the zonal-mean vertical integral of the horizontal dry enthalpy advection anomaly δcpvT¯ (W m−2) between 110° and 120°E in JJA. Gray shading represents latitudes where EASM precipitation changes most (27°–37°N), which is the region of interest here. Crosses represent anomalies that are statistically significant at a 95% confidence level.

Fig. 9.

Eddy decomposition of the zonal-mean vertical integral of the horizontal dry enthalpy advection anomaly δcpvT¯ (W m−2) between 110° and 120°E in JJA. Gray shading represents latitudes where EASM precipitation changes most (27°–37°N), which is the region of interest here. Crosses represent anomalies that are statistically significant at a 95% confidence level.

The response of the stationary eddy–mean zonal flow interaction can be further decomposed as follows:

 
cpδ([v¯]T*¯)=cpδ[v¯]Tc*¯+cp[v¯]cδT*¯+cpδ[v¯]δT*¯,
(2)
 
cpδ(v*¯[T¯])=cpδv*¯[T¯]c+cpvc*¯δ[T¯]+cpδv*¯δ[T¯],
(3)

where the subscript c stands for the climatology in the 100L + 100E simulation. The [v¯]T*¯ term approximately equals the [u¯]xT*¯ term, while the v*¯[T¯] term equals the υ*¯y[T¯] term. The total difference in cpδ([v¯]T*¯) (Fig. 10a, black line) is mainly due to decrease in zonal-mean zonal wind in the troposphere, δ[u¯] in cpδ[v¯]Tc*¯ (Fig. 10a, red line). Figure 11a shows that the reduction in cpδ[v¯]Tc*¯ occurs at all levels below 200 hPa, suggesting the importance of the middle and upper troposphere in addition to midtropospheric temperature advection anomaly mentioned by Sampe and Xie (2010). The term xTc*¯ is negative in the troposphere because of warmer land than sea in summer (Fig. 11b). Therefore, the decrease in zonal-mean zonal wind in the troposphere (δ[u¯]; Fig. 11c) contributes to the negative anomaly of cpδ[v¯]Tc*¯. The total difference in cpδ(v*¯[T¯]) (Fig. 10b, black line) is mainly due to decrease in meridional stationary eddy wind, δυ*¯ in cpδv*¯[T¯]c (Fig. 10b, red line), in the lower troposphere (Figs. 11d,f).

Fig. 10.

Decomposition of the zonal-mean stationary eddy–mean flow interaction terms (W m−2) between 110° and120°E in JJA. Decomposition of (a) the advection of the stationary eddy dry enthalpy by the zonal-mean flow cpδ([v¯]T*¯) and (b) the advection of the zonal-mean dry enthalpy by the stationary eddy velocity cpδ(v*¯[T¯]). Gray shading and crosses are as in Fig. 9.

Fig. 10.

Decomposition of the zonal-mean stationary eddy–mean flow interaction terms (W m−2) between 110° and120°E in JJA. Decomposition of (a) the advection of the stationary eddy dry enthalpy by the zonal-mean flow cpδ([v¯]T*¯) and (b) the advection of the zonal-mean dry enthalpy by the stationary eddy velocity cpδ(v*¯[T¯]). Gray shading and crosses are as in Fig. 9.

Fig. 11.

Vertical profile of variables over 110°–120°E, 27°–37°N in JJA: (a) cpδ[u¯]Tc*/x¯, (b) Tc*/x, (c) δ[u¯], (d) cpδυ*¯[Tc/y¯], (e) [Tc/y¯], and (f) δυ*¯. The subscript c stands for the climatology in the 100L + 100E simulation; δ(·) is defined as the difference of (0L + 0E) − (100L + 100E). The star denotes the deviation from the global zonal mean; e.g., u* = u − [u].

Fig. 11.

Vertical profile of variables over 110°–120°E, 27°–37°N in JJA: (a) cpδ[u¯]Tc*/x¯, (b) Tc*/x, (c) δ[u¯], (d) cpδυ*¯[Tc/y¯], (e) [Tc/y¯], and (f) δυ*¯. The subscript c stands for the climatology in the 100L + 100E simulation; δ(·) is defined as the difference of (0L + 0E) − (100L + 100E). The star denotes the deviation from the global zonal mean; e.g., u* = u − [u].

In short, the lowering of NH ice sheets decreases EASM precipitation through two mechanisms: the jet shifting mechanism and the stationary wave mechanism. In the jet shifting mechanism, the removal of ice sheet topography shifts the jet northward and reduces the mean westerly jet [u¯] in the subtropical region, which leads to a reduction of the dry enthalpy advection in the whole troposphere over East Asia, and in turn a reduction of the EASM precipitation. In the stationary wave mechanism, the lowering of ice sheet topography induces an anomalous stationary wave response, such that the meridional stationary eddy velocity υ*¯ is reduced over East Asia, also leading to a reduction of dry enthalpy advection and in turn rainfall of EASM. This stationary eddy mechanism has been discussed by Chen and Bordoni (2014) in their study of the role of Tibetan Plateau on the formation of the MB front. The role of the westerly jet in the EASM has also been discussed in general recently (Molnar et al. 2010; Chiang et al. 2015), although how the ice sheet affects the jet and in turn downstream EASM has not been discussed specifically.

6. Separate influences of the Laurentide and Eurasian Ice Sheets on EASM

To identify the separate effects of LIS and EIS on EASM, we also conducted two experiments in SLAB denoted as 0L + 100E and 100L + 0E in which either topography of LIS or EIS is removed, respectively. At LGM, we can consider the effect of the removal of LIS only as the experiment difference of (0L + 100E) − (100L + 100E) or noLIS_withEIS, and the removal of EIS only as (100L + 0E) − (100L + 100E), or noEIS_withLIS. (Table 1).

Our experiments show that the lowering of LIS and EIS both play an important role in the weakening of EASM. The lowering of LIS and EIS both contribute to the equivalent-barotropic low response and reduced lower-tropospheric southerlies over East Asia (Figs. 12a–d and 13a,c ). In the meantime, only the lowering of LIS induces the northward shift of the westerly jet and decreased zonal-mean zonal wind over the subtropics (Figs. 12a,b), because the LIS is in the latitude range of the jet and therefore has a strong impact on the jet. Therefore, the lowering of LIS weakens the EASM through shifting the jet and altering stationary waves, while the lowering of EIS weakens the EASM only through altering stationary waves.

Fig. 12.

Anomalies of eddy geopotential height (shading; gpm) and zonal wind (contours; m s−1) at (left) 200 hPa and (right) 850 hPa in JJA. Shown are anomalies induced by (a),(b) removal of LIS topography in the presence of EIS (noLIS_withEIS), (c),(d) removal of EIS topography in the presence of LIS (noEIS_withLIS), (e),(f) removal of LIS topography without EIS existing (noLIS_withoutEIS), and (g),(h) removal of EIS topography without LIS existing (noEIS_withoutLIS). Contour intervals are 4 m s−1 in (a), (c), (e), and (g) and 2 m s−1 in (b), (d), (f), and (h); dashed contours indicate negative values.

Fig. 12.

Anomalies of eddy geopotential height (shading; gpm) and zonal wind (contours; m s−1) at (left) 200 hPa and (right) 850 hPa in JJA. Shown are anomalies induced by (a),(b) removal of LIS topography in the presence of EIS (noLIS_withEIS), (c),(d) removal of EIS topography in the presence of LIS (noEIS_withLIS), (e),(f) removal of LIS topography without EIS existing (noLIS_withoutEIS), and (g),(h) removal of EIS topography without LIS existing (noEIS_withoutLIS). Contour intervals are 4 m s−1 in (a), (c), (e), and (g) and 2 m s−1 in (b), (d), (f), and (h); dashed contours indicate negative values.

Fig. 13.

Anomalies of (a),(c),(e),(g) low-level wind (vectors; m s−1) and (b),(d),(f),(h) precipitation (shading; mm day−1) in JJA. The effect represented by each row is the same as in Fig. 11. Pink shadings and black dots represent, respectively, meridional wind and precipitation anomalies that are statistically significant at a 95% confidence level.

Fig. 13.

Anomalies of (a),(c),(e),(g) low-level wind (vectors; m s−1) and (b),(d),(f),(h) precipitation (shading; mm day−1) in JJA. The effect represented by each row is the same as in Fig. 11. Pink shadings and black dots represent, respectively, meridional wind and precipitation anomalies that are statistically significant at a 95% confidence level.

Yin et al. (2008) and Sundaram et al. (2012) showed that EASM precipitation decreases mainly due to the removal of Eurasian ice sheet (EA case), contrary to our model results. However, there is a difference in the experimental setup. To make a more qualitative comparison with their studies, we also consider the separate effects of the removal of LIS and EIS without the presence of the other ice sheet as the experiment difference of (0L + 0E) − (100L + 0E) and (0L + 0E) − (0L + 100E), or noLIS_withoutEIS and noEIS_withoutLIS (Table 1). The experiment noEIS_withoutLIS is a purely idealized experiment with no real world relevance, which is similar to the EA case in Yin et al. (2008) and Sundaram et al. (2012). It is seen that the anomalies in the effects of noLIS_withoutEIS (comparing Figs. 12e,f with Figs. 12a,b, and Figs. 13e,f with Figs. 13a,b) are featured with decreased zonal-mean zonal wind and less precipitation (similar to noLIS_with EIS). However, stationary wave patterns east of EIS are different in noLIS_withoutEIS (absence of equivalent-barotropic low response and reduced lower-tropospheric southerlies). The anomalies in the effects of noEIS_withoutLIS are not significant (Figs. 13g,h), which is different from the dominant role of Eurasian ice sheet shown in Yin et al. (2008) and Sundaram et al. (2012). Several reasons may account for this difference, as follows. First is the model dynamics. The model they employed—LOVECLIM—is of intermediate complexity with quasigeostrophic atmospheric dynamics, and with coarse vertical resolution (three levels) and imposed stratification. CCSM3, on the other hand is a full general circulation model with finer resolution in the vertical. Second, the ice sheet configuration is different. Their North American ice sheet is located more north and its height is only about half of our LIS. The position of their Eurasian ice sheet is about 20° east of our EIS.

7. Role of the ocean

The indirect dynamic effect associated with the AMOC change caused by the lowering of ice sheets on EASM is not important. This can be seen from the similar responses in fully coupled CCSM3 and SLAB in section 3. Previous studies show that the lowering of ice sheets tends to weaken the AMOC (Eisenman et al. 2009; Zhang et al. 2014; Zhu et al. 2014; Sherriff-Tadano et al. 2018) and a decrease in the AMOC strength can lead to a decrease in the EASM (Fallah et al. 2016; Wen et al. 2016). Thus, the lowering of ice sheets may weaken EASM through the effect of AMOC change. We also simulated weakened AMOC in response to the lowering of the NH ice sheets: maximum AMOC decreases from 11.56 Sv (1 Sv ≡ 106 m3 s−1) (100%) to 7.64 Sv (0%) in the fully coupled sensitivity experiments, which is consistent with Zhu et al. (2014), who showed that AMOC weakened significantly (by 41%) in response to the deglacial ice sheet retreat. A weakened AMOC can also be validated from a typical dipole SST anomaly: cooling in the North Atlantic and weaker warming at the equator and the South Atlantic (Fig. 14c). We use differences of the SST anomalies in the sensitivity experiments (Fig. 14c) between the fully coupled experiments (0%–100%; Fig. 14a) and the SLAB experiments [(0L + 0E) − (100L + 100E); Fig. 14b] to represent the SST changes induced by ocean dynamics. However, the weakened AMOC seems to have limited impacts on the EASM compared to the direct orographic effect of the ice sheets.

Fig. 14.

Annual mean SST anomalies (K) in (a) fully coupled CCSM3: 0%–100%, (b) SLAB: (0L + 0E) − (100L + 100E), and (c) CCSM3 − SLAB: (0%–100%) − [(0L + 0E) − (100L + 100E)].

Fig. 14.

Annual mean SST anomalies (K) in (a) fully coupled CCSM3: 0%–100%, (b) SLAB: (0L + 0E) − (100L + 100E), and (c) CCSM3 − SLAB: (0%–100%) − [(0L + 0E) − (100L + 100E)].

Indeed, the total ocean effect, including both thermal and dynamic, is not important. This can be seen in the AGCM experiment (CAM3 with fixed SST and sea ice) without the ocean effect at all (Figs. 3e,f and 4c). We repeat the sensitivity experiments in the atmosphere alone model, with the atmosphere general circulation model CAM3 forced by the seasonal cycle of SST and sea ice cover prescribed from the 100% fully coupled simulation as 100L_100E_AGCM, 0L_0E_AGCM. Therefore, the difference between SLAB and AGCM simulations shows the effect of local ocean thermodynamics, and the difference between CCSM3 and AGCM simulations gives the total effect of the ocean. In all the three model configurations, the lowering of ice sheets induces weakened EASM monsoon wind (Fig. 4) and reduced precipitation (Figs. 3b,d,f). In contrast, Sundaram et al. (2012) considered the effect of ocean as a significant link in how ice sheets affect EASM. They argued that the presence of ice sheets over North America and Eurasia during MIS-13 induces a positive phase of the winter North Atlantic Oscillation (NAO)-like feature. The ocean allows the NAO like feature to persist until summer and excite a wave train. In our model, these oceanic effects do not seem to be important.

In summary, our results suggest that the direct ice sheet orographic effect on the EASM is dominant while the indirect effects of ocean dynamics and thermodynamics are not important.

8. Summary

Our CCSM3 model study suggests that receding ice sheets weaken the EASM with the ice sheet topography playing a major role. Two mechanisms are implicated in the connection between the ice sheet and EASM, as shown schematically in Fig. 15. In the jet shifting mechanism, the removal of ice sheet topography in the Northern Hemisphere induces a northward shift of westerly jet and decreased zonal-mean zonal wind over the subtropics in summer; the reduced jet decreases the advection of dry enthalpy into East Asia, and therefore reduces EASM rainfall there. Meanwhile, the lowering of ice sheet induces an anomalous stationary Rossby wave train from North America all the way to East Asia, along the westerly jet waveguide; the wave propagation causes an equivalent-barotropic low geopotential height anomaly over East Asia accompanied by reduced lower-tropospheric southerlies, which in turn reduces the advection of dry enthalpy and hence EASM rainfall.

Fig. 15.

Schematic of the influence of the removal of NH continental ice sheet topography on EASM. The orange color represents a weakened EASM.

Fig. 15.

Schematic of the influence of the removal of NH continental ice sheet topography on EASM. The orange color represents a weakened EASM.

There are still outstanding questions about the role of the ice sheet topography on climate and the EASM. We note that in a recent study (Roberts et al. 2019), the global effect of the Laurentide Ice Sheet on summer stationary waves is contributed significantly not only by the topography effect, but also by the surface albedo effect. Our experiments show a similar impact of the combined topography and albedo effect compared to their results (figures not shown). However, in our experiments, the topographic effect is dominant, while the albedo effect plays a minor role. Further studies are needed to understand the difference among different models. In addition to stationary wave change, there may remain alternative mechanisms which could explain the reduced lower-tropospheric southerlies over East Asia in response to the removal of NH continental ice sheet topography. For example, it is possible some disturbance caused by LIS can propagate into the tropical Atlantic first and then to the Asian sector in the equatorial Kelvin wave, which can also affect EASM (Ueda et al. 2009). We plan to further study these mechanisms in the future.

Acknowledgments

We thank Prof. X. Wen for helpful discussions and three anonymous reviewers and the editor for insightful suggestions. This work is supported by NSFC41630527, NSF1656907, NSF1810682, and the China Scholarship Council. The authors declare that there is no conflict of interests regarding the publication of this article.

Data availability. The NCEP–DOE Reanalysis II can be obtained from https://psl.noaa.gov/data/gridded/data.ncep.reanalysis2.html. The ICE-5G reconstruction dataset is available at https://pmip2.lsce.ipsl.fr/design/ice5g. The data underlying the figures from this manuscript are available at https://doi.org/10.7910/DVN/1ZHRSD.

REFERENCES

REFERENCES
Abe-Ouchi
,
A.
,
T.
Segawa
, and
F.
Saito
,
2007
:
Climatic conditions for modelling the Northern Hemisphere ice sheets throughout the ice age cycle
.
Climate Past
,
3
,
423
438
, https://doi.org/10.5194/cp-3-423-2007.
Anderson
,
P. M.
, and et al
,
1988
:
Climatic changes of the last 18,000 years: Observations and model simulations
.
Science
,
241
,
1043
1052
, https://doi.org/10.1126/science.241.4869.1043.
Chen
,
J.
, and
S.
Bordoni
,
2014
:
Orographic effects of the Tibetan Plateau on the East Asian summer monsoon: An energetic perspective
.
J. Climate
,
27
,
3052
3072
, https://doi.org/10.1175/JCLI-D-13-00479.1.
Cheng
,
H.
,
R. L.
Edwards
,
W. S.
Broecker
,
G. H.
Denton
,
X.
Kong
,
Y.
Wang
,
R.
Zhang
, and
X.
Wang
,
2009
:
Ice age terminations
.
Science
,
326
,
248
252
, https://doi.org/10.1126/science.1177840.
Chiang
,
J. C. H.
, and
C. M.
Bitz
,
2005
:
Influence of high latitude ice cover on the marine intertropical convergence zone
.
Climate Dyn.
,
25
,
477
496
, https://doi.org/10.1007/s00382-005-0040-5.
Chiang
,
J. C. H.
,
M.
Biasutti
, and
D. S.
Battisti
,
2003
:
Sensitivity of the Atlantic intertropical convergence zone to last glacial maximum boundary conditions
.
Paleoceanography
,
18
,
1094
, https://doi.org/10.1029/2003PA000916.
Chiang
,
J. C. H.
, and et al
,
2015
:
Role of seasonal transitions and westerly jets in East Asian paleoclimate
.
Quat. Sci. Rev.
,
108
,
111
129
, https://doi.org/10.1016/j.quascirev.2014.11.009.
Chou
,
C.
, and
J. D.
Neelin
,
2003
:
Mechanisms limiting the northward extent of the northern summer monsoons over North America, Asia, and Africa
.
J. Climate
,
16
,
406
425
, https://doi.org/10.1175/1520-0442(2003)016<0406:MLTNEO>2.0.CO;2.
Clark
,
P. U.
,
R. B.
Alley
, and
D.
Pollard
,
1999
:
Northern Hemisphere ice-sheet influences on global climate change
.
Science
,
286
,
1104
1111
, https://doi.org/10.1126/science.286.5442.1104.
Clemens
,
S. C.
, and
W. L.
Prell
,
1990
:
Late Pleistocene variability of Arabian Sea summer monsoon winds and continental aridity: Eolian records from the lithogenic component of deep-sea sediments
.
Paleoceanography
,
5
,
109
145
, https://doi.org/10.1029/PA005i002p00109.
Clemens
,
S. C.
, and
W. L.
Prell
,
2003
:
A 350,000 year summer-monsoon multi-proxy stack from the Owen Ridge, Northern Arabian Sea
.
Mar. Geol.
,
201
,
35
51
, https://doi.org/10.1016/S0025-3227(03)00207-X.
Collins
,
W. D.
, and et al
,
2006
:
The Community Climate System Model version 3 (CCSM3)
.
J. Climate
,
19
,
2122
2143
, https://doi.org/10.1175/JCLI3761.1.
Eisenman
,
I.
,
C. M.
Bitz
, and
E.
Tziperman
,
2009
:
Rain driven by receding ice sheets as a cause of past climate change
.
Paleoceanography
,
24
,
PA4209
, https://doi.org/10.1029/2009PA001778.
Fallah
,
B.
,
U.
Cubasch
,
K.
Prömmel
, and
S.
Sodoudi
,
2016
:
A numerical model study on the behaviour of Asian summer monsoon and AMOC due to orographic forcing of Tibetan Plateau
.
Climate Dyn.
,
47
,
1485
1495
, https://doi.org/10.1007/s00382-015-2914-5.
Galbraith
,
E.
, and
C.
de Lavergne
,
2019
:
Response of a comprehensive climate model to a broad range of external forcings: Relevance for deep ocean ventilation and the development of late Cenozoic ice ages
.
Climate Dyn.
,
52
,
653
679
, https://doi.org/10.1007/s00382-018-4157-8.
He
,
F.
,
J. D.
Shakun
,
P. U.
Clark
,
A. E.
Carlson
,
Z.
Liu
,
B. L.
Otto-Bliesner
, and
J. E.
Kutzbach
,
2013
:
Northern Hemisphere forcing of Southern Hemisphere climate during the last deglaciation
.
Nature
,
494
,
81
85
, https://doi.org/10.1038/nature11822.
Hofer
,
D.
,
C. C.
Raible
,
A.
Dehnert
, and
J.
Kuhlemann
,
2012
:
The impact of different glacial boundary conditions on atmospheric dynamics and precipitation in the North Atlantic region
.
Climate Past
,
8
,
935
949
, https://doi.org/10.5194/cp-8-935-2012.
Hoskins
,
B. J.
, and
D. J.
Karoly
,
1981
:
The steady linear response of a spherical atmosphere to thermal and orographic forcing
.
J. Atmos. Sci.
,
38
,
1179
1196
, https://doi.org/10.1175/1520-0469(1981)038<1179:TSLROA>2.0.CO;2.
Hoskins
,
B. J.
, and
T.
Ambrizzi
,
1993
:
Rossby wave propagation on a realistic longitudinally varying flow
.
J. Atmos. Sci.
,
50
,
1661
1671
, https://doi.org/10.1175/1520-0469(1993)050<1661:RWPOAR>2.0.CO;2.
Kanamitsu
,
M.
,
W.
Ebisuzaki
,
J.
Woollen
,
S.-K.
Yang
,
J. J.
Hnilo
,
M.
Fiorino
, and
G. L.
Potter
,
2002
:
NCEP–DOE AMIP-II Reanalysis (R-2)
.
Bull. Amer. Meteor. Soc.
,
83
,
1631
1643
, https://doi.org/10.1175/BAMS-83-11-1631.
Klockmann
,
M.
,
U.
Mikolajewicz
, and
J.
Marotzke
,
2016
:
The effect of greenhouse gas concentrations and ice sheets on the glacial AMOC in a coupled climate model
.
Climate Past
,
12
,
1829
1846
, https://doi.org/10.5194/cp-12-1829-2016.
Kutzbach
,
J. E.
,
1981
:
Monsoon climate of the early Holocene: Climate experiment with the Earth’s orbital parameters for 9000 years ago
.
Science
,
214
,
59
61
, https://doi.org/10.1126/science.214.4516.59.
Kutzbach
,
J. E.
, and
F. A.
Street-Perrott
,
1985
:
Milankovitch forcing of fluctuations in the level of tropical lakes from 18 to 0 kyr BP
.
Nature
,
317
,
130
134
, https://doi.org/10.1038/317130a0.
Lee
,
S. Y.
,
J. C. H.
Chiang
, and
P.
Chang
,
2015
:
Tropical Pacific response to continental ice sheet topography
.
Climate Dyn.
,
44
,
2429
2446
, https://doi.org/10.1007/s00382-014-2162-0.
Li
,
C.
, and
D. S.
Battisti
,
2008
:
Reduced Atlantic storminess during Last Glacial Maximum: Evidence from a coupled climate model
.
J. Climate
,
21
,
3561
3579
, https://doi.org/10.1175/2007JCLI2166.1.
Liu
,
Z.
,
Z.
Lu
,
X.
Wen
,
B. L.
Otto-Bliesner
,
A.
Timmermann
, and
K. M.
Cobb
,
2014
:
Evolution and forcing mechanisms of El Niño over the past 21,000 years
.
Nature
,
515
,
550
553
, https://doi.org/10.1038/nature13963.
Löfverström
,
M.
,
R.
Caballero
,
J.
Nilsson
, and
J.
Kleman
,
2014
:
Evolution of the large-scale atmospheric circulation in response to changing ice sheets over the last glacial cycle
.
Climate Past
,
10
,
1453
1471
, https://doi.org/10.5194/cp-10-1453-2014.
Lu
,
Z.
,
Z.
Liu
, and
J.
Zhu
,
2016
:
Abrupt intensification of ENSO forced by deglacial ice-sheet retreat in CCSM3
.
Climate Dyn.
,
46
,
1877
1891
, https://doi.org/10.1007/s00382-015-2681-3.
Lu
,
Z.
,
Z.
Liu
,
G.
Chen
, and
J.
Guan
,
2019
:
Prominent precession band variance in ENSO intensity over the last 300,000 years
.
Geophys. Res. Lett.
,
46
,
9786
9795
, https://doi.org/10.1029/2019GL083410.
Manabe
,
S.
, and
A. J.
Broccoli
,
1985
:
The influence of continental ice sheets on the climate of an ice age
.
J. Geophys. Res.
,
90
,
2167
2190
, https://doi.org/10.1029/JD090iD01p02167.
Molnar
,
P.
,
W. R.
Boos
, and
D. S.
Battisti
,
2010
:
Orographic controls on climate and paleoclimate of Asia: Thermal and mechanical roles for the Tibetan Plateau
.
Annu. Rev. Earth Planet. Sci.
,
38
,
77
102
, https://doi.org/10.1146/annurev-earth-040809-152456.
Neelin
,
J. D.
,
2007
: Moist dynamics of tropical convection zones in monsoons, teleconnections and global warming. The Global Circulation of the Atmosphere, T. Schneider and A. Sobel, Eds., Princeton University Press, 267–301.
Overpeck
,
J.
,
D.
Anderson
,
S.
Trumbore
, and
W.
Prell
,
1996
:
The southwest Indian monsoon over the last 18 000 years
.
Climate Dyn.
,
12
,
213
225
, https://doi.org/10.1007/BF00211619.
Pausata
,
F. S. R.
,
C.
Li
,
J. J.
Wettstein
,
M.
Kageyama
, and
K. H.
Nisancioglu
,
2011
:
The key role of topography in altering North Atlantic atmospheric circulation during the last glacial period
.
Climate Past
,
7
,
1089
1101
, https://doi.org/10.5194/cp-7-1089-2011.
Peltier
,
W. R.
,
2004
:
Global glacial isostasy and the surface of the ice-age Earth: The ICE-5G (VM2) model and GRACE
.
Annu. Rev. Earth Planet. Sci.
,
32
,
111
149
, https://doi.org/10.1146/annurev.earth.32.082503.144359.
Petit
,
J. R.
, and et al
,
1999
:
Climate and atmospheric history of the past 420,000 years from the Vostok ice core, Antarctica
.
Nature
,
399
,
429
436
, https://doi.org/10.1038/20859.
Roberts
,
W. H. G.
,
C.
Li
, and
P. J.
Valdes
,
2019
:
The mechanisms that determine the response of the Northern Hemisphere’s stationary waves to North American ice sheets
.
J. Climate
,
32
,
3917
3940
, https://doi.org/10.1175/JCLI-D-18-0586.1.
Ruddiman
,
W. F.
,
2006
:
What is the timing of orbital-scale monsoon changes?
Quat. Sci. Rev.
,
25
,
657
658
, https://doi.org/10.1016/j.quascirev.2006.02.004.
Sampe
,
T.
, and
S.-P.
Xie
,
2010
:
Large-scale dynamics of the meiyu-baiu rainband: Environmental forcing by the westerly jet
.
J. Climate
,
23
,
113
134
, https://doi.org/10.1175/2009JCLI3128.1.
Shackleton
,
N. J.
,
2000
:
The 100,000-year ice-age cycle identified and found to lag temperature, carbon dioxide, and orbital eccentricity
.
Science
,
289
,
1897
1902
, https://doi.org/10.1126/science.289.5486.1897.
Sherriff-Tadano
,
S.
,
A.
Abe-Ouchi
,
M.
Yoshimori
,
A.
Oka
, and
W.-L.
Chan
,
2018
:
Influence of glacial ice sheets on the Atlantic meridional overturning circulation through surface wind change
.
Climate Dyn.
,
50
,
2881
2903
, https://doi.org/10.1007/s00382-017-3780-0.
Sundaram
,
S.
,
Q. Z.
Yin
,
A.
Berger
, and
H.
Muri
,
2012
:
Impact of ice sheet induced North Atlantic oscillation on East Asian summer monsoon during an interglacial 500,000 years ago
.
Climate Dyn.
,
39
,
1093
1105
, https://doi.org/10.1007/s00382-011-1213-z.
Ting
,
M.
,
1994
:
Maintenance of northern summer stationary waves in a GCM
.
J. Atmos. Sci.
,
51
,
3286
3308
, https://doi.org/10.1175/1520-0469(1994)051<3286:MONSSW>2.0.CO;2.
Ueda
,
H.
,
M.
Ohba
,
S.-P.
Xie
,
H.
Ueda
,
M.
Ohba
, and
S.-P.
Xie
,
2009
:
Important factors for the development of the Asian–northwest Pacific summer monsoon
.
J. Climate
,
22
,
649
669
, https://doi.org/10.1175/2008JCLI2341.1.
Ullman
,
D. J.
,
A. N.
Legrande
,
A. E.
Carlson
,
F. S.
Anslow
, and
J. M.
Licciardi
,
2014
:
Assessing the impact of Laurentide Ice Sheet topography on glacial climate
.
Climate Past
,
10
,
487
507
, https://doi.org/10.5194/cp-10-487-2014.
Wen
,
X.
,
Z.
Liu
,
S.
Wang
,
J.
Cheng
, and
J.
Zhu
,
2016
:
Correlation and anti-correlation of the East Asian summer and winter monsoons during the last 21,000 years
.
Nat. Commun.
,
7
,
11999
, https://doi.org/10.1038/ncomms11999.
Yin
,
Q. Z.
,
A.
Berger
,
E.
Driesschaert
,
H.
Goosse
,
M. F.
Loutre
, and
M.
Crucifix
,
2008
:
The Eurasian ice sheet reinforces the East Asian summer monsoon during the interglacial 500 000 years ago
.
Climate Past
,
4
,
79
90
, https://doi.org/10.5194/cp-4-79-2008.
Yin
,
Q. Z.
,
A.
Berger
, and
M.
Crucifix
,
2009
:
Individual and combined effects of ice sheets and precession on MIS-13 climate
.
Climate Past
,
5
,
229
243
, https://doi.org/10.5194/cp-5-229-2009.
Zhang
,
X.
,
G.
Lohmann
,
G.
Knorr
, and
C.
Purcell
,
2014
:
Abrupt glacial climate shifts controlled by ice sheet changes
.
Nature
,
512
,
290
294
, https://doi.org/10.1038/nature13592.
Zhu
,
J.
,
Z.
Liu
,
X.
Zhang
,
I.
Eisenman
, and
W.
Liu
,
2014
:
Linear weakening of the AMOC in response to receding glacial ice sheets in CCSM3
.
Geophys. Res. Lett.
,
41
,
6252
6258
, https://doi.org/10.1002/2014GL060891.

Footnotes

Publisher’s Note: This article was revised on 29 July 2020 to correct the affiliation of the second author.

For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).