1. Introduction
Volcanic eruptions are one of the most important natural causes of climate change and have many climatic effects on ocean dynamics, the hydroclimate, monsoons, the large-scale atmospheric circulation and the associated Northern Hemisphere winter climate, and so on (Robock 2000; Trenberth and Dai 2007; D’Arrigo et al. 2009; Stenchikov et al. 2006; Driscoll et al. 2012; Timmreck 2012; Iles et al. 2013; Man et al. 2014; Man and Zhou 2014; Zambri and Robock 2016). Injections of sulfate aerosols into the lower stratosphere will reduce the incoming solar radiation reaching the surface and absorb the outgoing longwave radiation, in turn influencing the climate system through radiative and dynamic processes. Previous studies have indicated that there is a relationship between volcanic eruptions and El Niño (Handler 1984; Robock et al. 1995; Adams et al. 2003; Mann et al. 2005; Emile-Geay et al. 2008; McGregor et al. 2010; McGregor and Timmermann 2011; Ohba et al. 2013; Li et al. 2013; Wahl et al. 2014; Pausata et al. 2015a; Maher et al. 2015; Stevenson et al. 2016; Lim et al. 2016; Le 2017; Predybaylo et al. 2017; Swingedouw et al. 2017). Since El Niño is one of the most important modes of internal climate variability and influences the global climate through atmospheric teleconnections (Alexander et al. 2002), it is of great significance to understand the influence of volcanic eruptions on El Niño.
The eastern equatorial Pacific tends to be warmer following low-latitude volcanic eruptions and colder after eruptions occurred at high latitudes (Handler 1984). The former phenomenon could be confirmed by model calculations, and a hypothesis that low-latitude volcanic eruptions could produce El Niño events was proposed (McCracken and Luther 1984). However, some studies opposed this idea and stated that volcanic eruptions were not the cause of El Niño events (Nicholls 1990; Self et al. 1997). These studies were based on instrumental period data, which may not reveal the true relationship between volcanic eruptions and El Niño events since the length of observational data is too short and few volcanic events can be selected (Graf et al. 2014). Long-period proxy data reveal an El Niño–like response to tropical volcanic eruptions (Adams et al. 2003) and suggest that the equatorial Pacific tends to be in an El Niño state after tropical volcanic eruptions.
Subsequently, a number of studies based on numerical experiments employing different models have supported an El Niño–like SST response to volcanic eruptions. The ENSO response to explosive tropical eruptions was reproduced in the Zebiak–Cane model (Zebiak and Cane 1987) and was explained by the ocean dynamical thermostat mechanism (Clement et al. 1996; Mann et al. 2005; Emile-Geay et al. 2008). These results showed that only eruptions larger than a certain threshold could increase the intensity and probability of El Niño events (Emile-Geay et al. 2008), which is consistent with the recent study by Lim et al. (2016).
The Zebiak–Cane model is an idealized model that does not consider cloud feedbacks or thermocline ventilation; therefore, the following studies have used a more comprehensive coupled general circulation model (CGCM). Based on the CCSM3 volcanic forcing experiment, the SST responses to small, moderate, and large volcanic eruptions were analyzed (McGregor and Timmermann 2011). The eastern equatorial Pacific was found to initially exhibit a La Niña–like cooling and strengthened with an increased eruption intensity. They attributed the initial La Niña–like response to Newtonian cooling and changes in the mixed layer depths. In terms of the mechanisms of the El Niño–like SST response, most of the studies supported the ocean dynamical thermostat mechanism proposed by Clement et al. (1996). However, a new mechanism was proposed by Ohba et al. (2013), wherein the westerly anomalies could be caused by the surface pressure gradient and an indirect effect from the changes of precipitation over the Indo-Maritime Continent and the Pacific Ocean. Furthermore, the westerly anomalies could be enhanced via the Bjerknes feedback (Bjerknes 1969). In a recent study, the contribution of surface cooling over the Maritime Continent to the anomalous warming of the eastern Pacific was confirmed based on a set of experiments (Wang et al. 2018). However, Khodri et al. (2017) implied that the initial westerly wind anomaly after the eruption was largely driven by the cooling over tropical African landmasses (Khodri et al. 2017).
The above studies focused on tropical eruptions; however, high-latitude eruptions induce several climatic effects as well (Oman et al. 2005). It was found that large high-latitude eruptions in the Northern Hemisphere could result in an El Niño–like anomaly over the Pacific in the NorESM1-M model (Pausata et al. 2015a). They indicated that the weakening of trade winds induced by the southward shift of the ITCZ, which is related to the cooling of the Northern Hemisphere, will lead to an El Niño–like SST anomaly. A variety of studies have indicated that extratropical thermal forcing has a considerable impact on the displacement of the ITCZ (Kang et al. 2008; Seo et al. 2014); among these, the response of the ITCZ to hemispherically asymmetrical volcanic forcing was studied by Colose et al. (2016), who arrived at a conclusion consistent with that of Pausata et al. (2015a).
A comparison of the climate effects induced by low- and high-latitude volcanic eruptions was also performed (Schneider et al. 2009), and the results indicated that the impact of high-latitude eruptions are restricted to a shorter term than that of tropical eruptions. Volcanic eruptions were further divided into three categories—northern, tropical, and southern volcanic eruptions—and the CESM-LME simulation indicated that both northern and tropical eruptions favored an El Niño–like response that was not observed after southern eruptions (Stevenson et al. 2016). They attributed this hemispheric dependence of the ENSO response to eruptions to the migration of the ITCZ. Based on the proxy ENSO index and model simulation, divergent El Niño responses to volcanic eruptions at different latitudes were studied (Liu et al. 2018). The reconstructions show an El Niño–like response to northern and tropical eruptions, and a La Niña–like response to southern eruptions. The simulation demonstrates an eastern-Pacific El Niño state after northern and tropical eruptions and a central-Pacific El Niño–like response to southern eruptions.
Although a previous study (Liu et al. 2018) showed different El Niño responses to volcanic eruptions at different latitudes, they used only one experiment member. From a modeling perspective, ensemble simulations are the most suited method to study volcano-forced responses (Zanchettin et al. 2012; Pausata et al. 2015a; Stevenson et al. 2017). Moreover, these studies emphasized the response at an interannual time scale. There have been few attempts to interpret the response on a monthly time scale. Thus, the present study aims to analyze the different ENSO response to different volcanic eruptions based on the CESM Last Millennium Ensemble simulations, which have the largest ensemble of LM simulations and thus a large number of realizations of the same events. We analyze the ENSO response from a monthly time scale since El Niño has strong seasonality. The main motivation of the study is to address the following questions: 1) What are the Pacific SST responses following northern, tropical, and southern volcanic eruptions by using a large ensemble? What are the differences between them? 2) How does the different response mechanism arise due to differences in the volcanic forcing structure?
The remainder of the paper is organized as follows. Section 2 provides a description of the model and data, as well as the criteria for volcano selection. Section 3 presents the results. The conclusions and discussions are given in section 4.
2. Model and data description
a. Model configuration
The results presented in this study use the Community Earth System Model Last Millennium Ensemble (CESM-LME; Otto-Bliesner et al. 2016). The model configuration used is the CESM1.1, which was run at a lower land and atmosphere resolution because of the more intensive computing time demands of the ensemble (Stevenson et al. 2016). Model resolution is roughly 2° resolution in its atmosphere and land components and ~1° resolution in its ocean and sea ice components. This model provides the largest ensemble of Last Millennium (LM) simulations from a single model currently available, and the simulations are forced by solar insolation, volcanic eruptions, land use, greenhouse gases and orbital forcing. The greenhouse gases, such as CO2 and CH4, are derived from high-resolution Antarctic ice cores (Schmidt et al. 2012); changes in the orbital modulation of insolation can be calculated using the equations in Berger (1978); the reconstructed total solar irradiance (TSI) data are derived from Vieira and Solanki (2010); and the land use/land cover changes are from the merged reconstruction data in Pongratz et al. (2009) and Hurtt et al. (2011). The only differences between the ensemble members are due to different perturbations of the initial states of the air temperature field. For volcanic forcing, LME uses version 1 of the Gao et al. (2008) ice-core-derived reconstruction, which is based on volcanic deposition signals from 54 ice-core records from both the Arctic and Antarctica. These data are an estimate of aerosol loadings as a function of latitude, altitude, and month. Stratospheric aerosols are designed to be distributed in three layers in the lower stratosphere, and are subject to a single size distribution. See Otto-Bliesner et al. (2016) for more details.
b. Data and criteria for volcano selection
We use the output data of volcanic-only simulations, which consist of five ensemble members, and we compute the multimember ensemble mean to remove the effects of internal variability; the volcanic-only simulations remove all influences other than volcanic aerosols. Here, the volcanic eruptions are divided into three types according to the distributions of volcanic aerosols in the Northern and Southern Hemispheres, as computed by Gao et al. (2008). The global total stratospheric sulfate aerosol injection from volcanic eruptions in CESM-LME is shown in Fig. 1a, where the colored bars represent the selected volcanic eruptions. Among them, the blue bars represent northern volcanic eruptions, the red bars indicate the eruptions occurred in the tropical regions, and the green bars represent southern volcanic eruptions. In addition, when selecting these cases, we use the volcanic forcing data in CESM-LME and select the volcanic eruptions that reach their peaks at the same time, as shown in Fig. 1b. The blue, red, and green lines represent the time evolutions of the composited Northern Hemispheric, tropical, and Southern Hemispheric volcanic aerosol masses, and the brown triangle indicates the time at which the volcanic eruptions peaked. It can be seen that the three types of volcanoes peaked in August at the same time, since some eruptions with unknown seasonality are assumed to begin and peak at around the same time in the CESM-LME. In this study, the year in which the aerosol mass peaks is regarded as year 0. The following two years are named year 1 and year 2, respectively. The eruptions included in the final sample are listed in Table 1. The SEA method is used when analyzing the climatic response triggered by the volcanic eruptions (Haurwitz and Brier 1981). The anomalies are calculated relative to the five pre-volcanic eruption years (Paik and Min 2017). The significance of the response to volcanic eruptions was assessed using a Monte Carlo method.
(a) Global total stratospheric sulfate aerosol injection from the volcanic eruptions (units: kg m−2) in the CESM-LME simulation. The blue bars represent northern eruptions, the red bars represent tropical eruptions, and the green bars represent southern eruptions. (b) The time evolution of the composite volcanic aerosol masses for the three types of volcanic eruptions; the brown triangle indicates the time at which the volcanic eruptions peaked. The blue line represents northern eruptions, the red line represents tropical eruptions and the green line represents southern eruptions.
Citation: Journal of Climate 31, 17; 10.1175/JCLI-D-17-0571.1
List of eruption years for northern, tropical, and southern volcanic eruption classes.
3. Results
We first present the time evolution of SST anomalies over the Pacific following northern, tropical, and southern volcanic eruptions and then compare the differences between them. Next, we note the underlying causes of the different SST responses to these three types of eruptions.
a. Response of the Pacific sea surface temperature to eruptions
To examine the SST responses to volcanic forcing, we display the composite patterns of the longitude–time sections of the anomalous SST and the 850-hPa zonal wind averaged within 10°S–10°N along the equator from the March (year 0) to November (year 3) following northern, tropical, and southern volcanic eruptions after removing the zonal mean (Fig. 2). The red and blue triangles represent the times at which a volcano erupted and peaked respectively. Following northern and tropical volcanic eruptions, the SST in the central and eastern equatorial Pacific features significant positive values 5–10 months after the eruption, which can be regarded as an El Niño–like SST response; this result is consistent with the study of Pausata et al. (2015a) and that of Mann et al. (2005). For southern eruptions, the warm SST anomaly is not that strong but is nonetheless significant, and the El Niño–like response appears since the SST features a zonal temperature gradient with cooling in the western Pacific and warming in the eastern Pacific. Previous studies proposed that the El Niño–like SST response was relatively weak or did not even exist following southern eruptions (Stevenson et al. 2017; Liu et al. 2018). Here, we isolate the intrinsic ENSO signal from the volcanically induced cooling by removing the climatology of SST response over the equatorial Pacific as that in Khodri et al. (2017). After doing this, the El Niño–like response to southern eruptions is significantly strengthened. Additionally, a strong anomalous westerly wind at 850 hPa is seen a few months after all three types of eruptions. Specifically, following northern eruptions, westerly anomalies start to appear in August of the year 0 (peak of eruption), corresponding to a weak warming over the central to eastern Pacific. In the following year, the positive SST anomalies weaken slightly and then strengthen in the winter of the year 1, at which point the westerly anomalies are also enhanced. For tropical eruptions, the westerly anomalies start to appear in April of year 1 and the warm SST anomaly reaches its peak in the winter of year 1. Compared with northern eruptions, the Indo-western Pacific exhibits a significant cooling, and within the first 8 months after the peak of eruption, the El Niño–like response over the central to eastern Pacific does not occur. Moreover, the warm SST anomaly is mainly confined to the eastern Pacific with a stronger intensity. Following southern eruptions, the significant westerly anomalies appear in September of year 0, accompanied by the warm SST anomaly over the eastern Pacific. After about half a year, the SST anomalies peak in the central to eastern Pacific (Fig. 2c). In contrast to northern and tropical eruptions, the Pacific shows a weaker warming anomaly over the eastern Pacific and a weaker cooling anomaly over the western Pacific, but the changes in the zonal SST gradient are obvious.
Longitude–time sections of SST (units: °C) and 850-hPa wind anomalies (m s−1) along the equatorial region (10°S–10°N) from March (year 0) to November (year 3) for (a) northern eruptions, (b) tropical eruptions, and (c) southern eruptions after removing the zonal mean. SST (units: °C) anomalies along the equatorial region following (d) northern, (e) tropical, and (f) southern eruptions. Only significant wind anomalies are plotted. The significance levels are determined according to the Monte Carlo test, and stippling indicates the values that are significant at the 95% confidence level.
Citation: Journal of Climate 31, 17; 10.1175/JCLI-D-17-0571.1
After the summer of year 2, the westerly anomalies weaken but the weak warm SST anomaly still dominates the central to eastern Pacific Ocean. It should be noted that at the beginning of year 3, the SST anomaly turns to a La Niña–like state over the Pacific following all three types of volcanic eruptions, among which the response to southern eruptions is the most significant. This result is consistent with previous studies, which suggests a La Niña–like response to explosive tropical volcanoes approximately 2–3 years after the eruption (Zanchettin et al. 2012; Pausata et al. 2015a).
To compare the El Niño–like SST patterns caused by the three different types of volcanic eruptions more directly, the temporal patterns of the Niño-3 index with zonal mean removed following northern, tropical, and southern eruptions are shown in Fig. 3a. Following northern and tropical eruptions, the Niño-3 index reaches a maximum value in the winter of year 1, and the magnitude of the Niño-3 index is about 0.5° and 0.7°C respectively. However, for northern eruptions, the Niño-3 index has two peaks that are mainly related to the individual case. Following southern eruptions, the maximum value of the Niño-3 index is less than 0.5°C, which does not meet the criterion of an El Niño event, but an El Niño–like SST response can still be seen since there exists a positive value of the Niño-3 index, and the time at which the Niño-3 index reaches its peak is about 4 months earlier than that after northern and tropical eruptions.
(a) Evolution of the composite Niño-3 index with zonal mean removed (units: °C) after northern eruptions (blue line), tropical eruptions (red line), and southern eruptions (green line). The spreads of the individual volcanic eruptions are denoted by the blue, red, and green shading, respectively. (b) The lead–lag correlation between the Niño-3 index (5°S–5°N, 150°–90°W) and the 850-hPa zonal wind in the western-to-central equatorial Pacific (5°S–5°N, 110°E–150°W) following northern (blue line), tropical (red line), and southern eruptions (green line). The positive value of the horizontal axis indicates that the Niño-3 index lags the 850-hPa zonal wind.
Citation: Journal of Climate 31, 17; 10.1175/JCLI-D-17-0571.1
The response of the Niño-3 index also exhibits widespread variability among the ensemble members (shaded), especially around the peak of the El Niño–like SST response; the spread should be attributed to the initial state of the climate before the eruptions. Previous studies have noted the importance of the initial state to the response of the climate system to volcanic eruptions (Zanchettin et al. 2013; Pausata et al. 2015b, 2016), particularly with regard to how the climate system may respond to a large volcanic eruption, based on the phase of El Niño–Southern Oscillation (Adams et al. 2003; Ohba et al. 2013) and the eruption season (Stevenson et al. 2017).
b. Mechanism of the SST formation
As mentioned above, an El Niño–like SST anomaly exists following northern, tropical, and southern eruptions, manifested through a warm SST anomaly over the central to eastern Pacific and westerly anomalies over the central Pacific. We further calculate the lead–lag correlation coefficients between the Niño-3 index and the 850-hPa zonal wind anomaly averaged across the western-to-central equatorial Pacific (5°S–5°N, 110°E–150°W) following northern, tropical, and southern eruptions (Fig. 3b). The positive lag indicates that the Niño-3 index lags the 850-hPa zonal wind. The correlation peaks at 2 months where the Niño-3 index lags the 850-hPa zonal wind following the northern and southern eruptions. The maximum correlations are observed at a lag time of 5 months for the tropical eruptions, which is consistent with that for the non-volcano El Niño events (figures omitted). The results imply that the westerly wind anomaly plays an essential role in favoring the development of an El Niño following all three types of eruptions. Thus, the key point of the question is to find the causes of the westerly wind enhancement. Previous studies have indicated that these El Niño–like SST responses to volcanic eruptions were caused by the movement of the ITCZ (Pausata et al. 2015a; Stevenson et al. 2016). We display the position of the ITCZ and the spatial patterns of the SST anomaly and 850-hPa winds in the following summer and winter after all three types of volcanic eruptions in Figs. 4 and 5, respectively. Following northern eruptions, the ITCZ shifts southward, especially in the winter (Figs. 4a and 5a). The westerly anomalies occur at the peak of the eruptions (Fig. 4d) and the El Niño–like response appears in the winter of year 0 (Fig. 5d). Because the easterly winds around the ITCZ are the weakest, this southward shift implies a weakening of the trade winds over the central to eastern Pacific. However, following tropical eruptions, the ITCZ does not move during year 0 but moves southward in the following year during both the summer and winter (Figs. 4b and 5b). In the meanwhile, there is no westerly anomaly at the peak of the eruptions (Fig. 4g). The westerly anomalies appear in the summer of year 1 with a warm SST anomaly over the central to eastern Pacific. Following southern events, the ITCZ shifts northward in year 0 and moves back to nearly its original location in year 1 (Figs. 4c and 5c). However, given the spatial distributions of SST and the 850-hPa wind (Fig. 4i), the westerly anomaly and El Niño–like response occur, which illustrates that the westerly anomalies are not directly related to the movement of the ITCZ. As such, the movement of the ITCZ is different following each of the three different types of eruptions. However, the westerly anomalies and El Niño–like SST responses emerge for each type of eruption. The southward shift of the ITCZ and the occurrence of the westerly anomaly match following northern eruptions but do not match following tropical and southern eruptions. Thus, we cannot attribute the El Niño–like SST responses following tropical and southern eruptions to the movements of the ITCZ. They must be determined by other mechanisms. The mechanism of the El Niño–like SST response to volcanic eruptions has led to heated discussions and is not yet well understood. Here, we suggest that the movement of the ITCZ can only explain the SST anomaly caused by northern eruptions, which has not been proposed before.
The positions of the ITCZ for a no-volcano summer (blue solid line) and the first (red dashed line) and second (green dashed line) summers after volcanic eruptions, sorted by the three types of volcanoes: (a) northern, (b) tropical, and (c) southern eruptions. The composite surface temperature (units: °C) and 850-hPa wind (m s−1) response to eruptions of varying types: (d),(e) northern, (f),(g) tropical, and (h),(i) southern eruptions. (d),(f),(h)The response during JJA of year 0, and (e),(g),(i) the response during JJA of year 1. The significance levels are determined according to the Monte Carlo test, and the values that are significant at the 95% confidence level are stippled.
Citation: Journal of Climate 31, 17; 10.1175/JCLI-D-17-0571.1
As in Fig. 4, but for the response during DJF with (d),(f),(h) the response during DJF of year 0+1 (D0JF+1) and (e),(g),(i) the response during DJF of year 1+2 (D1JF+2).
Citation: Journal of Climate 31, 17; 10.1175/JCLI-D-17-0571.1
The ocean dynamical thermostat mechanism proposed in previous studies (Clement et al. 1996) demonstrated that the SST responds unevenly to uniform radiative forcing. Figure 6 shows the temporal patterns of SST anomalies over the western and eastern Pacific following each of the three types of volcanic eruptions. Within the first few months after the eruption, the eastern Pacific favors a stronger cooling, since the net radiation flux over the eastern Pacific favors a stronger reduction than that over the western Pacific following all three types of eruptions, in which the changes in shortwave radiation play a dominant role (Fig. 7). It is partly because the different type of clouds over the western and eastern Pacific. However, around the peak of the volcanic eruptions, the SST features a zonal temperature gradient with stronger cooling in the western Pacific than that in the eastern Pacific for all three types of eruptions. In terms of radiation flux, the reduction in the net flux over the western Pacific is less than that over the eastern Pacific after the peak of the eruption (Figs. 7c,d), so the change in the SST gradient is not a direct response to the radiative forcing but instead emerges through dynamic processes. Specifically, this is due to the ocean dynamical thermostat mechanism proposed by Clement et al. (1996). When the Pacific was uniformly cooled following the eruptions, the forced changes could be partially balanced by vertical advection over the eastern Pacific. For the western Pacific, the radiative forcing due to volcanic eruptions is only balanced by the surface heat flux. This favors a zonal temperature gradient with weaker cooling in the eastern Pacific than that in the western Pacific around the peak of volcanic eruptions, which in turn weakens the equatorial easterlies and causes the thermocline to deepen in the east. Compared with Northern Hemispheric and tropical eruptions, the SST gradient over the Pacific is not that strong following southern eruptions (Fig. 6c); thus, the El Niño–like response is relatively weak (Fig. 2c). From the time evolution of the SST anomaly, we can identify that the cooling over the eastern Pacific is stronger than that over the western Pacific around the peak of eruptions in the Northern and Southern Hemispheres. However, this transition in the zonal SST gradient appears later following tropical eruptions, which corresponds to the spatial SST pattern in Fig. 2b.
The evolution of the composite sea surface temperature anomalies (units: °C) over the western Pacific Ocean and eastern Pacific Ocean following (a) northern, (b) tropical, and (c) southern eruptions after removing the zonal mean. The dashed lines represent confidence intervals of 95% derived from 1000 Monte Carlo simulations.
Citation: Journal of Climate 31, 17; 10.1175/JCLI-D-17-0571.1
Decomposition of the surface radiation flux anomalies following northern, tropical, and southern eruptions for the period of March (year 0) to August (year 0) (eruption to peak) and the period September (year 0) to July (year 1) over the (a) western Pacific (eruption to peak), (b) eastern Pacific (eruption to peak), (c) western Pacific [September (year 0) to July (year 1)], and (d) eastern Pacific [September (year 0) to July (year 1)].
Citation: Journal of Climate 31, 17; 10.1175/JCLI-D-17-0571.1
To further test the ocean dynamical thermostat mechanism in the CESM-LME, we choose the first and the second year after volcanic eruptions that induce negative radiative forcing over the tropical ocean. The response of the basin mean SST anomaly over the tropical Pacific (30°S–30°N, 130°E–80°W) and the Niño-3 index are compared (figure omitted). The basin mean SST decreases under the negative radiative forcing, whereas the temperature anomaly in the Niño-3 region displays an opposite sign to that of the forcing. This indicates that the imposed cooling in the eastern equatorial Pacific is balanced by anomalous horizontal and vertical advection, so that the resulting SST changes are less than expected.
The sea level pressure (SLP) changes following each of the three types of eruptions are further shown in Fig. 8. The corresponding SLPs feature a lowering SLP in the eastern Pacific and rising SLP in the western Pacific (Figs. 8a–c). However, the SLP changes following northern and tropical eruptions are much stronger and have longer durations, from the winter of year 0 to the summer of year 2. For southern eruptions, the anomalies are relatively weaker and have shorter durations, but the decrease of the zonal SLP gradient still occurs in the winter of year 0. To depict this west–east asymmetry in the SLP fields, Figs. 8d–f show the time evolutions of the SLP anomalies over the western and eastern Pacific. Corresponding to the time evolution of the SST anomalies (Fig. 6), the Pacific favors a decrease in the zonal SLP gradient, with a positive SLP anomaly over the western Pacific and negative SLP anomaly over the eastern Pacific after the peak of all three types of eruptions. Moreover, the transition point of the zonal SLP gradient is consistent with that of the zonal SST gradient. The changes in SLP will lead to westerly anomalies, which can further amplify the SST anomaly over the central to eastern Pacific and form a positive feedback (Bjerknes 1969). Hence, the fundamental cause of the anomalous westerly wind following all three types of eruptions is the ocean dynamical thermostat mechanism, through the uneven response of the western and eastern Pacific to the radiative forcing.
Longitude–time sections of the sea level pressure anomalies (hPa) along the equatorial region (10°S–10°N) from March (year 0) to November (year 3) for (a) northern eruptions, (b) tropical eruptions, and (c) southern eruptions after removing the zonal mean. The significance levels are determined according to the Monte Carlo test, and values that are significant at the 95% confidence level are stippled. Also shown is the evolution of the composite sea level pressure anomalies over the western Pacific Ocean and eastern Pacific Ocean following (d) northern, (e) tropical, and (f) southern eruptions. The dashed lines represent confidence intervals of 95% derived from 1000 Monte Carlo simulations.
Citation: Journal of Climate 31, 17; 10.1175/JCLI-D-17-0571.1
In conclusion, the westerly wind anomaly over the western-to-central Pacific plays an essential role in favoring the development of an El Niño following all three types of eruptions. To find the cause of the anomalous westerly winds, we check the movements of the ITCZ and the ocean dynamical thermostat mechanism. The shift of the ITCZ can only explain the El Niño–like response to northern eruptions, which is not applicable for tropical and southern eruptions. The ocean dynamical thermostat mechanism can explain the SST response following all three types of eruptions.
4. Summary and discussion
a. Summary
The latitude dependence of volcanic eruption impacts on tropical Pacific SST in the last millennium is not well understood; hence, the Pacific SST response to northern, tropical, and southern volcanic eruptions over the past millennium and the different response mechanisms arising due to differences in the volcanic forcing structure was investigated by analyzing the CESM Last Millennium Ensemble simulations. The main results are summarized below:
There is a significant El Niño–like SST anomaly over the equatorial Pacific 5–10 months after northern and tropical eruptions, and a weaker such response following southern eruptions. After removing the background state of the tropical Pacific, the El Niño–like response is more obvious. The warm SST anomaly appears at the peak of northern eruptions and lags by one year and half a year after tropical and southern eruptions, respectively. The warm SST anomaly is mainly confined to the eastern Pacific following tropical eruptions, which is different from those following northern and southern eruptions. The Niño-3 index reaches its peak value in the winter of year 1 following both northern and tropical eruptions, and appears 4 months earlier following southern eruption.
An anomalous westerly wind at 850 hPa is observed after all three types of eruptions. Both northern and southern eruptions favor westerly winds during year 0; for tropical eruptions, the westerly anomalies occur at the beginning of year 1. Two years after the eruptions, this anomaly converts into a La Niña–like state over the equatorial Pacific following all three types of volcanic eruptions.
The weakening of the trade wind in the western-to-central equatorial Pacific induces the El Niño–like warming over the eastern Pacific following all three types of eruptions. The shift in the ITCZ can only explain the El Niño–like SST response to northern eruptions and is not applicable for tropical and southern eruptions. The decrease in the zonal SST gradient along the equatorial Pacific around the peak of an eruption through the ocean dynamical thermostat mechanism can trigger the westerly anomalies over the equatorial Pacific. Subsequently, the decrease in the zonal SST gradient (which features a warm anomaly over the eastern Pacific and a cold anomaly over the western Pacific) and the westerly anomalies is enhanced via the Bjerknes feedback.
b. Discussion
Based on the CESM Last Millennium Ensemble, which has the largest ensemble of LM simulations, we find that there is an El Niño–like response following northern and tropical eruptions. Liu et al. (2018) suggested that ENSO responds differently to Northern Hemisphere, tropical, and Southern Hemisphere eruptions, with some aspects in common with this study, others instead differing. The SST response to different eruptions in Liu et al. (2018) was studied using only one experiment member. Since it is not the best way to distinguish externally forced change and internal variability in a single experiment, ensemble simulations are useful method to study volcano-forced responses. A strong point of this study is that we use a large ensemble, hence there is a large number of realizations of the same events. As for the mechanism, the El Niño–like response following eruptions at different latitudes was attributed to the movement of ITCZ in Stevenson et al. (2016); however, our results illustrate that the movement of ITCZ can only explain the El Niño–like SST response following northern eruptions. The ocean dynamical thermostat mechanism is applicable for the SST responses after all three types of eruptions. The main difference between our study and Stevenson et al. (2016) is that we try to have improved understanding of how different response mechanisms arise due to differences in the volcanic forcing structure in this study. Finally, the above studies investigated the SST response on the interannual time scales, whereas we focus on the monthly SST responses since El Niño has strong seasonality.
The different SST response to different volcanic forcing may be not only related to latitude but also magnitude. Three sets of experiments based on CCSM3 with different intensity of the volcanic forcing, named small, moderate, and large volcanic forcing experiments, were conducted to study the different effects of volcanic forcing with different magnitude on the tropical Pacific Ocean (McGregor et al. 2010). The results revealed that the larger the magnitude of the volcanic forcing, the larger the magnitude of the SST changes. The threshold of the volcanic forcing that leads to the El Niño–like warming was further examined based on the ERIK simulation (Lim et al. 2016). The results demonstrated that when the forcing was above a threshold, the El Niño–like warming appeared and the intensity of the SST response would become stronger as the forcing became stronger, but the spatial pattern would remain similar. The effect of volcanic aerosols with different magnitude on China’s monsoon precipitation has also been studied using two independently compiled histories of volcanism (Zhuo et al. 2014). Results show that as more sulfate aerosol is injected into the NH stratosphere, the drying trend over China becomes more severe. They also demonstrated that, although larger in magnitude, the overall spatial pattern in drying is similar. We further analyze the SST response to the 1258 Samalas eruption (8°24′36″S, 116°24′30″E) and 1815 Tambora eruption (8°15′S, 118°0′E), both of which are tropical eruptions but with different magnitude. The SST response indicates that the spatial pattern is similar but with different intensity. We thus suppose that the magnitude of volcanic forcing only affects the intensity of SST response, and the spatial pattern would remain similar.
As is mentioned above, we have revealed the short-term response of the tropical Pacific to volcanic forcing, as well as the different dynamical mechanisms due to differences in the volcanic forcing structure. The results are consistent with previous modeling studies; however, differences are also seen among different models. The difference may result from the uncertainties in the reconstruction of external forcing data, model bias, and also the uncertainties in reconstruction. There are two common aerosol forcing input datasets provided for the last millennium simulations in phase 5 of the Coupled Model Intercomparison Project (CMIP5), which are the Gao et al. (2008) (GRA) and the Crowley et al. (2008) (CEA) datasets. CESM-LME used the GRA volcanic forcing data for their simulations. The GRA forcing assumes a linear relationship between total stratospheric aerosol load and global aerosol optical depth (AOD), whereas CEA forcing provides an estimate of AOD from a 2/3 power scaling for eruptions larger than the 1991 Pinatubo eruption (Crowley and Unterman 2013). Different events are captured in the two datasets, and common events often have different amplitudes based on the different methods of conversion. Thus multimodel intercomparison based on the output of the last millennial climate simulation coordinated by CMIP5/PMIP3 should help us to reduce modeling uncertainties. To understand the sources of these model diversities, the Model Intercomparison Project on the climatic response to volcanic forcing (VolMIP) has defined a coordinated set of idealized volcanic perturbation experiments to be carried out in alignment with the CMIP6 protocol (Zanchettin et al. 2016). VolMIP provides a common volcanic forcing dataset for each experiment to minimize differences in the applied volcanic forcing. It defines a set of initial conditions to assess how internal climate variability contributes to determining the response. VolMIP will assess to what extent volcanically forced responses of the coupled ocean–atmosphere system are robustly simulated by coupled climate models and identify the causes that limit robust simulated behavior. The tropical Pacific responses to different volcanic forcing and the physical processes will be further examined using the VolMIP experiments in future studies.
It should also be noted that the mechanism behind the SST response to three types of volcanic eruptions remains an open question. As shown in Fig. 2, the cooling over the Maritime Continent (10°S–10°N, 100°–150°E) is stronger than that over the surrounding ocean due to the different heat capacities of land and ocean. The surface cooling caused by the eruptions can reduce the precipitation over the Maritime Continent. Thus, we further analyze the response of precipitation to the three types of eruptions; the longitude–time sections of the precipitation anomalies are shown in Fig. 9. It is evident that the central to eastern Pacific favors more precipitation after the peak of the eruptions following northern, tropical, and southern eruptions, which is consistent with the response of the SST to the three types of eruptions (Fig. 2). Moreover, the precipitation shows a more pronounced response than that of the SST following southern eruptions (Fig. 9c). It is noted that the precipitation over the Indo-Maritime Continent and its surrounding ocean (100°–150°E, 10°S–10°N) shows a significant decrease from the time of the eruption to the peak after northern and tropical eruptions, and with a weaker such tendency after southern eruptions.
Longitude–time sections of precipitation (units: mm day−1) and 850-hPa wind anomalies (m s−1) along the equatorial region (10°S–10°N) from March (year 0) to November (year 3) for (a) northern eruptions, (b) tropical eruptions, and (c) southern eruptions after removing the zonal mean. The significance levels are determined according to the Monte Carlo test, and slashes indicate the values that are significant at the 95% confidence level.
Citation: Journal of Climate 31, 17; 10.1175/JCLI-D-17-0571.1
The reduction of precipitation over the Maritime Continent can drive the anomalous equatorial westerly anomalies via the divergent winds to the east of the Maritime Continent. This response is consistent with the mechanism proposed by Ohba et al. (2013), but they only focused on tropical eruptions. The land–sea thermal contrast between the Maritime Continent and the surrounding ocean, combined with the divergent winds induced by the reduction in precipitation over the Maritime Continent, can further amplify the westerly anomalies triggered by the decreased zonal SST gradient over the equatorial Pacific through the ocean dynamical thermostat mechanism.
We emphasize the fundamental cause of the anomalous westerly wind is the ocean dynamical thermostat mechanism in this study, which is amplified by the surface cooling over the Maritime Continent. We propose the role of the surface cooling around the Maritime Continent based on some previous studies, which demonstrated that the cooling over the Maritime Continent weakens the equatorial Walker circulation and contributes to the westerly anomalies (Ohba et al. 2013). In a recent study, Wang et al. (2018) also suggested an important role of the surface cooling over the Maritime Continent and surrounding ocean in shaping the anomalous westerly wind over the central to eastern equatorial Pacific.
Khodri et al. (2017) implied that the initial westerly wind anomaly after the eruption was largely driven by land–sea temperature contrasts. These land–sea temperature contrasts induced a strong cooling over tropical continental surfaces, which resulted in reduced convective activity and drying of tropical America, Africa, and the Maritime Continent. However, the experiment including only cooling over tropical Africa indicated that it contributes to more than 80% of the initial westerly wind anomaly, rather than the Maritime Continent, Southeast Asia, or the extratropics. The reduced precipitation and tropospheric heating in the equatorial latitudes drove a Matsuno–Gill response where atmospheric equatorial Kelvin waves induce westerly wind anomalies over the western Pacific. Inspired by this study, we also examined role of the surface cooling over tropical Africa in CESM-LME. The results indicate that the strong surface cooling in the tropical Africa leads to a reduction of the tropical African precipitation. However, we cannot assess a direct effect of the reduced tropical African precipitation on the westerly wind anomaly over the western Pacific based on our coupled system model results, which need to be verified by the sensitivity experiments in our future studies.
Acknowledgments
This work was jointly supported by the National Natural Science Foundation of China (Grants 41675082 and 41330423), the National Program on Key Basic Research Project of China (Grant 2017YFA0604600), and the Jiangsu Collaborative Innovation Center for Climate Change.
REFERENCES
Adams, J. B., M. E. Mann, and C. M. Ammann, 2003: Proxy evidence for an El Niño-like response to volcanic forcing. Nature, 426, 274–278, https://doi.org/10.1038/nature02101.
Alexander, M. A., I. Bladé, M. Newman, J. R. Lanzante, N.-C. Lau, and J. D. Scott, 2002: The atmospheric bridge: The influence of ENSO teleconnections on air–sea interaction over the global oceans. J. Climate, 15, 2205–2231, https://doi.org/10.1175/1520-0442(2002)015<2205:TABTIO>2.0.CO;2.
Berger, A. L., 1978: Long-term variations of daily insolation and quaternary climatic changes. J. Atmos. Sci., 35, 2362–2367, https://doi.org/10.1175/1520-0469(1978)035<2362:LTVODI>2.0.CO;2.
Bjerknes, J., 1969: Atmospheric teleconnections from the equatorial Pacific. Mon. Wea. Rev., 97, 163–172, https://doi.org/10.1175/1520-0493(1969)097<0163:ATFTEP>2.3.CO;2.
Clement, A. C., R. Seager, M. A. Cane, and S. E. Zebiak, 1996: An ocean dynamical thermostat. J. Climate, 9, 2190–2196, https://doi.org/10.1175/1520-0442(1996)009<2190:AODT>2.0.CO;2.
Colose, C. M., A. N. LeGrande, and M. Vuille, 2016: Hemispherically asymmetric volcanic forcing of tropical hydroclimate during the last millennium. Earth Syst. Dyn., 7, 681–696, https://doi.org/10.5194/esd-7-681-2016.
Crowley, T. J., and M. B. Unterman, 2013: Technical details concerning development of a 1200 yr proxy index for global volcanism. Earth Syst. Sci. Data, 5, 187–197, https://doi.org/10.5194/essd-5-187-2013.
Crowley, T. J., G. Zielinski, B. Vinther, R. Udisti, K. Kreutzs, J. Cole-Dai, and E. Castellano, 2008: Volcanism and the Little Ice Age. PAGES Newsletter, Vol. 16, No. 2, PAGES International Project Office, Bern, Switzerland, 22–23, https://doi.org/10.22498/pages.16.2.22.
D’Arrigo, R., R. Wilson, and A. Tudhope, 2009: The impact of volcanic forcing on tropical temperatures during the past four centuries. Nat. Geosci., 2, 51–56, https://doi.org/10.1038/ngeo393.
Driscoll, S., A. Bozzo, L. J. Gray, A. Robock, and G. Stenchikov, 2012: Coupled Model Intercomparison Project 5 (CMIP5) simulations of climate following volcanic eruptions. J. Geophys. Res., 117, D17105, https://doi.org/10.1029/2012JD017607.
Emile-Geay, J., R. Seager, M. A. Cane, E. R. Cook, and G. H. Haug, 2008: Volcanoes and ENSO over the past millennium. J. Climate, 21, 3134–3148, https://doi.org/10.1175/2007JCLI1884.1.
Gao, C., A. Robock, and C. Ammann, 2008: Volcanic forcing of climate over the past 1500 years: An improved ice core-based index for climate models. J. Geophys. Res., 113, D23111, https://doi.org/10.1029/2008JD010239; Corrigendum, 114, D09103, https://doi.org/10.1029/2009JD012133; Corrigendum, 117, D16112, https://doi.org/10.1029/2012JD018052.
Graf, H.-F., D. Zanchettin, C. Timmreck, and M. Bittner, 2014: Observational constraints on the tropospheric and near-surface winter signature of the Northern Hemisphere stratospheric polar vortex. Climate Dyn., 43, 3245–3266, https://doi.org/10.1007/s00382-014-2101-0.
Handler, P., 1984: Possible association of stratospheric aerosols and El Niño type events. Geophys. Res. Lett., 11, 1121–1124, https://doi.org/10.1029/GL011i011p01121.
Haurwitz, M. W., and G. W. Brier, 1981: A critique of the superposed epoch analysis method: Its application to solar–weather relations. Mon. Wea. Rev., 109, 2074–2079, https://doi.org/10.1175/1520-0493(1981)109<2074:ACOTSE>2.0.CO;2.
Hurtt, G. C., and Coauthors, 2011: Harmonization of land-use scenarios for the period 1500–2100: 600 years of global gridded annual land-use transitions, wood harvest, and resulting secondary lands. Climatic Change, 109, 117–161, https://doi.org/10.1007/s10584-011-0153-2.
Iles, C. E., G. C. Hegerl, A. P. Schurer, and X. Zhang, 2013: The effect of volcanic eruptions on global precipitation. J. Geophys. Res. Atmos., 118, 8770–8786, https://doi.org/10.1002/jgrd.50678.
Kang, S. M., I. M. Held, D. M. W. Frierson, and M. Zhao, 2008: The response of the ITCZ to extratropical thermal forcing: Idealized slab-ocean experiments with a GCM. J. Climate, 21, 3521–3532, https://doi.org/10.1175/2007JCLI2146.1.
Khodri, M., and Coauthors, 2017: Tropical explosive volcanic eruptions can trigger El Niño by cooling tropical Africa. Nat. Commun., 8, 778, https://doi.org/10.1038/s41467-017-00755-6.
Le, T., 2017: ENSO response to external forcing in CMIP5 simulations of the last millennium. Global Planet. Change, 148, 105–112, https://doi.org/10.1016/j.gloplacha.2016.12.002.
Li, J., and Coauthors, 2013: El Niño modulations over the past seven centuries. Nat. Climate Change, 3, 822–826, https://doi.org/10.1038/nclimate1936.
Lim, H.-G., S.-W. Yeh, J.-S. Kug, Y.-G. Park, J.-H. Park, R. Park, and C.-K. Song, 2016: Threshold of the volcanic forcing that leads the El Niño-like warming in the last millennium: Results from the ERIK simulation. Climate Dyn., 46, 3725–3736, https://doi.org/10.1007/s00382-015-2799-3.
Liu, F., J. Li, B. Wang, J. Liu, T. Li, G. Huang, and Z. Wang, 2018: Divergent El Niño responses to volcanic eruptions at different latitudes over the past millennium. Climate Dyn., 50, 3799–3812, https://doi.org/10.1007/s00382-017-3846-z.
Maher, N., S. McGregor, M. H. England, and A. Sen Gupta, 2015: Effects of volcanism on tropical variability. Geophys. Res. Lett., 42, 6024–6033, https://doi.org/10.1002/2015GL064751.
Man, W., and T. Zhou, 2014: Response of the East Asian summer monsoon to large volcanic eruptions during the last millennium. Chin. Sci. Bull., 59, 4123–4129, https://doi.org/10.1007/s11434-014-0404-5.
Man, W., T. Zhou, and J. H. Jungclaus, 2014: Effects of large volcanic eruptions on global summer climate and East Asian monsoon changes during the last millennium: Analysis of MPI-ESM simulations. J. Climate, 27, 7394–7409, https://doi.org/10.1175/JCLI-D-13-00739.1.
Mann, M. E., M. A. Cane, S. E. Zebiak, and A. Clement, 2005: Volcanic and solar forcing of the tropical Pacific over the past 1000 years. J. Climate, 18, 447–456, https://doi.org/10.1175/JCLI-3276.1.
McCracken, M. C., and F. M. Luther, 1984: Preliminary estimate of the radiative and climatic effects of the El Chichon eruption. Geofis. Int., 23, 385–401.
McGregor, S., and A. Timmermann, 2011: The effect of explosive tropical volcanism on ENSO. J. Climate, 24, 2178–2191, https://doi.org/10.1175/2010JCLI3990.1.
McGregor, S., A. Timmermann, and O. Timm, 2010: A unified proxy for ENSO and PDO variability since 1650. Climate Past, 6, 1–17, https://doi.org/10.5194/cp-6-1-2010.
Nicholls, N., 1990: Low-latitude volcanic eruptions and the El Niño/Southern Oscillation: A reply. Int. J. Climatol., 10, 425–429, https://doi.org/10.1002/joc.3370100410.
Ohba, M., H. Shiogama, T. Yokohata, and M. Watanabe, 2013: Impact of strong tropical volcanic eruptions on ENSO simulated in a coupled GCM. J. Climate, 26, 5169–5182, https://doi.org/10.1175/JCLI-D-12-00471.1.
Oman, L., A. Robock, G. Stenchikov, G. A. Schmidt, and R. Ruedy, 2005: Climatic response to high-latitude volcanic eruptions. J. Geophys. Res., 110, D13103, https://doi.org/10.1029/2004JD005487.
Otto-Bliesner, B. L., and Coauthors, 2016: Climate variability and change since 850 CE: An ensemble approach with the Community Earth System Model. Bull. Amer. Meteor. Soc., 97, 735–754, https://doi.org/10.1175/BAMS-D-14-00233.1.
Paik, S., and S.-K. Min, 2017: Climate responses to volcanic eruptions assessed from observations and CMIP5 multi-models. Climate Dyn., 48, 1017–1030, https://doi.org/10.1007/s00382-016-3125-4.
Pausata, F. S. R., L. Chafik, R. Caballero, and D. S. Battisti, 2015a: Impacts of high-latitude volcanic eruptions on ENSO and AMOC. Proc. Natl. Acad. Sci. USA, 112, 13 784–13 788, https://doi.org/10.1073/pnas.1509153112.
Pausata, F. S. R., A. Grini, R. Caballero, A. Hannachi, and Ø. Seland, 2015b: High-latitude volcanic eruptions in the Norwegian Earth System Model: The effect of different initial conditions and of the ensemble size. Tellus, 67B, 26728, https://doi.org/10.3402/tellusb.v67.26728.
Pausata, F. S. R., C. Karamperidou, R. Caballero, and D. S. Battisti, 2016: ENSO response to high-latitude volcanic eruptions in the Northern Hemisphere: The role of the initial conditions. Geophys. Res. Lett., 43, 8694–8702, https://doi.org/10.1002/2016GL069575.
Pongratz, J., T. Raddatz, C. H. Reick, M. Esch, and M. Claussen, 2009: Radiative forcing from anthropogenic land cover change since A.D. 800. Geophys. Res. Lett., 36, L02709, https://doi.org/10.1029/2008GL036394.
Predybaylo, E., G. Stenchikov, A. T. Wittenberg, and F. Zeng, 2017: Impacts of a Pinatubo-size volcanic eruption on ENSO. J. Geophys. Res. Atmos., 122, 925–947, https://doi.org/10.1002/2016JD025796.
Robock, A., 2000: Volcanic eruptions and climate. Rev. Geophys., 38, 191–219, https://doi.org/10.1029/1998RG000054; Corrigendum, 45, RG3005, https://doi.org/10.1029/2007RG000232.
Robock, A., K. E. Taylor, G. Stenchikov, and Y. Liu, 1995: GCM evaluation of a mechanism for El Niño triggering by the El Chichón ash cloud. Geophys. Res. Lett., 22, 2369–2372, https://doi.org/10.1029/95GL02065.
Schmidt, G. A., and Coauthors, 2012: Climate forcing reconstructions for use in PMIP simulations of the last millennium (v1.1). Geosci. Model Dev., 5, 185–191, https://doi.org/10.5194/gmd-5-185-2012.
Schneider, D. P., C. M. Ammann, B. L. Otto-Bliesner, and D. S. Kaufman, 2009: Climate response to large, high-latitude and low-latitude volcanic eruptions in the Community Climate System Model. J. Geophys. Res., 114, D15101, https://doi.org/10.1029/2008JD011222.
Self, S., M. R. Rampino, J. Zhao, and M. G. Katz, 1997: Volcanic aerosol perturbations and strong El Niño events: No general correlation. Geophys. Res. Lett., 24, 1247–1250, https://doi.org/10.1029/97GL01127.
Seo, J., S. M. Kang, and D. M. W. Frierson, 2014: Sensitivity of intertropical convergence zone movement to the latitudinal position of thermal forcing. J. Climate, 27, 3035–3042, https://doi.org/10.1175/JCLI-D-13-00691.1.
Stenchikov, G., K. Hamilton, R. J. Stouffer, A. Robock, V. Ramaswamy, B. Santer, and H. F. Graf, 2006: Arctic Oscillation response to volcanic eruptions in the IPCC AR4 climate models. J. Geophys. Res., 111, D07107, https://doi.org/10.1029/2005JD006286.
Stevenson, S., B. Otto-Bliesner, J. Fasullo, and E. Brady, 2016: “El Niño like” hydroclimate responses to last millennium volcanic eruptions. J. Climate, 29, 2907–2921, https://doi.org/10.1175/JCLI-D-15-0239.1.
Stevenson, S., J. T. Fasullo, B. L. Otto-Bliesner, R. A. Tomas, and C. Gao, 2017: Role of eruption season in reconciling model and proxy responses to tropical volcanism. Proc. Natl. Acad. Sci. USA, 114, 1822–1826, https://doi.org/10.1073/pnas.1612505114.
Swingedouw, D., J. Mignot, P. Ortega, M. Khodri, M. Menegoz, C. Cassou, and V. Hanquiez, 2017: Impact of explosive volcanic eruptions on the main climate variability modes. Global Planet. Change, 150, 24–45, https://doi.org/10.1016/j.gloplacha.2017.01.006.
Timmreck, C., 2012: Modeling the climatic effects of large explosive volcanic eruptions. Wiley Interdiscip. Rev.: Climate Change, 3, 545–564, https://doi.org/10.1002/wcc.192.
Trenberth, K. E., and A. Dai, 2007: Effects of Mount Pinatubo volcanic eruption on the hydrological cycle as an analog of geoengineering. Geophys. Res. Lett., 34, L15702, https://doi.org/10.1029/2007GL030524.
Vieira, L. E. A., and S. K. Solanki, 2010: Evolution of the solar magnetic flux on time scales of years to millenia [sic]. Astron. Astrophys., 509, A100, https://doi.org/10.1051/0004-6361/200913276.
Wahl, E. R., H. F. Diaz, J. E. Smerdon, and C. M. Ammann, 2014: Late winter temperature response to large tropical volcanic eruptions in temperate western North America: Relationship to ENSO phases. Global Planet. Change, 122, 238–250, https://doi.org/10.1016/j.gloplacha.2014.08.005.
Wang, T., D. Guo, Y. Gao, H. Wang, F. Zheng, Y. Zhu, J. Miao, and Y. Hu, 2018: Modulation of ENSO evolution by strong tropical volcanic eruptions. Climate Dyn., https://doi.org/10.1007/s00382-017-4021-2, in press.
Zambri, B., and A. Robock, 2016: Winter warming and summer monsoon reduction after volcanic eruptions in Coupled Model Intercomparison Project 5 (CMIP5) simulations. Geophys. Res. Lett., 43, 10 920–10 928, https://doi.org/10.1002/2016GL070460.
Zanchettin, D., C. Timmreck, H. F. Graf, A. Rubino, S. Lorenz, K. Lohmann, K. Krüger, and J. H. Jungclaus, 2012: Bi-decadal variability excited in the coupled ocean–atmosphere system by strong tropical volcanic eruptions. Climate Dyn., 39, 419–444, https://doi.org/10.1007/s00382-011-1167-1.
Zanchettin, D., O. Bothe, H. F. Graf, S. J. Lorenz, J. Luterbacher, C. Timmreck, and J. H. Jungclaus, 2013: Background conditions influence the decadal climate response to strong volcanic eruptions. J. Geophys. Res. Atmos., 118, 4090–4106, https://doi.org/10.1002/jgrd.50229.
Zanchettin, D., and Coauthors, 2016: The Model Intercomparison Project on the climatic response to volcanic forcing (VolMIP): Experimental design and forcing input data for CMIP6. Geosci. Model Dev., 9, 2701–2719, https://doi.org/10.5194/gmd-9-2701-2016.
Zebiak, S. E., and M. A. Cane, 1987: A model El Niño–Southern Oscillation. Mon. Wea. Rev., 115, 2262–2278, https://doi.org/10.1175/1520-0493(1987)115<2262:AMENO>2.0.CO;2.
Zhuo, Z., C. Gao, and Y. Pan, 2014: Proxy evidence for China’s monsoon precipitation response to volcanic aerosols over the past seven centuries. J. Geophys. Res. Atmos., 119, 6638–6652, https://doi.org/10.1002/2013JD021061.