Slow Preconditioning for the Abrupt Convective Jump over the Northwest Pacific during Summer

Wenyu Zhou Scripps Institution of Oceanography, University of California, San Diego, La Jolla, California

Search for other papers by Wenyu Zhou in
Current site
Google Scholar
PubMed
Close
,
Shang-Ping Xie Scripps Institution of Oceanography, University of California, San Diego, La Jolla, California, and Physical Oceanography Laboratory/Qingdao Collaborative Innovation Center of Marine Science and Technology, Ocean University of China, Qingdao, China

Search for other papers by Shang-Ping Xie in
Current site
Google Scholar
PubMed
Close
, and
Zhen-Qiang Zhou Physical Oceanography Laboratory/Qingdao Collaborative Innovation Center of Marine Science and Technology, Ocean University of China, Qingdao, China, and Scripps Institution of Oceanography, University of California, San Diego, La Jolla, California

Search for other papers by Zhen-Qiang Zhou in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

The rapid intensification of convective activity in mid-July over the northwest Pacific marks the final stage of the Asian summer monsoon, accompanied by major shifts in regional rainfall and circulation patterns. An entraining plume model is used to investigate the physical processes underlying the abrupt convective jump. Despite little change in sea surface temperature (SST), gradual lower-troposphere mixing leads to a threshold transition in the model as follows. Before mid-July, although SST is already high (29°C), the convective plume is inhibited by the capping inversion above the trade cumulus boundary layer. As the lower troposphere is gradually mixed, the boundary layer top rises with reduced atmospheric stability and increased humidity in the lower troposphere. These factors weaken the inhibition effect of the inversion on the entraining plume. As soon as the plume is able to overcome the inversion barrier, it can rise all the way to the upper troposphere. This marks an abrupt threshold transition to a deep convection regime with heavy rainfall. The convective available potential energy (CAPE) of the entraining plume is found to be a better indicator of the rainfall intensity compared to the conventional undiluted CAPE. The latter fails to capture the onset by neglecting interactions between convective clouds and the environment. Current general circulation models (GCMs) fail to capture the abrupt convective jump and instead simulate a rather smooth seasonal evolution of rainfall. Compared to observations, GCMs simulate a higher trade cumulus top with excessive mixing in the lower troposphere. Convection is no longer inhibited by the inversion barrier, and rainfall simply follows the smooth variation of SST.

Corresponding author address: Wenyu Zhou, Scripps Institution of Oceanography, University of California, San Diego, 9500 Gilman Dr., La Jolla, CA 92093-0206. E-mail: zhouwy1128@gmail.com

Abstract

The rapid intensification of convective activity in mid-July over the northwest Pacific marks the final stage of the Asian summer monsoon, accompanied by major shifts in regional rainfall and circulation patterns. An entraining plume model is used to investigate the physical processes underlying the abrupt convective jump. Despite little change in sea surface temperature (SST), gradual lower-troposphere mixing leads to a threshold transition in the model as follows. Before mid-July, although SST is already high (29°C), the convective plume is inhibited by the capping inversion above the trade cumulus boundary layer. As the lower troposphere is gradually mixed, the boundary layer top rises with reduced atmospheric stability and increased humidity in the lower troposphere. These factors weaken the inhibition effect of the inversion on the entraining plume. As soon as the plume is able to overcome the inversion barrier, it can rise all the way to the upper troposphere. This marks an abrupt threshold transition to a deep convection regime with heavy rainfall. The convective available potential energy (CAPE) of the entraining plume is found to be a better indicator of the rainfall intensity compared to the conventional undiluted CAPE. The latter fails to capture the onset by neglecting interactions between convective clouds and the environment. Current general circulation models (GCMs) fail to capture the abrupt convective jump and instead simulate a rather smooth seasonal evolution of rainfall. Compared to observations, GCMs simulate a higher trade cumulus top with excessive mixing in the lower troposphere. Convection is no longer inhibited by the inversion barrier, and rainfall simply follows the smooth variation of SST.

Corresponding author address: Wenyu Zhou, Scripps Institution of Oceanography, University of California, San Diego, 9500 Gilman Dr., La Jolla, CA 92093-0206. E-mail: zhouwy1128@gmail.com
Keywords: Convection; Monsoons

1. Introduction

The Asian summer monsoon (ASM) is the dominant feature of Asian climate and provides much-needed rain over the world’s most populated regions. Although it is ultimately driven by the smooth seasonal variation of solar insolation, the ASM exhibits discontinuous evolution in various elements (Wu and Wang 2001; Ding and Chan 2005), suggesting complicated interactions among atmospheric, oceanic, land, and geographic factors (Li and Yanai 1996; Wu and Zhang 1998; Xie et al. 2006). The ASM is divided into three distinct stages based on its climatological onset dates in different monsoon regions. In mid-May, the first transition of the ASM manifests rainfall over the Bay of Bengal and the South China Sea. Subsequently, in mid-June, the Indian summer monsoon begins and the East Asian rainy season arrives with mei-yu in China and baiu in Japan (mei-yu–baiu). Finally, after mid-July, convection over the western Pacific abruptly expands northeastward, forming a subtropical rainband centered at 20°N, 150°E (Fig. 1c).

Fig. 1.
Fig. 1.

Climatological mean precipitation (color shading) and surface wind (vector) averaged over (left) 3–12 July, (center) 18–27 July, and (right) the difference between these two periods for (a)–(c) observations and (d)–(f) AGCMs. The red contours are the 29°C SST isotherm in (a),(b) and the difference in (c). The hatched area indicates the CJ area that we are interested in.

Citation: Journal of Climate 29, 22; 10.1175/JCLI-D-16-0342.1

This last monsoon transition has been first recognized by Ueda et al. (1995) as the so-called convective jump. Over the northwest Pacific, the atmosphere undergoes a rapid regime transition from suppressed trade cumulus to active deep convection, with rainfall increasing by about 6 mm day−1 over 10 days (Fig. 2a). This abrupt transition is tightly linked with the withdrawal of the mei-yu–baiu rainband (Fig. 1c) and the eastward retreat of the western Pacific subtropical high (Fig. 1b), marking the final stage of the ASM development. The associated large-scale patterns also significantly affect the distribution of tropical cyclones over the western tropical Pacific (Ueda et al. 1995; Xu and Lu 2016).

Fig. 2.
Fig. 2.

(a) Time evolution of the mean cloud water content (color shading), SST (black solid line for SST > 29°C, gray dashed line for SST < 28°C, and black dashed line for 28° < SST < 29°C), and pentad precipitation (red line) over the CJ area from 15 May to 15 August. (b) Time evolution of the mean relative humidity (color shading) and the lower-troposphere-integrated relative humidity (; red line). (c) Difference in air temperature relative to that on 1 July (color shading) and time evolution of the atmospheric instability Δθυ, measured by the virtual potential temperature difference between 750 and 950 hPa (red line).

Citation: Journal of Climate 29, 22; 10.1175/JCLI-D-16-0342.1

Despite its profound impact on the Asian summer climate, our understanding of the underlying physical mechanisms for the convective jump remains elusive. It has been speculated that an increase in SST causes the onset of convection (Ueda and Yasunari 1996) since the subtropical rainband is roughly collocated with the northeastward expansion of warm sea surface temperature (SST). However, a closer examination finds that local SST is well above the SST threshold for tropical deep convection (~27°C; Johnson and Xie 2010) in early July and varies very little through the process of the regime transition (Fig. 2a). Moreover, general circulation model (GCM) simulations show that the convective jump is largely unaffected by artificially removing the SST variation after 1 July, implying that non-SST factors like atmospheric transient effects, solar insolation variation, or land memory are the dominating contributors (Ueda et al. 2009). Indeed, although the increase of SST has stopped, the lower troposphere continues to evolve. Before early July, the atmosphere is in a typical trade cumulus regime with thick clouds capped in the boundary layer. From early to mid-July, there is a noticeable rise in the trade cumulus top (Fig. 2a). At the same time, the lower troposphere is moistened (Fig. 2b) and destabilized (Fig. 2c). These changes are generally recognized as favorable for deep convection triggering. However, it is unclear how such gradual preconditioning can lead to an abrupt transition and what determines the onset.

In this study, we apply an entraining plume model, which can be seen as a simple idealization for an ensemble of convective clouds, to study the response of convection to the gradual preconditioning in the environment. We show that an updraft plume loses its buoyancy through entraining the dry environmental air and is initially intercepted by the capping inversion at the top of the planetary boundary layer (PBL), although SST has been sufficiently high to support deep convection. The gradual mixing of the lower troposphere raises the PBL top, moistens and destabilizes the lower troposphere, and eventually leads to a threshold transition to the deep convection regime as soon as the plume can marginally overcome the inversion barrier. Our focus on local convective processes is motivated by the observation that the large anomalies of precipitation and circulation are confined to the northwest Pacific while the South Asian monsoon is at the mature stage with little change across the mid-July transition (Fig. 1c).

The same analysis is applied to evaluate the performance of atmosphere-only GCMs (AGCMs). Different from the observed convective jump, AGCMs commonly simulate a smooth seasonal evolution of rainfall over the northwest Pacific. We find that the trade cumulus inversion top is higher in AGCMs as a result of the excessive mixing in the lower troposphere. Convective clouds are no longer intercepted by the inversion barrier but simply follow the smooth variation of SST.

The rest of this paper is organized as follows. Section 2 describes the datasets and the entraining plume model used in this study. Section 3 illustrates the abrupt regime transition as a response of the plume buoyancy to the gradual mixing of the lower troposphere. Section 4 investigates the biases in AGCMs and explores the underlying reasons. Section 5 offers a summary with discussion.

2. Methods

a. Datasets

Datasets used in this study include the observed and reanalyzed datasets (for simplicity referred to as observations hereafter) as well as outputs of AGCMs. Particularly, we have utilized the pentad and monthly precipitation data from the Global Precipitation Climatology Project (Adler et al. 2003) to examine the convective jump. The daily and monthly SST, air temperature, humidity, and cloud water content are obtained from the ERA-Interim dataset (Dee et al. 2011). To evaluate the performance in AGCMs, we have also examined the daily and monthly output of the historical run from 16 CMIP5 (Taylor et al. 2012) AGCMs as listed in Table 1.

Table 1.

List of 16 CMIP5 AGCMs used in this study. (Expansions of acronyms are available online at http://www.ametsoc.org/PubsAcronymList.)

Table 1.

In this study, we focus on the climatological and area mean by taking the average from 1979 to 2005 and over the convective jump (CJ) area (15°–25°N, 140°–160°E), as previously defined by Ueda et al. (1995, 2009) and marked in Fig. 1c. All datasets have been regridded to a common 2.5° latitude × 2.5° longitude grid before taking the area mean.

b. An entraining plume model

The convective available potential energy (CAPE) is a widely used index to measure the thermodynamic instability of the atmosphere and thus the favorability for deep convection. It computes the vertically integrated buoyancy of an undiluted plume originated from the surface. By assuming an undiluted plume, however, the effect of the ambient atmospheric humidity on the plume buoyancy through entrainment is not represented, such that the undiluted CAPE is overwhelmingly controlled by the updraft parcel from the surface and thus SST. This could lead to misinterpretation of the atmospheric favorability for deep convection under certain conditions.

Here, in order to represent the interaction between convective clouds and the environment, we use an entraining plume model (Kain and Fritsch 1990; Singh and O’Gorman 2013). The plume interacts with the environment by entraining the ambient air as it rises, such that its thermodynamic properties gradually change:
e1
e2
where ε is the entrainment rate, h is the total moist static energy, and q is the total water content. The subscripts u and e refer to the entraining updraft plume and the environment, respectively. Here, we assume the entrainment rate ε to be inversely proportional to the height z:
e3
This is based on the assumption that entrainment rate is proportional to the cloud diameter, which further scales with the height z. Similar formulas have been used in popular convective parameterizations (Bretherton et al. 2004; Kain 2004). In this study, we choose C = 0.45 to represent a “mean” entrainment strength of an ensemble of convective clouds. Increasing (decreasing) C will suppress (favor) convective plume and thus delay (accelerate) the onset time, but the abrupt transition behavior discussed in the following section will not be affected.
At each level, given its moist static energy hu and total water content qu, one can estimate the plume’s virtual temperature and further compute the plume buoyancy Bu:
e4
The vertical velocity of the plume wu is integrated from the level of free convection:
e5
where α is a virtual mass coefficient, and β is a drag coefficient (Simpson and Wiggert 1969). Here we use α = 1 and β = 2 according to Bretherton et al. (2004). The plume detrains when wu = 0 and the level is marked as the level of detrainment (LD).
Finally, we define an entraining CAPE (CAPE-ent) as the vertical integral of the entraining plume buoyancy from the surface to LD:
e6
Molinari et al. (2012) find that such entraining CAPE is more accurate compared to the conventional undiluted CAPE when used to predict the distribution of vigorous deep convection in tropical cyclones. Zhang (2009) shows that the entraining CAPE has some advantages over the undiluted CAPE when applied to convective mass flux closure in convective parameterizations. Despite this pioneering work, the acceptance of the entraining CAPE is still limited compared to the widely used undiluted CAPE.

3. A threshold regime transition by gradual lower-troposphere mixing

To investigate the regime transition, we have applied the entraining plume model to the observed daily thermodynamic profiles from mid-May to mid-August. Figure 3a shows the evolution of the entraining plume, which represents the behavior of an ensemble of convective clouds. The abrupt regime transition from trade cumulus to deep convection is reflected by an abrupt jump of the plume LD from the lower to upper troposphere around mid-July (white line in Fig. 3a).

Fig. 3.
Fig. 3.

Time evolution of the plume buoyancy (color shading), the LD (white line), and the CAPE (black line) computed using (a),(c) an entraining plume and (b),(d) an undiluted plume. The zero contour is highlighted with a thin pink line in (a). Please note that the upper branch of this zero contour indicates the LNB. The four stages mentioned in section 3 are denoted by the triangle symbols in (a). The climatological pentad precipitation (Pr) over the CJ area is plotted as the red line in (c),(d) for reference.

Citation: Journal of Climate 29, 22; 10.1175/JCLI-D-16-0342.1

According to its buoyancy and LD, the evolution of the plume can be categorized into four stages (denoted by triangle symbols in Fig. 3a): 1) On 10 June, the plume detrains in the lower troposphere as it quickly loses buoyancy by entraining dry air and is inhibited by the capping inversion (stable layer at the PBL top). The atmosphere is in a typical regime of trade cumulus. 2) On 1 July, SST has been further warmed up to 29°C and the plume can potentially reach a higher altitude, as indicated by the level of neutral buoyancy (LNB). However, there is a layer of negative buoyancy at about 750 hPa because of the capping inversion. The plume is thus intercepted and only detrains in the lower troposphere. The inversion layer of negative plume buoyancy, however, only works as a temporary barrier to inhibit convection. 3) Indeed, once the plume is able to marginally overcome the inversion barrier on 11 July, it rises all the way up to the upper troposphere, leading to the abrupt transition to the deep convection regime. 4) On 20 July, as the free troposphere is further moistened by convection (Fig. 2b), the plume is able to reach even higher altitude with reduced dry air dilution, marking the mature stage of deep convection.

As schematically illustrated in Fig. 4, the threshold transition at stage 3 is caused by a gradual preconditioning in the lower troposphere. From early to mid-July, as the lower troposphere is gradually mixed, the trade cumulus top rises with reduced atmospheric stability and increased humidity in the lower troposphere (Fig. 2). The moister lower troposphere reduces the buoyancy reduction due to entrainment, leading to a larger buoyancy when the plume reaches the capping inversion. Moreover, the rising inversion top leaves more distance (from the level of free convection to the PBL top) for the plume to gain a larger vertical velocity before it reaches the inversion top. These factors help the plume to overcome the inversion barrier and eventually lead to a threshold transition.

Fig. 4.
Fig. 4.

Schematic diagram illustrating the four stages during the abrupt regime transition from trade cumulus to deep convection. The blue line shows the vertical profile of the virtual equivalent potential temperature for the environment air, while the red line shows that for the updraft entraining plume. The positive plume buoyancy is indicated by the hatched area between. The abrupt jump of the detrainment level is indicated by the thick purple line.

Citation: Journal of Climate 29, 22; 10.1175/JCLI-D-16-0342.1

Rainfall intensity variation is well captured by the entraining CAPE (CAPE-ent; Fig. 3c), whereas the undiluted CAPE fails to predict the onset by tightly following SST (Fig. 3d). Moreover, the mismatch between CAPE and rainfall casts doubts on the widely used CAPE (short for the undiluted CAPE) closure for deep convection (e.g., Zhang and McFarlane 1995), which assumes that convective activity is controlled by CAPE as follows:
e7
where the subscript cu denotes CAPE change due to convective-scale processes, CAPEo is the threshold value of CAPE above which convection is triggered, and τ is the relaxation time scale. While CAPE has already reached 2000 J kg−1 by 1 July (“moderate instability” according to the categorization by the NCEP/Storm Prediction Center), convective activity is still strongly suppressed. Moreover, during the rainfall jump from 1 to 20 July, convection activity [lhs of Eq. (7)] increases significantly from trade cumulus to deep convection, while CAPE [rhs of Eq. (7)] is almost unchanged. Both errors essentially come from the overestimated sensitivity of CAPE to SST by neglecting the dry air entrainment.

4. A biased smooth evolution in AGCMs

Given its profound impact on the Asian summer climate, it is crucial for AGCMs to capture the abrupt transition over the northwest Pacific. AGCMs, however, simulate a rather smooth seasonal evolution of rainfall, with the rainy season starting as early as June (Fig. 5a). Thus, the abrupt convective jump around mid-July is missed, leading to a very different large-scale climate over the northwest Pacific and East Asia as compared to observations (Fig. 1).

Fig. 5.
Fig. 5.

Monthly variation of the mean (a) precipitation and (b) large-scale surface wind speed over the CJ area simulated by AGCMs. The black line shows the observation and the red line shows the multimodel mean of AGCMs.

Citation: Journal of Climate 29, 22; 10.1175/JCLI-D-16-0342.1

As we show in section 3, the evolution of the trade cumulus boundary layer plays an important role in the abrupt transition. We then compare the trade cumulus simulated in AGCMs with observations. AGCMs simulate a higher trade cumulus top compared to observations before and after the rainy season (Fig. 6), suggesting a raised inversion top with excessive lower-troposphere mixing. The raised inversion top can also be seen from the relative humidity forecast in GCMs. Under the trade cumulus region, the contour of 60% relative humidity, as an approximate indicator of the dry inversion, is raised from approximately 840 hPa in observations to approximatedly 800 hPa in GCMs (Fig. 7).

Fig. 6.
Fig. 6.

Monthly variation of cloud water content in (a) observations and (b) AGCMs. The black dashed line indicates the pressure level of 840 hPa in (a) and 800 hPa in (b), where the cloud content is approximately maximum in observations and AGCMs, respectively. The red dashed line indicates the month of June.

Citation: Journal of Climate 29, 22; 10.1175/JCLI-D-16-0342.1

Fig. 7.
Fig. 7.

Monthly variation of relative humidity in (a) observations, (b) AGCMs, and (c) the difference. The black dashed line indicates the pressure level of 840 hPa in (a) and 800 hPa in (b), where the contour of 60% relative humidity is approximately located in observations and GCMs, respectively. The red dashed line indicates the month of June.

Citation: Journal of Climate 29, 22; 10.1175/JCLI-D-16-0342.1

Given that the lower-troposphere mixing is mostly driven by the surface under the regime of trade cumulus, we further check the surface wind speed. To isolate the precipitation feedback on surface gust wind, only the large-scale wind speed is computed as the root-mean-square of the mean zonal and meridional wind over the CJ region. As shown in Fig. 5b, the large-scale wind is significantly stronger in AGCMs compared to observations, which can enhance the lower-troposphere mixing and raise the inversion top in GCMs.

The excessive lower-troposphere mixing along with a higher inversion top weakens the barrier effect of the capping inversion, helping a convective plume overcome the barrier. To illustrate this point, we compare the month of June between observations and AGCMs (Fig. 8). The excessive lower-troposphere mixing can be seen from the reduced vertical gradient of air temperature and humidity in the lower troposphere (Figs. 8a and 8b, respectively). The near-surface air temperature and humidity are also higher in AGCMs because of the stronger surface wind. Instead of being intercepted by the capping inversion as in observations, the plume is able to rise beyond the lower troposphere in AGCMs (Fig. 8c). The impact of excessive lower-troposphere mixing can be evaluated through an idealized computation. By using the observed lower-tropospheric stratification instead of that simulated in AGCMs, the plume buoyancy is significantly reduced to below zero at the corrected capping inversion. The fact that the plume buoyancy is still larger than that in observations reflects the higher initial energy of the plume due to the higher near-surface air temperature and humidity.

Fig. 8.
Fig. 8.

The June difference in (a) air temperature, (b) specific humidity, and (c) plume buoyancy between AGCMs and observations. The red line represents the multimodel mean of AGCMs and the shading indicates the intermodel standard deviation. The red dashed line in (c) shows the plume buoyancy for AGCMs computed with the observed lower-tropospheric thermal stratification in June.

Citation: Journal of Climate 29, 22; 10.1175/JCLI-D-16-0342.1

As schematically concluded in Fig. 9, the stronger large-scale surface winds in AGCMs induce excessive lower-troposphere mixing and raise the inversion top. This changes the atmosphere from a regime of trade cumulus to deep convection by weakening the barrier effect of the capping inversion, similar to what we see from stage 2 to 3 in observations (Fig. 4).

Fig. 9.
Fig. 9.

Schematic diagram illustrating the regime difference in June between the observation and AGCMs. The stronger surface wind in AGCMs results in excessive mixing in the lower troposphere with a raised boundary layer top, which changes the convective regime from the observed trade cumulus to deep convection in AGCMs.

Citation: Journal of Climate 29, 22; 10.1175/JCLI-D-16-0342.1

Figure 10 shows the monthly evolution of the plume buoyancy computed from the monthly thermodynamic profile for both observations and AGCMs. Both CAPE-ent and LD jump abruptly from June to July in observations but vary smoothly in AGCMs. This is consistent with the abrupt rainfall jump in observations but rather smooth evolution in AGCMs. Such a difference is due to the different barrier effect of the capping inversion in observations and AGCMs (Fig. 9). Generally, the convective plume will detrain shortly after it reaches its LNB. But its journey could be intercepted by the inversion barrier where the plume undergoes a local minimum buoyancy (Figs. 3a and 10a). In observations, although LNB is mainly controlled by SST and thus varies smoothly (Fig. 11a), the plume is intercepted by the capping inversion at the PBL top in June with a negative minimum plume buoyancy (Fig. 11c) and thus detrains in the lower troposphere before reaching LNB (Fig. 11a). As a result, LD and CAPE-ent no longer follow the smooth variation of SST but jump abruptly from June to July when the minimum plume buoyancy turns positive. In AGCMs, by contrast, the minimum plume buoyancy turns positive early in April (Fig. 11c); the plume is no longer inhibited by the raised weak inversion top in AGCMs and detrains after reaching LNB (Fig. 11b). Thus, LD and CAPE-ent then simply follow the smooth variation of SST.

Fig. 10.
Fig. 10.

Monthly variation of the precipitation (red line; right red y axis), the plume buoyancy (color shading), the detrainment level (white dash line; left y axis), and CAPE-ent (black line; right y axis) computed using an entraining plume model for (a) observations and (b) AGCM multimodel mean.

Citation: Journal of Climate 29, 22; 10.1175/JCLI-D-16-0342.1

Fig. 11.
Fig. 11.

Monthly variation of SST (black), LD (blue), and LNB (red) for (a) observations and (b) multimodel mean of AGCMs. (c) The monthly variation of the minimum buoyancy as the plume goes through the capping inversion at the PBL top (thickened line for positive values). The red dashed line indicates the month of June.

Citation: Journal of Climate 29, 22; 10.1175/JCLI-D-16-0342.1

5. Summary and discussion

We have investigated the abrupt regime transition associated with the convective jump in the northwest Pacific around mid-July using an entraining plume model. The abrupt regime transition can be seen as a combined result of two conditions: 1) sufficiently high SST that is able to support deep convection and 2) the trade wind inversion barrier at the PBL top, which initially intercepts convective clouds but is later erased by gradual mixing in the lower troposphere. The combination of these two conditions makes the northwest Pacific during summer a unique place that allows for the convective jump. The transition processes are as follows. Although SST has already reached its maximum of 29°C by 1 July and stays nearly constant thereafter, the atmosphere undergoes substantial evolution. The lower troposphere is gradually mixed, raising the trade cumulus top and moistening the lower troposphere. As predicted by the entraining plume model, such slow preconditioning weakens the barrier effect of the capping inversion on the convective plume. When the plume can marginally overcome the inversion barrier, it rises all the way up to the upper troposphere, leading to an abrupt threshold transition from suppressed trade cumulus to active deep convection.

Different from the abrupt convective jump in observations, GCMs commonly simulate a rather smooth seasonal evolution, with rainfall starting as early as June. It is found that AGCMs simulate stronger large-scale surface winds over the CJ area, which leads to enhanced lower-troposphere mixing with a raised trade cumulus top. These biases shift the atmosphere from a regime of suppressed trade cumulus to active deep convection in June and have a profound impact on the simulation of the northwest Pacific and East Asian summer climate.

Our result highlights the sensitivity of the convective regime to the subtle change in the lower-troposphere mixing, especially when the necessary condition of SST for deep convection has already been satisfied and no longer works as the limiting factor. Under this condition, the conventional undiluted CAPE fails to predict the convective jump because it is overwhelmingly controlled by SST by neglecting the interaction between the plume and the environment. In the case of convective jump, the effects of the inversion barrier and moisture dominate over the SST effect. The entraining CAPE proves to work well to predict the convective jump.

The mid-July convective jump is associated with major shifts in regional rainfall and circulation patterns over the northwest Pacific and East Asia. We have focused on the local gradual preconditioning in the lower troposphere and identified it as the triggering mechanism for the convective jump. This is motivated by the following: 1) local SST is well above the convective threshold and changes little leading to the convective jump, and 2) the neighboring Asian monsoon is at its mature stage with little change around mid-July. It is possible that slow large-scale circulation changes caused by other factors, such as slow insolation variation and land effect, could contribute to this gradual preconditioning. Studies suggest that anomalous upper-level PV anomalies might initiate northwest Pacific convection by destabilizing the lower troposphere (Lu et al. 2007; Wu et al. 2009; Wu and Chou 2012). In addition, the remote SSTs in the tropical Indian Ocean can cause convective anomalies in the northwest Pacific (Hu 2015; Xie et al. 2009, 2016). Further studies are needed.

We find that the GCMs’ bias in the simulation of the convective jump can be traced back to their poor representation of trade cumulus cloud. We recall the long-standing difficulty for GCMs to well represent low clouds. Previous studies focused on their effect on cloud feedbacks and identify low clouds as a key source of uncertainty for climate sensitivity (e.g., Bony and Dufresne 2005). Our study highlights that the bias in the trade cumulus cloud before the rainfall season can lead to biases in the transition to a heavy rainfall season.

Acknowledgments

We acknowledge the WCRP Working Group on Coupled Modelling, which is responsible for CMIP, and the climate modeling groups for producing and making available the model outputs. This work is supported by the U.S. National Science Foundation (NSF Grant 1305719).

REFERENCES

  • Adler, R. F., and Coauthors, 2003: The Version-2 Global Precipitation Climatology Project (GPCP) monthly precipitation analysis (1979–present). J. Hydrometeor., 4, 11471167, doi:10.1175/1525-7541(2003)004<1147:TVGPCP>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Bony, S., and J.-L. Dufresne, 2005: Marine boundary layer clouds at the heart of tropical cloud feedback uncertainties in climate models. Geophys. Res. Lett., 32, L20806, doi:10.1029/2005GL023851.

    • Search Google Scholar
    • Export Citation
  • Bretherton, C. S., J. R. McCaa, and H. Grenier, 2004: A new parameterization for shallow cumulus convection and its application to marine subtropical cloud-topped boundary layers. Part I: Description and 1D results. Mon. Wea. Rev., 132, 864882, doi:10.1175/1520-0493(2004)132<0864:ANPFSC>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Dee, D. P., and Coauthors, 2011: The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553597, doi:10.1002/qj.828.

    • Search Google Scholar
    • Export Citation
  • Ding, Y., and J. C. L. Chan, 2005: The East Asian summer monsoon: An overview. Meteor. Atmos. Phys., 89, 117142, doi:10.1007/s00703-005-0125-z.

    • Search Google Scholar
    • Export Citation
  • Hu, K., 2015: Contrasting impacts of south and north tropical Indian Ocean sea surface temperature anomalies on East Asian summer climate. Atmos. Ocean. Sci. Lett., 8, 327332, doi:10.3878/AOSL20150026.

    • Search Google Scholar
    • Export Citation
  • Johnson, N. C., and S.-P. Xie, 2010: Changes in the sea surface temperature threshold for tropical convection. Nat. Geosci., 3, 842845, doi:10.1038/ngeo1008.

    • Search Google Scholar
    • Export Citation
  • Kain, J. S., 2004: The Kain–Fritsch convective parameterization: An update. J. Appl. Meteor., 43, 170181, doi:10.1175/1520-0450(2004)043<0170:TKCPAU>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Kain, J. S., and J. M. Fritsch, 1990: A one-dimensional entraining/detraining plume model and its application in convective parameterization. J. Atmos. Sci., 47, 27842802, doi:10.1175/1520-0469(1990)047<2784:AODEPM>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Li, C. F., and M. Yanai, 1996: The onset and interannual variability of the Asian summer monsoon in relation to land–sea thermal contrast. J. Climate, 9, 358375, doi:10.1175/1520-0442(1996)009<0358:TOAIVO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Lu, R., H. Ding, C.-S. Ryu, Z. Lin, and H. Dong, 2007: Midlatitude westward propagating disturbances preceding intraseasonal oscillations of convection over the subtropical western North Pacific during summer. Geophys. Res. Lett., 34, L21702, doi:10.1029/2007GL031277.

    • Search Google Scholar
    • Export Citation
  • Molinari, J., D. M. Romps, D. Vollaro, and L. Nguyen, 2012: CAPE in tropical cyclones. J. Atmos. Sci., 69, 24522463, doi:10.1175/JAS-D-11-0254.1.

    • Search Google Scholar
    • Export Citation
  • Simpson, J., and V. Wiggert, 1969: Models of precipitating cumulus towers. Mon. Wea. Rev., 97, 471489, doi:10.1175/1520-0493(1969)097<0471:MOPCT>2.3.CO;2.

    • Search Google Scholar
    • Export Citation
  • Singh, M. S., and P. A. O’Gorman, 2013: Influence of entrainment on the thermal stratification in simulations of radiative-convective equilibrium. Geophys. Res. Lett., 40, 43984403, doi:10.1002/grl.50796.

    • Search Google Scholar
    • Export Citation
  • Taylor, K. E., R. J. Stouffer, and G. A. Meehl, 2012: An overview of CMIP5 and the experiment design. Bull. Amer. Meteor. Soc., 93, 485498, doi:10.1175/BAMS-D-11-00094.1.

    • Search Google Scholar
    • Export Citation
  • Ueda, H., and T. Yasunari, 1996: Maturing process of the summer monsoon over the western North Pacific: A coupled ocean/atmosphere system. J. Meteor. Soc. Japan, 74, 493508.

    • Search Google Scholar
    • Export Citation
  • Ueda, H., T. Yasunari, and R. Kawamura, 1995: Abrupt seasonal change of large-scale convective activity over the western Pacific in the northern summer. J. Meteor. Soc. Japan, 73, 795809.

    • Search Google Scholar
    • Export Citation
  • Ueda, H., M. Ohba, and S.-P. Xie, 2009: Important factors for the development of the Asian–northwest Pacific summer monsoon. J. Climate, 22, 649669, doi:10.1175/2008JCLI2341.1.

    • Search Google Scholar
    • Export Citation
  • Wu, C.-H., and M.-D. Chou, 2012: Upper-tropospheric forcing on late July monsoon transition in East Asia and the western North Pacific. J. Climate, 25, 39293941, doi:10.1175/JCLI-D-11-00343.1.

    • Search Google Scholar
    • Export Citation
  • Wu, C.-H., W.-S. Kau, and M.-D. Chou, 2009: Summer monsoon onset in the subtropical western North Pacific. Geophys. Res. Lett., 36, L18810, doi:10.1029/2009GL040168.

    • Search Google Scholar
    • Export Citation
  • Wu, G., and Y. Zhang, 1998: Tibetan Plateau forcing and the timing of the monsoon onset over South Asia and the South China Sea. Mon. Wea. Rev., 126, 913927, doi:10.1175/1520-0493(1998)126<0913:TPFATT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Wu, R., and B. Wang, 2001: Multi-stage onset of the summer monsoon over the western North Pacific. Climate Dyn., 17, 277289, doi:10.1007/s003820000118.

    • Search Google Scholar
    • Export Citation
  • Xie, S.-P., H. Xu, N. H. Saji, Y. Wang, and W. T. Liu, 2006: Role of narrow mountains in large-scale organization of Asian monsoon convection. J. Climate, 19, 34203429, doi:10.1175/JCLI3777.1.

    • Search Google Scholar
    • Export Citation
  • Xie, S.-P., K. Hu, J. Hafner, H. Tokinaga, Y. Du, G. Huang, and T. Sampe, 2009: Indian Ocean capacitor effect on Indo–western Pacific climate during the summer following El Niño. J. Climate, 22, 730747, doi:10.1175/2008JCLI2544.1.

    • Search Google Scholar
    • Export Citation
  • Xie, S.-P., Y. Kosaka, Y. Du, K. Hu, J. S. Chowdary, and G. Huang, 2016: Indo-western Pacific Ocean capacitor and coherent climate anomalies in post-ENSO summer: A review. Adv. Atmos. Sci., 33, 411432, doi:10.1007/s00376-015-5192-6.

    • Search Google Scholar
    • Export Citation
  • Xu, K., and R. Lu, 2016: Change in tropical cyclone activity during the break of the western North Pacific summer monsoon in early August. J. Climate, 29, 24572469, doi:10.1175/JCLI-D-15-0587.1.

    • Search Google Scholar
    • Export Citation
  • Zhang, G. J., 2009: Effects of entrainment on convective available potential energy and closure assumptions in convection parameterization. J. Geophys. Res., 114, D07109, doi:10.1029/2008JD010976.

    • Search Google Scholar
    • Export Citation
  • Zhang, G. J., and N. A. McFarlane, 1995: Sensitivity of climate simulations to the parameterization of cumulus convection in the Canadian Climate Centre general circulation model. Atmos.–Ocean, 33, 407446, doi:10.1080/07055900.1995.9649539.

    • Search Google Scholar
    • Export Citation
Save
  • Adler, R. F., and Coauthors, 2003: The Version-2 Global Precipitation Climatology Project (GPCP) monthly precipitation analysis (1979–present). J. Hydrometeor., 4, 11471167, doi:10.1175/1525-7541(2003)004<1147:TVGPCP>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Bony, S., and J.-L. Dufresne, 2005: Marine boundary layer clouds at the heart of tropical cloud feedback uncertainties in climate models. Geophys. Res. Lett., 32, L20806, doi:10.1029/2005GL023851.

    • Search Google Scholar
    • Export Citation
  • Bretherton, C. S., J. R. McCaa, and H. Grenier, 2004: A new parameterization for shallow cumulus convection and its application to marine subtropical cloud-topped boundary layers. Part I: Description and 1D results. Mon. Wea. Rev., 132, 864882, doi:10.1175/1520-0493(2004)132<0864:ANPFSC>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Dee, D. P., and Coauthors, 2011: The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553597, doi:10.1002/qj.828.

    • Search Google Scholar
    • Export Citation
  • Ding, Y., and J. C. L. Chan, 2005: The East Asian summer monsoon: An overview. Meteor. Atmos. Phys., 89, 117142, doi:10.1007/s00703-005-0125-z.

    • Search Google Scholar
    • Export Citation
  • Hu, K., 2015: Contrasting impacts of south and north tropical Indian Ocean sea surface temperature anomalies on East Asian summer climate. Atmos. Ocean. Sci. Lett., 8, 327332, doi:10.3878/AOSL20150026.

    • Search Google Scholar
    • Export Citation
  • Johnson, N. C., and S.-P. Xie, 2010: Changes in the sea surface temperature threshold for tropical convection. Nat. Geosci., 3, 842845, doi:10.1038/ngeo1008.

    • Search Google Scholar
    • Export Citation
  • Kain, J. S., 2004: The Kain–Fritsch convective parameterization: An update. J. Appl. Meteor., 43, 170181, doi:10.1175/1520-0450(2004)043<0170:TKCPAU>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Kain, J. S., and J. M. Fritsch, 1990: A one-dimensional entraining/detraining plume model and its application in convective parameterization. J. Atmos. Sci., 47, 27842802, doi:10.1175/1520-0469(1990)047<2784:AODEPM>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Li, C. F., and M. Yanai, 1996: The onset and interannual variability of the Asian summer monsoon in relation to land–sea thermal contrast. J. Climate, 9, 358375, doi:10.1175/1520-0442(1996)009<0358:TOAIVO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Lu, R., H. Ding, C.-S. Ryu, Z. Lin, and H. Dong, 2007: Midlatitude westward propagating disturbances preceding intraseasonal oscillations of convection over the subtropical western North Pacific during summer. Geophys. Res. Lett., 34, L21702, doi:10.1029/2007GL031277.

    • Search Google Scholar
    • Export Citation
  • Molinari, J., D. M. Romps, D. Vollaro, and L. Nguyen, 2012: CAPE in tropical cyclones. J. Atmos. Sci., 69, 24522463, doi:10.1175/JAS-D-11-0254.1.

    • Search Google Scholar
    • Export Citation
  • Simpson, J., and V. Wiggert, 1969: Models of precipitating cumulus towers. Mon. Wea. Rev., 97, 471489, doi:10.1175/1520-0493(1969)097<0471:MOPCT>2.3.CO;2.

    • Search Google Scholar
    • Export Citation
  • Singh, M. S., and P. A. O’Gorman, 2013: Influence of entrainment on the thermal stratification in simulations of radiative-convective equilibrium. Geophys. Res. Lett., 40, 43984403, doi:10.1002/grl.50796.

    • Search Google Scholar
    • Export Citation
  • Taylor, K. E., R. J. Stouffer, and G. A. Meehl, 2012: An overview of CMIP5 and the experiment design. Bull. Amer. Meteor. Soc., 93, 485498, doi:10.1175/BAMS-D-11-00094.1.

    • Search Google Scholar
    • Export Citation
  • Ueda, H., and T. Yasunari, 1996: Maturing process of the summer monsoon over the western North Pacific: A coupled ocean/atmosphere system. J. Meteor. Soc. Japan, 74, 493508.

    • Search Google Scholar
    • Export Citation
  • Ueda, H., T. Yasunari, and R. Kawamura, 1995: Abrupt seasonal change of large-scale convective activity over the western Pacific in the northern summer. J. Meteor. Soc. Japan, 73, 795809.

    • Search Google Scholar
    • Export Citation
  • Ueda, H., M. Ohba, and S.-P. Xie, 2009: Important factors for the development of the Asian–northwest Pacific summer monsoon. J. Climate, 22, 649669, doi:10.1175/2008JCLI2341.1.

    • Search Google Scholar
    • Export Citation
  • Wu, C.-H., and M.-D. Chou, 2012: Upper-tropospheric forcing on late July monsoon transition in East Asia and the western North Pacific. J. Climate, 25, 39293941, doi:10.1175/JCLI-D-11-00343.1.

    • Search Google Scholar
    • Export Citation
  • Wu, C.-H., W.-S. Kau, and M.-D. Chou, 2009: Summer monsoon onset in the subtropical western North Pacific. Geophys. Res. Lett., 36, L18810, doi:10.1029/2009GL040168.

    • Search Google Scholar
    • Export Citation
  • Wu, G., and Y. Zhang, 1998: Tibetan Plateau forcing and the timing of the monsoon onset over South Asia and the South China Sea. Mon. Wea. Rev., 126, 913927, doi:10.1175/1520-0493(1998)126<0913:TPFATT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Wu, R., and B. Wang, 2001: Multi-stage onset of the summer monsoon over the western North Pacific. Climate Dyn., 17, 277289, doi:10.1007/s003820000118.

    • Search Google Scholar
    • Export Citation
  • Xie, S.-P., H. Xu, N. H. Saji, Y. Wang, and W. T. Liu, 2006: Role of narrow mountains in large-scale organization of Asian monsoon convection. J. Climate, 19, 34203429, doi:10.1175/JCLI3777.1.

    • Search Google Scholar
    • Export Citation
  • Xie, S.-P., K. Hu, J. Hafner, H. Tokinaga, Y. Du, G. Huang, and T. Sampe, 2009: Indian Ocean capacitor effect on Indo–western Pacific climate during the summer following El Niño. J. Climate, 22, 730747, doi:10.1175/2008JCLI2544.1.

    • Search Google Scholar
    • Export Citation
  • Xie, S.-P., Y. Kosaka, Y. Du, K. Hu, J. S. Chowdary, and G. Huang, 2016: Indo-western Pacific Ocean capacitor and coherent climate anomalies in post-ENSO summer: A review. Adv. Atmos. Sci., 33, 411432, doi:10.1007/s00376-015-5192-6.

    • Search Google Scholar
    • Export Citation
  • Xu, K., and R. Lu, 2016: Change in tropical cyclone activity during the break of the western North Pacific summer monsoon in early August. J. Climate, 29, 24572469, doi:10.1175/JCLI-D-15-0587.1.

    • Search Google Scholar
    • Export Citation
  • Zhang, G. J., 2009: Effects of entrainment on convective available potential energy and closure assumptions in convection parameterization. J. Geophys. Res., 114, D07109, doi:10.1029/2008JD010976.

    • Search Google Scholar
    • Export Citation
  • Zhang, G. J., and N. A. McFarlane, 1995: Sensitivity of climate simulations to the parameterization of cumulus convection in the Canadian Climate Centre general circulation model. Atmos.–Ocean, 33, 407446, doi:10.1080/07055900.1995.9649539.

    • Search Google Scholar
    • Export Citation
  • Fig. 1.

    Climatological mean precipitation (color shading) and surface wind (vector) averaged over (left) 3–12 July, (center) 18–27 July, and (right) the difference between these two periods for (a)–(c) observations and (d)–(f) AGCMs. The red contours are the 29°C SST isotherm in (a),(b) and the difference in (c). The hatched area indicates the CJ area that we are interested in.

  • Fig. 2.

    (a) Time evolution of the mean cloud water content (color shading), SST (black solid line for SST > 29°C, gray dashed line for SST < 28°C, and black dashed line for 28° < SST < 29°C), and pentad precipitation (red line) over the CJ area from 15 May to 15 August. (b) Time evolution of the mean relative humidity (color shading) and the lower-troposphere-integrated relative humidity (; red line). (c) Difference in air temperature relative to that on 1 July (color shading) and time evolution of the atmospheric instability Δθυ, measured by the virtual potential temperature difference between 750 and 950 hPa (red line).

  • Fig. 3.

    Time evolution of the plume buoyancy (color shading), the LD (white line), and the CAPE (black line) computed using (a),(c) an entraining plume and (b),(d) an undiluted plume. The zero contour is highlighted with a thin pink line in (a). Please note that the upper branch of this zero contour indicates the LNB. The four stages mentioned in section 3 are denoted by the triangle symbols in (a). The climatological pentad precipitation (Pr) over the CJ area is plotted as the red line in (c),(d) for reference.

  • Fig. 4.

    Schematic diagram illustrating the four stages during the abrupt regime transition from trade cumulus to deep convection. The blue line shows the vertical profile of the virtual equivalent potential temperature for the environment air, while the red line shows that for the updraft entraining plume. The positive plume buoyancy is indicated by the hatched area between. The abrupt jump of the detrainment level is indicated by the thick purple line.

  • Fig. 5.

    Monthly variation of the mean (a) precipitation and (b) large-scale surface wind speed over the CJ area simulated by AGCMs. The black line shows the observation and the red line shows the multimodel mean of AGCMs.

  • Fig. 6.

    Monthly variation of cloud water content in (a) observations and (b) AGCMs. The black dashed line indicates the pressure level of 840 hPa in (a) and 800 hPa in (b), where the cloud content is approximately maximum in observations and AGCMs, respectively. The red dashed line indicates the month of June.

  • Fig. 7.

    Monthly variation of relative humidity in (a) observations, (b) AGCMs, and (c) the difference. The black dashed line indicates the pressure level of 840 hPa in (a) and 800 hPa in (b), where the contour of 60% relative humidity is approximately located in observations and GCMs, respectively. The red dashed line indicates the month of June.

  • Fig. 8.

    The June difference in (a) air temperature, (b) specific humidity, and (c) plume buoyancy between AGCMs and observations. The red line represents the multimodel mean of AGCMs and the shading indicates the intermodel standard deviation. The red dashed line in (c) shows the plume buoyancy for AGCMs computed with the observed lower-tropospheric thermal stratification in June.

  • Fig. 9.

    Schematic diagram illustrating the regime difference in June between the observation and AGCMs. The stronger surface wind in AGCMs results in excessive mixing in the lower troposphere with a raised boundary layer top, which changes the convective regime from the observed trade cumulus to deep convection in AGCMs.

  • Fig. 10.

    Monthly variation of the precipitation (red line; right red y axis), the plume buoyancy (color shading), the detrainment level (white dash line; left y axis), and CAPE-ent (black line; right y axis) computed using an entraining plume model for (a) observations and (b) AGCM multimodel mean.

  • Fig. 11.

    Monthly variation of SST (black), LD (blue), and LNB (red) for (a) observations and (b) multimodel mean of AGCMs. (c) The monthly variation of the minimum buoyancy as the plume goes through the capping inversion at the PBL top (thickened line for positive values). The red dashed line indicates the month of June.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 1652 912 116
PDF Downloads 242 88 16