Info

Figure 5. Schematic diagrams describing effects on continental convection zones (for northern hemisphere summer) of (a) soil moisture feedbacks, (b) ventilation and the interactive Rodwell-Hoskins mechanism, and (c) land-ocean contrast associated with ocean heat transport and land surface condition (adapted from CNS).

Figure 5. Schematic diagrams describing effects on continental convection zones (for northern hemisphere summer) of (a) soil moisture feedbacks, (b) ventilation and the interactive Rodwell-Hoskins mechanism, and (c) land-ocean contrast associated with ocean heat transport and land surface condition (adapted from CNS).

diagram is used to summarize the soil moisture effect [Fig. 5(a)]. Most summer monsoon rain zones over continents are in the descending branch of the Hadley cell, so soil moisture is relatively low prior to the summer season. This low soil moisture tends to decrease evaporation over land. With less moisture supply, the convective available potential energy (CAPE) reduces, so convection becomes less frequent. In other words, if soil moisture can be supplied indefinitely, such as in the saturated experiments discussed in CNS, the summer monsoon rain zones should spread across much wider areas. Soil moisture can also change the partitioning of evaporation and sensible heat fluxes, and then affects surface temperature. This change in the partitioning of surface heat fluxes does not affect Fnet since the net surface heat flux Fsnet over continents is near zero on a time scale longer than one day. Thus, the monsoon circulation is modified very little by soil moisture, and so the feedback of soil moisture to the monsoon rainfall via the monsoon circulation is much weaker than the feedback of Fnet discussed above.

4.2. Ventilation

The summer monsoon rainfall shown in Fig. 1 does not extend as far poleward as the corresponding Fnet indicates in Fig. 3. Even with saturated soil moisture, the monsoon rain zone still cannot extend as far poleward as Fnet indicates. This implies that mechanisms other than direct solar heating are working to limit the poleward extent of the summer monsoon rain zones. According to (3), the horizontal advection of MSE, —v ■ V(T + q), is a possible cause. At midlatitudes, the dominant winds in the column average are westerly, and these westerly winds are largely set by global-scale dynamics. In summer, these cross-continental westerlies bring cold and low moist static-energy air from ocean regions due to ocean heat storage in the western part of the continent. The transport of the cold and low moist static-energy air limits the poleward extent of the summer monsoon rain zones. This import of low moist static-energy air into the continental regions from ocean regions is defined as ventilation (CNS). In the study by CNS, strong negative —v ■ V(T + q) dominates the summer continent. This indicates that the ventilation mechanism removes the heating of Fnet efficiently and limits the poleward extent of the summer monsoon rain zone. The advection of low moist static-energy air suppresses convection mostly over the western part of the continent, so the ventilation mechanism creates an east-west asymmetry in the summer monsoon precipitation [Fig. 5(b)].

4.3. Interactive Rodwell-Hoskins mechanism

Another mechanism controlling the distribution of the summer monsoon rainfall is the interactive Rodwell-Hoskins (IRH) mechanism. Rodwell and Hoskins (1996, 2001) proposed a monsoon-desert-like pattern to explain the association of the Asian summer monsoon and the Sahara desert. Monsoon heating induces a Rossy wave to the west associated with subsidence, which disfavors convection in this region. Thus, this process not only limits the poleward extent of the monsoon rain zone but also creates an east-west asymmetry of the monsoon rain zone. The term "interactive" emphasizes the interaction of the monsoon heating and the monsoon circulation via —v ■ V(T + q), which further enhances the associated subsidence (CNS). The summer monsoon circulation is a cyclonic circulation with a continental scale associated with the convective heating of the summer monsoon. The eastern branch of the summer monsoon circulation transports moist and high MSE air from the south to the eastern part of the continent, and the convection over this region is enhanced. Meanwhile, the western branch of the circulation transports dry and low MSE air from the north to the western part of the continents, which suppresses convection over this region. The enhanced monsoon rainfall over the eastern part of the continent further strengthens the Rossby wave subsidence to the western part the continent, so the convection over the western part of the continent is suppressed more. This interaction with the monsoon circulation further enhances the east-west asymmetry [Fig. 5(b)]. This circulation-induced feedback to the asymmetry has also been discussed by Xie and Saiki (1999).

4.4. Land-ocean heating contrast

As discussed above, the feedback of the summer monsoon circulation plays an important role in the poleward extent of the summer monsoon rain zones. The strength of the monsoon circulation is determined by the horizontal pressure gradient between the continent and the neighboring oceans. This pressure gradient is strongly linked to the tropospheric temperature gradient induced by the land-ocean heating contrast, so the tropospheric temperature gradient is often used to represent the strength of the summer monsoon circulation (Li and Yanai, 1996). Thus, processes that control Fnet over land and ocean, such as land surface conditions and ocean heat transport, are important in determining how strong the land-ocean heating contrast is. Land surface conditions, such as surface albedo associated with surface type and vegetation, can affect atmospheric temperature over land via Fnet and change the pressure gradient between continental regions and ocean regions. For instance, small land surface albedo reflects less solar radiation into the space, and then Fnet increases. The increased Fnet not only enhances convection through the local processes discussed in Subsec. 4.1, but also increases the land-ocean heating contrast and the corresponding pressure and temperature gradients. Thus, the large-scale convergence over the continent is enhanced and the low-level cyclonic summer monsoon circulation is strengthened, and so the summer monsoon rainfall increases and the rain zones also extend farther poleward through the IRH mechanism [Fig. 5(c)]. The east-west asymmetry also becomes more significant.

Ocean heat transport, on the other hand, can also modify the land-ocean heating contrast via Fsnet. It is controlled by ocean circulation. Usually, tropical oceans are a heat sink to the atmosphere because of ocean heat transport from the tropics to higher latitudes and ocean heat storage, so tropical oceans need energy from the atmosphere to balance the loss, i.e. Fsnet > 0. Due to Fsnet « 0 over land, Fnet over continental regions is usually positive and is larger than over ocean regions according to (5), and so a land-ocean heating contrast is created. The more efficient ocean heat transport at lower latitudes tends to induce the stronger cooling effect over ocean regions. The land-ocean heating contrast is then enhanced [Fig. 5(c)]. The enhanced land-ocean contrast favors convection over continental regions through the large-scale convergence. Thus, climate variations that can/change/Fsnet over the neighboring oceans, such as in ENSO events, affect the poleward extent of the summer monsoon rain zones via the feedback of the monsoon circulation. This change of ocean heat transport can also directly modify local convection via the change of SST. Warmer SST implies less ocean heat transport out of the region. In other words, warm SST, such as in the western Pacific, tends to enhance convection via Fnet and heats the troposphere above, and so the local precipitation is enhanced.

5. Examination on the Asian Summer Monsoon System

Based on the mechanisms discussed in the previous section, we next examine the impacts of those mechanisms on the Asian summer monsoon, which is the biggest and strongest summer monsoon in the world. The Asian continent is the largest continent in the world and contains the highest mountains, namely the Tibetan Plateau. To the south of the Asian continent, the Indian Ocean covers the tropical region. This land-sea configuration is different from those in the idealized monsoon study (CNS) that was used to discuss these summer monsoon mechanisms in the previous section. Thus, the impacts of those mechanisms on the summer monsoon discussed in the idealized case may be modified when applied to the Asian summer monsoon case. Before using QTCM for studying a more realistic summer monsoon, we note that due to the simplicity of QTCM, the summer monsoon simulated by QTCM might not be as good as those in GCMs, especially at local scales. For instance, QTCM does not have distinct maximum precipitation over the Bay of Bengal. However, the large-scale aspects of the rainfall pattern, including the northward extent along the east coast, tend to be simulated, so QTCM is good enough for studying the mechanisms discussed in the previous section for the Asian summer monsoon.

5.1. Local processes

Without the feedback of large-scale circulation, two local effects that can affect the summer monsoon rainfall are Fnet and soil moisture via evaporation. Unlike the example of the African case discussed in Subsec. 4.1, the effect of Fnet cannot be easily separated from the feedback of the monsoon circulation. In the African case, the observed Fnet is close to zero because of high surface albedo. Modifying Fnet can create an impact that is stronger than the feedback via the monsoon circulation, so the effect of Fnet can be examined. However, Fnet over the summer Asian continent is much larger than that over Africa, so as the associated land-ocean contrast. Thus, modifying Fnet in Asia produces a dominant effect from the feedback of the Asian summer monsoon circulation, and the direct Fnet effect via convection in the MSE equation is weaker. Therefore, we discuss only the local effect of soil moisture here, not the Fnet effect.

The existence of the Tibetan Plateau is important for producing a realistic Asian summer monsoon (e.g. Flohn, 1968; Hahn and Manabe, 1975; He et al., 1987; Murakami, 1987; Yanai et al., 1992; Ye, 1981). However, experiments without the topographic effect can still give us some idea of the mechanisms for the Asian summer monsoon. Thus, QTCM without topography is used to examine the soil moisture effect, and the result of the control run is shown in Fig. 6(a). Comparison with the observed Asian summer monsoon (Fig. 1), excluding the Tibetan Plateau, may produce some caveats, such as no distinct feature of maximum precipitation over the Bay of Bengal and the west of the Indian peninsula and a little higher precipitation over the northeast coast of Asia. When the soil moisture over the Asian continent is saturated, the corresponding precipitation not only increases but also extends farther northward. Only the very center of the Asian continent is still dry. In the saturated soil moisture experiment, the evaporation over the Asian continent is enhanced due to larger moisture supply from soil and the local precipitation is then increased. The stronger local evaporation also cools the surface temperature over the Asian continent. Note that the colder surface temperature shows little influence on changing the monsoon circulation because of the near-zero net surface heat flux Fsnet. Meanwhile, the tropospheric temperature over South Asia becomes warmer due to the convective heating of the enhanced summer monsoon rainfall. This warmer tro-pospheric temperature creates a low pressure anomaly, and the associated anomalous cyclonic circulation over South Asia. This anomalous circulation transports moisture from the south (not shown) and further enhances the rainfall over the southern and southeastern parts of Asia. Both local soil moisture and the feedback of the convection-induced anomalous circulation affect the Asian summer monsoon rainfall and its northward extension.

5.2. Ventilation and interactive

Rodwell-Hoskins mechanisms

To examine the ventilation mechanism, the horizontal advection of moisture associated with v$ is suppressed in the QTCM simulations. We note that such experiments suppress not only the effect of ventilation but also the partial effect of the IRH mechanism that includes the effect of the moisture and MSE transports by the monsoon circulation. The summer monsoon rainfall is increased substantially, and the rain zone extends much farther northward [Fig. 6(c)]. However, the rainfall over the northeast coast of the Asian continent disappears when v$ ■ V (T+q) is suppressed. The ventilation mechanism does suppress the Asian summer monsoon rainfall in most Asian monsoon regions, except for East Asia including Korea and Japan. Over this region, the moisture transport is associated with the monsoon circulation in the IRH mechanism, so the horizontal moisture transport is a moisture source for the monsoon rainfall, not a moisture sink. Thus, the precipitation over East Asia is weakened when suppressing v.$ ■ V(T + q).

(a) Control run

0 0

Post a comment