Diurnal variation of tropical oceanic convection

Electronics Repair Manuals

Schematic Diagrams and Service Manuals

Get Instant Access

The diurnal variation of tropical oceanic convection is one of most important phenomena in tropical variability, and plays a crucial role in regulating tropical hydrological and energy cycles. The dominant diurnal signal is the nocturnal peak in precipitation that occurs in the early morning (see the review in Sui et al., 1997a). Kraus (1963) emphasized the role of radiative forcing in the diurnal variation, and suggested that solar heating and IR cooling tend to suppress convection during daytime and enhance convection during nighttime respectively. Gray and Jacobson (1977) suggested that the diurnal variation of convection is a result of a synoptic-scale dynamic response to cloud radiative forcing (the radiational differences between cloudy regions and clear-sky regions). The cloud radiative forcing causes upward motion and convection during nighttime through the low-level convergence.

Sui et al. (1997a) conducted an analysis using the observational data from TOGA COARE. The data are first categorized into the disturbed and undisturbed periods by calculating the standard deviation of brightness temperature measured by the Geostationary Meteorological Satellite (GMS), operated by the Japanese Meteorological Agency. Over the disturbed periods, the total surface rain rate as well as convective and stratiform rain rates reach the maxima at 0300 local standard time (LST). Fractional cover for stratiform clouds has a maximum at 0300 LST whereas fractional cover for convective clouds does not show a significant diurnal variation. Diurnal variation of the rain rate histogram shows that the evolution of nocturnal rainfall has a growing phase from 2200 to 0300 LST, when a wide range of convection (the rain rate is larger than 0.5mmh_1) becomes enhanced with most occurrences within 0.5-5mmh~1. The nocturnal rainfall is associated with anomalous ascending motion in the layer between 500 and 200 mb at 0400 LST. Over the undisturbed periods, the surface rain rate is very small, but shows a maximum from 1200 to 1800 LST. Diurnal variation of the rain rate histogram shows that the evolution of afternoon rainfall has a growing phase from 1200 to 1800 LST, while most occurrences of rain rates are within 0.5 to 5mmh~1. The afternoon rainfall peak is associated with the maximum SST after the solar radiation flux reaches the maximum. A schematic summary of the nocturnal maximum convection in the disturbed period and afternoon clouds and showers in the undisturbed period is shown in Fig. 1. Based on the observational analysis, the nocturnal rainfall peak is suggested to be related to the destabilization by radiative cooling during nighttime and the falling temperature that makes more precipitable water available for the surface precipitation.

Xu and Randall (1995) performed a study using a cumulus ensemble model (CEM), and suggested that nocturnal convection results from a direct radiation-convection interaction in which solar absorption by clouds stabilizes the atmosphere (Randall et al., 1991). Tao et al. (1996) performed a study of cloud-radiation mechanisms in the tropics and mid-latitudes using the Goddard cumulus ensemble (GCE) model. They emphasized the increase of surface precipitation by IR cooling as a result of increased relative humidity.

Sui et al. (1998a) conducted the cloud-resolving simulations to test their nocturnal rainfall mechanism. An experiment with the

Oceanic Convection

Local time [hour]

Figure 1. Schematic diagram of diurnal variations of convection during the disturbed (upper panel) and undisturbed (lower panel) periods. The dashed curve in the lower panel indicates the time rate of change of the saturation columnar water vapor amount, —dW*/dt, corresponding to the diurnal cycle of temperature distribution. This quantity represents a direct effect of the radiative cooling/heating cycle on available precipitable water (APW), or a change of APW in the first step. The convective response to the direct forcing can induce further changes in temperature and moisture that lead to a corresponding change of APW in the second step. Since observed and simulated diurnal variations of convection are evidently in phase with the idealized cycle, the curve is regarded as a good theoretical limit for diurnal rainfall. The dashed curve in the lower panel indicates the diurnal cycle of sea surface temperature.

Local time [hour]

Figure 1. Schematic diagram of diurnal variations of convection during the disturbed (upper panel) and undisturbed (lower panel) periods. The dashed curve in the lower panel indicates the time rate of change of the saturation columnar water vapor amount, —dW*/dt, corresponding to the diurnal cycle of temperature distribution. This quantity represents a direct effect of the radiative cooling/heating cycle on available precipitable water (APW), or a change of APW in the first step. The convective response to the direct forcing can induce further changes in temperature and moisture that lead to a corresponding change of APW in the second step. Since observed and simulated diurnal variations of convection are evidently in phase with the idealized cycle, the curve is regarded as a good theoretical limit for diurnal rainfall. The dashed curve in the lower panel indicates the diurnal cycle of sea surface temperature.

imposed large-scale ascending motion and a time-invariant SST generated a positive rainfall anomaly in the night and a negative rainfall anomaly in the day. The simulated maximum rain rate occurs around 0200 LST. Two additional experiments are carried out: one experiment with a zero imposed vertical velocity and a time-invariant SST, and the other with the cloud-radiation interaction suppressed. All three experiments show a dominant nocturnal rainfall maximum despite the experiments with very different external forcings and interaction processes. The results imply that cloud-radiation interaction does not play a crucial role in the formation of the nocturnal rainfall peak. The common feature in all of these experiments is the falling temperature induced by the nocturnal radiative cooling in the absence of the solar radiative heating. Thus, these numerical experiments support the suggestion of Sui et al. (1997a) that the nocturnal rainfall peak is related to more (less) available precipitable water in the night (day) as a result of the diur nal cooling/heating cycle (Fig. 1, lower panel). Sui et al. also conducted the experiment with zero imposed vertical velocity and a zonally uniform, diurnally varied SST and found that the simulated diurnal variations still have a nocturnal rainfall maximum but with a weaker magnitude and a secondary rainfall peak in the afternoon. This indicates that the maximum SST in the afternoon induces the unstable atmosphere that eventually leads to the rainfall peak (Fig. 1, upper panel).

Was this article helpful?

0 0

Post a comment