Issues and motivations: the TTL a transition layer between the troposphere and the stratosphere

Get Complete Project Material File(s) Now! »

TROPICAL CONVECTIVE STRUCTURES AND MECHANISMS

Theoretical model of the air parcel

The convective process of an air parcel can be evaluated by a theoretical model of the parcel (Morton, 1957; Turner, 1973) presented in an tephigram (Figure 2.6). The parcel is characterized by its temperature and its dew point temperature (i.e. the temperature when water vapour starts to condense). So the dew point temper-ature is a function of the amount of water vapour. The model estimates that the air parcel rises following firstly a dry adiabatic (T decreases whilst θ is constant) up to the point where its temperature equal its dew point temperature : the lifting con-densation level (LCL), (i.e. the base of the cloud). The air parcel follows a humid adiabatic (T decreases and θe is constant). When the temperature of the air parcel reaches the environment temperature, the level of the free convection (LFC) is reached. The air parcel continues to rise up to the level where its temperature is less than the ambient temperature: this is the level of neutral buoyancy (LNB). From the theoretical model of the air parcel, the convective available potential energy (CAPE) is used to evaluate the intensity of the convection. The CAPE is calculated by the integration of the area between the parcel trajectory and the environment profile. The convective inhibition (CIN) quantifies the energy that an air parcel has to reach the LFC in order to start the deep convection. Thus, in figure 2.6, the temperature of an ascending parcel is compared to the dew temper-ature (Td) profile of this parcel and to the speudo-adiabatic potential temperature (Tw) profile of this parcel.

Convective initiators and ascending processes

For deep convection to rise to its LFC, high energy combined with high mechan-ical action is required. Air mass convergences, for example promoted by breeze from surface heterogeneities (Land-Sea or Mountain-valley), generate the lifting of air masses (Wakimoto and Atkins, 1994). The formation of anabatic winds on warming landforms, or the collision of a breeze front with a gust front (Kingsmill, 1995), also promotes convergences along with the rise of air masses. Further-more, large-scale convergences such as in the Inter-Tropical Convergence Zone (ITCZ, see more in section 2.3.2), have also a strong role on the air rising into deep convection. The convergence flux of humidity at the surface determines the location of the convective cell. The vertical wind shear in the lowest layer of deep convective clouds also has a strong impact on the deep convective structure in shifting the precipitation and the downdrafts out of the center of the convec-tive systems, where humid and hot updrafts provide the convection. This process makes the convective systems become larger because of its propagation, which form multi-cells and increasing considerably the time life of the deep convection.
Once the convection is generated, convective ascents transport air masses up in altitude by updrafts. In general, convective updrafts are less than 500 m in size (Yang et al., 2016). However, in the tropics, deep convective updrafts can reach about 4-7 km in size (Yang et al., 2016) and more than 17 km in altitude (Fueglistaler et al., 2009a). The updraft speed can reach 11 m s-1 in median value (Varble et al., 2014), while shallow convections reach vertical speed of about twice less (Schumacher et al., 2015). In comparison, stratiform clouds (composing the convective anvil), has updraft speeds only about 1 m s-1. Some studies have shown that, in deep convection, the updraft is composed by a succession of bubbles rising one above the others up to the top (Ludlam and Scorer, 1953; Zhao and Austin, 2005; Heus et al., 2009; Sherwood et al., 2013). Some study also suggests that the air masses transported in deep convective updraft come from a layer very close to the surface, within the planetary boundary layer (McGee and van den Heever, 2014).

Radiative processes

Solar radiation has an impact on the convective processes and on the stratospheric-tropospheric interaction processes (Hartmann et al., 2001; Fueglistaler et al., 2009a; Tzella and Legras, 2011). Solar radiation impacts heating of the air masses. Clouds in the UT and TTL have got a strong impact on radiation. Solar radiation penetrating into the stratosphere and the troposphere depends on the absorption of each constituent present in these layers (Mohanakumar, 2008). Greenhouse gases have a warm effect on the atmospheric temperature by the atmospheric emission of infra-red radiation such as water vapour in the troposphere, controlling the overall radiative balance of the troposphere.
In region of the atmosphere where clear sky condition occur, horizontal transport can facilitate air masses to cross the TTL. To reach the stratosphere, an air parcel of air must go through the level of Zero Radiative Heat (LZRH, about 360 K or 125 hPa (Gettelman et al., 2004a; Fueglistaler et al., 2009a)), which can be a bar-rier to the transport of air parcels of air to the stratosphere (Folkins et al., 1999). Below this level, as radiative heating is negative, air parcels tend to descend to the troposphere. Above this level, radiative heating is positive and air parcels tend to rise towards the stratosphere. Thus, the lower the LZRH will be, the easier the air parcel will ascend up to the TTL or even the LS (Corti et al., 2005). Figure 2.7 presents the vertical profile of the different fluxes into the tropical troposphere during convective activity.

TROPICAL CONVECTIVE STRUCTURES AND MECHANISMS

Microphysical processes

The water convective microphysical processes are characterized by the water phase changes and the absorption or the release of latent heat flux in an air mass. Condensation or freezing release latent heat while evaporation or melting absorbs latent heat. Furthermore, condensation and evaporation release and absorbs re-spectively, the vapour latent heats (whilch varies strongly with the temperature). However, freezing and melting, release and absorbe respectively, the fusion latent heats (which varies little with the temperature). Water vapour can also be de-posited on ice crystals, which also releases heat into the environment. Conversely, the ice sublimation into water vapour absorbs latent heat. In this case, the sub-limation latent heat varies little with the temperature. According to Seigel et al. (2013), the release and the absorption of the latent heat can appear alternatively by cycles, following the ice crystal formation, fall, melting and then, training for a new ascent, which can freeze again. Figure 2.8 illustrates the water microphysical processes.

Turbulent processes

Convective wind shears promote the convective turbulence by formation of ed-dies in the convective cell. An energy cascade is built among the various vortices composing the convective system, helping the air masses to exchange with the surrounding environment. Unsaturated air transported by non-turbulent flow from the environment (dry wind) can be captured by the convective turbulent flow and transported into the convective system. This phenomenon is called the entrain-ment phenomenon and happens on the sides of the convective system and not on its top (Heus and Jonker, 2008). Usually, the entrainment occurs in the lower part of the cloud and can be transported by updrafts. However, the higher in altitude the entrainment occurs, the drier are air masses transported by updraft up to the cloud top. In contrast, the turbulent phenomenon ejecting air masses from the convective system to the environment is called the detrainment phenomenon. The detrainment occurs usually at the top of the deep convection and participates in the formation of the anvil. These two phenomena induce mixing of air in the convective system, decreasing the buoyancy and promoting a decrease of water vapour in the convective system because the air transported from the environment is drier than the air inside the cloud.

Overshoots

Deep convective overshoots are the part of the cloud overtaking the flat top of the cumulonimbus clouds (see illustration in Figure 2.5). Overshoots can measure several kilometres of height over the top of the cumulonimbus, even crossing the TTL and reaching the lower stratosphere. The overshoot time life is short (30-60 minutes according Chaboureau et al. (2007) and Frey et al. (2015)) and can cover large horizontal surfaces (tens to hundreds of kilometers depending on the altitude, Hassin and Lane, 2010). Overshoots are an important way for the tropo-sphere to stratosphere transport of air masses and water. However, the frequency of overshoots is low (Liu and Zipser, 2005; Fu et al., 2007)). Dauhut et al. (2015) have estimated to 18% the amount of overshoot compared to the total deep con-vective events. According to a study from Luo et al. (2008), using datasets from the space-borne CloudSat, only 1.3% of convective clouds reach surfaces higher than the CPT. Fu et al. (2007) have determined from datasets from the CALIOP lidar that, over all the tropics, 0.5% of clouds reach 18.5 km whereas 5% reach 17 km. But only overshoots would not be enough to explain the amount of air crossing the tropopause and maintaining the global circulation controlled by the Brewer-Dobson circulation (circulation detailed in section 2.3.3). Many other studies have tried to quantify the impact of deep convective overshoots on the transport of air masses up to the LS at global and local scales (Zhang, 1993; Get-telman and Fothers, 2002; Liu and Zipser, 2005; Rossow and Pearl, 2007; Dauhut, 2016). Although the convective overshoots are still difficult to represent in mod-els, measurements from high altitude aircraft, groundbased radar or space-borne observations of brightness temperatures (highlighting overshoots because of their very cold temperature) are currently being used to better understand the impact of overshoots on the troposphere to stratosphere transport (Schmetz et al., 1997; Romps and Kuang, 2009; Bedka et al., 2010).

Subsidence, precipitation and cold pockets

Precipitation is classified into three main types: stratiform rainfall, shallow con-vective rainfall and deep convective rainfalls, as presented in Figure 2.9. Strati-form and convective precipitations are the two main forms of precipitation with the strongest amount of precipitation.
Stratiform precipitation (or stable precipitation) results from slow and large-scale lifting of a moist air mass that condenses uniformly. Associated clouds are strat-iform clouds (grey and uniform slick appearance: nimbostratus, stratocumulus, stratus). Stratiform precipitations have low or moderate intensity (less than 10 mm-1) and long duration due to the slow displacement or orographic blockages. These precipitations are well simulated in models. Convective precipitations (or unstable precipitations) result from the quick rise of moisture-laden air masses by Archimedes’ principle. This type of precipitation comes from cumuliform clouds able to reach 10 to 18 km of altitude. Convective precipitations are characterised by very strong intensity and short duration (between 30 minutes and few hours). This precipitation can be accompanied by thunderstorms, ice pellets or hail and gusts of wind. Stratiform and convective precipitations can be observed together, resulting into unstablilities into stratiform rain or snow masses and on heavier showers. Conversely, mature or aged storm systems often have a more or less extensive stratiform zone in which precipitation is regular and less intense.
Convective subsidences are controlled by downdrafts, which are formed as a result of the evaporation of some of the hydrometeors. In fact, because of the cooling effect of evaporation, the air masses buoyancy is decreasing, and the air masses heading towards the surface. Usually, downdrafts form a sort of shell surrounding the updrafts. When the downdrafts reach the surface, cold pockets of dense and cold air masses are formed and spread at the surface. The cold pockets are respon-sible for cold front causing surface gust winds. Sometimes, gust winds can carry again hot and humid air masses a little further away, towards a new deep convec-tion cell. Thus, while the convection is still active, some new deep convective cells can appear near the cold pockets of the convection, and make the convection become deeper.

Table of contents :

1 Introduction 
1.1 Introduction (English)
1.2 Introduction (Français)
2 Fundamentals in atmospheric physics and chemistry 
2.1 Structure of the atmosphere and tropopause definitions
2.1.1 Global atmospheric structure
2.1.2 Tropopause, a level: thermal and dynamical definitions
2.1.3 Tropical Tropopause, a transition layer: the TTL
2.2 Tropical convective structures and mechanisms
2.2.1 Deep convective structures, diurnal variability and mechanisms
2.2.2 Diabatic processes
2.2.2.1 Theoretical model of the air parcel
2.2.2.2 Convective initiators and ascending processes
2.2.2.3 Radiative processes
2.2.2.4 Microphysical processes
2.2.2.5 Turbulent processes
2.2.2.6 Overshoots
2.2.2.7 Subsidence, precipitation and cold pockets
2.2.3 Geographical and seasonal variability of deep convection
2.3 Tropical tropospheric and stratospheric dynamic processes
2.3.1 Tropospheric zonal transport of energy: the Walker cells
2.3.2 Tropospheric meridional transport of energy : the Hadley circulation
2.3.3 Tropical stratospheric dynamics: the Brewer and Dobson cirulation
2.3.4 Equatorial Waves
2.3.5 Times scales in the tropics: annual and diurnal cycles
2.3.6 Madden Julian Oscillation
2.3.7 El Nino Southern Oscillation
2.3.8 Transport across the tropopause
2.4 Water Budget in the tropics: Troposphere to Stratosphere
2.4.1 Water Vapour and Ice
2.4.2 Water Saturation and Relative Humidity
2.4.3 Tropospheric Water Budget in the tropics
2.4.4 TTL Water Budget
2.4.5 Stratospheric water budget in the tropics
2.5 Atmospheric characteristics over the Maritime Continent
2.5.1 Precipitations over the MariCont
2.5.2 Troposphere and stratosphere air masses exchanges over
the MariCont and « Stratospheric fountain »
2.6 Chapter summary and Thesis objectives
3 Instruments and models used
3.1 Satellite remote sensing observations: definitions and motivations
3.1.1 Orbital and viewing modes
3.1.2 Overview of the satellites measuring Water vapour in the UTLS
3.1.3 Overview of satellites measuring the IceWater Content in the UTLS
3.2 Instruments used
3.2.1 MLS
3.2.1.1 MLS instrument on the Aura platform
3.2.1.2 MLS components characteristics
3.2.1.3 WV and IWC diurnal cycles from MLS in tropics: the Day-Night method
3.2.2 SMILES
3.2.3 TRMM
3.2.3.1 Deep convection estimation from TRMM
3.2.3.2 TRMM – 3B42
3.2.3.3 TRMM – LIS
3.3 Meso-scale models and meteorological reanalysis used
3.3.1 ERA5
3.3.2 Meso-NH
3.3.3 NCEP/NCAR Reanalysis
3.4 Chapter summary
4 Ice injected into the TTL in the austral tropics 
4.1 Article-1: Ice injected into the tropopause by deep convection
4.3 Chapter summary
5 Ice injected into the TTL over the Maritime Continent
5.1 Article-2: Ice injected into the tropopause by deep convection 
5.2 Abstract
5.3 Introduction
5.4 Datasets
5.4.1 MLS Ice Water Content
5.4.2 TRMM-3B42 Precipitation
5.4.3 TRMM-LIS number of Flashes
5.4.4 ERA5 Ice Water Content
5.5 Methodology
5.6 Horizontal distribution of IWC estimated from Prec over the  MariCont
5.6.1 Prec from TRMM related to IWC from MLS
5.6.2 Convective processes compared to IWC measurements
5.6.3 Horizontal distribution of ice injected into the UT and TL estimated from Prec
5.7 Relationship between diurnal cycle of Prec and Flash over Mari-Cont land and sea
5.7.1 Flash distribution over the MariCont
5.7.2 Prec and Flash diurnal cycles over the MariCont
5.7.3 Prec and Flash diurnal cycles and small-scale processes
5.8 Horizontal distribution of IWC from ERA5 reanalyses
5.9 Ice injected over a selection of island and sea areas
5.9.1 IWC deduced from observations
5.9.2 IWC deduced from reanalyses
5.9.3 Synthesis
5.10 Discussion on small-scale convective processes impacting IWC over a selection of areas
5.10.1 Java island, Sulawesi and New Guinea
5.10.2 West Sumatra Sea
5.10.3 North Australia Sea and seas with nearby islands
5.11 Conclusions
5.12 Author contribution
5.13 Acknowledgement
5.14 Data availability
6 Impact of large-scale oscillations on the ice injected in the TTL over the Maritime Continent 
6.1 Context
6.2 Datasets, study zones and methodology
6.2.1 Datasets
6.2.2 Study zones
6.2.3 Methodology
6.3 Diurnal cycle of Prec over the Maritime Continent: comparison between the study periods
6.4 Horizontal distribution of Prec and IWCMLS measured at 01:30 LT and 13:30 LT
6.4.1 Prec during the increasing phase of the convection
6.4.2 IWCMLS in the UT during the increasing phase of the convection
6.5 IWC injected in the UT during DJF, MJO active in MariCont and La Niña
6.5.1 Horizontal distribution of IWCPrec
6.5.2 IWC during MJO over island and sea of the MariCont
6.5.3 IWCPrec during the study periods over island and sea of the MariCont
6.6 Synthesis and discussion
6.7 Acknowledgement
7 Further works: Application of the methodology to the Asian monsoon region 
7.1 Introduction
7.2 Instruments and methodology
7.3 Horizontal distributions of Prec, Flash and IWC over Asia
7.3.1 Horizontal distributions
7.4 Diurnal cycle of Prec and Flash over the Asian study zones
7.5 Diurnal cycle of IWC over the Asian study zones
7.6 Horizontal distribution of IWC over Asia
7.7 IWC over Asian study zones
7.8 Synthesis
7.9 Author contribution
8 Conclusions and perspectives 
8.1 Conclusions (English)
8.1.1 Issues and motivations: the TTL a transition layer between the troposphere and the stratosphere
8.1.2 Objectives and strategy
8.1.3 Method used
8.1.4 Validation
8.1.5 Main results of the thesis
8.2 Perspectives (English)
8.2.1 Tropical deep convection from space-borne observations
8.2.2 Diurnal cycle of water budget in TTL
8.2.3 Ice injection up to the lower stratosphere
8.2.4 Impact of large-scale oscillation of the ice injected into the TTL
8.2.5 Integration of the results into current research projects
8.3 Conclusions (Français)
8.3.1 Enjeux et motivations : la TTL, une couche de transition entre la troposphère et la stratosphère
8.3.2 Objectifs et stratégie
8.3.3 Méthode utilisée
8.3.4 Validation
8.3.5 Résultats principaux de la thèse
8.4 Perspectives (Français)
8.4.1 Convection profonde tropicale à partir d’observations spatiales
8.4.2 Cycle diurne du bilan hydrique dans la TTL
8.4.3 Injection de la glace jusqu’à la basse stratosphère
8.4.4 Impact de l’oscillation à grande échelle de la glace injectée dans le TTL
8.4.5 Intégration des résultats dans les projets de recherche actuels226

GET THE COMPLETE PROJECT