Geochemistry and petrology of the ferropicrite dikes and associated rocks of Vestfjella, western Dronning Maud Land, Antarctica

This study provides insights into the composition and origin of ferropicrite dikes (FeOtot = 13 17 wt. %; MgO = 13 19 wt. %) and associated meimechite, picrite, picrobasalt, and basalt dikes found at Vestfjella, western Dronning Maud Land, Antarctica. The dikes crosscut Jurassic Karoo continental flood basalts (CFB) that were emplaced during the early stages of the breakup of the Gondwana supercontinent ~180 Ma ago. Selected samples (31 overall from at least eleven dikes) were analyzed for their mineral chemical, major element, trace element, and Sr, Nd, Pb, and Os isotopic compositions. The studied samples can be divided into two geochemically distinct types: (1) The depleted type (24 samples from at least nine dikes) is relatively depleted in the most incompatible elements and exhibits isotopic characteristics (e.g., initial εNd of +4.8 to +8.3 and initial 187Os/188Os of 0.1256 0.1277 at 180 Ma) similar to those of mid-ocean ridge basalts (MORB); (2) The enriched type (7 samples from at least two dikes) exhibits relatively enriched incompatible element and isotopic characteristics (e.g., initial εNd of +1.8 to +3.6 and initial 187Os/188Os of 0.1401 0.1425 at 180 Ma) similar to those of oceanic island basalts. Both magma types have escaped significant contamination by the continental crust. The depleted type is related to the main phase of Karoo magmatism and originated as highly magnesian (MgO up to 25 wt. %) partial melts at high temperatures (mantle potential temperature >1600 °C) and pressures (~5 6 GPa) from a sublithospheric, water-bearing, depleted peridotite mantle source. The enriched type sampled pyroxene-bearing heterogeneities that can be traced down to either recycled oceanic crust or melt-metasomatized portions of the sublithospheric or lithospheric mantle. The source of the depleted type represents a sublithospheric end-member source for many Karoo lavas and has subsequently been sampled by the MORBs of the Indian Ocean. These observations, together with the purported high temperatures, indicate that the Karoo CFBs were formed in an extensive melting episode caused mainly by internal heating of the upper mantle beneath the Gondwana supercontinent. My research supports the view that ferropicritic melts can be generated in several ways: the relative Fe-enrichment of mantle partial melts is most readily achieved by (1) relatively low degree of partial melting, (2) high pressure of partial melting, and (3) melting of enriched source components (e.g., pyroxenite and metasomatized peridotite). Ferropicritic whole-rock compositions could also result from accumulation, secondary alteration, and fractional crystallization, however, and caution is required when addressing the parental magma composition.

The prehistory of Suomussalmi, eastern Finland; the first billion years as revealed by isotopes and the composition of granitoid suites

The four papers summarized in this thesis deal with the Archean and earliest Paleoproterozoic granitoid suites observed in the Suomussalmi district, eastern Finland. Geologically, the area belongs to the Kianta Complex of the Western Karelian Terrane in the Karelian Province of the Fennoscandian shield. The inherited zircons up to 3440 Ma old together with Sm Nd and Pb Pb data confirm the existence of previously anticipated Paleoarchean protocrust in Suomussalmi. The general timeline of granitoid magmatism is similar to that of the surrounding areas. TTG magmatism occurred in three distinct phases: ca 2.95 Ga, 2.83 2.78 Ga and 2.76 2.74 Ga. In Suomussalmi the TTGs sensu stricto (K2O/Na2O less than 0.5) belong to the low-HREE type and are interpreted as partial melts of garnet amphibolites, which did not significantly interact with mantle peridotites. Transitional TTGs (K2O/Na2O more than 0.5), present in Suomussalmi and absent from surrounding areas, display higher LILE concentrations, but otherwise closely resemble the TTGs sensu stricto and indicate that recycling of felsic crust commenced in Suomussalmi 200 Ma earlier than in surrounding areas. The youngest TTG phase was coeval with the intrusion of the Likamännikkö quartz alkali feldspar syenite (2741 ± 2 Ma) complex. The complex contains angular fragments of ultrabasic rock, which display considerable compositional heterogeneity and are interpreted as cumulates containing clinopyroxene (generally altered to actinolite), apatite, allanite, epidote, and albite. The quartz alkali feldspar syenite cannot be regarded as alkaline sensu stricto, despite clear alkaline affinities. Within Likamännikkö there are also calcite carbonatite patches, which display mantle-like O- and C-isotope values, as well as trace element characteristics consistent with a magmatic origin, and could thus be among the oldest known carbonatites in the world. Sanukitoid (2.73 2.71 Ga) and quartz diorite suites (2.70 Ga) overlap within error margins and display compositional similarities, but can be differentiated from each other on the basis of higher Ba, K2O and LREE contents of the sanukitoids. The Likamännikkö complex, sanukitoids and quartz diorites are interpreted as originating from the metasomatized mantle and mark the diversification of the granitoid clan after 200 Ma of evolution dominated by the TTG suite. Widespread migmatization and the intrusion of anatectic leucogranitoids as dykes and intrusions of varying size took place at 2.70 2.69 Ga, following collisional thickening of the crust. The leucogranitoids and leucosomes of migmatized TTGs are compositionally alike and characterized by high silica contents and a leucocratic appearance. Due to compositional overlap, definitive discrimination between leucogranitoids and transitional TTGs requires isotope datings and/or knowledge of field relationships. Leucogranitoids represent partial melts of the local TTGs, both the sensu stricto and transitional types, mostly derived under water fluxed conditions, with possible fluid sources being late sanukitoids and quartz diorites as well as dehydrating lower crust. The Paleoproterozoic 2.44 2.39 Ga A-type granitoids of the Kianta Complex emplaced in an extensional environment are linked to the coeval and more widespread mafic intrusions and dykes observed over most of the Archean nucleus of the Fennoscandian shield. The A-type intrusions in the Suomussalmi area are interpreted as partial melts of the Archean lower crust and display differences in composition and magnetite content, which indicate differences in the composition and oxidation state of the source.

Neoarchean sanukitoid series in the Karelian Province, Finland

Sanukitoid series intrusions can be found throughout the Archean Karelian Province of the Fennoscandian shield. All sanukitoids share the same controversial elemental characteristics: they have high content of incompatible elements such as K, Ba, and Sr as well as high content of the compatible elements Mg, Cr, and Ni, and high Mg#. This composition is explained by an enriched mantle wedge origin in a Neoarchean subduction setting. This study concentrates on sanukitoid intrusions and tonalite-trondhjemite-granodiorite series (TTGs) from Finnish part of the Karelian Province. The collected rock samples have been studied in the field and under microscope as well as for their whole-rock (including isotopes) and mineral compositions. The new data together with previously published analyses help us to better understand the petrogenesis, tectonic setting and reworking of the Archean rock units. TTGs from the Karelian Province form a voluminous series of granitoids and reworked migmatites. This study divides TTG series into two subgroups based on their elemental composition: low-HREE (heavy rare earth element) TTGs and high-HREE TTGs indicating pressure differences in their source. Sanukitoid series is a minor, divergent group of intrusions. These intrusions are variable sized, and the texture varies from even-grained to K-feldspar porphyritic. The elemental composition differentiates sanukitoids from more voluminous TTG groups, the SiO2 in sanukitoids varies to include series of gabbro, diorite, and granodiorite. U Pb age determinations from sanukitoid series show temporally limited emplacement between ~ 2745 2715 Ma after the main crust forming period in the area. Hafnium, neodymium, common lead, and oxygene isotopes indicate well homogenized characteristics. Recycled crust has made a variable, yet minor, contribution to sanukitoids, as evidenced by oxygene isotopes and inherited zircon cores. A proposed tectonic setting for the formation of the sanukitoid series is slab breakoff of oceanic lithosphere in subduction setting, with sanukitoids deriving from an enriched mantle wedge. The proposed setting explains some of the peculiar features of sanukitoids, such as their temporally limited occurrence and controversial elemental composition. Sanukitoids would occur after cessation of the regional growth of Archean crust, and they could be derived from mantle wedge previously enriched by melts and fluids from oceanic crust and sediments. A subsequent event during the Paleoproterozoic Svecofennian orogeny at ~1.9 Ga affected the appearance and microstructures of the rocks as well as caused redistribution of lead between minerals and whole rock. However, the deformation was not able to obliterate the original geochemical characteristics of these sanukitoids.

Precursor Dependent Microstructure Evolution and Nonstoichiometry in Nanostructured Cerium Oxide Coatings Using the Solution Precursor Plasma Spray Technique

A solution precursor plasma spray (SPPS) technique has been used for direct deposition of cerium oxide nanoparticles (CNPs) from various cerium salt solutions as precursors. Solution precursors were injected into the hot zone of a plasma plume to deposit CNP coatings. A numerical study of the droplet injection model has been employed for microstructure development during SPPS. The decomposition of each precursor to cerium oxide was analyzed by thermogravimetric-differential thermal analysis and validated by thermodynamic calculations. The presence of the cerium oxide phase in the coatings was confirmed by X-ray diffraction studies. Transmission electron microscopy studies confirmed nanocrystalline (grain size <14 nm) characteristic of the coatings. X-ray photoelectron spectroscopy studies indicated the presence of a high concentration of Ce3+ (up to 0.32) in the coating prepared by SPPS. The processing and microstructure evolution of cerium oxide coatings with high nonstoichiometry are reported.

Desorptio/fotoionisaatio ilmanpaineessa takavarikoitujen huumausaineiden, steroidien ja räjähdysaineiden tutkimuksessa

Tutkimuksen tarkoituksena oli selvittää desorptio/fotoionisaatio ilmanpaineessa tekniikan (engl. desorption atmospheric pressure photoionization, DAPPI) soveltuvuutta rikosteknisen laboratorion näytteiden analysointiin. DAPPI on nopea massaspektrometrinen ionisaatiotekniikka, jolla voidaan tutkia yhdisteitä suoraan erilaisilta pinnoilta. DAPPI:ssa käytetään lämmitettyä mikrosirua, joka suihkuttaa höyrystynyttä liuotin- ja kaasuvirtausta kohti näytettä. Näytteen pinnan komponentit desorboituvat lämmön vaikutuksesta, jonka jälkeen ionisoituminen tapahtuu VUV-lampun emittoimien fotonien avulla.DAPPI:lla tutkittiin takavarikoituja huumausaineita, anabolisia steroideja ja räjähdysaineita sekä niiden jäämiä erilaisilta pinnoilta. Lisäksi kartoitettiin DAPPI:n mahdollisuuksia ja rajoituksia erilaisille näytematriiseille ilman näytteiden esikäsittelyä. Takavarikoitujen huumausaineiden tutkimuksessa analysoitiin erilaisia tabletteja, jauheita, kasvirouheita, huumekasveja (khat, oopium, kannabis) ja sieniä. Anabolisia steroideja tunnistettiin tableteista sekä ampulleista, jotka sisälsivät öljymäistä nestettä. Jauheet ripoteltiin kaksipuoliselle teipille ja analysoitiin siltä. Muut näytteet analysoitiin sellaisenaan ilman minkäänlaista esikäsittelyä, paitsi nestemäisten näytteiden kohdalla näyte pipetoitiin talouspaperille, joka analysoitiin DAPPI:lla. DAPPI osoittautui nopeaksi ja yksinkertaiseksi menetelmäksi takavarikoitujen huumausaineiden ja steroidien analysoimisessa. Se soveltui hyvin rikoslaboratorion erityyppisten näytteiden rutiiniseulontaan ja helpotti erityisesti huumekasvien ja öljymäisten steroidiliuosten tutkimusta. Massaspektrometrin likaantuminen pystyttiin ehkäisemään säätämällä näytteen etäisyyttä sen suuaukosta. Likaantumista ei havaittu huolimatta näytteiden korkeista konsentraatioista ja useita kuukausia jatkuneista mittauksista. Räjähdysaineiden tutkimuksessa keskityttiin seitsemän eri räjähdysaineen DAPPI-MS-menetelmän kehitykseen; trinitrotolueeni (TNT), nitroglykoli (NK), nitroglyseriini (NG), pentriitti (PETN), heksogeeni (RDX), oktogeeni (HMX) ja pikriinihappoä Nämä orgaaniset räjähteet ovat nitraattiyhdisteitä, jotka voidaan jakaa rakenteen puolesta nitroamiineihin (RDX ja HMX), nitroaromaatteihin (TNT ja pikriinihappo) sekä nitraattiestereihin (PETN, NG ja NK). Menetelmäkehityksessä räjähdysainelaimennokset pipetoitiin polymetyylimetakrylaatin (PMMA) päälle ja analysoitiin siitä. DAPPI:lla tutkittiin myäs autenttisia räjähdysainejäämiä erilaisista matriiseista. DAPPI:lla optimoitiin jokaiselle räjähdysaineelle sopiva menetelmä ja yhdisteet saatiin näkymään puhdasaineina. Räjähdysainejäämien analysoiminen erilaisista rikospaikkamateriaaleista osoittautui haastavammaksi tehtäväksi, koska matriisit aiheuttivat itsessään korkean taustan spektriin, josta räjähdysaineiden piikit eivät useimmiten erottuneet tarpeeksi. Muut desorptioionisaatiotekniikat saattavat soveltua paremmin haastavien räjähdysainejäämien havaitsemiseksi.

Impact of physical processes on the chlorophyll distribution in the Bay of Bengal

Satellite-derived chlorophyll a concentration (chl a) maps show three regions with high chl a in the Bay of Bengal. First among these is close to the coast, particularly off river mouths, with high values coinciding with the season of peak discharge; second is in the southwestern bay during the northeast monsoon, which is forced by local Ekman pumping; and the third is to the east of Sri Lanka in response to the summer monsoon winds. Chlorophyll-rich water from the mouths of rivers flows either along the coast or in an offshore direction, up to several hundred kilometers, depending on the prevailing ocean current pattern. The Irrawady River plume flows toward offshore and then turns northwestward during October–December, but it flows along the coast into the Andaman Sea for the rest of the year. From the Ganga-Brahmaputra river mouth, chl a–rich water flows directly southward into the open bay during spring but along the Indian coast during summer and winter. Along the Indian coast, the flow of chl a–rich water is determined by the East India Coastal Current (EICC). Whenever the EICC meanders off the Indian coast, it leads to an offshore outbreak of chl a–rich water from the coastal region into open ocean. The EICC as well as open ocean circulation in the bay is made up of several eddies, and these eddies show relatively higher chl a. Eddies near the coast, however, can often have higher chl a because of advection from the coastal region rather than generation within the eddy itself. The bay experiences several cyclones in a year, most of them occurring during October–November. These cyclones cause a drop in the sea surface temperature, a dip in the sea level, and a local increase in chl a. The impact of a cyclone is weaker in the northern part of the bay because of stronger stratification compared to the southern parts.

Observations on transition in plane bubble plumes

Flow visualization studies of plane laminar bubble plumes have been conducted to yield quantitative data on transition height, wavelength and wave velocity of the most unstable disturbance leading to transition. These are believed to be the first results of this kind. Most earlier studies are restricted to turbulent bubble plumes. In the present study, the bubble plumes were generated by electrolysis of water and hence very fine control over bubble size distribution and gas flow rate was possible to enable studies with laminar bubble plumes. Present observations show that (a) the dominant mode of instability in plane bubble plumes is the sinuous mode, (b) transition height and wavelength are related linearly with the proportionality constant being about 4, (c) wave velocity is about 40 % of the mean plume velocity, and (d) normalized transition height data correlate very well with a source Grashof number. Some agreement and some differences in transition characteristics of bubble plumes have been observed compared to those for similar single-phase flows.

Propagating intensity disturbances in polar corona as seen from AIA/SDO

Context. Polar corona is often explored to find the energy source for the acceleration of the fast solar wind. Earlier observations show omni-presence of quasi-periodic disturbances, traveling outward, which is believed to be caused by the ubiquitous presence of outward propagating waves. These waves, mostly of compressional type, might provide the additional momentum and heat required for the fast solar wind acceleration. It has been conjectured that these disturbances are not due to waves but high speed plasma outflows, which are difficult to distinguish using the current available techniques. Aims. With the unprecedented high spatial and temporal resolution of AIA/SDO, we search for these quasi-periodic disturbances in both plume and interplume regions of the polar corona. We investigate their nature of propagation and search for a plausible interpretation. We also aim to study their multi-thermal nature by using three different coronal passbands of AIA. Methods. We chose several clean plume and interplume structures and studied the time evolution of specific channels by making artificial slits along them. Taking the average across the slits, space-time maps are constructed and then filtration techniques are applied to amplify the low-amplitude oscillations. To suppress the effect of fainter jets, we chose wider slits than usual. Results. In almost all the locations chosen, in both plume and interplume regions we find the presence of propagating quasi-periodic disturbances, of periodicities ranging from 10-30 min. These are clearly seen in two channels and in a few cases out to very large distances (approximate to 250 `') off-limb, almost to the edge of the AIA field of view. The propagation speeds are in the range of 100-170 km s(-1). The average speeds are different for different passbands and higher in interplume regions. Conclusions. Propagating disturbances are observed, even after removing the effects of jets and are insensitive to changes in slit width. This indicates that a coherent mechanism is involved. In addition, the observed propagation speed varies between the different passpands, implying that these quasi-periodic intensity disturbances are possibly due to magneto-acoustic waves. The propagation speeds in interplume region are higher than in the plume region.

Natural Convection Heat Transfer From an Isothermal Vertical Surface to a Stable Thermally Stratified Fluid

Natural convection from an isothermal vertical surface to a thermally stratified fluid is studied numerically. A wide range of stratification levels is considered. It is shown that at high levels of ambient thermal stratification, a portion at the top of the plate absorbs heat, while a horizontal plume forms around a location where the plate temperature equals the ambient temperature. The plume is shown to be inherently unsteady, and its transient nature is investigated in detail. The effect of the temperature defect in striating the plume is discussed. Average Nusselt number data are presented for Pr = 6.0 and 0.7.

Hydrography and circulation in the western Bay of Bengal during the northeast monsoon

The Bay of Bengal, a semienclosed tropical basin that comes under the influence of monsoonal wind and freshwater influx, is distinguished by a strongly stratified surface layer and a seasonally reversing circulation. We discuss characteristics of these features in the western Bay during the northeast monsoon, when the East India Coastal Current (EICC) flows southward, using hydrographic data collected during December 1991. Vertical profiles show uniform temperature and salinity in a homogeneous surface layer, on average, 25 m deep but shallower northward and coastward. The halocline, immediately below, is approximately 50 m thick; salinity changes by approximately 3 parts per thousand. About two thirds of the profiles show temperature inversions in this layer. Salinity below the halocline hardly changes, and stratification is predominantly due to temperature variation, The halocline is noticeably better developed and the surface homogeneous layer is thinner in a low-salinity plume that hugs the coastline along the entire east coast of India, The plume is, on average, 50 km wide, with isohalines sloping down toward the coast. Most prominent in the geostrophic velocity field is the equatorward EICC. Its transport north of about 13 degrees N, computed with 1000 dbar as the level of reference, varies between 2.6 and 7.1 x 10(6) m(3) s(-1); just south of this latitude, a northwestward flow from offshore recurves and merges with the coastal current. At the southern end of the region surveyed, the transport is 7.7 x 10(6) m(3) s(-1). Recent model studies lead us to conclude that the EICC during the northeast monsoon is driven by winds along the east coast of India and Ekman pumping in the interior bay. In the south, Ekman pumping over the southwestern bay is responsible for the northwestward flow that merges with the EICC.

Role of silver doping in oxygen incorporation of oxide thin film

A distinctive characteristic of silver in oxygen incorporation of oxide thin films during pulsed laser ablation has been discovered. Optical emission spectroscopy studies of laser-induced plume of Ag-target indicates the presence of AgO species whose concentration increases with an increase in oxygen partial pressure. The formation of AgO in laser-plume has been found to be very useful for the realization of high temperature superconducting YBa2Cu3O7-delta (YBCO) and giant magnetoresistive La0.7MnO3-delta (LMO) thin films with dramatically superior quality if the target materials contained a small amount of silver. The improvement in the quality of these films is brought about by the supply of atomic oxygen to oxide lattices during their formation. This becomes possible due to the fact that Ag, after it is ablated with other constituent materials in the target, gets moderately oxidized in an oxygen atmosphere and the oxidized species dissociate back into Ag and nascent O at the substrate surface. The nascent oxygen is very highly reactive and is easily assimilated into the lattice of these compounds. (C) 1997 Elsevier Science S.A.

Longwave radiative forcing of Indian Ocean tropospheric aerosol

A spectrally resolved discrete-ordinates radiative transfer model is used to calculate the change in downwelling surface and top-of-the-atmosphere (TOA) outgoing longwave (3.9-500 mum) radiative fluxes induced by tropospheric aerosols of the type observed over the Indian Ocean during the Indian Ocean Experiment (INDOEX). Both external and internal aerosol mixtures were considered. Throughout the longwave, the aerosol volume extinction depends more strongly on relative humidity than in most of the shortwave (0.28-3.9 mum), implying that particle growth factors and realistic relative humidity profiles must be taken into account when modeling the longwave radiative effects of aerosols. A typical boundary layer aerosol loading, with a 500-nm optical depth of 0.3, will increase the downwelling longwave flux at the surface by 7.7 W m(-2) over the clean air case while decreasing the outgoing longwave radiation by 1.3 W m(-2). A more vertically extended aerosol loading, exhibiting a high opacity plume between 2 and 3 km above the surface and having a typical 500-nm optical depth of 0.7, will increase the downwelling longwave flux at the surface by 11.2 W m(-2) over the clean air case while decreasing the outgoing longwave radiation by 2.7 W m(-2). For a vertically extended aerosol profile, approximately 30% of the TOA radiative forcing comes from sea salt and approximately 60% of the forcing comes from the combination of sea salt and dust. The remaining forcing is from anthropogenic constituents. These results are for the external mixture. For an internal mixture, TOA longwave forcings can be up to a factor of two larger. Therefore, to complete our understanding of this region's longwave aerosol radiative properties, more detailed information is needed about aerosol mixing states. These longwave radiative effects partially offset the large shortwave aerosol radiative forcing and should be included in regional and global climate modeling simulations.

Laboratory simulations show diabatic heating drives cumulus-cloud evolution and entrainment

Clouds are the largest source of uncertainty in climate science, and remain a weak link in modeling tropical circulation. A major challenge is to establish connections between particulate microphysics and macroscale turbulent dynamics in cumulus clouds. Here we address the issue from the latter standpoint. First we show how to create bench-scale flows that reproduce a variety of cumulus-cloud forms (including two genera and three species), and track complete cloud life cycles-e.g., from a ``cauliflower'' congestus to a dissipating fractus. The flow model used is a transient plume with volumetric diabatic heating scaled dynamically to simulate latent-heat release from phase changes in clouds. Laser-based diagnostics of steady plumes reveal Riehl-Malkus type protected cores. They also show that, unlike the constancy implied by early self-similar plume models, the diabatic heating raises the Taylor entrainment coefficient just above cloud base, depressing it at higher levels. This behavior is consistent with cloud-dilution rates found in recent numerical simulations of steady deep convection, and with aircraft-based observations of homogeneous mixing in clouds. In-cloud diabatic heating thus emerges as the key driver in cloud development, and could well provide a major link between microphysics and cloud- scale dynamics.

Length of near-wall plumes in turbulent convection

We present planforms of line plumes formed on horizontal surfaces in turbulent convection, along with the length of line plumes measured from these planforms, in a six decade range of Rayleigh numbers (10(5) < Ra < 10(11)) and at three Prandtl numbers (Pr = 0.7, 5.2, 602). Using geometric constraints on the relations for the mean plume spacings, we obtain expressions for the total length of near-wall plumes on horizontal surfaces in turbulent convection. The plume length per unit area (L(p)/A), made dimensionless by the near-wall length scale in turbulent convection (Z(w)), remains constant for a given fluid. The Nusselt number is shown to be directly proportional to L(p)H/A for a given fluid layer of height H. The increase in Pr has a weak influence in decreasing L(p)/A. These expressions match the measurements, thereby showing that the assumption of laminar natural convection boundary layers in turbulent convection is consistent with the observed total length of line plumes. We then show that similar relationships are obtained based on the assumption that the line plumes are the outcome of the instability of laminar natural convection boundary layers on the horizontal surfaces.

Planform structure of turbulent free convection on horizontal surfaces

The planform structure of turbulent free convection over a heated horizontal surface has been visualized and analyzed for different boundary conditions at the top and for different aspect ratios, for flux Rayleigh numbers ranging from 10 exp 8 - 10 exp 10. The different boundary conditions correspond to Rayleigh-Benard convection, open convection with evaporation at the top and with an imposed external flow on the heated boundary. Without the external flow the planform is one randomly oriented line plume. At large Ra, these line plumes seem to align along the diagonal, persumably due to a large-scale flow along as visualized in the side view. When the external flow is imposed, the line plumes clearly align in the direction of external flow. Flow visualization reveals that at these Ra, the shear tends to break the plumes which otherwise would reach the opposite boundary. (Author)