Introduction: Instrumented gait analysis and inverse dynamics are commonly used in research and clinical practice to calculate lower extremity joint kinetics, such as power and work. However, multisegment foot (MSF) model kinetics have been limited by ground reaction force (GRF) measurements. New technology enables simultaneous capture of plantar pressure and shear stress distributions but has not yet been used with motion capture. Integrating MSF models and pressure/shear measurements will enhance the analysis of foot joint kinetics. The purpose of this study was to develop methodology to integrate these systems, then analyze the effects of speed on foot joint kinetics. Methods: Custom software was developed to synchronize motion capture and pressure/shear data using measured offsets between reference frame origins and time between events. Marker trajectories were used to mask pressure/shear data and construct segment specific GRFs. Inverse dynamics were done in commercial software. Demonstrative data was from 5 healthy adults walking unshod at 3 fixed speeds (1.0, 1.3, and 1.6 m/s, respectively) wearing retroreflective markers according to an MSF model. Plantar shear forces and ankle, midtarsal, and first metatarsophalangeal (MTP) joint kinetics were reported. Speed effects on joint net work were evaluated with a repeated measures ANOVA. Results: Plantar shear forces during stance showed some spreading effects (directionally opposing shear forces) that relatively were unaffected by walking speed. Midtarsal joint power seemed to slightly lag behind the ankle, particularly in late stance. Net work at the ankle (p = 0.024), midtarsal (p = 0.023), and MTP (p = 0.009) joints increased with speed. Conclusions: Functionally, the ankle and midtarsal joints became more motorlike with increasing speed by generating more energy than they absorbed, while the MTP joint became more damperlike by absorbing more energy than it generated. System integration appears to be an overall success. Limitations and suggestions for future work are presented.

<p>Development of complex aerospace systems often takes decades of research and testing. High  performing propellants are important to the success of rocket propulsion systems. Development  and testing of new propellants can be expensive and dangerous. Full scale tests are often required  to understand the performance of new propellants. Many industries have started using data science  tools to learn from previous work and conduct smarter tests. Material scientists have started using  these tools to speed up the development of new materials. These data science tools can be used to  speed up the development and design better propellants. I approach the development of new solid  propellants through two steps: Prediction of delivered performance from available literature tests,  prediction of ideal performance using physics-based models. Random Forest models are used to  correlate the ideal performance to delivered performance of a propellant based on the composition  and motor properties. I use Parsimonious Neural Networks (PNNs) to learn interpretable models  for the ideal performance of propellants. I find that the available open literature data is too biased  for the models to learn from and discover families of interpretable models to predict the ideal  performance of propellants. </p>

No description available.

Introduction: Recent evidence in isolated mitochondria and permeabilized muscle fibers in ex vivo using simultaneous measurements of O2 consumption and ATP production suggest that mitochondrial efficiency provides an additional mechanism to fine-tune oxidative phosphorylation rate to ATP demand in skeletal muscle. However, in the absence of a direct measurement of both VO2 and ATP synthesis from the same region of the contracting muscle, whether this mechanism plays a role in the skeletal muscle in vivo is still unknown. Purpose: Using a noninvasive approach combining phosphorus and proton magnetic resonance spectroscopy (31P/1H-MRS), the present study aimed to determine skeletal muscle ATP synthesis rate and muscle VO2 during a graded dynamic plantar flexion exercise to determine mitochondrial efficiency in contracting skeletal muscle contraction. Method: To measure mitochondrial efficiency under physiological conditions, we applied a recently developed methodological approach in ex vivo to human gastrocnemius muscle in vivo using 31P/1H-MRS noninvasive techniques. We conducted a series of constant workloads and ischemic protocols to assess oxidative ATP synthesis (ATPox) rate and Myoglobin-derived oxygen consumption (Mb-derived VO2). Specifically, during two separate visits, in 12 healthy, sedentary to recreationally active young male adults, we determined the ATPox rate by measuring the initial phosphocreatine (PCr) resynthesis rate during recovery and Mb-derived VO2 during 30 seconds of occlusion at the end of each given exercise workload. Results: The calculated mean power output during constant load performed by all subjects increased linearly at each incremental workload for both 1H and 31P visits. The mean percent coefficient of variation (CV%) at all exercise workloads (25% of WRmax= 6.0 ± 6.6, 50% of WRmax = 4.5 ± 2.7, 75% of WRmax = 4.9 ± 3.9, 100% = 7.0 ± 4.5) demonstrated high reliability and reproductivity in power output between each visit. The mean concentration of PCr consumption at the steady state remained constant between 25 and 50% of WRmax (28.6 ± 1.7 and 28.2 ± 1.4, respectively) but increased linearly beyond 50% of WRmax (75% = 22.9 ± 1.4 mM and 100%= 18.0 ± 1.4 mM, respectively). Mean intracellular pH was not significantly different between 25 and 50% of submaximal workloads (6.98 ± 0.02 and 6.97 ± 0.02, respectively). Intracellular pH dropped to 6.94 ± 0.02 during the last min of exercise at 75% of WRmax and further decreased at 100% of WRmax, (6.87 ± 0.03). The percentage of Mb oxygenation level and partial pressure of oxygen(PO2) at the steady state exhibited a consistent linear decline with increasing workload. Accordingly, a distinct and strong linear relationship was found between the MbO2 and workload (r2 = 0.71). Similarly, as the exercise workload increased, the ATPox synthesis rate also increased linearly throughout all exercise workloads (r2 = 0.45). In contrast, there was no significant change in Mb-derived VO2 with increased exercise workload averaged over a 30 s (r2 = 0.36) and 10 s time frame (r2 = 0.17). As a result, there was no significant correlation between ATPox synthesis rate and Mb-derived VO2 across exercise intensities. However, at rest, the P/O ratio in the gastrocnemius muscle was 1.95 ± 0.68, consistent with theoretical values and previous studies in mice. Conclusion: Despite the decrease in MbO2 (%), Mb-derived VO2 from both 30 and 10 s averages remained relatively constant during the ischemic protocol, likely due to O2 availability limitation induced by the prolonged occlusion and the slow time-resolution for measuring the dMb signal, which precluded the quantification of mitochondrial efficiency during exercise. However, mitochondrial efficiency calculated at rest was in agreement with previously documented values using other methodologies and thus can provide an additional parameter to more comprehensively evaluate mitochondrial function in vivo.

<p>The ignition of energetic materials, and specifically solid propellants, is a complex process</p> <p>that must be safe, consistent, and precisely controlled. There is a wide range of applications with</p> <p>specific ignition requirements for solid propellants including inflation of airbags, propulsion</p> <p>systems (including rockets), as well as arm and fire devices. Currently, electrical or percussion</p> <p>pyrotechnic igniters are most the commonly used ignition systems. These systems must be</p> <p>carefully designed to deliver the proper amount of energy to a specified surface area of the</p> <p>propellant. A photon light source (i.e. flash or laser-based, ranging from UV to IR wavelengths)</p> <p>can potentially be used to ignite energetic materials with lower input energy and more precise</p> <p>spatial and temporal control, thereby improving safety and reliability by eliminating electrical</p> <p>systems used in pyrotechnic igniters. In addition, they could be potentially safer from stray</p> <p>electrical charges causing unintentional ignition.</p> <p>The purpose of this work is to further explore the potential of optical ignition for energetic</p> <p>systems and identify ideal materials that can be used for optical ignition. In order to identify</p> <p>optically sensitive materials, we will study ignition energies, ignition delays, flame temperatures,</p> <p>and other combustion characteristics for possible energetic materials. This research addresses a</p> <p>gap in understanding of optical ignition for energetic materials, as finding and integrating materials</p> <p>that are optically sensitive while still being practical can be extremely challenging. These</p> <p>challenges include: (1) a lack of absorptivity to optical wavelengths in the UV to low-IR range,</p> <p>and subsequently, a very high sensitivity to input energy at the absorptive wavelengths that makes</p> <p>sustained ignition difficult, (2) a need for full density materials in practical energetic systems,</p> <p>while optically sensitive materials are exceedingly difficult to ignite as packing density increases</p> <p>due to heat transfer, and (3) the lack of research regarding novel fuels/oxidizers for the specific</p> <p>purpose of optical ignition.</p> <p>Metal/fluoropolymer energetic materials have been of interest to the energetic materials</p> <p>community for many years. Due to fluorine’s excellent oxidizing ability, they can be used in</p> <p>composite materials with metal fuels to produce energetic materials for a wide variety of</p> <p>applications. Polytetrafluoroethylene (PTFE), polyvinylidene fluoride (PVDF), polycarbon</p> <p>13</p> <p>monofluoride (PMF), and terpolymers such as tetrafluoroethylene, hexafluoropropylene, and</p> <p>vinylidene fluoride (THV) have already seen extensive use in applications ranging including</p> <p>protective coatings, strain gauges, and electronics. However, when combined with metals such as</p> <p>lithium, magnesium, aluminum, or titanium, they also present an opportunity for a wide variety of</p> <p>energetic materials. For this study, metal/fluoropolymer composites present a novel opportunity</p> <p>for exploring optical ignition of widely absorptive, full-density energetic materials. This work will</p> <p>characterize the combustion and sensitivity of metal/fluoropolymer composites to provide novel</p> <p>materials for optical ignition of energetics.</p> <p>Specifically, this work will begin with finding a suitable energetic composite that is optically</p> <p>sensitive. Once this material has been identified, research will be done to thoroughly characterize</p> <p>the optically sensitive composite by looking at additive manufacturability, flame temperatures, and</p> <p>ignition sensitivities from various methods and formulations. Once the material has been</p> <p>thoroughly characterized, it will be implemented into solid propellants to test the feasibility of the</p> <p>material in practical energetic systems. Finally, the lessons learned from this work will be applied</p> <p>to novel formulations to identify new optically sensitive energetic composites.</p>

No description available.

Particularly lacking in the current body of sexual selection literature are studies based on reptile species and intrasexually selected mating systems. Because the life history traits and ecology of reptiles are dramatically different from other animal taxa, current models of sexual selection are insufficient for predicting how sexual selection should influence the behavior and mating systems of lizards. Similarly, intersexually selected mating systems (i.e., based on female choice) are inappropriate predictive models for examining species with intrasexually selected mating systems (i.e., based on consexual contests). I investigated three aspects of Anolis carolinensis behavior and mating system (communication signals, male mate preference, and reproductive energetics) to contribute to a theoretical model for sexual selection based on a lizard with an intrasexually selected, polygynous mating system. In my first study, I quantified the structure and use of signals exchanged by both sexes, compared signal structure and use during heterosexual interactions to that of other social contexts (e.g., male-alone, male-male, female-female), then related signal structure and use to the species mating system. During heterosexual interactions, both sexes performed three kinds of stereotypic headbob displays with equal precision that were essentially identical to those previously documented for other social contexts. Thus, there is no courtship-specific headbob display for A. carolinensis. However, male and female signal use was extremely dimorphic. For the purpose of indicating sexual identity, the sexually dimorphic patterns of signal use were excessively redundant, yet equivocal. Although the male pattern of signal use reliably conveys sexual identity, the female pattern of signal use conveys ambiguous sexual identity. Based on circumstantial evidence from other studies, I propose the hypothesis that the female pattern of signal use may permit female-sized, nonterritorial males to mimic female signals. Small males may be selected to use female mimicry to gain access to the territories of larger males and mating opportunities with resident females, while females may be indirectly selected to use a signaling pattern that provides them with an alternative mating option. From field and laboratory data on A. carolinensis signal behavior during other social contexts and the species' female-defense mating system, I evaluate proposed functions for heterosexual signaling from a perspective of intrasexual selection. In my second study, I tested the prediction that males should include a preference for mating with novel females (PNF) as part of their mating strategy. This prediction was supported by both laboratory and field manipulations. Compared to their encounters with resident females, males during laboratory encounters with novel females significantly increased their display rate, volley frequency, volley length, and significantly decreased the distance and number of movements traveled away from the female. My laboratory data also suggest that males discriminated novel females from resident females independently of female behavioral or chemical cues. Similarly, compared to their interactions with resident females, free-ranging males responded to introduced novel females by significantly increasing the proportion of time spent in female-directed activities and the proportion of displays directed toward novel females, and significantly decreasing the proportion of time spent in territorial activities and the proportion of displays used in territorial activities. Data from both experiments indicate that males appear to distinguish among individual females, and use this ability to increase reproductive success by identifying and preferentially pursuing novel females over previously inseminated resident females. I suggest that males are able to cognitively identify individual resident females, and use this ability to control mating decisions within their territories. In my third study, I examined the energy expenditure of males and females during breeding and postbreeding seasons. I used laboratory respirometry to determine resting metabolic rates, and the doubly-labeled water technique to determine field metabolic rates in free-ranging lizards. Resting metabolic rates were significantly influenced by body mass and season, but not sex. Field metabolic rates were significantly influenced by body mass, but not sex or season. I attributed the ~40% seasonal increase in resting metabolic rates to a seasonal increase in feeding rates and the effect of specific dynamic action. Resting and field metabolic rates were used to calculate energy budgets for each sex during breeding and postbreeding seasons, and to calculate the energy expended by each sex for reproduction. Despite having 40% smaller body mass, females expended 46% more energy for reproduction than males, and a similar amount of total maintenance energy as males. The total maintenance energy of males was similar during both seasons, however that of females decreased 44% from breeding to postbreeding season. I found both seasonal and sexual differences in the amount of energy lizards allocated to resting and activity. Anolis carolinensis had field metabolic rates that were similar to tropical and temperate species of lizards, and higher than lizards from arid/semiarid environments. Anolis carolinensis also expended more energy on eggs, and more total energy during the breeding season, than lizards from arid/semiarid habitats. / Ph. D.

<p dir="ltr">Tailorable energetics are energetic materials that can be modified to alter their performance and sensitivity. Examples of tailoring energetic materials include additive manufacturing, manufactured hot spots, switchable energetics, and cocrystallization. Developing novel energetic material is a difficult and cost intensive process, because of this, tailoring the performance and sensitivity of existing energetic materials is critical for continued improvement. Additive manufacturing has provided new methods for generating complex geometries of composite materials. Additive manufacturing of composite materials through direct-ink-write (DIW) experiences extrusion limitations due to the high viscosities of highly solids loaded mixtures; the limitations being more severe with smaller syringe tip diameter. A novel printing technique called vibration-assisted printing (VAP) was developed as a method to extend the extrudability limits and extrusion speeds observed with direct-ink-write systems. Printability envelopes were shown in previous work to extend extrudability of monomodal glass bead composites in VAP systems over conventional DIW systems. This study compares the mass flowrates and extrudability limits for bimodal mixtures of glass beads suspended in a hydroxyl-terminated polybutadiene (HTPB) binder for both VAP and DIW printing as a function of volume percent solids loading. The bimodal glass bead mixtures showed a linear response in extrusion rate versus solids loading for both VAP and DIW systems. The VAP system was able to print higher volume loadings than the direct-ink-write system. In addition to extending the extrudability limits, the mass flowrate for the VAP system was significantly higher at all volume loadings tested compared to the DIW. Interestingly, bimodal mixtures were shown to extrude quicker than the monomodal mixtures at all volume loadings and across both printing systems.</p><p><br></p><p dir="ltr">Inhomogeneities within explosives affect the sensitivity and detonation wave shape of energetic materials. The influence of voids on explosive initiation has been well documented; however, the effects that voids between 0.1 mm and 10 mm have on a propagating detonation wave remains largely unexplored. The effect of single cylindrical voids on detonation wave shape and re-initiation was examined here using manufactured voids in a rubberized 1,3,5-trinitro-1,3,5-triazinane (RDX) explosive. Two streak imaging techniques were fielded to investigate void influence. For the first, back-surface streak imaging, the location of the void on the samples was changed and the resulting change in detonation wave shape at the downstream breakout was captured using a streak camera in cut-back experiments. The results from this experiment showed the effects of an initial jet form for short cut-back distances and as shock propagation progressed, the jet formation was absorbed by the unaffected portions of the wave front. The second method, top-surface streak imaging, was used to investigate the re-initiation/downstream propagation of the detonation front and the detonation velocity of the rubberized explosive. These experiments were compared to similar experimental results from machined voids in PBX 9501, a 1,3,5,7-tetranitro-1,3,5,7-tetrazocane (HMX)-based explosive, to investigate the interaction of a detonation wave with a 0.5 mm void for different explosives. The experiments were also compared to simulations using a multi-dimensional and multi-material hydrodynamic code (CTH). These results showed the influence that small features can have on detonation wave shaping and how explosive properties play a key role in that interaction. In addition to air-filled voids, this study examined the effects of 0.5 mm diameter voids filled with different inert metals on the detonation wave shape for an RDX-based rubberized explosive. The metals selected for experiments were 1066 aluminum, brass, copper, and tungsten. Experimental results showed that the extent of detonation wave shaping was closely tied to the density differential between the bulk explosive and metal insert. Simulations were performed using CTH to further analyze material inclusions. Forty-four different filler materials were simulated to isolate the driving factors for wave shaping of the detonation front. The main factors of interest were bulk sound speed, shock impedance, and filler material density. Understanding the influence of material inclusions on detonation performance and wave shape allows for tailoring of detonations as well as characterizing how unintentional defects will alter the explosive.</p><p><br></p><p dir="ltr">Improving the safety of explosive materials through the synthesis of insensitive explosives has been studied extensively. However, little work has focused on creating switchable explosives. A switchable explosive is normally insensitive to detonation, and therefore safe to handle and transport, but can be sensitized when needed to create a functional explosive. Similarly, it may be desired to desensitize an explosive to prevent its function. This study examined the ability to create a switchable RDX-based rubberized explosive using thermally-expandable microspheres (TEMs). The addition of TEMs to the explosive formulation allowed for microstructural changes and potential hot spot locations such as voids to form as the microspheres expanded. Small voids (less than about 10 µm) are more likely to be critical hot spots when shocked, and likewise larger voids are less likely to ignite successfully (sub-critical) when shocked. Consequently, both sensitization and desensitization are possible. The rubberized explosive considered here with unexpanded microspheres was unable to sustain a detonation for the size used, but after specific heating followed by cooling to produce small voids, a detonation was achieved. That is, the TEMs addition to the RDX-based rubberized explosive resulted in an explosive that is detonation insensitive when unheated but becomes a functional explosive after it is sensitized through heating. This paves the way to create insensitive explosive formulations with on-demand switchable detonation function through the incorporation of thermally-expandable microspheres. Desensitization was also demonstrated with specific heating of TEMs in an initially detonable explosive charge. And finally, we also demonstrated that deflagration can be affected by heating TEMs.</p><p><br></p><p dir="ltr">Energetic cocrystallization is a technique that produces a cocrystal that is formed using two known explosives to potentially gain the benefits of one or both without the drawbacks for a particular application. A comparison of cocrystals to a physical mixture of the same coformers can be considered. Cocrystals have unique material properties and crystal structure, whereas a physical mixture is just a mixed combination of the known materials at the same molar ratio. This study used photon Doppler velocimetry (PDV) to compare the particle velocity for 2,4,6,8,10,12-hexanitro-2,4,6,8,10,12-hexaazaisowurtzitane (CL-20) and 1-methyl-3,5-dinitro-1,2,4-triazole (MDNT) at a 1:1 molar ratio for both a cocrystal and a physical mixture of the two energetic materials. This cocrystal was previously shown to detonate faster than a physical mixture. However, the PDV results here were not consistent with this result. In addition to measuring output particle velocity with PDV, the cocrystal was characterized to examine phase purity and possible signs of deterioration of the material over time. Evaluating the cocrystal with Fourier-transform infrared spectroscopy (FT-IR), bomb calorimetry, and powder X-ray diffraction (PXRD) allowed for more accurate comparison and greater confidence in the particle velocity measurements obtained in these experiments. The most significant difference in the material characterization results was the difference in enthalpy of formation, as the material tested in this study had a substantially lower enthalpy of formation than previously measured for a CL-20/MDNT cocrystal.</p>

Interest in non-traditional fuel sources, carbon dioxide sequestration, and cleaner combustion has brought attention on gasification to supplement fossil fueled energy, particularly by a fluidized bed. Developing tools and methods to predict operation and performance of gasifiers will lead to more efficient gasifier designs. This research investigates bed fluidization and particle decomposition for fluidized materials. Experimental methods were developed to model gravimetric and energetic response of thermally decomposing materials. Gravimetric, heat flow, and specific heat data were obtained from a simultaneous thermogravimetric analyzer (DSC/TGA). A method was developed to combine data in an energy balance and determine an optimized heat of decomposition value. This method was effective for modeling simple reactions but not for complex decomposition. Advanced method was developed to model mass loss using kinetic reactions. Kinetic models were expanded to multiple reactions, and an approach was developed to identify suitable multiple reaction mechanisms. A refinement method for improving the fit of kinetic parameters was developed. Multiple reactions were combined with the energy balance, and heats of decomposition determined for each reaction. From this research, this methodology can be extended to describe more complex thermal decomposition. Effects of particle density and diameter on the minimum fluidization velocity were investigated, and results compared to empirical models. Effects of bed mass on pressure drop through fluidized beds were studied. A method was developed to predict hydrodynamic response of binary beds from the response of each particle type and mass. Resulting pressure drops of binary mixtures resembled behavior superposition for individual particles. / Master of Science

We use four current meter mooring records and quasi-synoptic hydrographic observations in conjunction with a one-dimensional quasi-geostrophic (QG) linear stability model to investigate the downstream changes in the Brazil Current (BC) System off the southeast Brazil (22°S-28°S) as well as its implications. The dataset depicts the downstream thickening of the BC: Its vertical extension increases from 350 m at 22.7°S to 800 m at 27.9°S. Most of this deepening occurs between 25.5°S and 27.9°S, and it is likely linked to the Santos bifurcation. To the south of that bifurcation, the BC transport is increased by at least 5 Sv. Moreover, the analysis of the water column average kinetic energy (IKE) and its barotropic/baroclinic partition show that the Santos bifurcation is associated with a substantial increase in the barotropic component of the BC System: The IKE is, on average, 70 % baroclinic to the north and becomes 63 % barotropic to the south of that bifurcation. The water column average eddy kinetic energy (IEKE) and its ratio to the IKE quantitatively reveal the conspicuous mesoscale activity associated to the BC off the southeast Brazil; accordingly, the IEKE accounts for (30-60)% of the IKE. The linear stability model predicts southwestward-propagating fastest-growing waves [~(180-190) km] within 25.5°S-27.9°S and quasi-standing most-unstable waves (~230 km) at 22.7°S, roughly consistent with observations and previous work. We also assess the ability of the QG modes and surface QG (SQG) solutions to represent the vertical structure of the sub-inertial time-varying flow in the southwestern Atlantic. At two moorings, which present a sharp near-surface decay in the vertical structure of the 1st empirical orthogonal function (EOF) of current meter time series, the SQG solutions are consistent with the data, accounting for up to 85 % of the 1st EOF variance. The SQG solutions are nonetheless indistinguishable from a four QG mode representation. In contrast, at a third mooring that do not present such sharp-decay, the vertical structure of the 1st EOF is fairly well-captured by the traditional barotropic/1st baroclinic mode combination, which accounts for 91 % of its variance. We argue that such vertical structures may be associated with the type of instability experimented by the mean flow in each region. \"Charney-like\" or surface-intensified \"Phillips-like\" instabilities may rationalize the observed SQG-like vertical structures depicted at two moorings. Mid-depthintensified \"Phillips-like\" instabilities are consistent with a two QG mode representation at a third mooring. / Séries temporais correntográficas, observações hidrográficas quase-sinóticas e um modelo linear quase-geostrófico (QG) são combinados com o propósito de investigar as transformações no Sistema Corrente do Brasil (CB) ao largo da costa sudeste (22°S-28°S) e suas implicações. O conjunto de dados revela o espessamento vertical da CB, que ocupa os 350 m superiores da coluna de água em 22,7°S e atinge 800 m em 27,9°S. Parte significativa deste espessamento ocorre entre 25,5°S e 27,9°S, provavelmente relacionado à Bifurcação de Santos. Ao sul desta bifurcação, o transporte da CB é pelo menos 5 Sv superior. Ademais, a análise da energia cinética média na coluna de água (ECM) e sua partição entre componentes barotrópica e baroclínica revela que a Bifurcação de Santos está associada ao aumento significativo da componente barotrópica do Sistema CB. A ECM é, em média, 70% baroclínica ao norte da bifurcação, tornando-se 63% barotrópica ao sul desta. A análise da energia cinética turbulenta média na coluna de água (ECTM) corrobora o importante papel da atividade de mesoescala do Sistema CB ao largo do sudeste do Brasil: A ECTM é responsável por (30-60)% da ECM. O modelo de estabilidade linear prevê ondas com maiores taxas de crescimento [~(180-190) km] que se propagam para sudoeste entre 25,5°S-27,9°S. Em 22,7°S, as ondas mais instáveis (~230 km) crescem essencialmente sem propagação, consistente com as observações e também com informações presentes na literatura. A habilidade dos modos QG e das soluções QG superficiais (QGS) em representar a variabilidade subinercial no Atlântico Sudoeste também é investigada. Em dois fundeios, a estrutura vertical da 1ª função empírica ortogonal (FOE) apresenta um decaimento agudo. Este decaimento é consistente com soluções QGS, que contêm até 85% da variância da 1ª FOE. No entanto, estas soluções convergem para uma representação por quatro modos QG. Por outro lado, a estrutura vertical da 1ª FOE em um terceiro fundeio não apresenta tal decaimento marcante. Consequentemente, a 1ª FOE é bem representada pela tradicional combinação dos modos barotrópico/1o baroclínico. Argumentamos que estas estruturas podem estar associadas ao tipo de instabilidade experimentada pelo escoamento médio em cada região. Instabilidades tipo \"Charney\" ou \"Phillips\" (intensificadas em superfície) são consistentes com estruturas verticais tipo QGS presentes em dois fundeios. Instabilidades tipo \"Phillips\" (intensificadas em meia água) são consistentes com a representação por dois modos QG em um terceiro fundeio