Numerical and experimental study of the dynamic thermal resistance of ventilated air-spaces behind passive and active façades

The variation of the thermo-hydrodynamic behavior of the airflow in the ventilated cavity behind external claddings in a wall structure could potentially affect the overall thermal resistance of the entire structure. Although the impact of an enclosed air-space in the wall on the total R-value of the assembly has been thoroughly investigated in the literature, it has not been addressed for a ventilated cavity. In the present study, as a first step, plausible definitions to determine the thermal resistance of the ventilated cavity behind external claddings in transient conditions are described. As the second step, the dynamic thermal resistance of the naturally ventilated air-spaces behind passive (i.e., brick and fiber cement) claddings and active (i.e., BIPV) facades are studied using two approaches. In the first approach, a 2-D numerical model validated against measurements is employed to compute the thermal resistance of the ventilated air gap under different conditions. Accordingly, the impact of the cladding type, seasonal variation, cavity thickness, and presence of the reflective insulation in the air gap on the dynamic change of the thermal resistance of the air-space is analyzed. In the second approach, in-field measurements are performed in a test facility to experimentally examine the contribution of the thermal resistance of the cavity to the total R-value of the wall structure. The results obtained from both approaches are compared with the design values calculated based on the ISO 6946:2017 and ISO 9869–1:2014 standards. The results reveal that the dynamic change of the thermal resistance of the air gap during a day could be captured using numerical simulations. It is shown that the daily averaged thermal resistance of the cavity could reach up to 47 times higher than the design R-value of the BIPV façade. The experimental measurements confirm that the thermal resistance of the ventilated air-space could converge to a steady-state value after a certain duration of time following the requirements provided in the standards, which could be practically used for the code-compliant analysis. It is also observed that the ventilated cavity could act as an insulation layer with higher thermal resistance compared to some of the solid materials used in the wall assembly.

Development of the Jungfraujoch UV DIAL Lidar to Observe the Vertical Ozone Distribution in the Context of Stratosphere Troposphere Exchange and Long Range Transport

Marcel Bartlome

Tropospheric ozone is a climate relevant greenhouse gas, as well as an atmospheric pollutant. Its abundance in the troposphere is mainly governed by the influx of stratospheric air masses and the photochemical production due to anthropogenic and biogenic ozone precursors. Precise model simulations and measurements are needed to assess the exact impact of these sources onto the total tropospheric ozone content. In the latter case, vertical tropospheric ozone profiles are commonly obtained from in-situ balloon-borne soundings. These routine observations are performed with a maximum frequency of several measurements per week. Therefore, they are not suited to resolve the rather fast changes in the tropospheric ozone concentration. However, accurate vertical ozone profiles with temporal and spatial resolutions suitable for studying those changes have been produced for decades using remote sensing by means of the ozone UV differential absorption lidar (DIAL) technique. The main goals of this thesis aimed to address some of the unresolved problems of the ozone origin and evolution in the upper troposphere and lower stratosphere and were: to develop an UV DIAL for the observation of the vertical ozone distribution in the free troposphere and at the tropopause region. to perform experimental measurements of vertical ozone profiles in the free troposphere and the tropopause in order to assess the ability of the ozone UV DIAL to observe fast variations in vertical ozone distribution caused by stratosphere-troposphere exchange (STE), long range transport, and advection from the boundary layer. to assess the feasibility of systematic ozone profile retrieval by the ozone UV DIAL technique from the High Altitude Research Station Jungfraujoch (HARSJ, 3580 m ASL, 7°59'2''E, 46°32'53''N). The HARSJ was chosen for the experiment because of its location, which allows ozone UV DIAL measurements of the tropopause region without perturbation of the lower troposphere. Furthermore, previous atmospheric studies have shown that because of the station's location, the upper tropospheric data taken at the HARSJ can be regarded as representative for central and western Europe. During the first two and a half years of the thesis project the ozone UV DIAL was designed and built. The ozone UV DIAL transmitter is based on a commercial, fourth harmonic Nd:YAG laser. The on- (284 nm) and off- (304 nm) ozone UV DIAL wavelengths are produced by stimulated Raman scattering in high-pressure nitrogen. The receiver uses the existing astronomical Cassegrain telescope (76 cm in diameter) at the HARSJ. Spectral separation of the backscattered ozone UV DIAL wavelengths is carried out by a polychromator based on an concave imaging diffraction grating. The entire ozone UV DIAL detection unit is integrated into the existing long-range multi-wavelength polychromator box optically coupled at the Cassegrain telescope's rear end. With the current design, the ozone UV DIAL system provides hourly averaged ozone profiles reaching from 6 km to 12 km ASL with a vertical resolution better than 400 m at 6 km ASL and 1000 m at 12 km ASL. The relative statistical error of the profiles is 10% at 12 km ASL. The experimental measurements started in the spring of 2008 but were interrupted several times because of serious laser failures. As a first step of the experiment, the ozone UV DIAL profiles were compared to balloon borne ozone profiles taken by electrochemical concentration cell (ECC) sondes. The sondes were launched by the Swiss Meteorological Institute – Payerne (SMI, 46°49'N, 6°56'E, 491 m ASL) situated 80 km in northwestern direction of the HARSJ. The ozone UV DIAL and the sonde measurements were taken quasi simultaneously. The relative differences between the ozone UV DIAL and the sonde profiles were found to be below ±25% in a horizontally homogeneous atmosphere. The intercomparison has shown that the ozone UV DIAL is capable to accurately reproduce the vertical ozone distribution. Intercomparison with vertical ozone profiles taken in the vicinity of the HARSJ by an airplane-borne UV-photometer confirmed the performance of the ozone UV DIAL. Time series with duration of up to 21 hours were taken in order to study the time evolution of tropospheric events characterized by elevated ozone concentrations. As a result of these measurements three events, demonstrating different processes have been identified. In the first case an air layer with elevated ozone concentration was observed in the altitude range between 6 and 7 km ASL. The lower than 10% relative humidity and the time evolution of this layer allowed to determine its stratospheric origin. The observations are in a good agreement with the EURopean Air Pollution Dispersion (EURAD) model. The model predictions also agree with the elevated ozone concentration measured by the UV-photometer of the National Air Pollution Monitoring Network (NABEL) operated at the HARSJ. The evolution of an ozone-enriched layer in the tropopause region was followed for 21 hours during the second case study. The time evolution, the ECC data, and the EURAD model predictions suggest the possibility of a short filamentation process developing at the tropopause region. The measurements of the third event were made according to a prediction of STE forecasted by a dynamical model developed at the ETHZ. A short living enhancement of the ozone concentration has been observed at an altitude of about 9.5 km ASL. There are two possible explanations for this event. The first one could be a short living reversible STE event. A second explanation is a possible long-range transport of ozone and ozone precursor rich air from the forest fires in the Athens region suggested by the Hybrid Single Particle Lagrangian Integrated Trajectory Model (HYSPLIT) backward trajectories. However, the available information does not allow to make a definite conclusion. The ozone UV DIAL has been built successfully and has shown its ability to observe fast variation in vertical ozone distribution with high accuracy and precision. In combination with additional water vapor and temperature lidar observations and supported by model forecasting, this system could become a powerful tool for the improving of the understanding of upper troposphere and lower stratosphere ozone related processes. However, to allow future systematic observations the ozone UV DIAL will need an essential upgrade to resolve the problems related to the long preparation time, and the limited by technical reasons time of observation.

EPFL2010

Moisture and energy dynamics in fields soils

Francesco Ciocca

Soil moisture is crucial to water-cycle as it affects infiltration, runoff and land-atmosphere interactions. In dry soils the role of water vapor fluxes becomes essential and causes a tight coupling between soil moisture dynamics and heat transfer, which makes measurements and modeling extremely difficult. Since advances in both theory and experimental techniques are required, we have focused on both aspects: we have performed theoretical-numerical investigations of soil moisture and energy dynamics at the land surface; tested novel measurements techniques, in laboratory and at field scale, to provide new insights into soil moisture dynamics and contribute to a comprehensive theory of coupled energy and moisture transport. As the dynamics of physical systems are dictated by conservation equations, it is common practice to estimate quantities that cannot be directly measured (e.g. vapor fluxes) from the residuals of balance equations. In laboratory as well as in field experiments, residuals are calculated from measurements that are sparse in space and discrete in time. We have theoretically and numerically shown how the use of discrete data may lead to large residuals even without measurements or model errors. We have demonstrated that residuals are very sensitive to the distance between the measurements and that they increase in presence of nonlinear processes, such as moisture transport. A careful error analysis is required to assess the reliability of residuals computed from discrete data and avoid misinterpretations of the underlying physical processes. Another crucial element dictating soil moisture dynamics and evaporation into the atmosphere is soil water-retention. Popular parametric models of the retention curve present a matric potential that becomes infinite at residual saturation, which is the water content below which liquid flow stops. Few extensions of the water-retention curves have been proposed, which remain rarely used in practice. We have studied the effects of different models on liquid and vapor transport. We have shown that parameterizations that allow vapor flux below residual saturation yield larger vapor fluxes that result in a drier soil surface, whereas deep soil remains relatively wet. When diurnal forcing is considered, potential evaporation is reduced in the energy-limited regime and enhanced in the moisture-limited regime. We have established that in this case extended water-retention models predict larger heat fluxes that might critically impact models at whole scales (possibly including global circulation models). Discrimination between different theoretical descriptions and parametric models requires experiments that monitor moisture and energy dynamics. This demands the use of accurate sensors and measurement techniques that are able to simultaneously monitor several quantities (e.g. soil moisture, temperature, soil thermal properties, solute concentration in soil water). As in general different probes are used, problems arise due to different footprints and sampled soil volumes. Multi functional heat pulse probes (MFHPPs) have been conceived to overcome these issues by allowing simultaneous measurements of all required variables. We have assessed MFHPPs ability to measure ground heat flux, which has to be accurately known to avoid significant imbalances in the surface energy budget. We have employed an experimental setup able to generate a non-uniform heat-flux field and we have demonstrated that MFHPPs can reliably capture the magnitudes and directions of the fluxes. This allows more complete information on ground heat flux than that obtained by the classic probes (e.g. ground heat flux plates). To monitor soil state in large-scale field applications we have investigated the possibility to infer distributed thermal conductivity and soil moisture profiles with an optical fiber. Information is obtained by monitoring soil response to an active heat pulse generated from the metal shield armoring the optic cable, and analyzing the temperature differences of the cooling phase, monitored each meter by a distributed temperature sensing system (DTS). We have shown that soil moisture measurements inferred from soil thermal response (recorded by the DTS) are in good agreement with independent measurements in wet and intermediately wet soils; however in dry soils moisture content is underestimated and further investigations are required to improve this technique for dry conditions. In this study we have significantly improved the knowledge of coupled energy and moisture transport on both theoretical-numerical and experimental sides. We have proven that reliable measurements can be obtained at different spatiotemporal scales, and provided both prognostic and diagnostic methods to optimize their use. Finally, we have suggested crucial improvements to numerical models which might have implications beyond field-soils dynamics.

EPFL2013

Numerical investigations of the dynamics of the full load vortex rope in a Francis turbine.

François Avellan, Olivier Braun, Andres Müller, Nicolas Ruchonnet, Adrien Taruffi

The draft tube vortex rope occurring in Francis Turbines at full load is a potential cause of severe system instabilities, limiting the operating range of concerned hydro power plants. Studies focusing on physical mechanisms [1], methods for stability assessment based on numerical flow simulation [2, 3], studies on simplified configurations [4] and detailed numerical simulations [5] have been published. The pulsations of the rope volume being the core element of the system instability, the size and shape and dynamic evolution of the vortex rope predicted by numerical methods are essential in the validation of these methods. More experimental results [6, 7] of the study case underlying former numerical simulations [3], allow a review of these cases with improved boundary conditions, based on measurements instead of assumptions. On the other hand, the completed experimental database provides more validation cases to better understand the shortcomings of the numerical simulation techniques and identify the potential for improvement of predication capabilities. Advances in computational resources and optimization of the numerical parameters have allowed completing the range of comparisons of numerical versus experimental results close to the onset of instability. Results are compared in terms of the globally unsteady behavior as well as local pressure and velocity measurements at different locations in the draft tube.