Mineral dust aerosols play an important role in climate and the Earth's energy budget. However, the nature and complexity of dust sources is poorly understood. Traditional techniques used for mapping the Sahara dust sources like the analysis of surface dust observations, back trajectory analysis of isobar data, and mineral tracers all tell a different story regarding the location of dust sources, they only agree that the Bodélé Depression and western Hoggar Mountains are important sources. Remote sensing techniques have more recently been used to identify Saharan dust sources and the different methods provide more agreement about more Saharan dust sources and also identify the Bodélé Depression and the western Hoggar Mountains as important source of dust. While there are several remote sensing techniques that can be used to identify desert dust sources yet no comprehensive comparison has yet been done to evaluate their utility. This thesis has evaluated the utility of nine different methods that can be employed to detect dust using MODIS have been investigated by comparing them to sun-photometer aerosol optical thickness (AOT) measurements at a wavelength of 1020μm from Banizoumbou (Niger). The five established techniques that were evaluated, these were: Ackerman (1989), Miller (2003), Handley (2004), Hansell et al., (2007) and the Deep Blue NASA aerosol product Hsu et al., (2003). Many of these methods employ brightness temperature differences (BTD). To determine the effectiveness of this approach all possible MODIS BTD’s were computed and evaluated, these were: BTD (12 μm-11 μm), BTD (3.7 μm-12 μm), BTD (8.6 μm-11 μm), and BTD (8.6 μm-12 μm). To evaluate the accuracy of these dust indices the correlation between the sun-photometer AOT and the result of each MODIS dust index were determined. The results show that the Deep Blue Algorithm gives the highest correlation (R²= 0.91), however the deep blue product has a 10km spatial resolution and thus is not good at locating dust sources, the ultimate aim of this project, furthermore the cloud mask applied to the product routinely masks out most of the dust. The other methods all have a spatial resolution of 1km and thus are more appropriate for this purpose. Of these methods the Ackerman (1989) shows a high R2 value (R²= 0.71) as do many other methods. The effect of different surface materials on dust detection was evaluate by studying five different backgrounds in order to see how the dust can be distinguished from these backgrounds using the M test. Over Limestone background, Miller (2003) has got the highest M value followed by Deep Blue and Ackerman (1989), excluding Deep Blue due to the poor cloud mask and visually difficult to trace the dust to their sources, Ackerman (1989) comes second and has an M value close to Miller (2003). However, Ackerman (1989) shows the best result visually. Based on the five tested backgrounds result using different techniques (visually, M test, and statistically) Deep Blue, Miller, and Ackerman present reasonable results. Based on these results Ackerman (1989) was selected to detect the dust sources over Cyrenaica (Libya) with using MODIS and SEVIRI data. SEVIRI images are also used to study the meteorology of the dust storms in order to provide more information on the wind direction, cause, and lifetime of dust storm. The highest percentage of the dust storms generated from Cyrenaica are caused by Anticyclone 70%. The majority last for six hours, starting at 8:00 am and ending at 14:00 pm. MODIS is used to detect the location of the dust sources and Landsat and Google Earth images are used to identify the geomorphology of the dust sources. Total of 53 dust sources are detected during 2008, 45% from alluvial fans, 15% from lake, 13% from alluvial plains, 6% from agriculture, 6% from river, 2% from multiple landforms in a single MODIS pixel and thus their nature could not be determined, while 13% were diffuse and no source could be detected. Alluvial fans were the most active sources and almost half of these dust sources are located at one large fan located south east of Benghazi.

Current information about the timing, location and characteristics of African biomass burning is primarily derived from two satellite sensors: the geostationary Spinning Enhanced Visible and Infrared Imager (SEVIRI) and the polar-orbiting Moderate Resolution Imaging Spectroradiometer (MODIS). Due to opposing limitations in their sampling design, however, SEVIRI and MODIS cannot independently detect and characterize all active fires from ignition to extinction. Instead SEVIRI (at 15 min temporal resolution) fails to detect relatively smaller and/or lower intensity fires within its coarser 3 km pixels while MODIS (at 1 km spatial resolution) fails to detect fires between the 3hr and 9hr gaps in overpasses. As part of an ongoing validation exercise, this work compares the SEVIRI and MODIS active fire products across a strong gradient of fire activity in the Central African Republic. The evaluation of SEVIRI’s errors of omission and commission were found to depend on MODIS scan angle, and were strongly affected by the SEVIRI cloud detection strategy and uncertainties in the background thermal characterization. , Despite instantaneous and localized differences in the SEVIRI and MODIS active fire products, the agreement between SEVIRI and MODIS derived maps of fire occurrence and the fire season duration increased as active fire pixels were aggregated into coarser ispatiotemporal grids. Beyond direct comparisons, this thesis also explores the ability to enhance active i detection and characterization across Africa by synthesizing the SEVIRI and 10DIS datasets. The first method developed here adjusts SEVIRI observations at 5.0° rid cell resolution and 15-minute temporal resolution to account for the undetected fire liative power (FRP) associated with SEVIRI’s errors of omission. Conversely, the ond method adjusts MODIS observations at 0.5° grid cell resolution and monthly aral resolution to account for the aforementioned undetected FRP emitted between ses. After assuaging a MODIS imaging artefact referred to as the ’bow-tie :," estimates of the fire radiative energy (FRE) emitted from Africa derived from ! two methods herein agree within 4% of each other and are ~1.46x greater than idously published estimates of FRE derived from MODIS. Nevertheless, continental-lie estimates of fuel consumption based on measurements of FRP alone remain ~ 2.5x I than recent bottom-up modelling approaches.

Measurements of wind speed were made within and above a 16 metre pine forest with sensitive anemometers linked to a computer controlled data acquisition system. The considerable interference of the supporting tower on the airflow was determined. The zero plane displacement, d, was about 12 metres. Profiles above the canopy approximated to a pure logarithmic form with a roughness length of about 1 metre. The applicability of the log-linear and power profile equations was investigated. One minute speed fluctuations showed a progressive decoupling with height with increasing stability. Swaying in strong gusts is thought to cause progressive loosening of the soil-root complex in extreme cases resulting in windthrow. The canopy wind profile followed the theoretical form fairly closely with the canopy top eddy approximating to 0.4 (tree height - d). A mean wind direction shift through the canopy of 270 was measured. The bulge often found in the trunk space wind profile is explained in terms of convective processes; smoke trails confirming the existence of convective plumes.

Entrainment, the dynamical incorporation of free air into the boundary layer, is important because it affects the temperature and moisture profiles of the boundary layer. Weather forecast models require a parameterization of entrainment so as to more accurately predict cloud formation at the top of the boundary layer; a process sensitive to small changes in relative humidity. Using a tethered-balloon based observing platform with multiple turbulence probes, observations of a convective planetary boundary layer were made to investigate the process of entrainment at the boundary layer's interface with the free atmosphere. Observations were taken throughout the boundary layer at a vertical resolution of about 100rn, enabling two-dimensional scalar fields to be constructed. Measurements made in situ within the entrainment zone were used to calculate the entrainment flux. By applying a filtering technique it was discovered that 50% of the entrainment flux could be attributed to processes occurring on scales larger than the boundary layer depth, Zi. This emphasised the importance of large-scale motions, e.g. thermals, to entrainment. At the other end of the scale, 20% of the entrainment flux was due to processes occurring on scales < zi/4. Hence, small-scale local mixing processes are also relevant to entrainment. Conditional sampling analysis showed that within the upper part of the entrainment zone, over 80% of the entrainment flux is carried by up draughts. The observations provide evidence for two previously postulated mechanisms for entrainment. Specific engulfment episodes are presented. These support recently reported Large-Eddy Simulation work that has resolved engulfment's within a model. Also observed is an instance of an upwelling thermal impinging on and distorting the inversion layer. This interaction results in a local reduction in stability across the inversion layer, allowing entrainment to occur as a result of wave-breaking.

The chemical and physical characteristics of aerosols are widely varying and the associated hygroscopic properties have far reaching impacts on our environment and the way we live. Whilst numerous inorganic models have been developed, it is clearly necessary to treat the influence of organics given the upsurge in data which has identified a ubiquitous contribution from such species to aerosol mass loadings. Similarly, the majority of equilibrium models available do not even treat the influence of curvature. In this report a thermodynamic model is presented which was designed to embrace both the chemical and physical characteristics of atmospheric aerosols on a fundamental level (ADDEM - Aerosol Diameter Dependent Model) and calculate the equilibrium composition of an aerosol of known composition and dry size. A direct minimisation of the Gibbs free energy, using a sequential quadratic programming algorithm, allows the calculation of equilibration between the gaseous, solid and aqueous phase components subject to thermodynamic data availability. Subsequently, a general inorganic equilibrium model was developed and found to compare excellently with other state-of-the-art inorganic equilibrium models available. The main crux of the problem of treating mixed inorganic/organic systems stems from not being able to treat the complex interactions taking place in solution between the two fractions. Given this, ADDEM combines two separate thermodynamic models and adds the water content associated with the inorganic and organic fractions. It was found that the additive approach for modelling mixed inorganic/organic systems worked well for a variety of mixtures, as has been found in previous studies for simple binary inorganic/organic systems. Whilst this is likely to remain the only option for some time, further work is still required to assess whether this approach reproduces errors which lie within the experimental uncertainty of observed hygroscopic behaviour for a variety of systems. With regards to the generalised contribution from water soluble organic species in the atmosphere, analysis of fractionated WSOC data from two ambient studies, using the modelling tools developed here, suggested that this portion of the atmospheric aerosol contributes little to the observed water uptake relative to the inorganic fraction. Indeed, whilst further studies are required from a wider range of locations, a combination with observed results that an additive approach for modelling mixed inorganic/organic aerosols works very well suggests that one representative growth factor for the WSOC fraction may be adequate for aged air masses if indeed the inorganic fraction dominates the water uptake behaviour in the sub-saturated humid environment. Whilst simple mixing rules for including organics may be relatively successful at reproducing water uptake characteristics, they rely on data from binary systems that may not be available and thus have to be calculated. This further validates the need for predictive models such as ADDEM. For calculation of non ideality in the organic model, the most widely used group contribution method UNIFAC is employed here. Results suggest that UNIFAC is quite capable of reproducing the water uptake characteristics of various complex species of atmospheric importance in the sub-saturated humid regime. ADDEM currently houses a general equilibrium model within a bisection iterative loop to find a solution to the definition of equilibrium given by the Köhler equation. In doing this, the complexities of including the surface term within the Gibbs energy summation are circumvented allowing any number of surface tension and density mixing rules to be used with ease. Comparisons with another diameter dependent model showed that ADDEM can reproduce diameter dependent phenomena including the onset of deliquescence. Similarly, it is possible to extend ADDEM into the super-saturated humid regime now that the curvature effect is included. ADDEM is subject to various sensitivities derived from the choice of parameters used to describe pure component and mixture properties. Surface tension sensitivity seems to increase as the aerosol dry size decreases and the RH increases. However, larger sensitivities are found for the value of dry density used. It is thus likely that the history of the aerosol studied in a HTDMA, specifically the nature of the drying process that will influence the final crystalline form, will create systematic uncertainties upon comparisons with theoretical predictions.

The Indian Monsoon is an extremely important and large-scale meteorological phenomenon that occurs with monotonous regularity every year in the northern hemisphere summer. India and the surrounding regions depend on the monsoon's rains which, although averaged over a season are relatively constant, can vary on timescales of days to weeks with devastating economic and social impact. This study was performed in an effort to gain a little more insight in to the intraseasonal variability of the Indian Monsoon. This was done with the help of an extensive observational study using ECMWF reanalysis data and also some modelling with a simple atmospheric model. In order to interpret results, degrees of simplification will be introduced by compositing and averaging processes that betray the complexities of the monsoon system. However, it will be shown that this allows well-defined atmospheric structures to be obtained for what are known as active and break states of the monsoon. Some observational aspects of the active and break monsoon states will be shown, highlighting the major differences between the two states. This will be followed by a series of modelling experiments with increasing sophistication, although remaining very simple. The results of these will show enough consistency with observations to allow an investigation in to the effects of ENSO on the intraseasonal variability of the Indian monsoon.

No description available.

No description available.

The propagation of Rossby waves with positive and negative frequency, corresponding to eastward and westward phase speeds respectively, is investigated. The techniques used are theoretical analysis, ray tracing, and initial value problems in barotropic and baroclinic numerical models. It is found that the characteristics of positive and negative frequency Rossby waves can differ significantly from each other andfrom those of stationary, zero frequency Rossby waves. However, general deductions from studies of stationary Rossby waves are still found to be valid. Using an analytic Gill-type model and a dry primitive equation model with only idealised vorticity or thermal forcing, a possible trigger mechanism for the Madden Julian Oscillation (MJO) has been studied. The results show that eastward moving forcing in the subtropics or extratropics can lead to a significant equatorial Kelvin wave response which tends to be a maximum in the African/Indian Ocean sector, and is enhanced by easterly winds in the upper troposphere. It is suggested . that one mechanism for initiating the MJO is for eastward moving extratropical waves to excite a large equatorial response, sufficient to trigger large-scale convection, in the presence of favourable easterly winds in the upper troposphere. The dry primitive equation model is used to study the possible interaction of atmospheric flow in the two hemispheres and the triggering of other equatorial waves. It is found that stationary and westward moving forcing in the Northern Hemisphere extratropics can give a significant Southern Hemisphere response. A westward moving forcing in the subtropics, with a period of several days, can trigger the equatorial mixed Rossby-gravity and n=l Rossby waves. The zonal basic flow is found to have a significant effect on these equatorial wave responses.

In this thesis two sensitive absorption techniques are used for the detection of trace concentrations of two different nitrogen oxides, which are of interest for atmospheric chemistry. These techniques are both based upon the interaction of mid-infrared laser radiation, provided by Quantum Cascade Lasers, with the molecules being studied. The Optical Feedback-Cavity Enhanced Absorption spectroscopy (OF-CEAS) technique combines the path length benefits of an optical cavity with injection locking of the Quantum Cascade Laser with light returned from the optical cavity. This optical feedback creates a comb of intense modes with spacing dependent on the length of the cavity whose intensity variations can be used to retrieve an absorption spectrum of an analyte gas. This thesis provides a novel method for the improvement of the spectral resolution of the spectrometer: by mounting one of the cavity mirrors upon a moveable stage the cavity mode comb can be translated. By interleaving a series of these spectra a higher resolution spectrum can be created . This is shown to give an improvement in the precision of both the concentration and linewidth retrieved from OF-CEAS measurements. The spectrometer showed the capability of detecting nitrous oxide at concentrations found in ambient air and with pressures as low as 10 Torr. The second technique is Faraday Modulation Spectroscopy. This technique which uses the magnetic splitting of ro-vibrational transitions of paramagnetic molecules to produce a sensitive . absorption technique. The magnetic field is modulated at a frequency much greater than that being used for the laser frequency sweeping and the signal is detected at multiples of the modulation frequency. This spectrometer was used for the detection of nitric oxide and showed an improvement in sensitivity relative to direct absorption measurements.