Small reservoirs have become increasingly widespread across semi-arid regions, due to their ability to reduce transport of eroded soil and harvest scarce and unreliable rainfall for local users. The scale and geographical dispersion of these multiple hydro-social systems restrict their investigation, leading to difficulties in assessing their agricultural potential, their cumulative influence on runoff, and in identifying strategies to support riparian farmers. This research sought to develop a multi-scalar interdisciplinary approach to assess water availability across multiple small reservoirs and understand hydrological and wider drivers’ influence on associated agricultural practices. An Ensemble Kalman Filter approach was developed to combine 30 m Landsat flooded surface area observations with a daily hydrological (GR4J + water balance) model on 7 gauged reservoirs. Data assimilation, providing near-real time corrections, reduced runoff uncertainties generated by highly variable and localised rainfall intensities and lowered daily volume root mean square errors (RMSE) by 50% compared to the initial rainfall-runoff model forecast. Compensating for Landsat’s reduced temporal resolution and correcting outliers, the method correctly reproduced flood dynamics of 5 ha lakes (R²=0.9). Validated against extensive field data over 1999-2014, the method notably establishes Landsat imagery’s ability to assess annual water availability on ungauged reservoirs as small as 1 ha (RMSE circa 25%). Applied to 48 small reservoirs and 546 Landsat 5-8 images, the treatment chain identified the significant water scarcity and unreliability that impedes agricultural development on 80% of lakes in the Merguellil upper catchment (Central Tunisia). In parallel, rapid surveys, quantitative questionnaires and semi-directed interviews confirmed minimal withdrawals, yet highlighted the diversification of practices and the peripheral benefits accompanying small reservoir development. Many farmers lack the capabilities to increase their withdrawals and suffer physical and economic water access difficulties, mismanagement, compounded through limited and short-term government assistance. Individual successes resulted from farmers’ economic resilience and means to secure alternate water supplies during dry spells. Faced with limited storage capacities and prolonged droughts, small reservoirs must in this climatic context retain their supplementary irrigation focus and not strive to support widespread intensification of irrigated practices.

Cirrus are thin upper tropospheric clouds that are composed entirely of ice particles. Due to their extensive coverage of the Earth’s surface, they have a significant impact on the Earth’s radiative balance. In the absence of a foreign surface, cirrus ice particles can form via homogeneous ice nucleation in liquid aqueous solution aerosol droplets. Ice can also nucleate heterogeneously when catalysed by the presence of solid particles, which can affect the physical properties of the cloud, potentially resulting in high in-cloud humidity and low optical depth. In order to better quantify the impact of cirrus on future climate it is vital to understand the microphysical processes that lead to their formation. This project focuses on the investigation of heterogeneous ice nucleation in the deposition mode under cirrus conditions. It can be divided into two sections: 1) The investigation of the ice nucleation behaviour of glassy aerosols. Atmospheric secondary organic aerosol (SOA) is likely to exist in a semi-solid or glassy state, particularly at low temperatures and humidities. Over the course of two experimental campaigns at the AIDA (Aerosol Interactions and Dynamics in the Atmosphere) cloud simulation chamber, the ice nucleation properties of aerosols of five different glass forming compositions (citric acid, raffinose, 4-hydroxy-3-methoxy-mandelic acid (HMMA), levoglucosan, and a mixture – raffinose/M5AS) were studied. These organic compounds have similar functionality to oxidised organic material found in atmospheric aerosol and have estimated temperature/humidity induced glass transition thresholds that fall within the range predicted for atmospheric SOA. A small fraction of aerosol particles of all compositions were found to nucleate ice heterogeneously in the deposition mode at temperatures < 200 K, which is relevant to the tropical tropopause layer (TTL). Raffinose and HMMA, which form glasses at higher temperatures, nucleated ice heterogeneously at temperatures as high as 214.6 and 218.5 K respectively. Using a 1D cirrus model, it was shown that nucleation on glassy aerosols may explain low ice crystal numbers and high in-cloud humidity in the TTL. It was also found that the ice nucleation efficiency of glassy aerosol particles became enhanced after having been frozen homogeneously at temperatures close to their glass transition thresholds. 2) The design and development of a benchtop instrument for the study of deposition mode ice nucleation. After characterisation, the deposition mode chamber was used to investigate ice nucleation by a natural sample of kaolinite clay mineral dust (KGa-1b, Clay Minerals Society). The surface area of kaolinite present was found to affect the onset humidity of ice nucleation. Evidence was also found for the enhancement of kaolinite particles as ice nuclei after they had nucleated ice in an initial experiment.

Mixed-phase Arctic stratocumulus clouds are ubiquitous to the region during the summer months. However, despite their prevalence, very little is known about the processes which maintain the cloud. Recent observations have shown that Arctic stratocumulus commonly extend into the temperature inversion which caps the Arctic boundary layer. This is atypical to sub-tropical stratocumulus where the cloud top is found in the vicinity of the inversion base, and unexpected as strong longwave radiative cooling would be expected to keep the cloud top and inversion base heights in equilibrium. Uniquely to the Arctic, inversions in speci�c humidity are also commonly observed coincident with temperature inversions, and this is thought to contribute to the clouds' subsistence in the strongly stable inversion layer. In this thesis, observations from the Arctic Summer Cloud Ocean Study (ASCOS) are used to characterize the lower Arctic atmosphere and provide the basis for simulations of stratocumulus cloud encroachment into the Arctic temperature inversion. Observations show that cloud extending into the inversion by more than 100 m was a common occurrence during ASCOS, which is consistent with measurements made during previous summer field campaigns. Simulations made with the Met Office Large Eddy Model (LEM) were used to model the encroachment, and results suggest that the depth of encroachment has a high correlation with the humidity inversion strength. A number of different cloud-inversion regimes were identi�ed from the model simulations. When specific humidity fell of inside the temperature inversion, the high relative humidity of the region just above the inversion base was found to allow encroachment of cloud up to 40 m into the inversion layer. While in the presence of a speci�c humidity inversion the encroachment was larger reaching a maximum of 200 m. The presence of specific humidity inversions and their relationship to the encroaching cloud was determined to be self-sustaining, and the cloud found to remain at a quasi-stable depth for as long as a moisture source is available to replenish the loss of water from ice precipitation. However, encroachment of cloud into the inversion was shown to cause a signi�cant reduction in the buoyant production of TKE at cloud top, which led to turbulence shutting off completely in the clouds with the largest encroachment depth. This caused a thermal adjustment of the inversion layer to the cloud which led a reduction in the encroachment depth. The overall impact of encroachment on boundary layer turbulence was found to be significant, with TKE reduced by up to 90% in the simulations with the largest encroachment depth.

This thesis describes spectrally-resolved radiance measurements, recorded using the Imperial College-built Tropospheric Airborne Fourier Transform Spectrometer (TAFTS), to validate the current parameterisation of the far-infrared foreign- broadened continuum of atmospheric water vapour in line-by-line radiative transfer models. The current parameterisation is not based on measurement through the centre of the far-IR spectral region, but is constructed by interpolation from measurements either side of the band. Accurate knowledge of this parameterisation is essential as the far-IR contribution to the planetary outgoing longwave radiation (OLR) is significant. The research takes the form of a radiative closure study, comparing the radiance measurements with simulated radiances modelled using the measured atmospheric state (temperature, humidity) profile as input. The profile information was recorded, alongside the radiance measurements, aboard the UK Met Office / FAAM (Facility for Airborne Atmospheric Measurements) BAe-146 research aircraft, during the first of two field campaigns as part of the CAVIAR (Continuum Absorption in the Visible and Infrared and its Atmospheric Relevance) project. This thesis provides an overview of the relevant atmospheric physics and Fourier Transform Spectroscopy, including a description of TAFTS; the construction of the atmospheric profiles from the various in-situ, radiosonde and model datasets; the calibration of the radiances measured by TAFTS and derivation of the foreign- broadened water vapour continuum between 114−540 cm−1 within the framework of the line-by-line radiative transfer model (LBLRTM v12.0). The resulting continuum is compared to other relevant recent studies. The measurements from this study suggest an increase in continuum strength up to 11% at 238.6 cm−1, decreasing steadily to a reduction of 55% at 114 cm−1 and 54% at 513 cm−1. However, the uncertainties in these measurements are considerable, and might equally suggest an increase in continuum strength.

No description available.

The scattering functions for oblate spheroidal raindrops of various eccentricities are evaluated by a perturbation method. The resulting attenuation and phase rotation due to rain-filled media is computed for frequencies ranging from 3 to 36 GHz, using empirical data on drop size distribution. The attenuation and phase variations are calculated for polarisations along the major and the minor axis of the raindrops. The difference in the attenuation and the phase between the two polarisations is used to evaluate the cross-polarisation level in the case of linear and circular polarisation. The attenuation and phase rotation is also considered for the case of slant path propagation. Comparison is made between the theory and experimental data obtained at 11 GHz (Bradford University link), and at other frequencies. Bearing in mind the complexity of the media, the theoretical results are found to be in a reasonable agreement with the experimental data. In addition it is seen from the theory that the linear orthogonal mode of propagation will give superior isolation to that of circular polarisation.

No description available.

Rainfall passing over a given area is a highly dynamic process; it changes constantly in form and intensity. It varies constantly on short spatial and temporal scales that makes real time measurements of the amount of rainfall challenging. Measuring and sensing rain is important to be able to understand and control our urban environment. Traditionally, rainfall analysis for hydrologic modelling use spatial measurements collected at various sparsely spread observation points using rain gauges working on various principles such as weighing type, tipping bucket, capacitive type etc. An accurate representation of spatial model of rainfall is essential for hydrological operational purposes such as forecasting of river flow, flood irrigation planning and modelling of catchment areas. Measurement of drop size distribution are also exploited to investigate microphysics of precipitation and to improve rainfall remote sensing estimation techniques. However, the high initial costs of convectional rain gauges prevent collection of data with high spatial resolution. The research looks at investigating the sensor stack to be a part of an integrated sensor approach to develop a device architecture for the development of low cost integrated rain sensing and measuring the rain. The device architecture consists of three main stacks – energy generation layer, sensing layer, processing layer. The raindrop on impact causes vibration on the device surface. This force exerted by the raindrop causes a deflection and is measured indirectly by the use of a thin film piezo sensor. As part of the work, we find there is a good correlation between the vibrations caused and the size or volume of the raindrop by indirectly measuring the impact force of the raindrop. The working range of the device is between 100hz and 2000hz, which includes the first modal peak of the impact that acts as an amplification to the drop's impact. Using this information, the device is able to calculate the raindrop size distribution and the rain intensity. Calibration of the device is key as we are measuring the impact force of the rain drops and correlating it to the size of the drop. Primary, not all rain drops will fall on the device at terminal velocity (the main assumption for calibration of the device), as the fall velocity of the droplet may also be affected by the wind. Secondly, the spatial variation of the frequency response function in Volts/Newton in decreasing order from the centre of the plate.

This thesis defines a physically based model describing the kinetic energy of throughfall from any vegetation canopy. Empirical measurements of the drop-size distribution of rainfall and sub-canopy throughfall were used to develop the model which was tested in the context of splash erosion. Comparisons are made for individual storms between rain falling in the open and through a canopy. Three canopies were used, one oak and two tropical rain forest differing in height. Through each storm raindrop sizes were frequently measured using the paper-staining technique. Kinetic energy/mm/m2 was calculated from the drop sizes, their velocities and amount of rain or throughfall. The velocities were assumed to depend on the height of fall. In the rain forest sites splash cups surrounded by uniform areas of sand were used to measure the material splashed. The oak canopy data was used to examine the validity of a working hypothesis relating qualitatively the size of throughfall drops to the saturation of the canopy. It was confirmed that the canopy changed the drop-size distribution of the rain and consequently changed the kinetic energy/mm/m2. The sequences of drop-size distribution change proposed by the hypothesis were related to the cumulated canopy storage. The tropical rain forest results confirmed these findings and extended them. Although rainfall kinetic energy/mm/m2 may be predicted from rainfall intensity, throughfall kinetic energy/mm/m2 was independent of intensity and the frequency distribution of the energy of throughfall samples was bimodal, with a high energy group which was commonly higher than that of the rainfall. The probability of a thoroughfall sample being in either energy group depended on the cumulated canopy storage or the percentage storm duration elapsed. The relative magnitude of rainfall and throughfall total kinetic energy depended on the saturation of the canopy and on the canopy height and for some storms the throughfall energy was higher than the rainfall. Soil splash increased with increasing kinetic energy. The model predicting throughfall energy requires inputs of canopy height, rainfall intensity and the frequency distribution of energy of discrete samples of throughfall. The model is most sensitive to canopy height.

There is growing interest in quantifying the impact of climate change on extreme hydrologic events where failing to integrate the effect of climate change in rainfall estimation will underestimate the severity of the events and the adequacy of current hydraulic structure designs. The purpose of this study aims is to assess the rainfall trend and frequency analysis with impact from climate change in Peninsular Malaysia using statistical methods. The thesis consists of two sections, where the statistics of rainfall trend are assessed by Mann-Kendall (MK) test and non-stationary tests while the frequency analysis illustrates the changes in distribution functions that fit full series and sub-series of annual maximum rainfall. The study area is delineated into five regions according to their distance to the nearest coast (the different extents of the influence of monsoon to the study area) to examine the spatial characteristic of the rainfall series. The MK test has detected changes for each delineated region during different monsoon seasons. At the same time, the result of non-stationary tests reveal that changes in rainfall trend have developed around year 1995 in most of the stations (41% to 50% annual rainfall over the west coast regions; more than 50% of the short duration annual maximum rainfall in the central west region have shown non-stationarity). Among the regions, the short duration rainfall in central west region show most significant increasing trend by both the MK test and the non-stationary tests. Thus, year 1995 served as trend change-point to split full series data into two sub-series data and frequency analyses are performed on these data sets. From the outcomes of the frequency analysis using two sub-series data sets, the estimated quantiles from most of the regions have increased when the sub-series posterior to 1995 is used compared to full series data, implying an overall upward rainfall trend. The results also indicate that the combination of Generalised Extreme Value distribution function and L-moments for parameters estimation (GEV-LM) outperforms the other choices. The GEV-LM is able to fit well to all regions for short-duration rainfall and three regions for long-duration rainfall. This study demonstrates the importance of incorporating climate change in rainfall assessment. There are two-fold implications of this study. First, there is considerable variability of rainfall patterns due to climate change and hence, it is important to divide the study area into regions based on the results of the MK trend and non-stationary tests. Then, the best fitted distribution function and parameter estimation method combination for frequency analysis should be tested for every region. Second, it is important to appreciate the non-stationarity of rainfall series due to climate change and the impact on how frequency analysis shall be carried out. As the warming trends in Peninsular Malaysia started around year 1995, rainfall series have shown significance upward trend, while the results of the frequency analysis (estimated quantiles) reflects the changes in the rainfall characteristics as well. Hence, in this case, it is important to concern the non-stationarity in data to achieve better estimation performance using frequency analysis.