No description available.

The variation of solar energy entering Earth’s atmosphere, the solar irradiance, is an important influence on the Earth’s climate. Total output of the Sun varies by 0.1% over an 11-year solar cycle and on longer, secular scales there still remains uncertainty. Global temperatures on Earth have increased over the last 160 years along with increasing anthropogenic impact on the environment. It is, therefore, important to fully understand how much of this increase can be attributed to changes in the Sun. This thesis makes use of a semi-empirical version of the Spectral And Total Irradiance REconstruction model, or SATIRE-S, to reconstruct both total and spectral irradiance variations. The physical basis of SATIRE-S is that all variations in solar irradiance are caused by changes in surface magnetic flux emergence. In this thesis, SATIRE-S is updated to accommodate the input of full-disk continuum image and magnetogram data from ground-based and satellite sources spanning three full solar cycles over the period 1974-2009. These changes are described in detail. The combination of terrestrial and space-based images allows for a reconstruction that, for the first time, has been produced independently to, and therefore provides an unbiased comparison of, the composites of direct radiometric observations of total solar irradiance (TSI), which began in 1978. The excellent agreement with, in particular, the PMOD composite supports the simple model assumptions. It also provides constraints on the influence of other mechanisms that may affect solar irradiance on these timescales. The reconstruction of TSI over three solar cycles provides the opportunity to produce a consistent long-term spectral solar irradiance (SSI) dataset that can be put to use by the climate and atmospheric physics communities. This has been one of the goals of this work. In this thesis, comparisons are made with the new SORCE/SIM broadband spectral observations, which include for the first time the visible and infrared above 400 nm. During the declining phase of the recent solar cycle it is found that there is almost no agreement between the model and instrument on a long-term, cycle-length period, but very good agreement on short-term rotational variation. The influence of the ultra-violet region below 400 nm is important in Earth-based stratospheric chemistry and is a strong influence on both temperature and ozone concentration. Therefore, the SATIRE-S spectral dataset is employed in a simple 2D atmospheric model to evaluate its sensitivity in light of recent spectral irradiance changes suggested by data from the SORCE satellite.

This research is mainly focussed on solar radiation in the UK. It can be divided into four main areas; evaluation of models, analysis of the relations between temperature and solar radiation, critical analysis of the projected future data for the UK and the improvement to the UKCP09 Weather Generator (WG). From the evaluations of models carried out, the Liu-Jordan model performs well for estimating the average hourly global and diffuse radiation. At the individual hourly level however, a number of problems were observed. Regarding clear- sky radiation models, for semi-arid climatic conditions Page model was found to be suitable and for humid climates Yang model is recommended. As for all-sky radiation models, the MRM and Yang model were selected. For the UK, the MRM was found to perform better than the Yang model. Furthermore, a study was carried out to analyse the relationship between temperature and solar radiation. The development of temperature-based mathematical models to obtain mean-daily irradiation was established. A procedure to decompose daily to hourly temperatures was evaluated with respect to world-wide locations and its performance found to be satisfactory. As part of the UKCP09/COPSE project, detailed analysis on the future projected data was carried out to critically evaluate sol-air temperature and the likely change that may occur in the key climatic variables, i.e. temperature, sunshine duration and solar irradiation. Drastic increase of sol-air temperatures and shifting trend of daylight illuminance were found. Furthermore, a sensitivity test was also carried out to analyse the effects of each input variables on sol-air temperature. As a result of the present investigations and communications with the UKCP personnel a new version of WG was released with appropriate modifications. A comparison of the now old- and new WG data sets has been made. Improvements in ratio of diffuse to global radiation and sunshine datasets were found.