This dissertation reports on three different yet related topics in light scattering computation, radiative transfer simulation, and remote sensing implementation, regarding the cloud properties and the retrieval of cloud properties from satellite-based infrared radiometric measurements. First, the errors associated with the use of circular cylinders as surrogates for hexagonal columns in computing the optical properties of pristine ice crystals at infrared (8-12 µm) wavelengths are investigated. It is found that the differences between the results for circular cylinders and hexagonal columns are on the order of a few percent at infrared wavelengths. Second, investigated in this dissertation are the outgoing broadband longwave and window channel radiances at the top-of-atmosphere under clear-sky conditions on the basis of the data acquired by the Cloud and the Earth's Radiant Energy System (CERES) instrument onboard the NASA Terra satellite platform. Based on the comparison of the observed broadband radiances with those obtained from rigorous radiative transfer simulations, it is found that the theoretical results tend to be larger than their measured counterparts. Extensive sensitivity studies regarding the uncertainties of various parameters were carried out. Within the considered uncertainties of various factors, the computed radiances are still larger than the observed radiances if thin cirrus clouds are excluded. Thus, a potential cause for the differences could be associated with the presence of thin cirrus clouds whose visible optical thickness is smaller than approximately 0.3. Third, presented in this dissertation is an illustration of the application of hyperspectral infrared channel observations to the retrieval of the cloud properties. Specifically, the hyperspectral measurements acquired from the Atmospheric Infrared Sounder (AIRS) aboard the NASA Aqua platform are used to infer cloud top pressure, effective cloud amount, cloud thermodynamic phase, cloud optical thickness, and the effective size of cloud particles. The AIRS-based retrievals are compared with the counterparts of the operational cloud products derived from the Moderate Resolution Imaging Spectroradiometer (MODIS). The two retrievals agree reasonably well except for the retrieved cloud effective particle size. Furthermore, the diurnal and seasonal contrasts of cloud properties are also investigated on the basis of the cloud properties retrieved from the AIRS data.

Ice particles in the atmosphere affect for example remote sensing measurements made for climate and atmospheric research. When these ice particles melt their single scattering properties alters, which in turn affect the remote sensing measurements. It is therefore of importance to understand how these scattering properties change due to the melting of the particles to be able to interpret the measurements.  The ARTS database contains single scattering data for a large set of ice particles in the microwave region, data which describes how electromagnetic radiation interacts with said particles. The database does not however contain any data for melting ice particles. A software package called RimeCraft has been used to simulate the melting process of ice particles, together with the discrete dipole approximation program ADDA, which simulated their single scattering properties at frequencies between 1 and 247.2 GHz. Neither of the two programs had been extensively tested on melting ice particles. This thesis therefore tests both RimeCraft and ADDA for their suitability and performance on melting ice particles as a preparation for extending the ARTS database with these types of particles.  The simulation results from ADDA showed that the single scattering properties, such as absorption and extinction cross sections, were greatly affected by the melting process, especially during the initial phase of melting (below meltfraction 0.1) where the increase was strong. At higher meltfractions the increase was slower, and sometimes even decreased for certain particles. Some unexpected results were seen, such as spikes and oscillations in the extinction and absorption cross sections.  Both RimeCraft and ADDA are suitable for generating models of melting ice particles, respectively to simulate their single scattering properties, as long as the output from ADDA undergoes quality controls first. / Ispartiklar i atmosfären påverkar fjärranalysmätningar gjorda för bland annat klimat- och atmosfärsforskning. När dessa ispartiklar smälter påverkar det deras spridningsegenskaper, vilket i sin tur påverkar fjärranalysmätningarna. Det är därför viktigt att förstå hur dessa spridningsegenskaper påverkas av att partikeln smälter för att kunna tolka mätningarna.  ARTS databasen innehåller spridningsdata för ett stort antal olika ispartiklar i mikrovågsområdet, data som beskriver hur elektromagnetisk strålning påverkas av dessa partiklar. Databasen innehåller dock inte någon data för smältande ispartiklar.   Ett mjukvarupaket kallat RimeCraft har använts för att simulera smältprocessen av ispartiklar, tillsammans med diskret dipolsapproximationsprogrammet ADDA som användes för att simulera deras spridningsegenskaper för frekvenser mellan 1 och 247.2 GHz. Inget av de två programmen har blivit utförligt testat med smältande ispartiklar. Detta examensarbete testar därför både RimeCrafts och ADDAs lämplighet och prestationsförmåga med smältande ispartiklar som en förberedelse för att utöka ARTS databasen med dessa partiklar.  Simulationsresultaten från ADDA visade att spridningsegenskaperna, så som absorptions- och extinktionstvärsnitt, påverkades starkt av smältprocessen, speciellt under den initiala fasen av smältningen. Vid högre smältgrader var ökningen långsammare och ibland till och med negativ för vissa partiklar. Några oväntade resultat sågs, som t.ex. spikar och oscillationer i extinktions- och absorptionstvärsnitten.  Både RimeCraft och ADDA är lämpade för att generera modeller av smältande ispartiklar, respektive för att simulera deras spridningsegenskaper, så länge utdatan från ADDA genomgår en kvalitetskontroll.