The Hindu-Kush Himalayan region possesses a large resource of snow and ice, which acts as a freshwater reservoir for irrigation, domestic water consumption or hydroelectric power for billions of people in South Asia. Monitoring hydrologic resources in this region is challenging because of the difficulty of installing and maintaining a climate and hydrologic monitoring network, limited transportation and communication infrastructure and difficult access to glaciers. As a result of the high, rugged topographic relief, ground observations in the region are extremely sparse. Reanalysis data offer the potential to compensate for the data scarcity, which is a barrier in hydrological modeling and analysis for improving water resources management. Reanalysis weather data products integrate observations with atmospheric model physics to produce a spatially and temporally complete weather record in the post-satellite era. This dissertation creates an integrated hydrologic modeling system that tests whether streamflow prediction can be improved by taking advantage of the National Aeronautics and Space Administration (NASA) remote sensing and reanalysis weather data products in physically based energy balance snow melt and hydrologic models. This study also enhances the energy balance snowmelt model by adding capability to quantify glacier melt. The novelty of this integrated modeling tool resides in allowing the user to isolate various components of surface water inputs (rainfall, snow and glacier ice melt) in a cost-free, open source graphical-user interface-based system that can be used for government and institutional decision-making. Direct, physically based validation of this system is challenging due to the data scarcity in this region, but, to the extent possible, the model was validated through comparison to observed streamflow and to point measurements at locations in the United States having available data

Earth is characterized by episodes of glaciations and periods of minimal or no ice through geologic time. Using the linear energy balance model (EBM), nonlinear EBM with empirical ice sheet schemes, the general circulation model coupled with an ice sheet model, this study investigates the occurrence and stability of glaciations in geologic time. The simulations since the last glacial maximum (LGM) suggest that the summertime thawline of ice sheets conforms closely to the equatorward edge of the ice sheets and implies the relative stability toward deglaciation. CO2 levels are indispensable in controlling the initiation of ice sheet in the Cretaceous. At low CO2 levels, ice sheets exist in all periods no matter LGM or the last interglacial (LIG) orbital elements; however, at high CO2 levels ice sheets rarely exist. The simulations agree well with recent geological evidence of the hysteresis of glaciations in the Permo-Carboniferous. Gondwanaland reached its glacial maximum when CO2 level was roughly the same or slightly higher than the preindustrial value. With a further increase of CO2, deglaciation dominates and results in an ice free state. Again, if CO2 decreased to the present level, Gondwanaland would be glaciated once more and start a new cycle of glaciation and deglaciation. Simulations from five paleogeography maps in Gondwanaland with a suite of CO2 levels and different orbital elements reveal that paleogeography, CO2 levels and the Milankovitch cycles all contribute to the glaciations of Gondwanaland. This study shows that orbital elements alone are insufficient to account for the evolution of ice sheets. Net radiative forcing caused by greenhouse gases, such as CO2 and solar constant change are the primary drivers to glacial inception or demise. Continental geography, CO2 levels, solar constant change, and the Milankovitch cycles complicate the glacial history of Earth.

This thesis focused on the modelling of a pulping process. The purpose was to see if an accurate model can be crated based on relatively simple premises and if the errors can be identified or analysed. To realise this, the authors conducted a literature study to identify the current state of the art regarding the chemical pulping process. In addition, flow charts and sample data from a case study were examined. Based on the literature review and case study, model assumptions were derived. The model is divided into sixteen components. Where mixing occurs, lumped conditions are assumed. The model has five validation points, four of which are temperatures and a mass flowrate. These are shown as deviations from the measured values. In conclusions, it was the model could produce stable results over a narrow time frame. More so if the transition period at the start of the simulation is overlooked. Several new model assumptions are presented with the purpose to increase accuracy e.g. account for the components ability to store mass.

Rubber granulates, used at artificial grass pitches, pose a threat to the environment when leaking into the nature. As the granulates leak to the environment through rain water and snow clearances, they can be transported by rivers and later on end up in the marine life. Therefore, reducing the snow clearances to its minimum is of importance. If the snow clearance problem is minimized or even eliminated, this will have a positive impact on the surrounding nature. The object of this project is to propose a method for deciding when to remove snow and automate the information dispersing upon clearing or closing a pitch. This includes finding low powered sensors to measure snow depth, find a machine learning model to predict upcoming snow levels and create an application with a clear and easy-to-use interface to present weather information and disperse information to the responsible persons. Controlled experiments is used to find the models and sensors that are suitable to solve this problem. The sensors are tested on a single snow quality, where ultrasonic and infrared sensors are found suitable. However, fabricated tests for newly fallen snow questioned the possibility of measuring snow depth using the ultrasonic sensor in the general case. Random Forest is presented as the machine learning model that predicts future snow levels with the highest accuracy. From a survey, indications is found that the web application fulfills the intended functionalities, with some improvements suggested.