Biological functions that use amino acids (AA) are limited by AA supply. This concept was likened to staves in a barrel, where the shortest stave determines the barrel's ability to hold water (Mitchell and Block, 1946). Inaccuracies in models that predict nutrient supply and requirements of dairy cows result in inefficient feeding, as under-prediction of requirements results in deficiency, and over-prediction results in excess. To avoid limitations in production due to AA deficiencies, protein is fed in quantities that likely exceed requirements. Overfeeding of AA results in increased expenses for producers and increased N excretion to the environment, providing economic and environmental incentives to increase N-efficiency. Work presented in the following chapters evaluated the impact of AA supplementation on milk production in dairy cattle, and evaluated the PREP10 and NRC (2001) nutrient requirement model predictions. In two feeding trials (Chapter 2 and Chapter 3), low protein diets did not result in decreased milk production, indicating that protein requirements were overestimated. Although supplementation of AA did not increase milk production, low protein diets resulted in greater N-efficiency, especially when supplemented with Histidine. Evaluation of the PREP10 and NRC (2001) models (Chapter 4) used production data from the literature to identify deficiencies in prediction equations, and found that correction of model bias would considerably reduce prediction errors. Model inaccuracies affect the inefficiency of dairy cow feeding, and must be evaluated to improve feed efficiency. Such improvements could act synergistically with AA supplementation to more closely match nutrient supply to requirements. / Master of Science / Biological functions that use amino acids (AA) are limited by AA supply. This concept was likened to staves in a barrel, where the shortest stave determines the barrel’s ability to hold water (Mitchell and Block, 1946). Inaccuracies in models that predict nutrient supply and requirements of dairy cows result in inefficient feeding, as under-prediction of requirements results in deficiency, and over-prediction results in excess. To avoid limitations in production due to AA deficiencies, protein is fed in quantities that likely exceed requirements. Overfeeding of AA results in increased expenses for producers and increased N excretion to the environment, providing economic and environmental incentives to increase N-efficiency. Work presented in the following chapters evaluated the impact of AA supplementation on milk production in dairy cattle, and evaluated the PREP10 and NRC (2001) nutrient requirement model predictions. In two feeding trials (Chapter 2 and Chapter 3), low protein diets did not result in decreased milk production, indicating that protein requirements were overestimated. Although supplementation of AA did not increase milk production, low protein diets resulted in greater N-efficiency, especially when supplemented with Histidine. Evaluation of the PREP10 and NRC (2001) models (Chapter 4) used production data from the literature to identify deficiencies in prediction equations, and found that correction of model bias would considerably reduce prediction errors. Model inaccuracies affect the inefficiency of dairy cow feeding, and must be evaluated to improve feed efficiency. Such improvements could act synergistically with AA supplementation to more closely match nutrient supply to requirements.

In this study a new wind gust estimate (WGE) method proposed by Brasseur (2001) is implemented in a limited area climate model (RCA, Rossby Centre regional Atmospheric model). The WGE method assumes that wind gusts develop when air parcels higher up in the boundary layer deflect down to the surface by turbulent eddies. The method also gives an interval of possible gusts by estimating an upper and lower bound of a bounding interval. Two separate storms (December 3-4, 1999 and January 8-9, 2005) and a three month period (November 1, 2004 - January 31, 2005) are simulated with RCA. The results are compared to direct observations and to gridded analysis (MESAN). The result is highly dependent on how well the meteorological fields are represented in RCA. Since the storm of December 1999 was not well captured by RCA, the wind gusts were consequently not correctly estimated. The storm of January 2005 was well captured by the RCA and the wind gusts relatively well described. Both the storm of January 2005 and the simulation over a three month period give rather good estimated gusts over sea areas, while over land there is an obvious overestimation of the calculated gusts. A correction to the estimated gust is necessary in order to make the parameterisation useful. Such a correction is tested in this study. It shows significant improvement over most land areas and also gives a certain underestimation in other areas. / Sammanfattning av ”Beräkning av byiga vindar i RCA” En ny metod (WGE-metoden) för att bestämma byvindar har i den här studien implementerats i en regional klimatmodell (RCA, Rossby Centre regional Atmospheric model). WGE-metoden utgår från att vindbyar genereras när luftpaket högre upp i gränsskiktet förs ner till marken av stora turbulenta virvlar. Ett intervall av möjliga byvindar erhålls genom att en övre och undre gräns för detta intervall beräknas. Två stormar (3-4 december 1999 och 8-9 januari 2005) och en tremånaders period (1 november 2004 – 31 januari 2005) har simulerats, och resultaten har jämförts med mätdata och MESAN. Resultatet är till stor del beroende av hur väl de meteorologiska fälten representeras av RCA. Stormen i december 1999 simulerades dåligt av RCA, vilket medförde att byvinden inte heller simulerades korrekt. Både stormen januari 2005 och simuleringen över tre månader ger en tämligen korrekt byvind över hav, samtidigt som man över land får kraftiga överskattningar av den beräknade byvinden. För att byvind-parametriseringen ska vara användbar krävs korrigeringar för att komma till rätta med överskattningen över land. En korrigering testades i den här studien med resultatet att ett förbättrat resultat över land samtidigt leder till en viss underskattning av byvinden i andra områden.

The aim of this study is to investigate how well the wind stress is described in the regional climate model RCA (Rossby Centre Atmospheric model). The model data is compared with measurements taken at Östergarnsholm, an island located east of Gotland in the Baltic Sea. The investigation covers the period October 1997 to September 2000. Compared with the measurements RCA underestimates the wind speed over ocean in most cases. When the wind speed is studied in monthly mean it was found that the wind speed is consistently underestimated in the model, especially during summer and early winter. The wind stress is also underestimated during the year except during late winter and in October when it is overestimated. When wind speed from the model and measurements were studied it was found that lower wind speed is overestimated in the model. This contributes to overestimation of the wind stress for lower wind speeds. For higher wind speeds both wind speed and wind stress is underestimated in RCA. If the model would have a correct wind speed the wind stress would be overestimated with 20-30 % for all wind speeds compared to the measured. This is probably due to miscalculations of wind stress in the model. RCA fail to capture the stability in a satisfactory way. The model describes most of the times a different stratification than the measurements do. Separate studies over modeled and measured wind speed and wind stress was made. RCA is found to represent stable condition well. During unstable stratification modeled wind speed and wind stress are underestimated. Investigation of the measured wave age shows that the model underestimates both wind speed and wind stress for growing sea and mature sea. For the situations when the measured wave age was defined as swell the model overestimated wind speed and wind stress.2 / Den här studien tar upp hur väl friktionen över havsytan beskrivs i den regionala klimatmodellen RCA (Rossby Centre Atmospheric model). Modelldata jämförs med mätningar gjorda på Östergarnsholm, en ö belägen öster om Gotland i Östersjön. Studien spänner över tidsperioden oktober 1997 till september 2000. RCA underskattar i många fall vindhastigheten på 10 m över havsytan jämfört med mätningarna. Då månadsmedelvärden studeras visar det sig att modellen genomgående underskattar vindhastigheten, särskilt på sommaren och på vintern. Friktionen överskattas under sen vinter och för oktober månad. För alla andra månader underskattas friktionen. Studier över vindhastigheterna visar på att de lägre vindhastigheterna överskattas i modellen. Detta bidrar till överskattning av friktionen för de lägre vindhastigheterna. För högre vindhastigheter underskattar RCA vindhastigheterna och därmed även friktionen. Om modellen skulle ge en korrekt vindhastighet skulle resultatet istället visa på en överskattning av friktionen med 20-30 % för alla vindhastigheter. Detta beror troligen på felaktigheter i beräkningen av friktionen i modellen. Stabiliteten visar sig vara dåligt beskriven i RCA. I de flesta fall är skiktningen i modellen och mätningarna helt olika. Jämförelser mellan separata studier mellan modelldata och mätningar visar att vindhastigheten och friktionen beskrivs bra under stabila förhållanden, men i samband med instabil skiktning underskattas vindhastigheten och friktionen i modellen. Studier över den uppmätta vågåldern visar att modellen underskattar både vindhastigheten och friktionen för växande och fullvuxen sjö. Då det råder dyning överskattar modellen både vindhastigheten och friktionen.

This thesis is analyzing customers of Geocart CZ a.s. company. Thesis is presenting new models of evaluating of customers. Weight mathematic model is created in MS Excel software. In software fuzzyTECH and MATLAB are designed fuzzy models for evaluating of customers with application of fuzzy logic. This method of evaluating should eliminate risk of mistakes with choosing of customers and minimize opportunity costs.

This thesis presents a proof of concept study for the better integration of the electrophysiological and modelling aspects of neuroscience. Members of these two sub-disciplines collaborate regularly, but due to differing resource requirements, and largely incompatible spheres of knowledge, cooperation is often impeded by miscommunication and delays. To reduce the model design time, and provide a platform for more efficient experimental analysis, a rapid iterative model design method is proposed. The main achievement of this work is the development of a rapid model evaluation method based on parameter estimation, utilising a combination of evolutionary algorithms (EAs) and graphics processing unit (GPU) hardware acceleration. This method is the primary force behind the better integration of modelling and laboratorybased electrophysiology, as it provides a generic model evaluation method that does not require prior knowledge of model structure, or expertise in modelling, mathematics, or computer science. If combined with a suitable intuitive and user targeted graphical user interface, the ideas presented in this thesis could be developed into a suite of tools that would enable new forms of experimentation to be performed. The latter part of this thesis investigates the use of excitability-based models as the basis of an iterative design method. They were found to be computationally and structurally simple, easily extensible, and able to reproduce a wide range of neural behaviours whilst still faithfully representing underlying cellular mechanisms. A case study was performed to assess the iterative design process, through the implementation of an excitability-based model. The model was extended iteratively, using the rapid model evaluation method, to represent a vasopressin releasing neuron. Not only was the model implemented successfully, but it was able to suggest the existence of other more subtle cell mechanisms, in addition to highlighting potential failings in previous implementations of the class of neuron.

Known problems with the air pollution model, TAPM’s meteorological predictions over the Canterbury Plains, New Zealand, has been its lack of ability to simulate low nocturnal stagnant wind speeds, certain wind field directions as well as greatly overestimating the sensible heat flux. The aim of this thesis has been to reduce these known disagreements along with other meteorological parameters in attempt to optimize TAPM’s predictions for New Zealand conditions. The methodology used to obtain better agreement with observational data during a four-day period (1-4 of August 2000) was to modify certain initial parameters. These parameters were the deep soil volumetric moisture content, the deep soil temperature, the sea surface temperature, the synoptic pressure gradient scaling factor, and to include rain processes or not in the simulations. It is found that including rain processes and increasing the deep soil volumetric moisture content (at least during wintertime) do improve the simulations performance. / Från tidigare studier med luftföroreningsmodellen TAPMs meteorologiska prognoser över Canterburys slätter, i Nya Zeeland, har man funnit problem. Dess brister har bestått i svårigheter att simulera de nattetid svaga och stagnerande vindhastigheter, vissa riktningar i vindfältet samt grov överskattning av det sensibla värmaflödet. Målet med denna rapport har varit att reducera dessa kända avvikelser tillsammans med andra meteorologiska parametrar i ett försök att optimera TAPMs prognoser för förhållandena över Nya Zeeland. Metodiken som används för att uppnå bättre överensstämmelse med observationsdatan från en fyradagarsperiod, 1-4 augusti år 2000, var att modifiera vissa specifika initieringparametrar. Dessa parametrar var den djupgående markfuktigheten, den djupgående marktemperaturen, havsytans temperatur, det synoptiska tryckets skalningsfaktor. Det undersöktes också om man skulle inkludera nederbördsprocesser eller inte i simuleringarna. Resultaten visade att en inkludering av nederbördsprocesser samt en ökning av den djupa markfuktigheten (åtminstone vintertid) verkar förbättra kvalitéten på simuleringarnas resultat.

"In the prediction of phenomenon behavior there is a presupposition that a similarity exists between model and phenomenon. Success of application is derived from that similarity. An example of this approach is the use of similarity conditions such as Reynolds number in flow problems or Fourier number in heat transfer problems. The advent of performance-based codes has opened up opportunities for many diverse avenues of fire model implementation. The reliability of models depends upon model correspondence uncertainty. Model correspondence uncertainty is incomplete and distorted information introduced into a simulation by a modeling scheme. It manifests itself as 1) the uncertainty associated with the mathematical relationships hypothesized for a particular model, and 2) the uncertainty of the predictions obtained from the model. Improving model implementation by providing a method for rank-ordering models is the goal of the Model Validity Criterion (MVC) method. MVC values can be useful as a tool to objectively and quantitatively choose a model for an application or as part of a model improvement program. The MVC method calculates the amount of model correspondence uncertainty introduced by a modeling scheme. Model choice is based upon the strategy of minimizing correspondence uncertainty and therefore provides the model that best corresponds to the phenomenon. The MVC value for a model is quantified as the sum of the length of two files. These files are individual measures of model structure correspondence uncertainty and model behavior correspondence uncertainty. The combination of the two uncertainty components gives an objective and structured evaluation of the relative validity of each model from a set of likely candidate models. The model with the smallest uncertainty files has the lowest MVC value and is the model with the most validity. Ultimately the value of such a method is only realized from its utility. Example applications of the MVC method are demonstrated. Examples evaluate the rank-ordering of plume physics options used within the computer zone model WPI-Fire when validated against upper layer temperature data from compartment-fire test scenarios. The results show how candidate models of a set may be discriminated against based on validity. These results are powerful in that they allow the user to establish a quantitative measure for level of model performance and/or choose the most valid model for an application."

High-quality short-term forecasts of wind speed are vital to making wind power a more reliable energy source. Gneiting et al. (2006) have introduced a model for the average wind speed two hours ahead based on both spatial and temporal information. The forecasts produced by this model are accurate, and subject to accuracy, the predictive distribution is sharp, i.e., highly concentrated around its center. However, this model is split into nonunique regimes based on the wind direction at an off-site location. This work both generalizes and improves upon this model by treating wind direction as a circular variable and including it in the model. It is robust in many experiments, such as predicting at new locations. This is compared with the more common approach of modeling wind speeds and directions in the Cartesian space and use a skew-t distribution for the errors. The quality of the predictions from all of these models can be more realistically assessed with a loss measure that depends upon the power curve relating wind speed to power output. This proposed loss measure yields more insight into the true value of each model's predictions. One method of evaluating time series forecasts, such as wind speed forecasts, is to test the null hypothesis of no difference in the accuracy of two competing sets of forecasts. Diebold and Mariano (1995) proposed a test in this setting that has been extended and widely applied. It allows the researcher to specify a wide variety of loss functions, and the forecast errors can be non-Gaussian, nonzero mean, serially correlated, and contemporaneously correlated. In this work, a similar unconditional test of forecast accuracy for spatial data is proposed. The forecast errors are no longer potentially serially correlated but spatially correlated. Simulations will illustrate the properties of this test, and an example with daily average wind speeds measured at over 100 locations in Oklahoma will demonstrate its use. This test is compared with a wavelet-based method introduced by Shen et al. (2002) in which the presence of a spatial signal at each location in the dataset is tested.

This paper exposes problems of the commonly used technique of splitting the available data in neural spatial interaction modelling into training, validation, and test sets that are held fixed and warns about drawing too strong conclusions from such static splits. Using a bootstrapping procedure, we compare the uncertainty in the solution stemming from the data splitting with model specific uncertainties such as parameter initialization. Utilizing the Austrian interregional telecommunication traffic data and the differential evolution method for solving the parameter estimation task for a fixed topology of the network model [ i.e. J = 9] this paper illustrates that the variation due to different resamplings is significantly larger than the variation due to different parameter initializations. This result implies that it is important to not over-interpret a model, estimated on one specific static split of the data. (authors' abstract) / Series: Discussion Papers of the Institute for Economic Geography and GIScience

A Tropical Temperate Trough (TTT) is a type of weather system that links the tropics and the extra-tropics across southern Africa. TTT events have been studied statistically in detail, but very little research has been done to study this phenomenon dynamically and especially on a seasonal scale. This study therefore focuses on the predictability of the characteristics of TTTs across southern Africa on a seasonal scale, by using a state-of-the-art seasonal forecasting model, namely the GloSea5 developed by the UK Met Office. Gridded hindcast data for the months of November, December, January and February from 1996/1997 to 2009/2010 are compared to observed data. The different ensemble members of the GloSea5 model (with lead-times of 1 week up to 2 months) are first compared separately to the observed data, after which the model average, with a 0-month, a 1-month and a 2-month lead-time, is calculated and also compared to the observed dataset. TTT events have distinctive characteristics during the formation and the development phases. Most prominent of these characteristics are the cloud bands associated with these weather systems, which have a north-west to south-east orientation and move from west to east across southern Africa. To identify the TTTs, daily outgoing long-wave radiation values are processed by a Meteorological Robot (MetBot), with a strict criterion to identify the cloud bands that characterise these systems. The MetBot’s algorithm produces the information needed to further investigate the different characteristics of TTTs, such as the frequency, the location and the intensity of these systems. Analysis of the MetBot output includes calculating the Root Mean Square Error, the percentage error and in some cases the percentage deviation of the number of cloud bands, as well as the anchor point, the centroid position, the area, the tilt and the minimum and maximum OLR values of the cloud bands. This investigation revealed that the characteristics of TTT events can to some extent be predicted on a seasonal scale for the summer rainfall season of southern Africa. The model used in this study fared particularly well with a 1-month lead-time forecast (compared to a 0-month and a 2-month lead-time forecast). The intensity and the location of the cloud bands associated with TTT events are forecast with a smaller percentage error than the frequency of these systems, as the frequency of TTTs tend to be significantly under-predicted by the model. For some predicted quantities, such as the area of the cloud bands, a bias-adjustment is necessary which produces significantly better results with smaller percentage errors. In the conclusions, suggestions are made on possible future studies, and how to develop this study further to create seasonal forecasts with higher skill with special regards to TTT events. / Dissertation (MSc)--University of Pretoria, 2019. / Geography, Geoinformatics and Meteorology / MSc / Unrestricted