Little research has been conducted on the effects of housing price cycles on preferences for environmental landscape attributes over time (Cho, Kim, & Roberts, 2011). If the economic value of scarce resources like water resource amenities depends on consumer preferences, then it is useful to address possible effects of cyclical variation in the housing market on these values. This issue is addressed in the primary research question for this thesis: Did the 2007-2009 recession and consequent real estate bust affect marginal willingness to pay for water resource amenities for properties in proximity to the lower St. Johns River (SJR) in Duval County, FL? Prior published studies on the most recent real estate cycle were used to evaluate the timing of housing market impacts during the most recent recession. Also, sales price and sales volume distributions for Duval County were evaluated to compare trends. Based on prior research and results, three separate hypotheses were generated and tested using the hedonic pricing method for residential properties in Duval County. The first hypothesis was that the recent recession impacted the implicit prices of water resource amenities for residential properties in proximity to the SJR. Two separate regression models were developed to test different recession periods (2007-2012 and 2008-2012) based on sample data. Time fixed effect binary variables were used to construct recession interaction effects with water related amenities (proximity to the SJR as well as tributary and riverfront properties). Results showed that during the recession period, sales prices for houses further away from the river experienced a greater negative impact than those closer to the river. This result is similar to research by Cohen, Coughlin, and Lopez (2012) who suggest that, higher priced or high tier residential houses (in this case, those closer to river) tend to hold their value more than low tier residential houses. Also, consistent with research by Bin, Czajkowski, Jingyuan, and Villarini (2015), sales prices for tributary and riverfront homes were not impacted by the recession. A second hypothesis was developed to test whether sales prices for houses in Duval County recovered to pre-recession levels. A regression model was constructed with a separate recession interaction effect variable for 2013-2015 and results indicated that the housing market did not make a full recovery from the recession. A final hypothesis was developed on the significance of interaction variables water quality indicator Chlorophyll-A and a recession effects binary variable. All water quality interaction variables introduced within the model were not significant at the 1% or 5% levels. Future research might include testing interactions with parcel land area and recession time effects and also examining other water quality indicators including Secchi Disk, dissolved oxygen, or turbidity. It may also be useful in the future to use an alternative method of measuring implicit prices of environmental characteristics, such as the repeat sales method.

No meio científico são bastante utilizados os modelos matemáticos para avaliar as características hidrodinâmicas de reatores, porém a literatura é pobre em informações relativas à aplicação do equacionamento matemático em regime não-permanente. Neste trabalho foi aplicado um modelo matemático simulando a hidrodinâmica de reatores tubulares com dispersão para avaliar o efeito da variação de vazão afluente sobre os parâmetros hidrodinâmicos. As simulações foram efetuadas considerando vazão e volume constantes, vazão e volume variáveis e vazão variável e volume constante. Foi investigada a influência de dois tipos de ensaios estímulo-resposta, pulso e degrau, para aplicação de modelos matemáticos e determinação das curvas de distribuição do tempo de residência (DTR) experimentais. Teoricamente ambos devem fornecer os mesmos resultados embora o ensaio em pulso costume apresentar maior sensibilidade experimental. Conforme esperado, ambos os ensaios apresentaram os mesmos resultados finais. Também foram avaliados os traçadores empregados nos dois tipos de ensaios estímulo-resposta, pulso e degrau. Foram empregados três traçadores diferentes - verde de bromocresol, azul de bromofenol e eosina Y que proporcionaram curvas com diferentes características. Como ferramenta auxiliar da modelação foram estudas a determinação das curvas DTR experimentais com auxílio de duas técnicas distintas. Para calibrar o modelo matemático proposto foram utilizados dados de reatores em escala de bancada com diferentes configurações submetidas a variações de vazões afluentes. Os dados de um reator UASB em escala piloto - submetido a variações cíclicas de vazão afluente de 40 e 60% - foram empregados para calibrar e verificar o modelo matemático proposto. Os resultados encontrados com o modelo matemático proposto nesta pesquisa demonstraram que a variação de vazão afluente não deve ser negligenciada. O modelo utilizado representou adequadamente o reator UASB. Seus resultados, quando comparados aos modelos matemáticos que não consideram a variação de vazão, mostraram que para flutuações de vazão elevadas, vazões com valores até 60% maiores do que a vazão média, os valores dos coeficientes de difusão diferem significativamente em função das hipóteses empregadas no desenvolvimento do modelo matemático. / Mathematical models for hydrodynamic characteristics evaluation of reactors are commonly used however there is a lack of information in the literature concerning the application of mathematical modeling for non-steady state flow. In this thesis, it is presented a non-steady mathematical model to simulate the hydrodynamic behavior of pipe flow reactors with diffusion in order to evaluate the effect of the variation of the influent flowrate in the hydrodynamic parameters. The simulations were performed considering constant flowrate and volume, variable flowrate and volume and, variable flowrate and constant volume. It was investigated the influence of two types of stimulus-response assays, pulse and step function, on the application of mathematical models and the determination of the experimental retention time distribution curves. Theoretically, both tests should provide the same results although the pulse stimulus assay usually presents higher experimental sensitivity. As it was expected, both tests presented the same final results. The tested tracers were also evaluated in relation to the pulse and step stimulus-response tests. Three tracers were tested (bromocresol green, bromophenol blue and eosin Y) and it was shown that they provided different curves. It was studied the construction of the experimental retention time distribution curves using two procedures as an auxiliary tool for modeling. The proposed mathematical model was calibrated with data of bench scale reactors submitted to different cyclical variations of flowrates. Moreover, data of UASB reactor in pilot scale - submitted to 40 and 60% of cyclical variations of flowrates were utilized to calibrate and verify the obtained mathematical model. The results found with the mathematical model proposed in this research showed that the variations in influent flow rate can not be neglected. The model adequately represented a pilot scale UASB reactor. The results showed that the diffusion coefficients differ significantly for high flowrate fluctuations, when compared to other mathematical models that do not incorporate variable flowrate, depending upon the hypothesis used to derive the model.

No meio científico são bastante utilizados os modelos matemáticos para avaliar as características hidrodinâmicas de reatores, porém a literatura é pobre em informações relativas à aplicação do equacionamento matemático em regime não-permanente. Neste trabalho foi aplicado um modelo matemático simulando a hidrodinâmica de reatores tubulares com dispersão para avaliar o efeito da variação de vazão afluente sobre os parâmetros hidrodinâmicos. As simulações foram efetuadas considerando vazão e volume constantes, vazão e volume variáveis e vazão variável e volume constante. Foi investigada a influência de dois tipos de ensaios estímulo-resposta, pulso e degrau, para aplicação de modelos matemáticos e determinação das curvas de distribuição do tempo de residência (DTR) experimentais. Teoricamente ambos devem fornecer os mesmos resultados embora o ensaio em pulso costume apresentar maior sensibilidade experimental. Conforme esperado, ambos os ensaios apresentaram os mesmos resultados finais. Também foram avaliados os traçadores empregados nos dois tipos de ensaios estímulo-resposta, pulso e degrau. Foram empregados três traçadores diferentes - verde de bromocresol, azul de bromofenol e eosina Y que proporcionaram curvas com diferentes características. Como ferramenta auxiliar da modelação foram estudas a determinação das curvas DTR experimentais com auxílio de duas técnicas distintas. Para calibrar o modelo matemático proposto foram utilizados dados de reatores em escala de bancada com diferentes configurações submetidas a variações de vazões afluentes. Os dados de um reator UASB em escala piloto - submetido a variações cíclicas de vazão afluente de 40 e 60% - foram empregados para calibrar e verificar o modelo matemático proposto. Os resultados encontrados com o modelo matemático proposto nesta pesquisa demonstraram que a variação de vazão afluente não deve ser negligenciada. O modelo utilizado representou adequadamente o reator UASB. Seus resultados, quando comparados aos modelos matemáticos que não consideram a variação de vazão, mostraram que para flutuações de vazão elevadas, vazões com valores até 60% maiores do que a vazão média, os valores dos coeficientes de difusão diferem significativamente em função das hipóteses empregadas no desenvolvimento do modelo matemático. / Mathematical models for hydrodynamic characteristics evaluation of reactors are commonly used however there is a lack of information in the literature concerning the application of mathematical modeling for non-steady state flow. In this thesis, it is presented a non-steady mathematical model to simulate the hydrodynamic behavior of pipe flow reactors with diffusion in order to evaluate the effect of the variation of the influent flowrate in the hydrodynamic parameters. The simulations were performed considering constant flowrate and volume, variable flowrate and volume and, variable flowrate and constant volume. It was investigated the influence of two types of stimulus-response assays, pulse and step function, on the application of mathematical models and the determination of the experimental retention time distribution curves. Theoretically, both tests should provide the same results although the pulse stimulus assay usually presents higher experimental sensitivity. As it was expected, both tests presented the same final results. The tested tracers were also evaluated in relation to the pulse and step stimulus-response tests. Three tracers were tested (bromocresol green, bromophenol blue and eosin Y) and it was shown that they provided different curves. It was studied the construction of the experimental retention time distribution curves using two procedures as an auxiliary tool for modeling. The proposed mathematical model was calibrated with data of bench scale reactors submitted to different cyclical variations of flowrates. Moreover, data of UASB reactor in pilot scale - submitted to 40 and 60% of cyclical variations of flowrates were utilized to calibrate and verify the obtained mathematical model. The results found with the mathematical model proposed in this research showed that the variations in influent flow rate can not be neglected. The model adequately represented a pilot scale UASB reactor. The results showed that the diffusion coefficients differ significantly for high flowrate fluctuations, when compared to other mathematical models that do not incorporate variable flowrate, depending upon the hypothesis used to derive the model.