Browsing by Author "Jerry Davis, Committee Member"

Now showing 1 - 4 of 4

Associations Between Gaussian Markov Random Fields and Gaussian Geostatistical Models with an Application to Model the Impact of Air Pollution on Human Health
(2006-02-19) Song, Hae-Ryoung; Sujit Ghosh, Committee Co-Chair; David Dickey, Committee Member; Jerry Davis, Committee Member; Montserrat Fuentes, Committee Co-Chair; Peter Bloomfield, Committee Member
Gaussian geostatistical models (GGMs) and Gaussian Markov random fields (GMRFs) are two distinct approaches commonly used in modeling point referenced and areal data, respectively. In this dissertation, the relations between GMRFs and GGMs are explored based on approximations of GMRFs by GGMs, and vice versa. The proposed framework for the comparison of GGMS and GMRFs is based on minimizing the distance between the corresponding spectral density functions. In particular, the Kullback-Leibler discrepancy of spectral densities and the chi-squared distance between spectral densities are used as the metrics for the approximation. The proposed methodology is illustrated using empirical studies. As a part of application, we model associations between speciated fine particulate matter (PM) and mortality. Mortality counts and PM are obtained at county and point levels, respectively. To combine the variables with different spatial resolutions, we aggregate PM to the county level. The aggregated PM are modeled using GMRFs, and associations between PM and mortality are investigated based on Bayesian hierarchical spatio-temporal framework. This model is applied to speciated PM[subscript 2.5] and monthly mortality counts over the entire U.S. region for 1999-2000. We obtain high relative risks of mortality associated to PM[subscript 2.5] in the Eastern and Southern California area. Particularly, NO₃ and crustal materials have greater health effects in the Western U.S., while SO₄ and NH₄ have more of an impact in the Eastern U.S. We show that the average risk associated with PM[subscript 2.5] is approximately twice what we obtained for PM₁₀.
Observations and Modeling of Evapotranspiration across North Carolina
(2005-07-19) Puryear, Margaret W. P.; Sethu Raman, Committee Chair; S. Pal Arya, Committee Member; Jerry Davis, Committee Member
Evapotranspiration (ET) is measured at 14 stations in North Carolina using an ET gage, which allows distilled water to evaporate through a waterproof surface acting as the reference crop alfalfa. These stations are part of the State Climate Office of North Carolina's Environment and Climate Observing Network (ECONet) and are distributed throughout the state's three broad climate regions: mountains, Piedmont and coastal plain. Observations from the ET gage at Lake Wheeler were compared with the gradient method to show that the ET gage observations are accurate and reliable. Averages of reference evapotranspiration, precipitation, soil moisture and temperature are analyzed for spatial trends. Daily reference evapotranspiration values are also compared with precipitation, soil moisture, temperature and humidity at one location in each of the three climate regions. Evapotranspiration decreases initially when precipitation occurs. This is due to the increase in atmospheric humidity and decrease in temperature. After precipitation ends, evapotranspiration increases due to the increase in the water supply. Daily pan evaporation values at two sites from years 2003 and 2004 were compared with evapotranspiration at the same locations. Effective precipitation was calculated for 2004 at each of the 14 evapotranspiration measuring stations. Two different empirical methods for estimating evapotranspiration were used for comparison with observations: Priestley-Taylor and Penman-Monteith. An appropriate value for the Priestley-Taylor parameter was estimated. Estimates of ET from these empirical methods indicated that the Penman-Monteith method overestimated ET as compared to observations and the Priestley-Taylor method gave better results. A mesoscale numerical simulation was performed for the period, 00Z July 18, 2004 to 00Z July 24, 2004 using the MM5 model with a domain centered over North Carolina. Spatial distribution of evapotranspiration was obtained and comparisons made between model predictions and observations of latent heat flux, soil moisture, soil temperature, wind speed, and air temperature. Latent heat flux was estimated using model parameters and empirical methods and compared with observations. A special focus on the simulations of days July 21 through July 24 included one day prior to a precipitation event, two days during the event, and one day after the precipitation ended. The model has a tendency to overestimate the latent heat flux in the coastal plain and eastern Piedmont, due to the overestimation of precipitation in this area. The mountains had the highest latent heat flux predicted by the model. Across the state, the Priestley-Taylor method using simulated parameters performed better than the Penman-Monteith method.
Space-time Modeling of Health Effects while Controlling for Spatially Varying Exposure Surfaces.
(2010-06-17) Kalendra, Eric; Montserrat Fuentes, Committee Chair; Brian Reich, Committee Chair; Huixia Wang, Committee Member; Jerry Davis, Committee Member
Using Hyperspectral Remote Sensing to Estimate Leaf Area Index of Loblolly Pine Plantations
(2003-07-25) Flores, Francisco Jose; Jerry Davis, Committee Member; Daniel Kelting, Committee Member; H. Lee Allen, Committee Chair; Heather Cheshire, Committee Member; Montserrat Fuentes, Committee Member
High spectral resolution (Hyperspectral) and multispectral imagery was used to examine the spectral response of loblolly pine with contrasting LAI and foliar nitrogen concentration. The research studies included very different stand structures (age, number and size of the trees). As a result of treatments applied, seasonal, stand and site variation, the range in LAI (0.4 — 3.6 m2 m2) and foliar nitrogen concentration (0.79 — 1.62%) on our study covered most of the range observed in midrotation loblolly pine plantations. Hyperspectral data was used to calculate narrowband vegetation indices (VIs) based on reflectance in the red (R) and near infrared (NIR). LAI was linearly related to the simple ratio (SR) vegetation index, the relationship was not affected by site, stand structure or season. We also found a strong relationship between SR calculated from hyperspectral data and SR calculated from readily available Landsat 7 ETM+ data. Using that relationship, we determine an equation to estimate LAI from Landsat 7 ETM+ data. Hyperspectral data was also used to estimate foliar nitrogen concentration and content of loblolly pine. We found a stronger relationship between nitrogen content and reflectance data than nitrogen concentration and reflectance data.