A Negative Binomial with a Non-Homogenous Gamma Distributed Mean for Robustifying Pseudo R2 Regression Values of Immature Vector and Nuisance Mosquito Count Data for Optimally Discerning Un-Geosampled Waste Tires in a Subtropical Oviposition Site in SASÃÂ®/GIS employing Worldview-3 Visible and Near Infra-Red Data in Hillsborough County, Florida
Dinh ETN* and Jacob BG
Department of Global Health, College of Public Health, University of South Florida, Tampa, FL 33612, USA
- Corresponding Author:
- Dinh ETN
Department of Global Health, College of Public Health
University of South Florida, 13201 Bruce B. Downs Blvd., MDC
56, Tampa, FL 33612, USA
E-mail: [email protected]
Received Date: April 25, 2016; Accepted Date: June 17, 2016; Published Date: June 24, 2016
Citation: Dinh ETN, Jacob BG (2016) A Negative Binomial with a Non- Homogenous Gamma Distributed Mean for Robustifying Pseudo R2 Regression Values of Immature Vector and Nuisance Mosquito Count Data for Optimally Discerning Un-Geosampled Waste Tire Oviposition Sites in a Subtropical Habitat in SAS®/GIS Using Worldview-3 Visble and Near Infra-Red Data in Hillsborough County, Florida. J Remote Sensing & GIS 5:167. doi: 10.4172/2469-4134.1000167
Copyright: © 2016 Dinh ETN, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Refuse vehicle tires on undeveloped land plots near human dwellings may be a public health threat, as they can provide a suitable habitat for vector and nuisance mosquito (Diptera: Culicidae) population growth. These tires are currently found only through ground-based searches, so interpolated spectral signature of a georefernceable, known, positive tire may help expedite discriminating unknown waste tire geolocations. However, frequentistic and nonfrequentistic quantification of bioenvironmental explanatorial time series covariates statistically significant to mosquito hyperproductivitt in waste tire habitats is needed to limit the search criteria of the signature. This study aimed to develop an iteratively interpolative geo-spectral biosignature for detecting unknown, un-geosampled waste tires conducive to mosquito propagation. After constructing various regression models, we found that the field geo-sampled mosquito count data featured deviations from the assumptions of regression modeling (i.e., collinear and heteroskedastic parameters) Thus, a negative binomial paradigm was utilized to assuage the violations of regression analysis and to robustify the model’s R2 value. Based on the results of the linear analyses, a spectral signature of a productive habitat was created from multispectral band imagery from WorldView-3 satellite sensor data. The signature was then applied in Hillsborough County, FL to remotely determine the eco-geographical geo-locations of anthropogenic waste tire habitats. The signature model exhibited a sensitivity of 83% and a specificity of 87%. In conclusion, the regression and signature models constructed here provided a parsimonious yet accurate estimation of undiscovered waste tire habitats that may yield many mosquitoes.