HR: 0830h
AN: B21E-0766 [PDF]
TI: Application of MODIS GPP to Forecast Risk of Hantavirus Pulmonary Syndrome Based on Fluctuations in
Reservoir Population Density
AU: * Loehman, R
EM: rachel@ntsg.umt.edu
AF: NTSG, College of Forestry and Conservation, University of Montana, Missoula, MT 59812 United States
AU: Heinsch, F A
EM: faithann@ntsg.umt.edu
AF: NTSG, College of Forestry and Conservation, University of Montana, Missoula, MT 59812 United States
AU: Mills, J N
EM: jmills@cdc.gov
AF: Special Pathogens Branch, Division of Viral and Rickettsial Diseases, Centers for Disease Control and
Prevention, Atlanta, GA 30333 United States
AU: Wagoner, K
EM: kew7@cdc.gov
AF: Special Pathogens Branch, Division of Viral and Rickettsial Diseases, Centers for Disease Control and
Prevention, Atlanta, GA 30333 United States
AU: Running, S
AF: NTSG, College of Forestry and Conservation, University of Montana, Missoula, MT 59812 United States
AB:
Recent predictive models for hantavirus pulmonary syndrome (HPS) have used remotely sensed spectral reflectance data to
characterize risk areas with limited success. We present an alternative method using gross primary production (GPP) from the
MODIS sensor to estimate the effects of biomass accumulation on population density of {\it Peromyscus maniculatus} (deer
mouse), the principal reservoir species for Sin Nombre virus (SNV). The majority of diagnosed HPS cases in North America are
attributed to SNV, which is transmitted to humans through inhalation of excretions and secretions from infected rodents.
A logistic model framework is used to evaluate MODIS GPP, temperature, and precipitation as predictors of {\it P.
maniculatus} density at established trapping sites across the western United States. Rodent populations are estimated using
monthly minimum number alive (MNA) data for 2000 through 2002. Both local meteorological data from nearby weather stations
and 1.25 degree x 1 degree gridded data from the NASA DAO were used in the regression model to determine the spatial
sensitivity of the response. MODIS eight-day GPP data (1-km resolution) were acquired and binned to monthly average and
monthly sum GPP for 3km x 3km grids surrounding each rodent trapping site.
The use of MODIS GPP to forecast HPS risk may result in a marked improvement over past reflectance-based risk area
characterizations. The MODIS GPP product provides a vegetation dynamics estimate that is unique to disease models, and
targets the fundamental ecological processes responsible for increased rodent density and amplified disease risk.
DE: 1620 Climate dynamics (3309)
DE: 1640 Remote sensing
SC: Biogeosciences [B]
MN: 2003 Fall Meeting