Geoscience Reference
In-Depth Information
11.3 Infectious Disease Vector Habitat Identification
and Monitoring
The potential utility of such resources to public health and safety is significant.
Studies using satellite data to investigate the spatial and seasonal dynamics of infec-
tious disease transmission were mostly initiated in the late 1970s and the 1980s.
Most of these studies focused on utilization of satellite data to identify habitats and
breeding sites of disease vectors, as well as other environmental parameters affect-
ing health (Barnes and Cibula
1979
; Wagner et al.
1979
; Jovanovic
1987
; Linthicum
et al.
1987
; Hugh-Jones
1989
). Conclusive results from these early studies revealed
the significant potential of remote sensing applications for disease vector monitoring
and related epidemiological applications in a wide array of site-specific geographic
locations.
Over the past three decades, the use of remote sensing techniques in epidemio-
logical studies has evolved significantly due to two factors: (i) the rapid development
in spatial information technologies (i.e. remote sensing, Geographical Information
Systems and Geo-positioning Systems) and (ii) the development of disease ecol-
ogy (Thomson et al.
1996
), a sub-discipline of epidemiology, which investigates the
biological, physical and anthropogenic links between the environment and disease,
consequently accounting for spatial variation in transmission (Graham et al.
2004
).
As shown in a number of comprehensive reviews on the use of remote sensing and
GIS for epidemiological applications (e.g. Thomson and Connor
2000
; Hay et al.
2000
; Kalluri et al.
2007
), a growing number of epidemiologists are now taking
advantage of the reductions in cost and increased ease of access.
A variety of techniques ranging from simple correlations to complex combi-
nations with models have been used to link satellite-derived variables to vector
biology. A typical approach involves classifying remotely sensed images into habi-
tats for disease vectors (Rogers
2000
; Randolph
2000
; Guo et al.
2005
; Lacaux
et al.
2007
; Leblond et al.
2007
) and creating environmental and land use maps
associated with vector-borne diseases (Hay et al.
2006
). Such methods often use the
satellite-derived Normalized Difference Vegetation Index (NDVI) to estimate the
amount of green biomass, which is an indicator of not only the abundance of vec-
tors (Rogers and Randolph
1991
; Malone et al.
2001
;Wuetal.
2002
; Odiit et al.
2005
; Peterson et al.
2005
;Rasoetal.
2006
), but also of other environmental factors
such as moisture, soil type, slope and elevation (Boone et al.
2000
). Among other
commonly used satellite-derived variables linked to vector biology are the modified
soil-adjusted vegetation index (MSAVI), the land surface temperature (LST) derived
from the Advanced Very High Resolution Radiometer (AVHRR) channels 4 and 5,
the middle infrared band derived from AVHRR channel 3 (detection of hot regions
associated with forests), the near-surface air temperature derived from LST and veg-
etation index measurements, atmospheric and near-surface humidity, vapor pressure
deficit, and various precipitation indexes. For example, Rogers (
2000
) found a good
correlation between the satellites derived index of LST and monthly mortality rates
and used the index as input for tsetse population models for the Yankari Game
