Introduction and Use of Remote Sensing to Predict Disease Outbreak

1
Introduction and Use of Remote Sensing to Predict
Disease Outbreak

• Pietro Ceccato
• Summer 2006 Colloquium on Climate and Health

2
(No Transcript)
3
International Research Institute for Climate and
Society The IRI mission is to enhance society's
capability to understand, anticipate and manage
the impacts of seasonal climate fluctuations, in
order to improve human welfare and the
environment, especially in developing countries.
4
Climate Variability
1930
1940
1950
1960
1970
1980
1990
2000
Climate variability will continue and may even
increase under climate change
5
Agenda

• Introduction to Remote Sensing
• Monitoring Rainfall
• Monitoring Vegetation
• Integration of Remote Sensing and Climate
  Information for Human Health Applications

6
Introduction to Remote Sensing
7
Sensing
8
Sensor
9
Image
An image is composed by pixels
10
(No Transcript)
11
Spatial Resolution

• Size of the pixel

12
Spatial Resolution

• 5 Km

13
Spatial Resolution

• SPOT-VEGETATION 1 Km

14
Spatial Resolution
LANDSAT 30 m
15
Spatial Resolution
1 m
16
(No Transcript)
17
Remote
18
Satellites

• Geostationary
• Polar-orbiting

Configuration of Global Observing System
19
Characteristics of Geostationary Satellites

• Located at about 35,800 km above the equator
• Orbit at the same speed as earths rotation
• Main Sensors VV IR
• Repeat coverage about 15 to 30 minutes
• Cover full earth disk
• Observes events and their evolution

20
An Example of Geostationary Satellite
The METEOSAT Satellite

• Located at 0oW and 0oN
• 30 min repeat time
• Channel Wavelength(?m)
• VIS 0.4-1.1
• IR 10.5-12.5
• WV 5.7-7.1
• Meteosat Second Generation
• 15 min repeat time
• 1km(VV) and 3km(IR) spatial resolutions
• 12 Channels

21
Examples of a Polar-Orbiting Satellites
1. The NOAA satellite

• Sensors of interest
• Advanced Very High Resolution Radiometer/3
  (AVHRR/3)
• Advanced Microwave Sounding Unit A /B
• High Resolution Infrared Radiation Sounder
  (HIRS/3)

22
Characteristics of Polar-Orbiting Satellites

• 350 to 1000 km height
• Main sensors MW,VV,IR
• Narrow spatial coverage
• Less frequent observations
• Observes events at fixed and infrequent times
• Repeat coverage about twice daily
• Some provides global coverage of all
  latitudes

23
Polar-Orbiting Satellites
2. Defense Meteorological Satellite Program (DMSP)

• Sensor of interest
• Special Sensor Microwave
• / Imager (SSM/I)
• Swath width 1400-km
• - Seven passive MW channels

24
Polar-Orbiting Satellites
3. Tropical Rainfall Measurement Mission (TRMM)
Sensors of interest - Precipitation radar (PR) -
215 km Swath - 250 m vertical resolution -
TMI - 9-channel MW - 760 km swath - VV/IR
trmm.gsfc.nasa.gov
25
Rainfall Estimation
26

• 1. What is Satellite Rainfall Estimation?
• There is no such thing as satellite rainfall
  measurement
• Satellite sensors just measure radiation emitted
  or reflected by hydrometeors and/or surface
• Satellite rainfall estimation techniques try to
  convert radiation measurements to precipitation
  information

27
What do satellite sensors see ?
VV, IR Thermal IR
MW (low frequency- emission by rain)
ICE
-10
MW (high frequency- scattering by ice)
Mixed
0
Radar (PR)
Liquid
28
Rainfall Estimation

• METEOSAT
• Thermal Infrared (Cold Cloud Duration )

29
Merging IR and Passive Microwave Rainfall
Estimates

• Combines the best features of both approaches
• Good space/time resolution of geostationary
  estimates
• Better accuracy of microwave estimates

30
Some Satellite Rainfall Products
31
CPC-Merged Analysis of Precipitation (CMAP)
CMAP December Climatology(1979-2005)

• - Merged satellites, numerical model
  predictions and gauge observations
• 2.5 spatial resolution
• monthly total rain
• Also 5-day total
• 1979-current

From IRI data library
32
TRMM

• - Active and passive microwave instruments
• 0.25 spatial resolution
• monthly total rain
• Also 3-hourly
• 1998-current

http//disc.sci.gsfc.nasa.gov/data/datapool/TRMM/0
1_Data_Products/02_Gridded/07_Monthly_Other_Data_S
ource_3B_43/index.html
33
Tropical Rainfall Measurement Mission (TRMM)
34
(No Transcript)
35
(No Transcript)
36
Rainfall Estimate (RFE)

• Merged satellites and gauge
• 0.1 spatial resolution
• Daily total rainfall
• RFE1 1995-2000
• RFE2 2002-current

http//www.cpc.ncep.noaa.gov/products/fews/rfe.htm
l
37
Monitoring Vegetation
38
Vegetation Monitoring

• Measure at sensor level

39
Vegetation Monitoring
BRDF Bidirectional Reflectance Distribution
Function
40
Red Channel
10
100
Chlorophyll absorption
41
Red Channel
50
100
Chlorophyll absorption
42
Near Infra Red (NIR)
100
43
Near Infra Red (NIR)
100
50
44
Vegetation Indices

• In the Red channel Low reflectance values
  Strong chlorophyll absorption green vegetation
• In the NIR channel
• High reflectance values High quantity of
  biomass
• Normalised Difference Vegetation Index NDVI
  (NIR-Red)/(NIRRed)

45
NDVI Study
(NIR-Red)/(NIRRed)
46
NDVI Study
(NIR-Red)/(NIRRed)
47
NDVI problems
Problem resolved using RGB composite image
NDVI shows the presence of vegetation when there
is no vegetation in the field
NDVI 0.14 0.16 0.20 0.30 0.40 0.60
48
Other Indices
49
(No Transcript)
50
NDVI 8km spatial resolution
51
Solution higher spatial resolution image
(TERRA-MODIS 250 m)
52
Water Bodies
Khartoum
Blue Nile
White Nile
53
(No Transcript)
54
Integration of Remote Sensing and Climate
Information for Human Health
P. Ceccato, M. Thomson, S. Connor, P. Graves, T.
Ghebremeskel, M. Jaiteh, M. Levy, S.
Ghebreselassie, T. Barnston, M. Bell, J. Del
Corral, I. Fessha, E. Brantly
55
I. Climate and Distribution of Malaria
Changing Malaria in West Africa ltEndemicity gtEpid
emicity
56
Epidemic Malaria in Africa in Semi-Arid and
Highland Areas
124 millions of people at risk
57
Developing and Establishing MEWS
Integrated MEWS gathering cumulative evidence for
early and focused epidemic preparedness and
response (WHO 2004)
Flag 1 Flag 2 Flag 3 Alert Response
58

• Stratification Map

59
(No Transcript)
60
Long-Term Forecasting
61
Short-Term Forecasting

• Monitoring Rainfall,Vegetation and Water Bodies
• Using measurements from meteorological stations
  or from remotely-sensed images available free of
  charge

62
IRI Data Library Map Room
IRI develops simple tools accessible via Internet
to analyze the relationship between Climate and
Human Health
63
(No Transcript)
64
Rainfall Estimates Product derived from
Satellites
Interface to provide additional information on
the current rainfall season compared with recent
seasons
65
Continued
Various summary information is available for the
specific geographic point queried
66
Create simple maps easy to understand for non
climate specialists
67
Comparison between rainfall estimate long-time
series with incidence data at each sub-district
68
Real-Time Monitoring

• Vegetation index
• Normalized Difference
• Vegetation Index (NDVI)

R2 for significant regression of NDVI anomalies
and concurrent malaria incidence anomalies
69
(No Transcript)
70
Asmara
Sanaa
71
Water and Vegetation Monitoring
Using TERRA-MODIS images (250m spatial
resolution), integrated into a Geographical
Information System (HealthMapper developed by UN
WHO)
72
(No Transcript)
73
(No Transcript)
74
Comparison between vegetation (NDVI) with
incidence data at each sub-district
75
NDVI
Rainfall
76
Asmara
Sanaa
77
Operational Use of MEWS
NMCP strengthen vector control measures, prepare
emergency containers with mobile treatment
centers and mobilize localized response
78
Provide Training
Eritrea, July 2005
79
Research Areas
80
Temperature
Vectorial Capacity
Needs further development to monitor temperature
using RS
81
NASA Land Surface Temperature
Daily products from 1995-2000 provided by NASA
GSFC (Pinheiro et al.)
82
Meningococcal Meningitis

• Transmission of the bacteria Neisseria
  meningitidis is by direct droplet contact
• 20-40 of the population in West Africa are
  symptomless carriers
• Meningococcal meningitis occurs when the bacteria
  penetrate the mucous membrane in dry and dusty
  conditions??

83
Monitoring Dust from Remotely-Sensed Data
NASA Global Monitoring of Dust Events from
Space One of the primary data sets for this study
is the Aerosol Indices (AI) derived from Ultra
Violet (UV) observations from the Total Ozone
Mapping Spectrometer (TOMS, 1978 to present)
calculated using the residue method of Herman et
al. (1997),
Products provided by Jay Herman NASA GSFC
84
The MDGs and climate

• Climate variability dominates rural food
  production, hunger and poverty (MDG1)
• Integrated water resource management is
  impossible without good information on climate
  variability (MDG7)
• Health and mortality rates from diseases such as
  malaria are highly sensitive to climate
  variability. (MDGs 4,5,6)

85
Climate Data Analyses
Meeting Abuja Targets
Areas which have been persistently wetter (2 or 3
years) or drier (-2 or -3 years) during 2001-2003
compared to the 2000 baseline time series of
rainfall variability for b) Niger, c) Eritrea and
d) Botswana
Figure 1a
Figure 1c
Figure 1b
Figure 1d
86
http//iri.columbia.edu/