1

Xiangming XiaoXiangming Xiao11

Marius GilbertMarius Gilbert22, Jan Slingenbergh, Jan Slingenbergh33, Fumin Lei, Fumin Lei44

11Institute for the Study of Earth, Oceans and SpaceInstitute for the Study of Earth, Oceans and Space

University of New Hampshire, Durham, NH, USAUniversity of New Hampshire, Durham, NH, USA

http://http://remotesensing.unh.eduremotesensing.unh.edu

22Free University of Brussels, BelgiumFree University of Brussels, Belgium33Animal Health Division, FAO, ItalyAnimal Health Division, FAO, Italy

44Institute of Zoology, Chinese Academy of Sciences, ChinaInstitute of Zoology, Chinese Academy of Sciences, China

May 30May 30--31, 2006, Rome, Italy31, 2006, Rome, Italy

FAO/OIE International Scientific Conference on Avian Influenza aFAO/OIE International Scientific Conference on Avian Influenza and nd Wild BirdsWild Birds

Remote Sensing, Ecological Variables and

Bird Migration

From Declan Butler

Global distribution of H5N1 outbreaks

2003-2006

45 countries

cumulative number of

confirmed human cases (as 6/4/2006)

192 cases/109 deaths

Bird migration

- important for understanding the role of wild waterbirds in the spread of H5N1 virus

- a long history of in-situ field observations, but limited geospatial databases at continental to global scales

A changing Planet Earth poses challenge for bird migration

- land use change

- loss of wetlands

- agricultural expansion and intensification

- post-harvest management

- water use change (e.g., fish and shrimp ponds, dams)

- climate change

Bird migration and climate change

Qinghai - Tibet plateau

the roof-of-the-world, 2.5 million km2

breeding sites for wild birds

temperature rise by ~0.9oC since 1980s

Increase in thawing of permafrost

Decrease in glacier area by 7% per year

���� increase in surface water bodies

Early onset of Indian summer monsoon

Climate variation/change calls for re-analysis of bird migration and climate data.

2

TRMM11/27/97

Terra12/18/99

Landsat 74/15/99

Aqua5/4/02

MODIS

Red/Green/Blue

(band 1-4-3) 5/25-

6/1, 2002

Tonle Sap, Mekong

Delta (Cambodia /

Viet Nam)

500-m spatial

resolution

Satellite remote sensing provides routine observations for quantifying ecological variables relevant to bird migration

Bird migration and satellite remote sensing

1. Phenology of waterbirds migration

timing of migration

Ecological variables that trigger migration

2. Land surface phenology and ecosystems

Habitats along migratory flyway

breeding, stop-over, moulting and wintering sites

3. Satellite remote sensing track ecological variables and habitats for migratory birds

Ecological variables that trigger bird migration in autumn (seasonal migration)

Temperature (e.g., frost)

Snow, Ice

Wind

Phenology of waterbirds migration

Lake Baykal (one MODIS pixel) (Terra/MODIS)

Year 2004 (8-day interval)

1/1/04 3/1/04 5/1/04 7/1/04 9/1/04 11/1/04 1/1/05

Wate

r surf

ace

te

mp

era

ture

(oC

)

-30

-20

-10

0

10

20

LSTday

LSTnight

Traveling wave of the first frost in July – Dec. 2005 (nighttime LST from MODIS/Aqua

Land surface temperature (LST) from MODIS sensor

(1-km resolution)

Ecological variables that trigger bird migration in spring (seasonal migration)

Temperature (e.g., frost)

Snow and Ice (melting)

Wind

Phenology of waterbirds migration

Lake Baykal (one MODIS pixel) (Terra/MODIS)

Year 2004 (8-day interval)

1/1/04 3/1/04 5/1/04 7/1/04 9/1/04 11/1/04 1/1/05

Wate

r surf

ace

te

mp

era

ture

(oC

)

-30

-20

-10

0

10

20

LSTday

LSTnight

Traveling wave of the last frost in Jan - June 2005 (nighttime LST from MODIS/Aqua)

Land surface temperature (LST) from MODIS

(1-km resolution)

3

Ecological variables that affect bird migration in autumn and spring (seasonal migration)

- onset of wet & dry seasons

Phenology of waterfowl migration

Tropical Rainfall Measuring Mission (TRMM)

5/2003

11/2003

Monthly rainfall at 0.5o

grid

Land surface phenology and ecosystems

In a human-dominated Earth, changes in land use and water use have substantially altered ecosystems both spatially and temporally, which may affect migratory flyway (breeding, stop-over, moulting and wintering sites). Wild waterbirds may use croplands as part of their habitats.

- Ecosystems along the “migration flyway”

- Land surface phenology- Rural landscape where wild birds interact and

mix with domestic poultry

Where do wild waterbirds fly to and stay?

Satellite mapping

H5N1 outbreaks

Free-range duck Paddy rice field

migratory waterbirds

one MODIS pixel in Bangkok area

Month in 2004

1 2 3 4 5 6 7 8 9 10 11 12

Veg

eta

tio

n I

ndic

es

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

NDVI

LSWI

EVI

Starting date Starting date

Southeast Asia in 2004

wet season ���� dry season

MODIS-based land surface phenology H5N1 outbreaks in Thailand

4

HPAI cases

Free-range ducks

Cropping intensity

Paddy rice

Spatial correlations among HPAI cases,

free-ranging ducks and paddy rice in

Thailand in 2004-2005.

H5N1 outbreaks

Free-range ducks

H5N1 outbreaks in Thailand and Vietnam were concentrated in those area with multiple cropping systems

MODIS-based land surface phenology

Double cropping pixels in Mekong area

Year 2004 (8-day interval)

1/1/04 3/1/04 5/1/04 7/1/04 9/1/04 11/1/04

Cro

pla

nd a

rea

(km

2)

0

1000

2000

3000

4000

Planting date

Harvesting date

Natural

wetlandsFish

ponds

Paddy rice

fields

domestic

poultry

Human

population

Migratory

waterbirds

Free-range

ducks/geeseAIVAIV

AIV

Weather / Climate Satellite remote sensing

AIV

AIV

Role of satellite remote sensing Satellite-based geospatial datasets

1. Agricultural land use

cropping intensity, calendar & irrigation

2. Wetlands

natural wetlands, water bodies,

3. Climate

water & land surface temperature, snow, ice,

wind, drought, rainfall

-�migration timing, survival rates of HPAIV

5

Satellite observatories

• MODIS, Landsat, PALSAR, Cloudsat, TRMM

International

FAO

Decision Support Tools

• Risk assessment

• Decision-support

System

• Scenario tools

National

Decision Support Tools

HPAI surveillanceOutbreak location & time

Control area

In-situ observatories

• cropping intensity & crop calendar• wetlands, rainfall• water bodies, irrigation,• land surface temperature, snow, ice

• land primary production

• bird surveillance • waterfowl migration time & flyway • poultry and livestock

• agricultural statistical data

Geospatial analysis

Spatially-explicit

epidemiological

model

Ecology of HPAI2003-2006

Patterns, dynamics, variables

Nowcast & Early Warning

2006-likely “hot-spots” & “hot-times”

Epidemic potentialDistribution of persistence

Framework of observations, analysis and modeling

Summary

1.Geospatial data of climate, agro-ecosystems,

natural wetlands are needed in order to better

understand bird migration.

2.Satellite remote sensing can provide such geospatial datasets in timely mode.

3. Integrating the geospatial databases with bird

migration data (in-situ, satellite-tracking) will

improve geospatial analysis of bird migration, which would help understand the role of migratory birds in the spread of HPAI virus, and identify likely “hot spots” and “hot times” for risk assessment and targeted surveillance.

Thank you!Thank you!

Lake

Qinghai