satellite based drought monitoring and early warning · proba-v follow up s3 follow up s3 follow up...
TRANSCRIPT
Satellite based drought monitoring
and early warning
Examples of EO data and applications
Flemish Institute for Technological Research (VITO)
Roel Van Hoolst, Bart Deronde
VITO – Remote Sensing UnitSENSOR/MISSION DESIGN
IMAGE PROCESSING
REMOTELY PILOTED AIRCRAFT SYSTEMS
IMAGE QUALITY
TRANING & CONSULTANCY
GLOBAL & LOCAL APPLICATIONS
vito-eodata.beland.copernicus.eu
Free of charge
WATER
CAL/VAL
AGRICULTURE
BIODIVERSITY
JULJUNMAY AUGAPRMARFEBJANDECNOVOCTSEP
ccrop establishment
10 14
greening up heading mature
14 14 21 13 425 7 9 521 29 34 28 108 days 50
26
Sow Tillering
Emergence
Dormancy Growth
resumesHeading
Boot
Dead
ripe
Hard doughSoft
dough
Harvest
Jointing
Maximum Coverage
Phenological cycle of wheatOne crop= different spectral responses over the season
Bare soil
Max greennessYellowing (ripening)
Straw, bare soil
Biomassdevelopment
NDVI
Vegetation monitoring – Time Series Analysis
Examples of existing EO datasetsFreely available at VITO and Copernicus Global Land
Service(historical & in near real time)
SPOT-VEGETATION 1SPOT-VEGETATION 2
METOP-AMETOP-B
PROBA-VSENTINEL-3A
98 99 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
METOP-C
SENTINEL-3B
PROBA-V follow upS3 follow up
S3 follow up
Proba-V (100m)/300m/1km products NDVI, fAPAR, Dry Matter Productivity,…
METOP-AVHRR 1km NDVI
Global Vegetation Monitoring Systems
Two examples:
Agricultural Stress Index System (ASIS) based on METOP-AVHRR
Proba-V Mission Exploitation Platform – Time Series Viewer
GlobalNear Real Time
Cropland & grasslandAgricultral Stress Index
Example 1
7
Example 1
Proba-V Mission Explotation Platform
http://proba-v-mep.esa.int/
Example 2
Capacity Building & Training on the use of
time series for vegetation monitoringSet-up of low cost
receiving stations
SPIRITS workshop for early warning
Support in generation of bulletins
and reports for food-security or
drought monitoring
• JRC-VITO development: Software for the Processing and
Interpretation of Remotely Sensed Image Time Series
• Several tools to exploit RS time series & to make agro-
meteorological bulletins
• Freely available, incl. Manual & Tutorials
• Users: VITO, JRC, FAO, WFP, Africa
• Training sessions (± 50 so far)
Post-Processing Software: SPIRITS
MAY JUNE JULY AUGUST SEPTEMBER OCTOBER
1 11 21 1 11 21 1 11 21 1 11 21 1 11 21 1 11 21
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
Short season with low NDVI
Important delay in SoS
Senegal, province DIOURBEL
Comparison of seasons: matrix display
Delay in March rains
below avg. rainfall May-June
with negative effect on NDVI
Above avg. rainfall
with positive effect on NDVI
2012 Meher season, West-Shewa, Ethiopia
Example of combining vegetation indices and rainfall data
Vegetation monitoring for food security
GMFS, AgriCab projects for JRC, FAO, WFP, national and regional
authorities
Analysis of long term time series of EO data
Vulnerabilty mapping – drought frequency returning period
Trend analysis
Phenology mapping
Probability of production deficit at end of season
Agricultural index based insurance
• Use of RS for crop monitoring, damage and risk assessment for the insurance sector: studies forESA and Swiss Re for parts of Ukraine, Russia, Morocco.
• Development of RS based index insurance products for IFAD-WFP in Senegal
15
“damage map” (drought)y = 0,1264x - 16,308
R² = 0,8289
0
5
10
15
20
25
30
100 150 200 250 300 350
cere
al
yie
ld
(qx
/h
a)
fAPARcum (D3 Feb - D2 Apr)
BEN SLIMANE
BEN SLIMANE
Linear (BEN SLIMANE)
“risk map” (drought)
trigger
exit
“damage map”
(winter kill)
Index insurance
contract
development
Agricultural insurances
EO based phenology
Analysis of global EO based phenology
Global results – GIMMS – ’83-’12
17
Satellite based drought monitoring
and early warning
Flemish Institute for Technological Research (VITO)
Roel Van Hoolst, Bart Deronde