Slide 1

Report
Global Monitoring for Food Security 3
26/10/2011, Nairobi, Kenya
http://www.gmfs.info
ESA’s Crop Monitoring
And
Early Warning Service
Outline
Nairobi 26/10/2011
http://www.gmfs.info
This presentation
• Background
• Partners
• Services
•
•
•
Early Warning
Agricultural Monitoring
CFSAM
Nairobi 26/10/2011
http://www.gmfs.info
ESA Stage 3 Background
•
•
•
•
GMFS started in 2003
New contract GMFS3 -> 2013
Continuation of services to Stage 2 users
Focus is on Sustainability
Partners
Nairobi 26/10/2011
http://www.gmfs.info
Nr.
Company
Country
General competences, tasks & responsibilities
01
VITO
Belgium
Project management, early warning, Medium/(high) resolution optical
and training , information packaging, promotion
Secondment to FAO through visiting scientists programme
02
Consorzio
ITA
Italy
Agriculture Mapping with optical remote sensing and ground statistics Validation methodology definition – Training: MALAWI
03
EARS
Netherlands
FAST Meteosat based Early Warning Services for African regions.
Generation of rainfall and evapotranspiration data fields. Production of
crop yield forecasts. Provision of dedicated user software (Imageshow
2) and training.
04
EFTAS
Germany
Agriculture Mapping with optical remote sensing and ground statistics,
with radar data - Validation data collection - Service network support:
SUDAN
05
SARMAP
Switzerland
Agriculture mapping with radar – Software programming – Training
package: RCMRD
06
ULg
Belgium
Early warning support and Service portfolio evolution for Early Warning
Services: AGRHYMET
07
GeoVILLE
Austria
Soil Moisture Indicator products
4
Nairobi 26/10/2011
http://www.gmfs.info
ESA
Technical officer
Nairobi 26/10/2011
Service Groups
7 GMFS3 Services
Early
Warning
Crop Yield and Vegetation Monitoring
Service
FAST Service
Soil Moisture Monitoring Service
Agricultural
Monitoring
Support to the Optimization of the
National Agricultural Survey Service
Agricultural Mapping Service
SAR Knowledge Transfer Service
CFSAM
Support
Support to Crop and Food Supply
Assessment Mission Service
User Board
Overall Management
VITO
West Africa
Regional coordinator
AGRHYMET(CRA)
East Africa
Regional coordinator
RCMRD
Southern Africa
Regional coordinator
East Africa region
Southern Africa region
West Africa region
Senegal
CSE
Mali
LaboSEP
Sudan
FMoAF
Ethiopia
MoARD
Scientific Board
http://www.gmfs.info
Malawi Mozambique Zimbabwe
MoA
INAM
MoAFS
Early Warning Service
Nairobi 26/10/2011
http://www.gmfs.info
•
•
•
Crop Yield and Vegetation Monitoring
Soil Moisture Analysis
FAST
Early Warning Service
Nairobi 26/10/2011
http://www.gmfs.info
 3 independant sources of information
 Convergence of evidence analysis
MERIS
Vegetation
MSG
Rainfall
Radiation
ASCAT
Soil moisture
SUPPORT EW ANALYSIS
TRAIN HOW TO USE THE DATA
Early Warning Service
Nairobi 26/10/2011
http://www.gmfs.info
Crop Yield and Vegetation Monitoring
Nairobi 26/10/2011
http://www.gmfs.info
Qualitative Analysis: Early or Late Start of the Growing Season
Context: In Sahel Region the start of Season is a first indicator of crop development success or
failure (Approach published in the CILLS bulletin)
Profile Matching Approach:
- Compare the fAPAR profile for the 3 first months of the 2011 season with the average 1999-2010
0,9
0,8
0,7
NDVI
0,6
0,5
0,4
0,3
Phase/shift
0,2
0,1
0
3
6
9
12
15
18
21
24
27
30
33
36
Decades
current year
historic year
-Display the shift that have the best fit with the
Long Term Average
- Give an overview of the anticipated or delayed area at pixel level
Crop Yield and Vegetation Monitoring
Nairobi 26/10/2011
http://www.gmfs.info
Qualitative Analysis of ongoing Growing Season for Agricultural Monitoring
based on low Resolution Data
Context: Standard Anomaly Maps based on comparison with LTA give a good spatial overview of anomalies,
this approach adds duration and intensity to the maps
Cluster analysis :
- Iso data classification based on the relative
difference between 10 day VI and
-Display classified map
with the corresponding classes profiles
Crop Yield and Vegetation Monitoring
Nairobi 26/10/2011
http://www.gmfs.info
Yield Estimation based on Low Resolution Monitoring
Context: Non Parametric Yield Forecast
Similar Years to Yield estimate
Similarty Analysis :
- CROP MAP !
- for each pixel the most similar year is found
-Display classified map
-Per ADMIN percentage
TABLE1: Percentage of similar year to 2009 per administrative area
Addis Ababa
Amhara
Harari
Oromiya
Somali
Southern
Tigray
ETH001
ETH003
ETH007
ETH008
ETH009
ETH010
ETH011
1999
10%
7%
36%
9%
6%
10%
4%
2000
9%
2%
3%
4%
4%
5%
2%
2001
0%
1%
9%
8%
8%
7%
1%
2002
8%
12%
2%
17%
9%
20%
11%
2003
14%
23%
25%
11%
9%
7%
20%
2004
8%
19%
4%
13%
10%
10%
17%
2005
6%
21%
0%
9%
4%
8%
17%
2006
10%
2%
8%
13%
15%
12%
6%
2007
10%
6%
12%
7%
4%
6%
15%
2008
23%
8%
2%
9%
31%
15%
6%
1999
2000
2001
2002
2003
2004
2005
2006
11.97 11.69 11.815 11.94 12.45 18.71 17.32 13.87
7.38
9.4 10.36
8.87 12.93
14.5 15.24 15.94
6.08
6.37 12.43
6.37 12.43 12.43 12.43 11.16
13.23 13.27 14.12 12.04 18.82 17.02 18.17 17.64
7.73
4.9
5.2
7.98 16.27
9.55
7.32
4.46
5.59
6.28
7.31
8.91
8.78
7.54
8.48 10.41
10.3
8.95
9.78
6.56 13.15
9.79 13.32 14.79
2007
17.32
15.24
12.43
18.17
7.32
8.48
13.32
2008
13.87
15.94
11.16
17.64
4.46
10.41
14.79
2007
1.81
0.98
1.43
1.33
0.26
0.49
2.04
2008
3.20
1.22
0.18
1.65
1.39
1.56
0.91
100%
100%
100%
100%
100%
100%
100%
TABLE2: WHEAT yield statistics from CFSA
Addis Ababa
Amhara
Harari
Oromiya
Somali
South Gonder
Tigray
TABLE3 = TABLE1 X TABLE2: Test of calculation of the estimated WHEAT yield for 2009
Addis Ababa
Amhara
Harari
Oromiya
Somali
Southern
Tigray
ETH001
ETH003
ETH007
ETH008
ETH009
ETH010
ETH011
1999
1.25
0.49
2.20
1.13
0.46
0.53
0.45
2000
1.07
0.15
0.18
0.57
0.22
0.34
0.18
2001
0.00
0.08
1.09
1.12
0.40
0.52
0.09
2002
0.98
1.10
0.14
2.07
0.72
1.80
0.71
2003
1.73
2.91
3.07
2.12
1.51
0.65
2.65
2004
1.48
2.71
0.48
2.14
0.91
0.77
1.71
2005
1.10
3.25
0.00
1.57
0.31
0.65
2.24
2006
1.45
0.32
0.92
2.27
0.68
1.21
0.90
Calculated yield
for 2009
14.08
13.21
9.70
15.97
6.85
8.53
11.87
Crop Yield and Vegetation Monitoring
Nairobi 26/10/2011
http://www.gmfs.info
Yield Forecasting
USERS
Participatory Development
Soil Moisture Monitoring
FAST Service
Nairobi 26/10/2011
http://www.gmfs.info
Agricultural Monitoring
Nairobi 26/10/2011
http://www.gmfs.info
•
•
•
Support to the Optimization of the National
Agricultural Survey Service
Agricultural Mapping
SAR Knowledge Transfer
ASO in Malawi
Nairobi 26/10/2011
http://www.gmfs.info
“ASO” service: “support to the optimisation of national surveys”.
In collaboration with JRC this service is focussed to consultancy on the introduction of area frame
sampling approaches , making use of EO data which i stechnically sound and sustainable in the
Malawian context
Point frame: 500 m spacing ,
- approx. 4 points per km2
1. EO data are used for the design of the sampling frame and
its realisation: location of sample points, interpretation and
classification, masking and stratification. Points vs. segments
to be interpreted and classified
Based on a simple LCLU
legend
Up to 58.000
points are visited
on the ground
(sampling rate
around 15 %)
2. EO data are used with GPS for survey optimisation and
execution, e.g. maps for identifying the sampled points, planning
of itineraries, control (synergies with field area measurements for
APES, etc. see also proposals by MoAFS on the use of IT , FAO)
3. EO data are also used as auxiliary variables (land cover/land
use classified images) to improve the accuracy in the
estimation of the crop acreage by means of ad hoc statistical
procedures.
ASO + AM North Sudan
Nairobi 26/10/2011
2 Scales, Medium Resolution, High Resolution
LCCS support
•
•
•
High Resolution Satellite Image coverage NKOR
Multi resolution segmentation of NKOR
Harmonized field work
•




Outputs
Data & processing support for current LCCS mapping
recent HR Satellite image mosaic on state level
Segmentation layer
Field work
http://www.gmfs.info
ASO + AM in North-Sudan
Nairobi 26/10/2011
GMFS fAPAR-EoG North Sudan 2010
http://www.gmfs.info
GMFS fAPAR-EoG North Sudan 2005
Change maps:
Context: there is a huge variability in the extent of
growth (EoG) in North Sudan, based on MERIS-FR
fAPAR images and analysis is made on this
difference. Maps can be used in support of the
Agricultural Survey
• Indication upon differences in growth activities
• Training on use of change maps is currently
ongoing (September 2011)
CFSAM support to FAO/WFP
http://www.gmfs.info
FAO publication explaining the importance of Environmental Remote Sensing indicators for
monitoring, using amongst others ESA-MERIS RR fAPAR (data processing + methodology = GMFS)
Nairobi 26/10/2011
http://www.gmfs.info
Thank you !
Questions?

similar documents