Quantifying agricultural and water Quantifying agricultural and water management practices from RS data management practices from RS data using GA based data assimilation using GA based data assimilation techniques techniques HONDA Kiyoshi HONDA Kiyoshi Asian Institute of Technology Asian Institute of Technology Mie University Mie University Amor V.M. Ines Amor V.M. Ines Texas A&M University Texas A&M University
26
Embed
HONDA Kiyoshi Asian Institute of Technology Mie University Amor V.M. Ines Texas A&M University
Quantifying agricultural and water management practices from RS data using GA based data assimilation techniques. HONDA Kiyoshi Asian Institute of Technology Mie University Amor V.M. Ines Texas A&M University. Introduction. Agriculture Monitoring acreage, sowing date, growth - PowerPoint PPT Presentation
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Quantifying agricultural and water management Quantifying agricultural and water management practices from RS data using GA based data practices from RS data using GA based data
assimilation techniquesassimilation techniques
HONDA KiyoshiHONDA KiyoshiAsian Institute of TechnologyAsian Institute of Technology
Mie UniversityMie University
Amor V.M. InesAmor V.M. InesTexas A&M UniversityTexas A&M University
Introduction• Agriculture
– Monitoring acreage, sowing date, growth– Monitoring impact of water availability to its impact– Optimize water use for higher yield
• Contents
– Crop Growth Dynamics observed by RS– Data Assimilation for SWAP model parameter
identification– Water use optimization– Mixed Pixel Modeling– High-Low RS Data Fusion for High Spatio -
Temporal Data– Future Plan
Fluctuation pattern of Non-irrigated rice
NDVI Fluctuation of Non-irrigated rice, Year 1999-2001
-1.00
-0.80
-0.60
-0.40
-0.20
0.00
0.20
0.40
0.60
0.80
1.00
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
NDVI Time Series (10 days composite)
ND
VI
Val
ue
Peak of Rainfall Peak of Rainfall Peak of Rainfall
2000 20011999
Non-irrigated/Rainfed rice field (20 th June 2003)
Landsat TM 08 Jan 2002: False Color Composite Non-irrigated area
(Map: 604632E, 1624227N)
Monitoring IrrigationPerformance through Crop Dynamics
Fluctuation pattern of Irrigated rice 2 crops/year
(Homogeneous)
NDVI Fluctuation of Irrigated rice 2 crops, Year 1999-2001
-1.00
-0.80
-0.60
-0.40
-0.20
0.00
0.20
0.40
0.60
0.80
1.00
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106NDVI Time Series (10 days composite)
ND
VI
Val
ue
Peak of Rainfall Peak of Rainfall Peak of Rainfall
2000 20011999
Irrigated rice, largecontinuous field (26 th April 2003)
Irrigated rice, large continuous field. (Map: 621930E, 1578132N)
Fluctuation pattern of Irrigated rice 3 crops/year
(Heterogeneous field)
NDVI Fluctuation of Irrigated rice 3 crops, Year 1999-2001
-1.00
-0.80
-0.60
-0.40
-0.20
0.00
0.20
0.40
0.60
0.80
1.00
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
NDVI Time Series (10 days composite)
ND
VI
Val
ue
Peak of Rainfall Peak of Rainfall Peak of Rainfall
2000 20011999
Irrigate rice 3 crops per year, discontinuous/small patchy
fields (Map: 611549E, 1620653N).
Irrigate rice 3 crops per year, growing stage (20 th June 2003)
Unclassified
Non-irrigated rice
Irrigated rice; 2 crops/year
Irrigated rice; 3 crops/year
Poor irrigated rice; 1 crop/year
Others
Provincial boundary
Irrigation zone
1999
2000
2001
Number of Cultivation in a Year
Suphanburi: 5 Classes
33
2211
Non Irri.Non Irri.
. Discrimination of Irrigated and . Discrimination of Irrigated and Rainfed Rice in a Tropical Rainfed Rice in a Tropical Agricultural System using SPOT-Agricultural System using SPOT-VEGETATION NDVI and Rainfall VEGETATION NDVI and Rainfall Data: Daroonwan Kamthonkiat, Data: Daroonwan Kamthonkiat, Kiyoshi Honda, Hugh Turral, Nitin K. Kiyoshi Honda, Hugh Turral, Nitin K. Tripathi, Vilas Wuwongse: Tripathi, Vilas Wuwongse: International Journal of Remote International Journal of Remote Sensing , pp.2527-2547, Vol. 26, No. Sensing , pp.2527-2547, Vol. 26, No. 12, 20 June, 200512, 20 June, 2005
Modeling and Simulation• RS is a useful tool to monitor the situation• Limitation: Only a snap shot• Modeling the phenomena on the ground
100x100 pixels will takes 7 months(30 min. * 100 * 100) -> Parallel computing
Mr. Shamim AkhtarMr. Shamim Akhtar
Future Development Future Development
• Expand the modeling from a few pixels to regional scale.Expand the modeling from a few pixels to regional scale.• Field Survey SupportField Survey Support
• Difficulty on field level calibration and validationDifficulty on field level calibration and validation• Field ServerField Server
• Soil MoistureSoil Moisture• Sowing and HarvestingSowing and Harvesting• R/C Flying MonitoringR/C Flying Monitoring
• Develop a flowDevelop a flow• local observationlocal observation• satellite observationsatellite observation• data collection/fusiondata collection/fusion• modeling & simulationmodeling & simulation• feed back to decision making feed back to decision making •( farmers to regional - national )( farmers to regional - national )
Develop a flow from monitoring, modeling, simulation and feed back to decision makings