Crop Yield Modeling using SAR data and RS.ppt
Download as PPT, PDF0 likes9 views
The document details a crop yield modeling approach using the WOFOST model and spatial simulation for generating leaf area index (LAI) in real-time, incorporating various data inputs like weather and soil characteristics. It emphasizes the model's necessity for being both process-based and user-friendly, with a capability to run spatially across multiple grids. Calibration and validation results for wheat and rice yields in specific regions like Punjab are also presented, highlighting the model's effectiveness in simulating crop productivity.
Crop Yield Modeling using SAR data and RS.ppt
- 1. Crop Yield Modeling through Spatial Simulation Model
- 2. Grid LAI Generation Real time LAI (56 m) Average grid LAI (5 km)
- 3. LAI Forcing in WOFOST model Computing the correction factor CF= observed LAI through remote sensing/Model derived LAI on RS observation date 0 2 4 6 8 10 12 0 20 40 60 80 100 120 WLV WST TAGP LAI Days after emergence WLV, WST,TAGP in kg/ha; LAI in m 2 /m 2 WLV, WST,TAGP in t/ha; LAI in m 2 /m 2 Days after emergence 0 2 4 6 8 10 12 0 20 40 60 80 100 120 WLV WST TAGP LAI Days after emergence WLV, WST,TAGP in kg/ha; LAI in m 2 /m 2 WLV, WST,TAGP in t/ha; LAI in m 2 /m 2 Days after emergence Before forcing After forcing Date of forcing: 60 days after emergence 68 days after emergence
- 4. Spatial Wheat Yield for 2009-10 (5 km) Input Data Interpolated Weather Data Calibrated Crop Coefficient Sowing Date from Remote sensing LAI from Remote Sensing Rajasthan Punjab < 2.5 2.5-3.5 3.5-4.5 >4.5 Non-wheat Non wheat < 2 t/ha 2-3 t/ha 3-4 t/ha >4 t/ha
- 5. Exploring WARM (Water Accounting Rice model) for rice yield simulation WARM Downloaded from: http://www.robertoconfalonieri.it/software_download.htm WARM version 1.9.6
- 6. Data used for calibration Daily weather data Station latitude Rain fall, Tmax, Tmin and solar radiation Crop data Date of sowing GDDs to reach emergence GDDs from emergence to flowering GDDs from flowering to maturity Periodical LAI (4 times) Dry biomass at harvest and grain yield at harvest Soil data Bulk density OC Clay Sand Field capacity PWP KS Variety: PR 118 Location: Punjab Agricultural Univ, Ludhiana, Punjab, India Climate: Semiarid subtropic
- 7. Calibration Result LAI (m2/m2) Validation Result 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 170 180 190 200 210 220 230 240 250 260 270 Simulated Observed LAI (m2/m2) DOY 0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 biomass (t/ha) yield (t/ha) Simulated Observed • N.B. Two days delay in flowering was observed, Harvesting date was same as observed
- 9. Converting Point WOFOST Model to Spatial Mode WOFOST-exe Spatial data for weather Spatial data for crop Spatial data for soil Spatial data for sowing date Batch mode for all grid Output for all grid FORTRAN
- 10. Data Source 1. Real time Weather Data Maximum & Minimum Temperature Rainfall Daily Incoming Solar Radiation Wind speed Relative humidity IMD website (~80 station) IMD website (~80 station) Computed from temperature* Climatic normal Climatic normal 2. Soil Data Soil texture Soil moisture constants Hydraulic properties FAO soil map (1: 5M) 3. Management Data Planting/sowing date Irrigation (Date & Amount) Fertilizer (Date & Amount) Remote sensing (SPOT-VGT/INSAT-CCD) Not required for potential simulation 4. Crop data •Phenology •Physiology •Morphology Derived for a major variety in each state through calibration Input Data and Source *Solar radiation Where, Ah and Bh are the empirical constants and Ra is the extra terrestrial radiation (Duffie and Beckman,1980) h h a s B T T A R R ) ( min max (Hargreaves, 1985)
- 11. Crop Growth Simulation Model Inputs Process Output Weather (Temperature, Rainfall, solar radiation) Soil Parameters (Texture, depth, soil moisture, soil fertility) Crop Parameters (Phenology, physiology, morphology) Management (DOS, irrigation, fertilizer) Phenological Development CO2 Assimilation Transpiration Respiration Partitioning Dry matter Format Biomass, LAI, Yield Water Use Nitrogen Uptake
- 12. Choice of Simulation Models in FASAL • The model needs to be sufficiently process based to simulate crop productivity over a range of environments, while being simple enough to avoid the need for large amounts location specific input data • It should be possible to run the model spatially, in large number of grids. • The user interface of the model should be simple enough for multi-disciplinary users. • There needs to be a scope for assimilation of in-season remote sensing derived parameters. • The source code should be open for any modification WOFOST model has been chosen because of the availability of source code and relatively less input requirement