Development and Application of DSSAT - Past, Current, and Future

Slide 1

Slide 1 text

Development and Application of the Decision Support System DSSAT “Past, Current, and Future” A tribute to Professor Jones

Slide 2

Slide 2 text

2 DSSAT: Decision Support System for Agrotechnology Transfer  How did we start?  Where have we come from?  Where are we now?  Where are we planning to go?

Slide 3

Slide 3 text

2 DSSAT: A Vision  Benchmark Soils Project  International Benchmark Sites Network for Agrotechnology Transfer  International Consortium for Agrotechnology Transfer  DSSAT Foundation

Slide 4

Slide 4 text

2 Benchmark Soils Project

Slide 5

Slide 5 text

2 International Benchmark Sites Network for Agrotechnology Transfer (IBSNAT) Project

Slide 6

Slide 6 text

2 IBSNAT Project  Funded by the U.S. Agency for International Development  1982 – 1993 • No continuation of financial support since 1993  University of Hawaii (Dr. Uehara & Dr. Tsuji) • Michigan State University • University of Florida • International Fertilizer Development Center (IFDC)

Slide 7

Slide 7 text

2 IBSNAT - Approach  To provide support for developing countries based on a systems analysis approach  To use computer models and data in agricultural sciences in contrast to a traditional agronomic approach (trial and error)  Understanding  Prediction Control & Manage (H. Nix, 1983)

Slide 8

Slide 8 text

2 IBSNAT - Outcomes  A global network of: • Crop model developers • Crop model users  Data standards for model applications  Data sets for model evaluations across many environments  A computerized product for decision support: DSSAT

Slide 9

Slide 9 text

2 Decision Support System for Agrotechnology Transfer  DSSAT:  A single software package that facilitates the application of crop simulation models in: Research Teaching Decision making Outreach & service Policy and Planning

Slide 10

Slide 10 text

2 DSSAT Crop Simulation Models  Based on understanding of plants, soil, weather and management interactions • Morphological and phenological development • Photosynthesis, respiration, partitioning and growth • Root water and nitrogen uptake • Stress effects on growth processes  Predict growth, yield, timing (Outputs)

Slide 11

Slide 11 text

2 Soil Conditions Weather data Model Simulation Crop Management Genetics Growth Development Yield

Slide 12

Slide 12 text

2 Soil Conditions Weather data Model Simulation Crop Management Genetics Growth Development Yield Net Income Environmental Impact Natural Resource Use

Slide 13

Slide 13 text

2 DSSAT - software  A software program that includes: • Crop simulation models • Utilities and tools for data handing  Experimental, soil, weather, economics • Application programs  Seasonal, crop rotational, and spatial analysis

Slide 14

Slide 14 text

2 DSSAT - components  A simple shell to access all programs, tools and utilities  15+ programs, utilities and tools • Mixture of languages, including Fortran, Visual Basic, Delphi, Excel, etc.  Experimental, crop, weather, soil, pest, genotype and economic data files  Developed by scientific programmers

Slide 15

Slide 15 text

2 DSSAT GRO family of models  SOYGRO : Soybean crop simulation simulation model

Slide 16

Slide 16 text

2 DSSAT GRO family of models  SOYGRO : Soybean simulation model  PNUTGRO : Peanut simulation model  BEANGRO : Common bean simulation model • 1989  CROPGRO : A generic grain legume model for soybean, peanut, common bean, chickpea, cowpea and other crops • 1994 + 1998

Slide 17

Slide 17 text

2 Modular Models

Slide 18

Slide 18 text

2 Cropping System Model (CSM) 2003  Modular Crop Simulation Model  CROPGRO module for soybean, peanut, common bean and other grain legumes

Slide 19

Slide 19 text

2 Cropping System Model (CSM) 2003  Modular Crop Simulation Model  CROPGRO module for soybean, peanut, common bean and other grain legumes  CROPGRO module for cotton  CROPGRO module for tomato, bell pepper, green beans and cabbage  CROPGRO module for bahia and brachiaria

Slide 20

Slide 20 text

2 Cropping System Model (CSM) - 2003  CROPGRO  CERES modules for maize, rice, sorghum and millet  CROPSIM-CERES for wheat and barley  Substor module for potato  CANEGRO module for sugarcane  IXIM module for maize  ORYZA2000 for rice  APSIM-Asseng for wheat

Slide 21

Slide 21 text

2 DSSAT Cropping System Model  CERES • Maize • Wheat • Sorghum • Rice • Barley • Millet  [Other crops] • Potato • Sweetcorn • Sugarcane • Cassava • Taro/Tanier • [Sunflower]  CROPGRO (Legumes) • Soybean • Peanut • Common bean • Faba bean • Chickpea • Cowpea • Velvet bean  CROPGRO (Other) • Cotton • Tomato • Bell Pepper • Cabbage • Green bean • Canola • Forages

Slide 22

Slide 22 text

2 Cropping System Model (CSM) – Genetic Coefficients  Species parameters and functions • Defines the response of a crop to environmental conditions, including temperature, solar radiation, CO2 and photoperiod, as well as plant composition and other functions and parameters.

Slide 23

Slide 23 text

2 Cropping System Model (CSM) – Genetic Coefficients  Ecotype coefficients • Defines coefficients for groups of cultivars that show similar behavior and response to environmental conditions.  Cultivar coefficients • Cultivar and variety specific coefficients, such as photothermal days to flowering & maturity, sensitivity to photoperiod, seed size, etc.

Slide 24

Slide 24 text

2 1.0 Temperature (°C) Tb TM T01 T02 Daylength (h) CSDL PPSEN Model 1/d =f(T) x f(D) Stagei = f(photothermal days) Cultivar Coefficients Species Coefficients Genetic Coefficients Response to Temperature and Photoperiod

Slide 25

Slide 25 text

2 DSSAT Cropping System Model  Google Scholar: cited by 495

Slide 26

Slide 26 text

2 Crop Simulation Models  Require information (Inputs) • Field and soil characteristics • Weather (daily) • Management • Cultivar characteristics  Model calibration for local variety  Model evaluation with independent data set  Can be used to perform “what-if” experiments

Slide 27

Slide 27 text

2 Linkage between experimental data and simulations  Model credibility and evaluation  Experimental data needs

Slide 28

Slide 28 text

2 DSSAT - Minimum Data Set

Slide 29

Slide 29 text

2 DSSAT - Minimum Data Set

Slide 30

Slide 30 text

2 DSSAT Minimum Data Set  Level 1 - Operate crop simulation models  Level 2 - Evaluate model performance - Calibrate, estimate parameters  Level 3 - Develop models (Maximum) 

Slide 31

Slide 31 text

2 DSSAT Minimum Data Set  Standard files, formats designed, documented, and implemented in DSSAT and its crop models  ICASA Data Standards for crop simulation models  AgMIP Data Base for crop model comparison, improvement and impact assessment

Slide 32

Slide 32 text

•Yield 0 2000 4000 6000 8000 D ry W eig h t ( k g /h a ) 175 200 225 250 275 300 Day of Year Grain - Irrigated Total Crop - Irrigated Total Crop - Not Irrigated Grain - Not Irrigated Simulated and Measured, Soybean Gainesville, FL 1978

Slide 33

Slide 33 text

2 G eorgia Variety Trial Soybean Crop M odel Predictions B ryan = 0.9255x + 249.76 H utcheson = 1.1099x - 194.79 500 1000 1500 2000 2500 3000 3500 4000 4500 500 1000 1500 2000 2500 3000 3500 4000 5-site Yield A verage (kg/ha) V ariety Yield (kg/ha) H utcheson B ryan

Slide 34

Slide 34 text

2 Observed Yields 0 500 1000 1500 2000 2500 3000 3500 4000 25 27 29 31 33 Max Temp Average (C) Yield (kg/ha) Simulated Yields 0 500 1000 1500 2000 2500 3000 3500 4000 25 27 29 31 33 Max Temp Average (C) Yield (kg/ha) Relationship between seasonal average max temperature and soybean yield: Observed (Georgia yield trials) and Simulated

Slide 35

Slide 35 text

2 Observed Yield vs. Rainfall (mm/d) 0 500 1000 1500 2000 2500 3000 3500 4000 0 2 4 6 8 Rainfall (mm/d) Yield (kg/ha) Simulated Yield vs. Rainfall (mm/d) 0 500 1000 1500 2000 2500 3000 3500 4000 0 2 4 6 8 Rainfall (mm/d) Yield (kg/ha) •Relationship between seasonal rainfall and soybean •yield; Observed (Georgia yield trials) and Simulated

Slide 36

Slide 36 text

2 DSSAT Applications  Diagnose problems (Yield Gap Analysis)  Precision agriculture • Diagnose factors causing yield variations • Prescribe spatially variable management  Water and irrigation management  Soil fertility management  Plant breeding and Genotype * Environment interactions  Yield prediction for crop management

Slide 37

Slide 37 text

2 DSSAT Applications  Adaptive management using climate forecasts  Climate variability  Climate change  Soil carbon sequestration  Land use change analysis  Targeting aid (Early Warning)  Risk insurance (rainfall)  Investment potential risks  Biofuel production

Slide 38

Slide 38 text

2 DSSAT - distribution  DSSAT v2.1 1989 589  DSSAT v3.0 1994 433  DSSAT v3.1 1996 138  DSSAT v3.5 1998 429+  DSSAT v4B 2002&2004 Workshops @ UGA  DSSAT v4.02 2005 293+  DSSAT v4.02 2006 Workshop @ UGA  DSSAT v4.5B 2008&2010 Workshops @ UGA  DSSAT v4.5B 2011 Workshop @ IFDC  DSSAT v4.X 2012 ?  ~2000 users in over 90 countries

Slide 39

Slide 39 text

2 Who uses DSSAT and its associated crop models?  Researchers  Educators  Extension Service and other farmer advisors  Private sector  Policy makers  Farmers?

Slide 40

Slide 40 text

2 Crop Simulations Farmers, consultants and extension www.AgroClimate.org

Slide 41

Slide 41 text

2 Concerns  Un-experienced users • Training workshops  ICRISAT 2011  Thailand 2011  DSSAT 2012 • User support  Manuals & Documentation  Listservers  Tutorials (?)  E-mail (?)

Slide 42

Slide 42 text

2 Concerns  Input data • Unavailable • Poor quality • Uncertainty  Model calibration and evaluation • Ignored • Poor input data • Poor observed data

Slide 43

Slide 43 text

2 Concerns  Model improvement • Existing literature • New data sets • Personnel support (?) • Grant support (?)

Slide 44

Slide 44 text

2 DSSAT Future?  Open Source? • Tools • Applications • Model source code & modules • Personnel support • Software support • Documentation

Slide 45

Slide 45 text

2 DSSAT Future?  Increasing interest in the application of models  Increasing interest by granting agencies to support model improvements and applications  Decrease in personnel due to retirements and career changes  Need for collaboration among modeling groups?