DSSAT 2021 Update - Speaker Deck

Slide 1

Slide 1 text

1 Improvement and Application of Agroecosystem Models: The DSSAT Experience 2021 ASA-CSSA-SSSA International Annual Meeting November 8, 2021 Gerrit Hoogenboom, Cheryl Porter, Vakhtang Shelia, Ken Boote, Upendra Singh, Willingthon Pavan, Jeff White and many others University of Florida & International Fertilizer Development Center [email protected] - www.GerritHoogenboom.com www.DSSAT.net

Slide 2

Slide 2 text

2 • IBSNAT Project; USAID, 1982-193 • Minimum Data Set Concept, 1983-1986 • Initial models included the CERES-Maize, CERES-Wheat and SOYGRO soybean models. • Data standards for compatibility of models (1986, 1994) • DSSAT v2.1 released in 1986 • ICASA, 1994 - 2003 • IDSSAT Version 3.5 released in 1998 (after project ended) • DSSAT Cropping System Model, DSSAT v4 released in early 2004 • DSSAT Version 4.02 in 2006, v4.5 in 2012, v4.6 in 2015 • DSSAT Version 4.7 in 2017, v4.7.5 in 2019 • DSSAT Version 4.8 in 2021 (?) Some Historical Notes on DSSAT

Slide 3

Slide 3 text

3 Original scientific publications

Slide 4

Slide 4 text

4 Original User’s Guides

Slide 5

Slide 5 text

5 Initial price: US $495 + shipping costs Updated price: US $195 + shipping costs Free download from DSSAT portal Free download & Open Source 3-clause BSD license Original Software

Slide 6

Slide 6 text

6 The DSSAT Crop Modeling Ecosystem www.DSSAT.net

Slide 7

Slide 7 text

7 DSSAT is not just a software program but an ecosystem of: • Crop model users • Crop model trainers • Crop model developers • Models for the most important food, feed, fiber, fuel, and vegetable crops • Tools and utilities for data preparation • Minimum data for model calibration and evaluation • ICASA Data standards • Application programs for assessing real-world problems

Slide 8

Slide 8 text

8 Country Downloads India 2493 China 1536 USA 1307 Pakistan 1127 Brazil 996 Iran 480 Argentina 473 Ethiopia 444 Indonesia 403 Thailand 328 Philippines 277 Germany 255 Peru 233 Colombia 212 Spain 197 Nigeria 179 Mexico 168 Italy 161 South Africa 156 United Kingdom 155 Total 15501 Countries 183

Slide 9

Slide 9 text

9 DSSAT Interface & Organization

Slide 10

Slide 10 text

10 Input and Output Files Format

Slide 11

Slide 11 text

11 Cropping System Model (CSM) Structure

Slide 12

Slide 12 text

12 Cropping System Model (CSM) Structure

Slide 13

Slide 13 text

Plant Modules

Slide 14

Slide 14 text

Cropping System Model (CSM) 14 Net Income Resource use Environmental Plant growth (grain, biomass, roots, etc.) Plant development (time to flowering, maturity, etc.) Yield Soil conditions (physical & chemical properties by layer) Weather (daily rainfall, solar radiation, max & min temperatures, …) Management events (sowing, irrigation, fertilizer, organic matter, tillage, harvest) Genetics (cultivar- specific parameters controlling growth and development) Crop Model Simulation

Slide 15

Slide 15 text

DSSAT Advancement • Improve model performance • Improve model functionality • Improve portability • Add new crops • Add new crop modules • Add new capabilities and process simulations • Add new tools and utilities • Develop new model applications • Add new experimental data that encompass new environments and/or new management scenarios

Slide 16

Slide 16 text

Genetics in Crop Models • Current crop models use empirical genotype specific parameters (GSPs) for cultivar environment interactions that are not linked to actual genes. These GSPs do not adequately include the genetic (G) and gene-by- environment interaction (G x E) effects on crop development, thus inherent limitations. • Genetics in the DSSAT Cropping System Model – Species coefficients – Ecotype coefficients – Cultivar coefficients • Bridging the gap between biotechnology, breeding and crop management

Slide 17

Slide 17 text

Simulation of plant responses to temperature and photoperiod 1.0 Temperature (°C) Temp base Temp Max Opt 1 Opt 02 Daylength (h) CSDL PPSEN Model 1/d =f(T) x f(D) Stagei = f(photothermal days) Cultivar Coefficients Species Coefficients CSM Genetic Coefficients

Slide 18

Slide 18 text

18 Predicting time to flowering for dry bean based on QTL and Environmental Variables Stand Alone Gene-Based Model CSM-CROPGRO-Dry Bean

Slide 19

Slide 19 text

19 Gene-Based Model Integration diagram

Slide 20

Slide 20 text

20 Input Files ● The reading of the file '.GEN' uses the line code (VAR#) and the TF (QTLs) with values of 1 and -1. For each experiment 13 RILs/genotypes were used. BNGRO047.GEN CTFL1101.BNX ● If GENF is equal to Y the Gene- Based model can be executed. Otherwise, it will not affect the simulation.

Slide 21

Slide 21 text

21 Predicted versus Observed Flowering Original Model Gene-based Model

Slide 22

Slide 22 text

22 Final Simulated Yield Gene-based module integrated into the Main Model Original Model

Slide 23

Slide 23 text

Crop simulation models as a tool for yield forecasting Yield Forecasting • DSSAT Crop Simulation Model • Input data requirements – Access to current and historical daily weather data – Local soil characteristics – Crop management • Benefits – Predict yield directly – No dependence on satellite data – Can be run locally – free!

Slide 24

Slide 24 text

Yield forecasting for wheat – Case Study Yield Forecasting • Model evaluation - Akmola region – Observed 1981 to 2015 – Simulated 1984 to 2018 Simulated & Observed Historical Yield Historical Regional & FAO Yield

Slide 25

Slide 25 text

2019 Wheat Yield Forecast • Ensemble yield forecast - May 1, 2019 • Current weather data January 1 to May 1, 2019 for Petropavlovsk • Historical weather data for 1984 to 2018 • Each line represents a different weather ensemble  2019 weather data  May 1  Historical weather data 

Slide 26

Slide 26 text

2019 Wheat Yield Forecast • Ensemble Forecast – Petropavlovsk • Monthly forecast dates from April 1 to August 1 • Predicted wheat yield variability and uncertainty is reduced for later forecast dates • Accurate prediction on July 1 • Significant change between June 1 and July 1 forecast dates • The forecast improves as more current weather information becomes available

Slide 27

Slide 27 text

2019 Wheat Yield Forecast • Monthly Ensemble Forecast – Petropavlovsk • Low rainfall in May & high rainfall in June

Slide 28

Slide 28 text

Slide 29

Slide 29 text

Crop Yield Forecasting Using the CCAFS Regional Agricultural Forecasting Toolbox (CRAFT) in Ethiopia Kindie Tesfaye, Esayas Lemma, Robel Takele, Vakhtang Shelia , Addisu Dabale, Pierre C. Sibiry Traore, Gerrit Hoogenboom, Dawit Solomon GHACOF 58, May 27, 2021

Slide 30

Slide 30 text

Slide 31

Slide 31 text

Slide 32

Slide 32 text

32 Perennial Forage Model-Alfalfa

Slide 33

Slide 33 text

33 CROPGRO-Strawberry

Slide 34

Slide 34 text

34 CROPGRO-Carinata

Slide 35

Slide 35 text

35 CROPGRO-Quinoa

Slide 36

Slide 36 text

36 MANIHOT-Cassava How do we model cassava?

Slide 37

Slide 37 text

37 MANIHOT-Cassava

Slide 38

Slide 38 text

38 Energy balance

Slide 39

Slide 39 text

39 Time Series Calibration

Slide 40

Slide 40 text

40 Coupling Pests and Diseases

Slide 41

Slide 41 text

Challenges DSSAT Future? • Free software • Mixed language programming • Open Source for the model and tools • New tools and applications • New platforms (Linux, iOS, Web) • Linkages with other models using wrappers and docking technologies • Driven by the availability of resources • Driven by the interest of the user community

Slide 42

Slide 42 text

Challenges DSSAT Future? • New crops: safflower, alfalfa, teff, sugarbeet, quinoa, chia, carinata, strawberry, hemp • New crop modules: CROPGRO-Perennial, NWheat-Teff, CERES-Rice-Teff, SAMUCA- Sugarcane • New processes: – Salinity – Plant P & Soil and plant K – 2D soil model • Greenhouse Gas Emissions • Pest and disease coupling

Slide 43

Slide 43 text

DSSAT Portal www.DSSAT.net

Slide 44

Slide 44 text

DSSAT Development Sprint July 26-30, 2021 @ International Fertilizer Development Center