Crop Growth Model (DSSAT/APSIM)
Decision Support System for Agrotechnology Transfer (DSSAT) / Agricultural Production Systems Simulator (APSIM) · Also known as: DSSAT, APSIM, Crop Simulation Model
Crop growth models are mechanistic simulation systems designed to predict crop development, biomass accumulation, and yield under varying environmental and management conditions. DSSAT (Decision Support System for Agrotechnology Transfer) and APSIM (Agricultural Production Systems Simulator) are the most widely used platforms, developed in the 1990s-2000s to support agronomic decision-making and climate adaptation research.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
Use crop models when: (1) you need to forecast yield under new varieties or management practices; (2) you lack long-term experimental data for a location or cultivar; (3) you want to stress-test adaptation strategies (irrigation, cultivar shift, planting date adjustment); (4) you integrate with regional climate scenarios for long-term planning. Requires good-quality soil and weather data. Less suitable for extreme climates or novel pests where mechanistic knowledge is incomplete.
Strengths & limitations
- Extrapolates beyond observed conditions using mechanistic first principles
- Integrates multiple stresses (water, nitrogen, temperature) in a single framework
- Enables rapid evaluation of management and variety options without field trials
- Supports long-term climate and policy impact assessments at regional scales
- Community-validated: thousands of publications and calibrations for major crops
- Requires extensive, accurate input data; garbage in → garbage out
- Parameter uncertainty can be large; model output sensitivity to poorly-measured soil properties is high
- Does not account for pests, diseases, or extreme events (hail, flooding) unless post-hoc adjustments applied
- Computationally intensive for large-area or ensemble simulations
- Mechanistic assumptions may fail under novel management or climate extremes not seen during development
Frequently asked
What is the difference between DSSAT and APSIM?
Both are mechanistic crop simulators with similar core physics (phenology, photosynthesis, water/nitrogen cycles), but differ in scope: DSSAT focuses on individual crops and is typically run field-by-field; APSIM is designed for crop sequences and rotations, includes livestock modules, and integrates soil organic matter cycling more explicitly. DSSAT uses a modular architecture (plug-and-play crop modules); APSIM is more integrated. Choice depends on whether you model a single crop or a farm system.
How sensitive are yield predictions to input data uncertainty?
Very sensitive. Uncertainty in soil water-holding capacity, cultivar parameters, and weather data can easily propagate to ±20-30% error in yield. Always run sensitivity analyses and ensemble runs (perturbed inputs) to quantify prediction bounds. Validation against independent field data from your region is essential before using forecasts for decisions.
Can these models predict crop failure or extreme events?
They predict stress indices (water or nitrogen limitation) that correlate with yield loss, but do not explicitly model pests, diseases, or catastrophic weather (hail, severe flooding). You can post-hoc apply damage functions (e.g., 'if excess water days exceed threshold, reduce yield by X%'), but the model itself assumes 'no pests' unless you modify code or calibrate empirical stress factors.
How much historical data do I need to calibrate a crop model?
At least 2-3 seasons of measured yield, phenology (anthesis, maturity), and biomass for the cultivar and soil type you model. If unavailable, use published parameter sets from similar cultivars and regions, then validate against any available local data. Regional parameter databases (ICASA standard) help fill gaps.
What is the computational cost of large-scale ensemble simulations?
A single-field, single-season DSSAT run takes seconds to minutes on a desktop. A 1000-pixel district-level ensemble under 100 climate scenarios (100,000 runs) might take hours to days on a standard server; distributed computing (cloud clusters) is practical for regional impact assessments.
Sources
- Jones, J. W., Hoogenboom, G., Porter, C. H., et al. (2003). The DSSAT cropping system model. European Journal of Agronomy, 18(3-4), 235-265. DOI: 10.1016/S1161-0301(02)00107-7 ↗
- Keating, B. A., Carberry, P. S., Hammer, G. L., et al. (2003). An overview of APSIM, a model designed for farming systems simulation. European Journal of Agronomy, 18(3-4), 267-288. DOI: 10.1016/S1161-0301(02)00108-9 ↗
- Passioura, J. B. (1996). Simulation models: science, snake oil, education, or engineering? Agronomy Journal, 88(5), 690-694. DOI: 10.2134/agronj1996.00021962008800050002x ↗
How to cite this page
ScholarGate. (2026, June 3). Decision Support System for Agrotechnology Transfer (DSSAT) / Agricultural Production Systems Simulator (APSIM). ScholarGate. https://scholargate.app/en/agronomy/crop-growth-model
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- Agrometeorological Yield ModelAgronomy↔ compare
- Penman-Monteith EquationAgronomy↔ compare
- Soil Moisture CurveAgronomy↔ compare