Agriculture Reference
In-Depth Information
Crop and Soil Water Models
The Decision Support System for Agrotechnology Transfer (DSSAT) (Tsuji et al.
1998) contains a suite of crop models widely used to simulate crop biomass and
yield as influenced by weather, soil, crop management, and crop genotype. The
primary crop models contained in DSSAT are CROPGRO for major grain legumes,
CERES for cereal crops, and SUBSTOR for crops with belowground storage
organs. The models were developed with a goal of minimizing the data needed
for prediction and control purposes. Simulations are executed on a daily time step
using solar radiation, temperatures (maximum and minimum), and precipitation,
thereby accounting for day-to-day variation that can be substantial. They are based
on empirical functions to estimate the soil water balance (runoff, drainage, evapo-
transpiration, soil storage) and biomass production. Input needs include soil physi-
cal and chemical properties for several depth increments as available in soil surveys.
Crop management input needs include date of sowing, plant population, dates and
quantities of nutrient and irrigation water applications, photoperiod, and crop geno-
type. Air temperature and photoperiod during critical phases of development deter-
mine plant ontogeny and biomass partitioning, and are based on plant genotype.
The DSSAT system has two options for simulating N balance and SOM: the origi-
nal SOM model (Godwin and Singh 1998) and a modified CENTURY model that
operates on a daily time increment and at soil depth increments that conform to
DSSAT (Gijsman et al. 2002).
The Environmental Policy Integrated Climate (EPIC) model was originally
designed to simulate soil erosion and its effects on soil fertility (Williams et al.
1984). EPIC has now evolved into a comprehensive agro-ecosystem model capable
of simulating biomass and yields of crops grown in complex rotations and under
diverse management practices such as tillage, irrigation, fertilization, and liming
(Williams 1995). The SOM module in EPIC uses processes similar to CENTURY
but with daily time increments and several soil depths. The soil water balance sub-
model is similar to that in DSSAT models.
The Agricultural Production Simulator (APSIM) model is another widely used
model (Keating et al. 2003) similar in detail to DSSAT and EPIC. APSIM was
developed with a modular structure to allow testing and use of various methods of
simulating several components of the soil, plant, and atmosphere system.
Rivington and Koo (2010), in a recent comprehensive meta-analysis of crop
modeling for climate change and food security, reported that DSSAT crop models
were the models most commonly used by various groups surveyed throughout the
world. The report revealed perceived model limitations and made suggestions for
model improvements based on user feedback.
Simulation of Crop Yield
Yield simulation in crop models is based on two processes: crop growth and devel-
opment. The fraction of total biomass partitioned into grain or other harvested bio-
mass is termed the economic yield. Crop simulations thus involve the two-step
