Agriculture Reference
In-Depth Information
tested, can provide an appropriate level of detail needed for assessing many aspects
of crop production. Functional type models are now routinely used in DSSs.
Examples of Models to Simulate Crop Ecosystems
Here, we provide a brief summary of the most widely adopted models to simulate
soil organic matter, soil water, and biogeochemical fluxes and how these variables
affect crop growth in response to land management.
Soil Organic Matter and Gas Emission Models
One of the most widely used soil organic matter (SOM) models is the CENTURY
model developed by Parton et al. (1988) to simulate long-term (10-1000 years)
patterns in surface SOM dynamics, plant production, and nutrient cycling (N, phos-
phorus [P], and sulfur [S]). The model uses a monthly time step with monthly aver-
age maximum air temperature (at 2 m height), monthly precipitation, soil texture
(sand, silt, and clay content), nutrient and lignin content of dead plant material, and
atmospheric and soil inputs of N. Plant material is divided into structural (difficult
to decompose) and metabolic (readily decomposable) fractions. Soil organic mat-
ter is divided into active, slow, and passive pools. Decomposition of plant material
and SOM is a function of soil water and temperature, and is influenced by soil type
and the C/N ratio of decomposing material. A complete description of the N and
soil C model is presented by Parton et al. (1987). The plant submodel is highly
simplified, using only inputs of stored water at planting, precipitation during plant
growth, a fixed water-use efficiency, and available soil N. Partitioning of C and N
into various plant components is performed using fixed partitioning coefficients.
While emphasizing long-term organic matter dynamics, the CENTURY model
lacks details important for short-term soil water and crop growth dynamics as well
as soil management other than N inputs.
A daily incrementing modification of CENTURY called NGAS-DAYCENT or
simply DAYCENT (Parton et al. 1996, 1998, 2001; Del Grosso et al. 2000a, b)
simulates trace gas fluxes of nitric oxide (NO), nitrous oxide (N
2
O), and dinitrogen
(N
2
) from soils as well as methane (CH
4
) formation and oxidation. The DAYCENT
model has been used to simulate national N
2
O emissions in the United States from
major cropped soil regions (Del Grosso et al. 2006). Soil water calculations are
performed at hourly time steps, which may not match other processes simulated at
daily time steps (Basso et al. 2010, 2011).
Another mechanistic SOM and gas emission model is the DeNitrification-
DeComposition (DNDC) model. The DNDC model has been used for estimating
N
2
O and CH
4
emissions from agricultural lands (Li 1995, 2000), but it requires
substantially more input detail than other models.
While providing much detail about soil greenhouse gas emissions and carbon
dynamics, these three models lack detail for estimating crop yield. Thus, they are
useful for simulating SOM and soil greenhouse gas dynamics but have limited util-
ity for evaluating the sustainable production of food, fuel, and fiber.
