Agriculture Reference
In-Depth Information
(Frank and Bauer 1982, 1984; Bauer et al., 1986;
Maas and Grieve 1990; Cutforth et al., 1992),
and this might be hypothesized to be the result
of lack of resources required for growth of the
organ.
As an illustration of this, leaf primordia initia-
tion showed little response to elevated CO
2
(Li
et al., 1997), yet a slight positive relationship
between CO
2
and the rate of leaf appearance
seems to exist (McMaster et al., 1999). Similarly,
differentiation and growth of the axillary bud,
resulting in the appearance of a new tiller, is
strongly regulated by water, nutrients, and carbo-
hydrate availability, as well as light intensity and
quality, although the relationship is sometimes
stronger for tillers than the main stem (Fraser
et al., 1982; Longnecker et al., 1993; Maas et al.,
1994). Further, younger tillers or secondary and
tertiary tillers often respond more to resource
limitations. One hypothesis is that these tillers are
most dependent on the whole plant for resources
that every shoot is competing for and are farthest
removed from access to the resource in terms of
vascular connections (water, nutrients, and so on)
or location within the canopy (i.e., light intensity
and quality).
CropSim (Ritchie and Otter 1985; Ritchie 1991;
Hunt and Pararajasingham 1995; Jones et al.,
2003; Hoogenboom et al., 2004), Sirius (Jamieson
et al., 1995, 1998), SUCROS (van Laar et al.,
1992), SWHEAT (van Keulen and Seligman
1987), and WINTER WHEAT (Baker et al.,
1985). A diversity of approaches exists in how
growth and development are simulated in these
wheat process-based models. However, for most
simulation models, the earliest approaches that
still remain popular today use the energy- or
carbon-driven approach, in which sunlight energy
is captured by the plant, converted to biomass,
and then partitioned within the plant. Develop-
mental detail in these models is usually minimal
and consists of calculating canopy LAI and growth
of general plant components of leaf, stem, roots,
and seeds. Often these models have quite accurate
grain yield predictions for variable environments
and cultivars.
Phenology modeling has been one of the most
successful components in existing wheat simula-
tion models, and the ability to simulate genotype
phenology across a broad range of environments
is quite reliable. Many alternative approaches are
available for use in predicting phenology, and
approaches differ in input requirements and
number of developmental stages simulated. All
models are based on the thermal time approach
(with the many variations that exist), with some
emphasizing the role of vernalization and photo-
period (e.g., DSSAT, AFRCWheat2, Sirius)
more than others. One area of divergence in phe-
nology submodels is whether leaf number or
strictly thermal time is used to estimate the time
interval between developmental stages. The
AFRCWHEAT2, Sirius, MODWht3 (Rickman
et al., 1996), and SHOOTGRO (McMaster et al.,
1992b; Zalud et al., 2003) models are based pri-
marily on the leaf number approach, while the
others use the more common approach of strict
thermal time.
Almost all models do not explicitly consider
the phenological responses to water defi cits
(and usually nutrient defi cits) for most, or all,
developmental stages (McMaster et al., 2009).
Exceptions include the SHOOTGRO model
DIGITAL TECHNOLOGIES FOR
WHEAT DEVELOPMENT
Probably more digital technologies are available
for modeling wheat growth and development
than for any other crop. Digital technologies
include simulation and regression models, deci-
sion support systems (DSS), web sites, and spe-
cifi c computer programs. This section will briefl y
focus on simulation models as related to wheat
development.
Many models simulate wheat production and
yield (McMaster 1993), and at the risk of omitting
many deserving models, several that have been of
historical signifi cance or played an infl uential role
in wheat modeling include AFRCWHEAT1/2
(Porter 1984, 1993; Weir et al., 1984), APSIM
(Asseng et al., 1998), DSSAT/CERES-Wheat/
