Simulations for predicting the effects of future climates on forests and grasslands have various
levels of complexity depending upon the scale of what is being modeled, such as BACROS, BIOMASS,
FORGRO, and MAESTRO as reviewed by Ågren [35]. The most basic of these simulations model plant
physiology and how plants respond to the environmental changes [35]. These simulations include
models that predict aspects such as conversion of light within the canopy to assimilated carbon, water and
nutrient uptake, and partitioning of dry weight in the form of growth. Other processes within the plant that
they model generally include respiration, partitioning, reproduction, and senescence. All of these pro-
cesses are affected by environmental factors such as temperature; herbivory (grazing and insects); avail-
ability of water, nitrogen, and phosphorus; and atmospheric levels of CO 2. All these factors are integrated
over time to make future predictions based on the combination of initial conditions and ongoing environ-
mental changes. Many simulations also include feedback from the growing plants to the environment in
such ways as litter accumulation and degradation of fallen matter. These models give a physiological ba-
sis upon which other models, which encompass larger scales, can be used to predict such things as how
much C can be stored in the earth’s forest over time given changes in temperature and the atmospheric
CO 2 levels.
Agricultural simulations tend to be oriented to a smaller scale than the ecological simulations, putting
more emphasis on plants within a single crop. Examples of some of these include SOYGRO for soybeans,
PNUTGRO for peanuts, BEANGRO for dry bean [36], and GOSSYM for cotton [34]. However, they are
not limited to that and are gaining use for estimating potential yields in a marketing application as well as
predicting fertilizer losses to the environment [13,21] and changing climate effects on crops [37]. They
have more variability in input because many factors are controlled by the farmer such as water, nutrients,
and, for those designed for glasshouses, climate to some extent. The physiological models within the
simulation are generally the same as those in ecological applications, conversion of light to assimilated
carbon, uptake of water and nutrients, partitioning, respiration, and reproduction. The environmental fac-
tors affecting the process tend to be more local and smaller in scale, such as day-to-day climate, plant
spacing, and specific pests. The predictions (output) are generally dry matter accumulation specific to the
crop—for instance, fruit size and quantity—and estimates for timing of harvest. But they are certainly not
limited to this, as demonstrated by the use of the Erosion-Productivity Impact Calculator (EPIC) crop
model used to predict the relationship between soil erosion and soil productivity [38].
V. CONCLUDING REMARKS
Unfortunately, even with all the marvels of new computing power, computer simulations are only as good
as the programs written for them. The programs, in turn, are ultimately a reflection of our biological
understanding of plant growth. Some processes are more clearly understood than others, but a simulation
relies on the combination of the models used to predict the outcome. If one or more of the models is not
reliable over the range of input, results may be misleading [11]. Thus, it is necessary that all plant pro-
cesses be modeled as accurately as possible. However, it is usually the case that we do not have a deep
enough understanding of all of the myriad of biological processes occurring in a growing plant to even
begin to adequately quantify the outcome with mathematical equations.
REFERENCES
- R Brouwer. In: FL Milthorpe, JD Ivins, eds. The Growth of Cereals and Grasses. London: Butterworths, 1965,
pp 153–166. - R Brouwer, CT De Wit. In: WJ Whittington, ed. Root Growth. New York: Plenum, 1968, pp 224–244.
- LFM Marcelis, E Heuvelink, J Goudriaan. Sci Hortic 74:83, 1998.
- IF Wardlaw. New Phytol 116:341, 1990.
- A Troughton. J Br Grassland Soc 6:56, 1956.
- R Brouwer. Neth J Agric Sci 10:399, 1962.
- RL Davidson. Ann Bot 40:561, 1969.
- JHM Thornley. Ann Bot 36:431, 1972.
- JF Reynolds, JHM Thornley. Ann Bot 49:585, 1982.
- IR Johnson, JHM Thornley. Ann Bot 60:133, 1987.
- JHM Thornley. Grassland Dynamics: An Ecosystem Simulation Model. Wallingford, UK: CAB International,
912 DUBAY AND MADORE