Systems Informatics and Analysis - Engineering and Science of Biomass Feedstock Production and Provision - page 200

on the plant and soil properties. The CENTURY model was later modifi ed into the
DAYCENT model which provided daily simulation time steps, primarily to study the
trace gases fl uxes (CO 2 , N 2 O, NO x , and CH 4 ) [ 46 ]. Lee et al. [ 47 ] used the DAYCENT
model to predict switchgrass yield in Central Valley, California, in the USA.
WIMOVAC is a mechanistic ecophysiological model that has been used to simu-
late various aspects of plant photosynthesis, especially the effects of global climate
change [ 21 ]. It differs from the other models in that it does not use the RUE approach
to calculate biomass. Such models are known as the enzyme-kinetic models that use
a semi-mechanistic understanding to calculate the photosynthesis and transpiration.
These models, by virtue of their modeling approach, can quantify the impact of
physiological trait improvement or ecosystem processes. WIMOVAC was adapted
and parameterized for Miscanthus × giganteus by Miguez et al. [ 48 ] and was shown
to realistically estimate the productivity at various sites in Europe. WIMOVAC has
a strong informatics component supporting the model. It allows the control of simu-
lation processes through a standardized Windows user interface and generates
results automatically. WIMOVAC is written in Visual Basic so that the user can
easily create user-friendly modules. Specifi cally, a number of controls are available
so that the user is able to handle automatic graphing, clipboard, and data-handling
facilities. In addition, the model uses the Windows Object Linking and Embedding
(OLE) technology for the transfer of simulation results from WIMOVAC to other
Windows-based applications. WIMOVAC can be installed with an optional Database
Management System (DBMS), which enables the exchange of experimental data
between the database fi les and the model modules for comparison and validation
purposes. WIMOVAC also includes a database of standard soil types, with the abil-
ity to enter information related to the characteristics of other “user-defi ned” soils.
Miguez et al. [ 49 ] have recently developed BioCro, which is also a mechanistic
model like WIMOVAC. However, it includes parameter estimation using optimiza-
tion routines and diagnostics and graphics that facilitate integration of fi eld experi-
mentation. Written in C and R with a number of user-friendly features built in, the
model has provided accurate yield predictions for Miscanthus × giganteus and
switchgrass. Feyereisen et al. [ 50 ] developed a plant-soil-atmosphere model RyeGro
to model the growth of cereal rye ( Secale cereale L.) as a winter cover crop and a
potential biomass feedstock. Recently, Feyereisen et al. [ 51 ] have used the model for
further analysis to estimate the potential yield in corn-soybean areas in the eastern
USA. However, instead of using the model itself, they have developed a quadratic
regression model using fi eld data for 30 sites across the region of interest. The inde-
pendent parameters in the regression model were precipitation, temperature, crop
rotation, and planting and harvest date. The simplifi ed regression model was then
used to determine yield for each county, thereby avoiding excessive simulations.

8.3.1.2 Importance of Data and Informatics

These biophysical crop growth models rely substantially on the availability of reli-
able data. Miguez et al. [ 26 ] have summarized the various useful data sources.