Geoscience Reference
In-Depth Information
commonly termed as a model. Thus, a model serves the pur-
pose of predicting the behaviour of crops in response to the
weather variations. The model simulates/limitates the behav-
iour of a real crop by predicting the growth of its components.
Crop growth is a very complex phenomenon and a product of
a series of complicated interactions of soil, plant and weather.
A crop growth model synthesises our insights into the physi-
ological and ecological processes that govern crop growth into
mathematical equations.
Several regression models have been developed by many
workers to predict the relationship with rice crop productivity
and its components. These models, when dealing with multi-
year time series, usually include a technology trend factor, thus
lumping everything other than climatic factors into one regres-
sor. In addition to climatic factors, there are a large number
of edaphic, hydrologic, biotic, agronomic and socio-economic
factors that influence crop growth and productivity. Crop mod-
els can accelerate inter-disciplinary knowledge utilisation in
agricultural research and development. These models present
an opportunity for assessing potential production in a region
and facilitate analysis of the sustainability options for agri-
cultural development, including planning of resource alloca-
tion. These approaches have been used in the recent past for
determining the production potential of a location knowing its
resources, germplasm and the level of available technology, in
matching agrotechnology with the resources of farmers and in
analysing the precise reasons for yield gap, in estimating crop
yield before the actual harvest and in studying short- and long-
term consequences of climatic variability and climatic change
on agriculture.
The use of various crop simulation models has been classi-
fied into three primary categories: (i) for research knowledge
synthesis, (ii) for crop system decision management and (iii)
and for policy analysis. Crop models have been used to assist
in the genetic improvement of crops by (i) determining optimal
genetic traits of plants for specific environment and (ii) predict-
ing the performance of new cultivars for specific environments,
thus reducing the number of locations or seasons of multilo-
cation breeding trials. The greater application of crop models
in agricultural research and development, however, requires a
simple, user-friendly modelling framework, whose inputs are
easily available/measurable. In addition, the framework should
provide a structure that can be easily integrated in the appli-
cation and not be very user-friendly. Preliminary results have
also indicated that some of them do not perform very well in
