Geoscience Reference
In-Depth Information
related to crop responses such as yield and yield contributing
characters. The independent variables are weather parameters
derived from agrometeorological variances. The weighting coef-
ficients in these equations are obtained in the statistical manner
using standard statistical procedures. Such variables are used as
multivariable regression analysis. This statistical approach does
not easily lead to an exploration of the cause and effect or rela-
tionship, but it is a very practical approach for the assessment or
prediction of yield and its related parameters. The coefficients in
the statistical model and the validity of the estimates depend to
a large extent on the design of the model, as well as on the rep-
resentations of the input data. If the soil and climate conditions
and the cropping practice are fairly homogeneous over a spe-
cific area represented by the input data, or if soil and geography
are properly weighted in the equations, then it can be expected
that the coefficients and the estimates have a practical signifi-
cance for the assessment of the crop conditions or predictions
of yield for any specific area in question. Regression models are
attractive because of their simple and straightforward relation
between yield and one or more environment factors, but these
are not accurate enough to be used for other areas and other
crops (Chou and Chen, 1995). Despite this limitation, they are
used extensively for the prediction of a single crop yield over
a large region with a variety of soils, agronomic practices and
insect-disease problems. A combination of such factors is still
beyond the success of dynamic simulation models.
The following points may be incorporated to provide the
accurate forecast of crop growth and development by a statisti-
cal model:
1. In a statistical model, each predictor for the regression
equation must have a significant value, and the year must
be included in all equations reflecting the impact of tech-
nology. Also, the equation predictor for border district
must be sorted out.
2. A model for different climatic conditions within the
districts and ensemble technique for crop yield forecast
should be developed.
3. The ecological level should be included in the crop sim-
ulation model. At any altitude, the weather data can be
taken by multiplying the lapse rate at this altitude.
4. For validation of forecast, the trail/experimental field
should be at a controlled condition and also at different
climatic conditions/ecological conditions/district levels
so that it represents the farmers' field condition.
Search WWH ::




Custom Search