Abstract: Crop yield prediction is a paramount issue in
agriculture. The main idea of this paper is to find out efficient
way to predict the yield of corn based meteorological records.
The prediction models used in this paper can be classified into
model-driven approaches and data-driven approaches, according to
the different modeling methodologies. The model-driven approaches are based on crop mechanistic
modeling. They describe crop growth in interaction with their
environment as dynamical systems. But the calibration process of
the dynamic system comes up with much difficulty, because it
turns out to be a multidimensional non-convex optimization problem.
An original contribution of this paper is to propose a statistical
methodology, Multi-Scenarios Parameters Estimation (MSPE), for the
parametrization of potentially complex mechanistic models from a
new type of datasets (climatic data, final yield in many situations).
It is tested with CORNFLO, a crop model for maize growth. On the other hand, the data-driven approach for yield prediction
is free of the complex biophysical process. But it has some strict
requirements about the dataset.
A second contribution of the paper is the comparison of these
model-driven methods with classical data-driven methods. For this
purpose, we consider two classes of regression methods, methods
derived from linear regression (Ridge and Lasso Regression, Principal
Components Regression or Partial Least Squares Regression) and
machine learning methods (Random Forest, k-Nearest Neighbor,
Artificial Neural Network and SVM regression).
The dataset consists of 720 records of corn yield at county scale
provided by the United States Department of Agriculture (USDA) and
the associated climatic data. A 5-folds cross-validation process and
two accuracy metrics: root mean square error of prediction(RMSEP),
mean absolute error of prediction(MAEP) were used to evaluate the
crop prediction capacity.
The results show that among the data-driven approaches, Random
Forest is the most robust and generally achieves the best prediction
error (MAEP 4.27%). It also outperforms our model-driven approach
(MAEP 6.11%). However, the method to calibrate the mechanistic
model from dataset easy to access offers several side-perspectives.
The mechanistic model can potentially help to underline the stresses
suffered by the crop or to identify the biological parameters of interest
for breeding purposes. For this reason, an interesting perspective is
to combine these two types of approaches.
Abstract: The Northeast China (NEC) was the most important
agriculture areas and known as the Golden-Maize-Belt. Based on
observed crop data and crop model, we design four simulating
experiments and separate relative impacts and contribution under
climate change, planting date shift, and varieties change as well
change of varieties and planting date. Without planting date and
varieties change, maize yields had no significant change trend at
Hailun station located in the north of NEC, and presented significant
decrease by 0.2 - 0.4 t/10a at two stations, which located in the middle
and the south of NEC. With planting date change, yields showed a
significant increase by 0.09 - 0.47 t/10a. With varieties change, maize
yields had significant increase by 1.8~ 1.9 t/10a at Hailun and Huadian
stations, but a non-significant and low increase by 0.2t /10a at Benxi
located in the south of NEC. With change of varieties and planting
date, yields presented a significant increasing by 0.53- 2.0 t/10a. Their
contribution to yields was -25% ~ -55% for climate change, 15% ~
35% for planting date change, and 20% ~110% for varieties change as
well 30% ~135% for varieties with planting date shift. It found that
change in varieties and planting date were highest yields and were
responsible for significant increases in maize yields, varieties was
secondly, and planting date was thirdly. It found that adaptation in
varieties and planting date greatly improved maize yields, and
increased yields annual variability. The increase of contribution with
planting date and varieties change in 2000s was lower than in 1990s.
Yields with the varieties change and yields with planting date and
varieties change all showed a decreasing trend at Huadian and Benxi
since 2002 or so. It indicated that maize yields increasing trend
stagnated in the middle and south of NEC, and continued in the north
of NEC.