Predicting Wheat Growth Using
the CSM-Cropsim-CERES -Wheat Crop Model
M. J. Ottman
Summary
CSM-Cropsim-CERES -Wheat is a crop growth model that predicts crop
development stages, among other things, using genetic coefficients for
vernalization and photoperiod. We used this model to predict flowering date for
12 durum varieties seeded in trials at Maricopa and Yuma from 1998 to 2006.
The difference between simulated and measured flowering date averaged 4 days
without genetic coefficients and improved to 3.5 days if genetic coefficients for
flowering and vernalization were included for each variety.
Introduction
Computer models of the development of small grains are used to predict when the crop reaches various stages of
development. The small grain advisory uses a model based on temperature developed at Maricopa. This model
works well at Maricopa but adjustments are required at other locations. A model that may predict crop growth
stages in Arizona more accurately is CSM-Cropsim-CERES Wheat. This model predicts crop development using a
physiologically more complete approach that includes effects of photoperiod and vernalization (cold requirement for
flowering) and considers differences in comparative earliness or lateness of cultivars.
Procedure
The ability of CSM-Cropsim-CERES Wheat v.4.5 (hereafter refered to as CERES Wheat) to predict flowering date
was tested using existing data from Arizona. Flowering dates for 12 durum varieties were obtained or estimated
from small grain variety evaluations at Maricopa and Yuma. These studies were conducted by the University of
Arizona, Western Plant Breeders (WPB), and World Wide Wheat (WWW). Flowering dates were estimated by
adding 7 days to heading date for the studies conducted by WPB and WWW. Weather from the nearest AZMET
station and crop production data was entered into the CERES-Wheat program.
Genetic coefficients for vernalization (P1V) and photoperiod (P1D) were estimated using GenCalc2 (Table 1), a tool
within the Decision Support System for Agtotechnology Transfer (DSSAT, version 4.5), while keeping the ecotype
coefficients constant (see Table 2). Genetic coefficients adjust for differences in variety response to vernalization
and day length. Ecotype coefficients adjust for heat units among growth phases in different growing regions.
Generic genetic coefficients for P1V and P1D were obtained by averaging these coefficients over varieties.
The model was then tested under three conditions: 1) No genetic coefficients, or strictly a heat unit approach,
where P1V and P1D were set to 0, and adjusting the ecotype coefficients until the mean simulated and measured
flowering dates were nearly equivalent, 2) Generic genetic coefficients, where P1V and P1D were equal for all
varieties, and the ecotype coefficients were adjusted until the mean simulated and measured flowering dates were
nearly equivalent, and 3) Genetic coefficients, P1V and P1D, were used for each variety.
CERES-Wheat calculates heat units using the mean daily temperature with 0 and 26° C as base and ceiling
Forage and Grain Report (P-156) October 2008
12
temperatures, respectively. These threshold temperatures are used when the average daily temperature is above or
below the thresholds. More specifically, the minimum and maximum daily heat unit accumulation is 0° and 26° C
day. In this exercise, we averaged the maximum and minimum temperature from AZMET to obtain average daily
temperature.
Discussion
Genetic coefficients for vernalization and photoperiod for 12 durum varieties is presented in Table 1. The genetic
coefficients for vernalization vary from about 18 to 24 and for photoperiod vary from 49 to 58. These coefficients
are a bit higher than what may be expected for spring wheat, but certainly within the realm of possibilities. Spring
wheat supposedly has no vernalization requirement, but most spring wheat respond to cold weather by heading
somewhat earlier. Photoperiod sensitivity is also relatively low in spring wheat because of shuttle breeding, that is,
selecting in a low latitude location in the winter/spring and a high latitude location in the spring/summer.
Nevertheless, development rate is slowed somewhat by photoperiods shorter than the threshold, which is assumed to
be 20 hours by CERES-Wheat.
CERES-Wheat was run with no genetic coefficients for vernalization or photoperiod, with generic genetic
coefficients, and with genetic coefficients specific to each durum variety (Table 3 and A1). When the model was
run with no or generic genetic coefficients, the average error, or difference between simulated and measued days
after planting to flower, was about 4 days. The inclusion of genetic coefficients specific to each variety decreased
the error by about half a day, to 3.5 days. The improvement in the ability of the model to predict flowering date by
including genetic coefficients for each variety is not great for three reasons, possibly. First, the durum varieties
tested do not differ much. The range in flowering dates for these varieties averages about 7 days. The range in
flowering dates for barley varieties grown in the state ranges about 1 month. Second, the range in planting dates is
relatively small for the data tested, about 1 month for each location. Under commercial conditions, the range of
planting times may differ by 2 months. Third, the environmental conditions between Maricopa and Yuma do not
differ as much as between Arizona other states in the US and the Mediterranean area, where some of these varieties
may also be grown.
Some bias in the model exists in prediction of flowering time at Maricopa and Yuma. The model overestimates the
time to flowering at Maricopa by 0.6 to 1.4 days, and underestimates the time to flowering at Yuma by 1.2 to 1.4
days. Using genetic coefficients narrows the sum of the bias over locations from 2.8 to 1.8 days.
One of the advantages of CERES-Wheat over other models may be that method it uses to calculate heat units. In
this model, heat units are calculated simply from an average temperature. Other models adjust minimum and
maximum temperatures relative to temperature thresholds, while CERES-Wheat makes adjustments based on mean
temperature which rarely approach the threshold. So, heat units may be over- or under-estimated depending on
method of calculation.
In conclusion, using genetic coefficients in CERES-Wheat improved the prediction of days to flowering in the data
tested, but the improvement was not great. We expect that the improvement would have been greater if a wider
range of planting dates and locations had been used in this analysis. We will continue testing the model for use in
Arizona and expect to use the model for future reporting on wheat development.
Acknowledgments
Financial support for this project was received from the Arizona Grain Research and Promotion Council. A special
work of thanks is due to Jeff White, Plant Physiologist at the USDA Arid Land Agricultural Research Center at
Maricopa, for instructing me on the use of CERES-Wheat and preparing many of the files for me. I wish to thank
the following individuals for testing the varieties at their respective sites: Kim Shantz of Western Plant Breeders
and the late Rex Thompson of World Wide Wheat. The technical assistance of Mary Comeau and Mike Sheedy is
greatly appreciated.
Forage and Grain Report (P-156) October 2008
13
Table 1. Genetic coefficients for vernalization and photoperiod for 12 durum varieties generated using GenCalc2 in
CERES-Wheat. The ecotype coefficients were set as follows: P1 = 400, P2 = 285, P3 = 240, P4 = 300 (Table 2).
These coefficients were generated from Maricopa and Yuma data from 1998-2007 (Table 3).
Variety
Vernalization coefficient (P1V)
Photoperiod coefficient (P1D)
Alamo
Crown
Duraking
Kofa
Kronos
Matt
Mohawk
Ocotillo
Orita
Platinum
Sky
WPB881
23.56
22.57
23.69
21.12
18.37
22.73
21.12
22.88
24.42
22.66
22.51
24.47
50.21
58.34
56.05
49.28
53.25
49.28
49.28
51.80
59.62
55.74
52.97
51.47
AVG
22.51
53.11
Table 2. Ecotype and genetic coefficients used when running CERES-Wheat with no genetic coefficients, generic
or average genetic coefficients, and genetic coefficients specific to each variety. When the model was run with no
or generic genetic coefficients, the ecotype coefficients were generated using suggested coefficients in CERESWheat and adjusting these coefficients until there was no bias in prediction of days to flowering. When the model
was run with genetic coefficients specific to each variety, ecotype coefficients suggested by the model were used,
and genetic coefficients were generated with GenCalc2, a tool within CERES-Wheat.
Coefficient
P1
P2
P3
P4
P1V
P1D
PHINT
Definition
Ecotype (specific to region)
End juvenile to double ridges (oC.d)
Double ridges to end leaf growth (oC.d)
End leaf growth to end spike growth (oC.d)
End spike growth to end grain fill lag (oC.d)
Genetic (specific to varieties)
Vernalization - days at optimum vernalizing
temperature required to complete vernalization
Photoperiod - % reduction in development rate
in a photoperiod 10 hour shorter than the
threshold relative to that at the threshold
Phyllochron interval - interval between
successive leaf tip appearances (oC.d)
Forage and Grain Report (P-156) October 2008
No
genetic
coefficients
Generic
genetic
coefficients
Genetic
coefficients
425
310
270
330
400
285
240
330
400
285
240
300
0
22.51
0
53.11
Variety specific
(see Table 1)
Variety specific
(see Table 1)
95
95
95
14
Table 3. Mean days after planting (DAP) for flowering of durum varieties as measured and averaged at various
locations and years, and as simulated by CERES-Wheat using no genetic coefficients in the model (P1V and P1D set
to 0), generic or average coefficients for all varieties, and genetic coefficients specific to each variety. The error is
the absolute error, or the difference between simulated and measured DAP for flowering. The data in this table is
averaged from Table A1.
Mean (DAP)
Absolute error (DAP)
Root mean square error (DAP)
Root mean square error (% of measured DAP)
Forage and Grain Report (P-156) October 2008
No genetic
coefficients
Measure Simulate
flower flower
Error
DAP
----- DAP -----
Generic genetic
Genetic
coefficients
coefficients
Simulate
Simulate
flower
flower
Error
Error
----- DAP --------- DAP -----
112.25
112.21
112.31
0.06
3.96
4.33
4.87
-0.04
3.95
4.34
4.87
111.95
-0.30
3.52
3.90
4.37
15
Appendix
Table A1. Days after planting (DAP) for flowering of durum varieties as measured at various locations and years,
and as simulated by CERES-Wheat using no genetic coefficients in the model (P1V and P1D set to 0), generic or
average coefficients for all varieties, and genetic coefficients specific to each variety. The error is the absolute error,
or the difference between simulated and measured DAP for flowering.
No genetic
coefficients
Harvest
Planting Measure Measure Simulate
Location year
Variety
date
flower flower flower
Error
DAP
----- DAP -----
Generic genetic
Genetic
coefficients
coefficients
Simulate
Simulate
flower
Error
flower
Error
----- DAP --------- DAP -----
Maricopa 1998 Crown
Duraking
Kofa
Kronos
Mohawk
Ocotillo
Platinum
WPB881
11/21
11/21
11/21
11/21
11/21
11/21
11/21
11/21
04/03
03/31
03/27
03/26
03/25
03/29
03/30
03/25
133
130
126
125
124
128
129
124
130
130
130
130
130
130
130
130
-3
0
4
5
6
2
1
6
128
128
128
128
128
128
128
128
-5
-2
2
3
4
0
-1
4
132
130
125
126
125
127
130
127
-1
0
-1
1
1
-1
1
3
Maricopa 1998 Crown
Duraking
Kofa
Kronos
Mohawk
Ocotillo
Platinum
WPB881
12/17
12/17
12/17
12/17
12/17
12/17
12/17
12/17
04/13
04/12
04/07
04/10
04/09
04/10
04/13
04/10
117
116
111
114
113
114
117
114
118
118
118
118
118
118
118
118
1
2
7
4
5
4
1
4
117
117
117
117
117
117
117
117
0
1
6
3
4
3
0
3
120
118
114
115
114
116
118
116
3
2
3
1
1
2
1
2
Maricopa 2002 Crown
Duraking
Kofa
Kronos
Matt
Mohawk
Ocotillo
Orita
Platinum
Sky
WPB881
11/29
11/29
11/29
11/29
11/29
11/29
11/29
11/29
11/29
11/29
11/29
04/01
03/29
03/29
03/29
03/28
03/25
03/27
04/01
04/01
03/27
03/30
123
120
120
120
119
116
118
123
123
118
121
123
123
123
123
123
123
123
123
123
123
123
0
3
3
3
4
7
5
0
0
5
2
122
122
122
122
122
122
122
122
122
122
122
-1
2
2
2
3
6
4
-1
-1
4
1
124
123
120
121
120
120
121
125
123
122
122
1
3
0
1
1
4
3
2
0
4
1
Maricopa 2003 Alamo
Crown
Duraking
Kofa
Kronos
Matt
Mohawk
11/26
11/26
11/26
11/26
11/26
11/26
11/26
03/22
03/23
03/22
03/22
03/19
03/22
03/22
116
117
116
116
113
116
116
118
118
118
118
118
118
118
2
1
2
2
5
2
2
118
118
118
118
118
118
118
2
1
2
2
5
2
2
116
120
119
115
116
116
115
0
3
3
-1
3
0
-1
Forage and Grain Report (P-156) October 2008
16
Ocotillo
Orita
Platinum
Sky
WPB881
11/26
11/26
11/26
11/26
11/26
03/21
03/24
03/22
03/21
03/23
115
118
116
115
117
118
118
118
118
118
3
0
2
3
1
118
118
118
118
118
3
0
2
3
1
117
122
119
117
118
2
4
3
2
1
Maricopa 2003 Duraking
Kofa
Kronos
Matt
Mohawk
Ocotillo
Orita
Platinum
Sky
WPB881
11/28
11/28
11/28
11/28
11/28
11/28
11/28
11/28
11/28
11/28
03/29
03/29
03/27
03/24
03/28
03/26
03/31
03/29
03/29
03/26
121
121
119
116
120
118
123
121
121
118
117
117
117
117
117
117
117
117
117
117
-4
-4
-2
1
-3
-1
-6
-4
-4
-1
117
117
117
117
117
117
117
117
117
117
-4
-4
-2
1
-3
-1
-6
-4
-4
-1
118
114
115
115
114
116
121
118
117
117
-3
-7
-4
-1
-6
-2
-2
-3
-4
-1
Maricopa 2004 Alamo
Crown
Duraking
Kofa
Kronos
Mohawk
Ocotillo
Orita
Platinum
Sky
WPB881
11/30
11/30
11/30
11/30
11/30
11/30
11/30
11/30
11/30
11/30
11/30
03/24
03/23
03/31
03/29
03/28
03/24
03/28
03/27
03/26
03/24
03/26
115
114
122
120
119
115
119
118
117
115
117
116
116
116
116
116
116
116
116
116
116
116
1
2
-6
-4
-3
1
-3
-2
-1
1
-1
116
116
116
116
116
116
116
116
116
116
116
1
2
-6
-4
-3
1
-3
-2
-1
1
-1
115
117
117
114
114
114
115
119
116
115
116
0
3
-5
-6
-5
-1
-4
1
-1
0
-1
Maricopa 2004 Alamo
Crown
Duraking
Kofa
Kronos
Mohawk
Ocotillo
Orita
Platinum
Sky
WPB881
12/03
12/03
12/03
12/03
12/03
12/03
12/03
12/03
12/03
12/03
12/03
03/19
03/20
03/20
03/19
03/21
03/19
03/20
03/24
03/20
03/20
03/19
107
108
108
107
109
107
108
112
108
108
107
114
114
114
114
114
114
114
114
114
114
114
7
6
6
7
5
7
6
2
6
6
7
114
114
114
114
114
114
114
114
114
114
114
7
6
6
7
5
7
6
2
6
6
7
113
115
115
111
112
111
113
117
114
113
113
6
7
7
4
3
4
5
5
6
5
6
Maricopa 2005 Alamo
Crown
Duraking
Kofa
Kronos
Matt
Mohawk
Ocotillo
Orita
11/20
11/20
11/20
11/20
11/20
11/20
11/20
11/20
11/20
03/19
03/23
03/21
03/17
03/12
03/17
03/16
03/20
03/31
119
123
121
117
112
117
116
120
131
118
118
118
118
118
118
118
118
118
-1
-5
-3
1
6
1
2
-2
-13
117
117
117
117
117
117
117
117
117
-2
-6
-4
0
5
0
1
-3
-14
116
121
120
114
115
115
114
116
122
-3
-2
-1
-3
3
-2
-2
-4
-9
Forage and Grain Report (P-156) October 2008
17
Platinum
Sky
WPB881
11/20
11/20
11/20
03/20
03/18
03/19
120
118
119
118
118
118
-2
0
-1
117
117
117
-3
-1
-2
119
117
117
-1
-1
-2
Maricopa 2005 Alamo
Crown
Duraking
Kofa
Kronos
Matt
Mohawk
Ocotillo
Orita
Platinum
Sky
WPB881
12/02
12/02
12/02
12/02
12/02
12/02
12/02
12/02
12/02
12/02
12/02
12/02
03/21
03/25
03/24
03/20
03/16
03/18
03/20
03/25
04/06
03/23
03/24
03/22
109
113
112
108
104
106
108
113
125
111
112
110
114
114
114
114
114
114
114
114
114
114
114
114
5
1
2
6
10
8
6
1
-11
3
2
4
114
114
114
114
114
114
114
114
114
114
114
114
5
1
2
6
10
8
6
1
-11
3
2
4
112
117
116
110
112
111
110
113
119
115
113
114
3
4
4
2
8
5
2
0
-6
4
1
4
Maricopa 2006 Alamo
Crown
Duraking
Kofa
Kronos
Matt
Mohawk
Orita
Platinum
Sky
WPB881
12/02
12/02
12/02
12/02
12/02
12/02
12/02
12/02
12/02
12/02
12/02
03/29
04/03
04/01
03/29
03/28
03/30
03/27
04/01
04/01
03/29
03/29
117
122
120
117
116
118
115
120
120
117
117
118
118
118
118
118
118
118
118
118
118
118
1
-4
-2
1
2
0
3
-2
-2
1
1
118
118
118
118
118
118
118
118
118
118
118
1
-4
-2
1
2
0
3
-2
-2
1
1
116
120
119
115
116
116
115
121
119
117
117
-1
-2
-1
-2
0
-2
0
1
-1
0
0
Yuma
1998 Alamo
Crown
Duraking
Kofa
Kronos
Mohawk
Ocotillo
Platinum
WPB881
12/04
12/04
12/04
12/04
12/04
12/04
12/04
12/04
12/04
04/04
04/06
04/06
04/03
04/01
04/02
04/02
04/05
04/02
121
123
123
120
118
119
119
122
119
112
112
112
112
112
112
112
112
112
-9
-11
-11
-8
-6
-7
-7
-10
-7
112
112
112
112
112
112
112
112
112
-9
-11
-11
-8
-6
-7
-7
-10
-7
112
115
115
110
110
110
112
113
113
-9
-8
-8
-10
-8
-9
-7
-9
-6
Yuma
1999 Crown
Duraking
Kofa
Kronos
Mohawk
Ocotillo
Orita
Platinum
01/06
01/06
01/06
01/06
01/06
01/06
01/06
01/06
04/13
04/14
04/11
04/07
04/10
04/12
04/16
04/12
97
98
95
91
94
96
100
96
105
105
105
105
105
105
105
105
8
7
10
14
11
9
5
9
105
105
105
105
105
105
105
105
8
7
10
14
11
9
5
9
107
108
102
101
102
105
110
106
10
10
7
10
8
9
10
10
Wellton
2001 Crown
12/16
Duraking 12/16
04/06
04/07
111
112
109
109
-2
-3
108
108
-3
-4
110
109
-1
-3
Forage and Grain Report (P-156) October 2008
18
Kofa
Kronos
Matt
Mohawk
Ocotillo
Orita
Platinum
Sky
12/16
12/16
12/16
12/16
12/16
12/16
12/16
12/16
04/04
03/30
04/03
04/04
04/01
04/05
04/06
04/01
109
104
108
109
106
110
111
106
109
109
109
109
109
109
109
109
0
5
1
0
3
-1
-2
3
108
108
108
108
108
108
108
108
-1
4
0
-1
2
-2
-3
2
106
106
106
106
107
111
108
107
-3
2
-2
-3
1
1
-3
1
Yuma
2002 Crown
Duraking
Kofa
Kronos
Matt
Mohawk
Ocotillo
Orita
Platinum
Sky
WPB881
12/07
12/07
12/07
12/07
12/07
12/07
12/07
12/07
12/07
12/07
12/07
04/02
04/01
03/31
03/31
04/01
03/31
03/31
04/01
03/31
04/01
04/01
116
115
114
114
115
114
114
115
114
115
115
109
109
109
109
109
109
109
109
109
109
109
-7
-6
-5
-5
-6
-5
-5
-6
-5
-6
-6
109
109
109
109
109
109
109
109
109
109
109
-7
-6
-5
-5
-6
-5
-5
-6
-5
-6
-6
111
110
106
106
107
106
108
113
110
108
109
-5
-5
-8
-8
-8
-8
-6
-2
-4
-7
-6
Yuma
2003 Alamo
Crown
Duraking
Kofa
Kronos
Matt
Mohawk
Ocotillo
Orita
Platinum
Sky
WPB881
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
03/29
04/01
03/30
03/30
03/28
03/28
03/28
03/28
04/04
03/30
03/29
03/30
107
110
108
108
106
106
106
106
113
108
107
108
106
106
106
106
106
106
106
106
106
106
106
106
-1
-4
-2
-2
0
0
0
0
-7
-2
-1
-2
107
107
107
107
107
107
107
107
107
107
107
107
0
-3
-1
-1
1
1
1
1
-6
-1
0
-1
106
109
109
103
103
105
103
106
112
108
106
108
-1
-1
1
-5
-3
-1
-3
0
-1
0
-1
0
Yuma
2004 Alamo
Crown
Duraking
Kofa
Kronos
Mohawk
Ocotillo
Orita
Platinum
Sky
WPB881
12/14
12/14
12/14
12/14
12/14
12/14
12/14
12/14
12/14
12/14
12/14
03/30
04/01
04/01
03/30
03/28
03/31
03/30
04/03
04/02
03/31
03/30
107
109
109
107
105
108
107
111
110
108
107
103
103
103
103
103
103
103
103
103
103
103
-4
-6
-6
-4
-2
-5
-4
-8
-7
-5
-4
103
103
103
103
103
103
103
103
103
103
103
-4
-6
-6
-4
-2
-5
-4
-8
-7
-5
-4
103
105
105
101
101
101
103
107
104
103
104
-4
-4
-4
-6
-4
-7
-4
-4
-6
-5
-3
Yuma
2005 Alamo
Crown
Duraking
Kofa
12/23
12/23
12/23
12/23
03/31
04/04
04/04
03/31
98
102
102
98
103
103
103
103
5
1
1
5
104
104
104
104
6
2
2
6
103
106
106
100
5
4
4
2
Forage and Grain Report (P-156) October 2008
19
Kronos
Matt
Mohawk
Ocotillo
Orita
Platinum
Sky
WPB881
12/23
12/23
12/23
12/23
12/23
12/23
12/23
12/23
03/28
03/29
03/30
04/03
04/09
04/03
03/31
03/30
95
96
97
101
107
101
98
97
103
103
103
103
103
103
103
103
8
7
6
2
-4
2
5
6
104
104
104
104
104
104
104
104
9
8
7
3
-3
3
6
7
99
102
100
103
109
105
103
105
4
6
3
2
2
4
5
8
Yuma
2006 Alamo
Crown
Duraking
Kofa
Kronos
Matt
Mohawk
Ocotillo
Orita
Platinum
Sky
WPB881
12/26
12/26
12/26
12/26
12/26
12/26
12/26
12/26
12/26
12/26
12/26
12/26
04/09
04/13
04/13
04/08
04/03
04/07
04/07
04/15
04/10
04/12
04/09
04/09
104
108
108
103
98
102
102
110
105
107
104
104
106
106
106
106
106
106
106
106
106
106
106
106
2
-2
-2
3
8
4
4
-4
1
-1
2
2
107
107
107
107
107
107
107
107
107
107
107
107
3
-1
-1
4
9
5
5
-3
2
0
3
3
107
109
109
104
103
105
104
107
112
108
107
108
3
1
1
1
5
3
2
-3
7
1
3
4
Yuma
2007 Alamo
Crown
Duraking
Kronos
Matt
Ocotillo
Orita
Platinum
Sky
WPB881
12/24
12/24
12/24
12/24
12/24
12/24
12/24
12/24
12/24
12/24
04/06
04/11
04/08
04/04
04/04
04/06
04/13
04/12
04/06
04/09
103
108
105
101
101
103
110
109
103
106
99
99
99
99
99
99
99
99
99
99
-4
-9
-6
-2
-2
-4
-11
-10
-4
-7
99
99
99
99
99
99
99
99
99
99
-4
-9
-6
-2
-2
-4
-11
-10
-4
-7
99
101
101
96
98
99
102
100
99
100
-4
-7
-4
-5
-3
-4
-8
-9
-4
-6
112.25
112.31
111.95
-0.30
3.52
3.90
4.37
Mean (DAP)
Absolute error (DAP)
Root mean square error (DAP)
Root mean square error (% of measured DAP)
Forage and Grain Report (P-156) October 2008
0.06
3.96
4.33
4.87
112.21
-0.04
3.95
4.34
4.87
20