Table Of ContentsNext Page

Variation in Phenology of Wheat Genotypes Grown at Different Temperatures under Similar Photoperiod in Southern NSW

M. Stapper

CSIRO Plant Industry, Canberra, ACT.

Abstract

Eighteen times of sowing were conducted over four years with 11 wheat genotypes differing in maturity at two locations of similar latitude but with a difference of 3C in average daily temperature. Variations in thermal times from emergence to anthesis (DC65) and from flag leaf stage (DC39) to anthesis were considerable across sowing dates and genotypes, and were some 9% higher in the warmer climate. However, despite these variations, at both locations DC39 occurred on average at 0.75 of the thermal time required to advance from emergence to anthesis, and spike emergence (DC50) occurred at 0.46 of the thermal time from DC39-65. Vernalisation was most effective for emergence in July. It is suggested that neither the method for calculating degree-days, choice of base temperature (2C), nor vernalisation caused the main variations in thermal time for the two phases. Rather, variations were caused by genotype photoperiod sensitivity being dependent on temperature and stage of development.

Key words

Crop development, maturity, crop models, simtag, ceres-wheat.

Introduction

Phenological development is an important component of crop simulation models as it determines crop duration (maturity), and regulates partitioning of dry matter between organs. In wheat models differences in maturity between genotypes are controlled by assigning to each genotype numeric parameters representing response to temperature, photoperiod and vernalisation. To establish these calibration parameters, comprehensive data sets are required for genotypes across agro-ecological zones. Such data are scarce but important to enable validation and systematic improvement of phenology models. Trials were conducted at Griffith, NSW (1983, 1984) and at Ginninderra, ACT (GES, 1990, 1991), a 3C cooler site of similar latitude (ie. photoperiod). There were nine sowing dates between late March and late September over two years at both locations each with 11 genotypes differing widely in maturity. They included early (Sonalika, Yecora70), medium-early (Takari, Mexipak), medium (Bindawarra, Egret), medium-late (Harrier) and late (UQ189) spring wheats, semi-winter (Osprey), early winter (Novi Sad 2568) and late winter (Maris Huntsman) wheats. The genotypes were sown in blocks of single rows with border rows. Phenological development was similar to that of matching genotypes in adjacent plots (3). Plots were well fertilised and irrigated when required.

Crop observations were made with the Decimal Code (DC; 4) once or twice per week. Leaf appearance was measured on selected genotypes. Dates for floral initiation, double ridge or terminal spikelet stages were determined by microscopic inspection, but not for all treatments. Flag leaf stage (DC 39), spike emergence (DC50) and anthesis (DC65) dates were known for all genotypes across sowings. Durations from emergence (DC10) were calculated in thermal time with a base temperature of 2C. This study reports the measured variation in these phases across locations, sowing dates and genotypes.

Results and discussion

Anthesis was generally about a month later at the cooler site GES for a given sowing date. There were marked genotypic differences in the patterns of thermal times against emergence date and daylength at emergence. The warmer climate of Griffith required some 9% more degree-days to reach anthesis under similar daylengths, which may indicate that photoperiod sensitivity varies with temperature both between and within (ie. over phases) genotypes. Degree-day calculations may also contribute to differences, as there were more days at GES with minimum temperature below the base temperature of 2C. Egret, whose duration from emergence to anthesis (E-A) was 6.5% longer at Griffith, had the smallest variation of all genotypes, while it was 11% for Harrier (Fig.1). The maximum difference was 12.7 % for both UQ189 and Novi Sad, which are spring and winter wheats, respectively. Hence, vernalisation cannot be a main cause of these variations. Within sowing dates durations were least variable across genotypes when expressed relative to the one of mid-April emergence (=100%). For earlier emergence spring types had shorter and those with vernalisation longer relative durations. The latter reached its shortest duration (ie. maximum vernalisation) at about 70%, when emerging after late-June. Duration generally showed a bi-linear response to date of emergence with inflection in late-June, that is, with emergence near the shortest day. Correlations before winter solstice for each variety explained 54 to 75% of variation in DC10-65 duration with date of emergence. Duration for the two early genotypes decreased by 2.6Cd per day delay in emergence before the winter solstice and by 1.1 oCd thereafter. This was -5.0/0.0Cd for Harrier and

-4.8/+0.2Cd for Egret, where late sowing started delaying anthesis because of vernalisation. Anthesis for the three (semi-) winter wheats was reached 5.6Cd earlier per day delay in emergence before the solstice. Thereafter, duration remained constant for about a month before increasing with 0.8, 1.5 and 2.2 oCd per day delay in emergence for Osprey, Novi Sad and Maris Huntsman, respectively. This was followed by a sudden cut off date from where vernalisation was not met and genotypes remained vegetative. This occurred at Griffith for Maris Huntsman sown after mid-August and for the other two sown in September. At GES only Novi Sad and Maris Huntsman remained vegetative when sown in September.

Figure 1: Thermal duration from flag leaf to anthesis (FL-A) and emergence to anthesis (E-A) for Egret and Harrier sown at GES (closed symbols) and Griffith (open symbols).

The DC39-65 phase (FL-A) had 8% longer durations for Griffith with averages of 240 and 221 degree-days for Griffith (22 d) and GES (25 d), respectively. Figure 1 shows that this duration decreased with later sowing towards the solstice, and remained generally constant thereafter. The mean for each sowing date decreased from 290 (51 d) to 170 degree-days (15 d) at GES and from 296 (36 d) to 220 degree-days (14 d) at Griffith. Durations also differed between genotypes with Osprey having the shortest, 8% below average, and Sonalika/Harrier the longest, 6% above average, which was consistent across locations. This phase has been simplified as lasting 2 phyllochrons (leaf appearance intervals) in crop simulation model CERES-Wheat (2). Mulholland et al. (1), evaluating a model with a DC39-65 duration of 1.8 phyllochron, found it to be around 3 for spring wheat. In this study the average duration for Griffith decreased with later sowing from 3.5 to 2.5 phyllochrons, while for GES it gradually decreased from 3.6 to 2.1. So it was variable and never reached the constant 2 phyllochrons of CERES-Wheat. Flag leaf stage was reached on average at 0.75 of the DC10-65 duration at both Griffith and GES, but there were considerable variations in means over genotypes (0.67-0.82) and sowing dates (0.73-0.78). The mean values were very similar for most genotypes at both locations and increased with maturity from 0.69 for early spring, 0.75 for medium spring, 0.79 for late spring and semi-winter, to 0.81 for the late winter wheat. Spike emergence (DC50) occurred at both locations when 46% of the DC39-65 degree-days were accumulated. The variability in DC39-50 and DC50-65 phases was similar with a CV for all data points of about 28% at both locations.

Conclusions

Though there was considerable variation in thermal time required to complete the two nominated phases, both due to genotype, sowing date and location, it did little to alter the relative timing of the average occurrence of stages DC39 and DC50 in the pre-anthesis period. Therefore, the degree-days method used, choice of base temperature, use of screen or soil temperature nor vernalisation will have caused the main variations in this study. Variations around the mean could be caused by genotype photoperiod sensitivity being dependent on temperature and stage of development.

References

1. Mulholland, B.J., Craigon, J., Black, C.R., Stokes, D.T., Zhang, P., Coli, J.J. and Atherton, J.G. 1997.

2. J. Agric. Sci., Camb., 129, 155-161.

3. Ritchie, J.T., Godwin, D.C. and Otter, S. 1985. CERES-Wheat: A User-Oriented Wheat Yield Model. AGRISTARS YM-U3-04442-JSC-18892.

4. Stapper, M. and Fischer, R.A. 1990. Aust. J. Agric. Res., 41, 997-1019.

5. Zadoks, J.S., Chang, T.T. and Konzak, C.F. 1974. Weed Res., 14, 415-421.

Top Of PageNext Page