PHENOLOGICAL RESPONSE OF AN ANNUAL AND A BIENNIAL OILSEED RAPE CULTIVAR TO SOUTHERN ITALY ENVIRONMENTAL CONDITIONS

Carmelo Santonoceto, Umberto Anastasi

Dipartimento di Agrochimica, Università di Reggio Calabria. 89061 Gallina (RC). Italy.

ABSTRACT

The influence of environmental factors in southern Italy on the phenological response of an annual and a biennial oilseed rape cultivar was studied in five consecutive seasons. The sowing dates ranged, independently of the year, between the end of September and the middle of January. This provided a large variation in terms of photoperiod, temperature and rainfall regimes. Changes in phase duration of both cultivars can be largely accounted for by the relationships with a number of meteorological factors. The pre-flowering phenology of both cultivars is closely related to temperature and photoperiod, and post-flowering phenology to the temperature and rainfall.

KEyWORDs: Brassica napus subsp. oleifera, phenology, temperature, photoperiod, rainfall

INTRODUCTION

          Southern Italian environmental conditions present very different patterns of temperature, daylength and moisture supply from those countries where the greatest part of the oilseed rape varieties are constituted. Studies on the phenology and optimal conditions for each phase of the crop cycle are essential in searching for the most suitable varieties and sowing times for particular regions (Myers et al, 1982).

          This paper reports the phenological responses of an annual and a biennial cultivar of oilseed rape to the environmental conditions of two localities in the south of Italy potentially suitable to the diffusion of this crop.

Materials and methods

          Twenty-three separate field experiments were carried out in five consecutive seasons, beginning in 1988/89, at Rocca di Neto (Lat. N 39°05', 40 m a.s.l.) and San Marco Argentano (Lat. N 39°33', 232 m a.s.l.) using two oilseed rape cultivars: Activ (annual) and Ceres (biennial).

          The sowing dates ranged, independently of the year, from the end of September to the middle of January (Table 1). This provided a large variation in terms of photoperiod, temperature and rainfall regimes.

TABLE 1. Trials carried out and relevant sowing dates listed according to day and month.

SOWING DATES SITE	30 09 92	0810 92	1610 92	2810 88	3010 92	3110 88	0811 92	1511 91	1611 92	2211 90	2811 89	3011 92	0412 89	0912 92	1512 92	2012 90	2212 90	2312 91	0301 93	1001 91	1401 92	1501 92	1601 93
ROCCA DI NETO	*		*		*	*		*	*			*	*		*	*	*	*					*
S.MARCO ARGENTANO		*		*			*			*	*			*					*	*	*	*

          Development stages were recorded according to the CETIOM growth-stage key (1987). Emergence (A), beginning of flowering (F1) and colour change of the seeds (G5) were established when 50% of the plants reached the stage.

          Daily photoperiod including twilight was calculated from Keisling (1982).

          The relationships between phase duration in days, daily temperature (mean or minimum or maximum), mean daily photoperiod and amount of rainfall recorded within each phase were studied by multiple regression. Linear and quadratic temperature, photoperiod and rainfall were each partially regressed on each phase duration separately for each cultivar (Hodgson, 1978a, b). All the terms which were not significant in each phase and cultivar, were sequentially excluded from the model.

          Figure 1 shows the monthly variations in mean daily temperature and amount of rainfall. During the five-year period the mean monthly temperature between October and January fell in both localities from 17.8°C to a minimum of 7.8°C and, successively, rose to 20.8°C in June. The daily photoperiod varied between 10h 20min in late December to 15h 46min in late June.

          The amount of rainfall from October to June in each year were 197, 149, 804, 280 and 881 mm at Rocca di Neto; 540, 393, 1169, 416 and 516 mm at S.Marco Argentano.

Results

Duration of the crop cycle

          In Activ and Ceres, the duration of S-G5 progressively decreased to the delaying of S, respectively, from 226 to 153 and from 243 to 165 days. In 1989-90, Ceres did not reach the F1 stage; in the last sowing, it failed to mature.

          The reduction of S-G5 can be ascribed, mainly, to the duration of the A-F1 interval which, with the delaying of S, underwent an even greater reduction compared with that of the whole cycle (from 140 to 55 days, in Activ; from 178 to 82 days, in Ceres). This increased reduction could be accounted for by the duration of the S-A interval which progressively increased with the delaying of S.

          S did not directly affect the duration of the F1-G5 interval (Fig. 2).

Relationships between phase duration and meteorological factors

Sowing-emergence (S-A). Since no significant differences appeared between the cultivars for this phase of growth, the data of both cultivars were pooled.

          Temperature, particularly the minimum temperature, was the only environmental factor which influenced the duration of S-A. This relationship is described by a second-order polynomial exponential. The partial regression coefficients were significant at p£0.01 and p£0.05 for linear and quadratic temperature respectively (Fig. 3).

Emergence-beginning of flowering (A-F1). The duration of A-F1 in both cultivars was affected by  photoperiod as well as mean temperature. Temperature accounted for only 7 % of the variation of A-F1 in Activ, and for 12 % in Ceres. The partial regression coefficients were highly significant for linear photoperiod and linear temperature and significant for quadratic photoperiod in Activ; they were highly significant in all three terms in Ceres (Fig. 4).

Onset of flowering-ripening (F1-G5). Both temperature (particularly maximum temperature) and rainfall influenced the duration of the post-F1 phase of the two cultivars. In Activ, the significant levels of the partial regression coefficients were 1% for linear temperature and rainfall. In Ceres, the significant levels were 1% for linear temperature and 5% for linear rainfall (Fig. 4).

Conclusions

The variation in phase duration in the field of the two oilseed rape cultivars can be largely accounted for by the influence of a number of meteorological parameters.

          The pre-flowering phenology of both cultivars is largely related to temperature and photoperiod and the post-flowering phenology to temperature and rainfall.

          In both varieties the period between sowing and emergence decreased as the minimum daily temperature increased until an optimum temperature was reached according to a quadratic polynomial exponential.

          The duration of the following phase, A-F1, was related to temperature and photoperiod, although the partial regression coefficients varied according to the cultivar. In fact, at the same range of temperature and photoperiod, the annual cultivar showed a shorter duration of the above-mentioned phase than the biennial one. This different behaviour allows to the former cultivar, in a climate characterised by hot and dry late springs and summers, to develop the subsequent phase, F1-G5, in a more favourable period than the biennial variety do. In fact, during this later phase of the crop growth, higher maximum temperatures and lower rainfall totals were recorded for Ceres than for Activ. Moreover, the relatively low R² value indicates that factors other than those studied may be influencing the duration of the F1-G5 phase of the biennial cultivar.

ACKNOWLEDGEMENTS

This work was funded by the Italian Research Council.

ReFERENCES

CETIOM, 1978. Les stades repères du colza d’hiver. Ed CETIOM, Paris.

Hodgson A.S., 1978a. Rapeseed adaptation in Northern New South Wales. I. Phenological responses to vernalisation. Temperature and photoperiod by annual and a biennial cultivars of Brassica campestris L., Brassica napus L. and weath cv Timgalen. Australian Journal of Agricultural Research, 29, 693-710.

Hodgson A.S., 1978b. Rapeseed adaptation in Northern New South Wales. II. Predicting plant development of Brassica campestris L., Brassica napus L. and its implications for planting time, designed to avoid water deficit and frost. Australian Journal of Agricultural Research, 29, 711-726.

Keisling T. C., 1982. Calculation of the length of day. Agronomy Journal , 74, 758-759.

Myers L. F., Christian K. R., Kirehner R. J., 1982. Flowering responses of 48 lines of oilseed rapes (Brassica spp.) to vernalisation and daylenght. Australian Journal of Agricultural Research, 33, 927-936.