New oilseed rape (Brassica napus L.) varieties – canopy development, yield components, and plant density

Abstract

Full exploitation of the potential of new varieties requires research, whose aim is to adapt the technology to their needs. A two-factor experiment evaluated the effect of row spacing (33, 44, and 55 cm) of three oilseed rape (OSR) varieties (conventional, hybrid, and “semi-dwarf” hybrid variety) on canopy area index (CAI) and yield components. At higher row spacing, OSR plants were characterized by a strong increase in the CAI at successive growth stages; thereby, the incomplete use of production area was compensated at the pod development stage. The differences in the CAI between row spacings were significant until the end of flowering, whereas differences in the CAI between varieties were significant until the flower bud development stage. In the next development phases, CAI of OSR plants was at a similar level to all plots. However, the statistical analysis showed a declining trend in seed yield and yield biomass (pods and straw weight) with increasing row spacing. A positive aspect of increased row spacing was a decrease in glucosinolates content in seeds. The differences in fat content were statistically insignificant. The hybrid varieties of OSR produced the highest seed yield at a row spacing of 33 cm, while the conventional – at a row spacing of 44 cm. These relationships are confirmed by high positive correlations of seed weight and pod weight per unit area with CAI. The results have important practical aspect, because it shows that it is possible to reduce the number of OSR plants per unit area, thereby reducing demand for expensive certified seeds for sowing but to certain limits. Too small plant density binds to the risk of decreased seed yield. It may be justified, e.g., in extensive or organic farming where wide row creates the possibility of mechanical weed control.

Keywords:

• biomass
• canopy area index
• glucosinolates
• fat
• seed yield
• semi-dwarf variety

Introduction

Oilseed rape (OSR) has been the most important oilseed crop for many countries (Hrivna et al. 2002). One of methods of optimizing the yielding ability of OSR and other crop plants is to adjust crop density. An increase in the number of plants per unit area is associated with a better use of arable land and better light interception, but this does not always result in higher yielding capacity (Leach et al. 1999; Momoh & Zhou 2001; Sidlauskas & Bernotas 2003; Hiltbrunner et al. 2007; Lin et al. 2009; Worku & Astatkie 2011; Zhang et al. 2012). An inappropriate selection of canopy architecture parameters (too high plant density) may cause irregularities in the pollination of flowers and assimilate transport to ovules (Wang et al. 2011). At the same time, too high values of plant density can mean a worsening of light conditions and CO₂ supply to plants inside the canopy and can increase their susceptibility to pathogen infection (Hirose et al. 1997; Seassau et al. 2012).

The maximum productivity of plants growing at optimal density depends on the species, environmental conditions, and agronomic factors. Low plant density (<X plants m⁻²) can compensate by superior productivity at per-plant scale. Some studies designed to evaluate the tolerance of crop plants to the effect of pesticides and to compare crop plant varieties and the productivity of OSR crops give the number of plants per unit area or the leaf area index (LAI) as the canopy density parameter. Other studies are focused on the actual leaf area as the canopy architecture parameter (López-Bellido et al. 2005; Behrens & Diepenbrock 2006; Hiltbrunner et al. 2007; Andruszczak et al. 2012; Zhang et al. 2012).

The introduction of new varieties (e.g. “semi-dwarf” ones) give rise to the need to conduct research on the optimization of plant density, light conditions, and distribution of radiation within the canopy.

The aim of the present study was to evaluate the rate of increase in canopy area of three different groups OSR varieties using LAI depending on plant density and growth stage of OSR. A hypothesis was made that it was possible to change the number of OSR plants per unit area by changing row spacing (not changing in-row plant density) without deterioration of the canopy architecture parameters and yielding ability. The pursuit of the research goal and verification of the research hypothesis also required a presentation of fuller characteristics in the form of statistically calculated correlations of canopy architecture parameters at particular phenological stages with seed yield components and OSR biomass. Statistically derived correlations allow the determination of optimal density of OSR plants belonging to different varietal groups. The possibility of adjusting the conditions within the canopy and the effectiveness of crop protection treatments and seed requirement, without deteriorating yields, would increase the cost-effectiveness of OSR cropping technology.

Material and methods

Site description

A field experiment was carried out in the period 2010–2012 at the Experimental Farm in Bezek located near the city of Chełm (N: 51°19′ E: 23°25′), belonging to the University of Life Sciences in Lublin. The experiment was set up in a randomized block design in three replications on mixed rendzina soil developed from chalk rock with the granulometric composition of medium silty loam. This soil was characterized by an alkaline pH (pH in 1 mol KCl was 7.35). Its sorption complex was characterized by a high content of phosphorus and potassium as well as a very low content of magnesium (117.8–P; 242–K; and 19–Mg mg kg⁻¹ of soil). The organic carbon content was at a level of 2.47%.

Experimental design

The two-factor experiment evaluated the effect of plant density and OSR varieties on canopy architecture parameters and yield components as well as on the correlation between these traits. The number of OSR plants per unit area was controlled by row spacing, which was 33, 44, and 55 cm; this corresponded to the actual average population size of 38.0, 29.4, and 22.9 plant m⁻². In-row spacing of OSR plants did not change and it was about 8 cm. The winter OSR varieties were as follows: A – conventional variety; B – hybrid variety (F1, father line – male fertile); and C – low-growth (“semi-dwarf”) hybrid variety (F1, father line – male fertile) with high winter hardiness.

After the harvest of winter wheat, phosphorus, potassium, and magnesium fertilizers were incorporated (35 kg P ha⁻¹ – superphosphate; 50 kg K ha⁻¹ – potassium chloride; 18 kg S ha⁻¹ and 13.5 kg Mg ha⁻¹ – magnesium sulfate) and subsequently pre-sowing plowing was done. Nitrogen fertilizer was applied (130 kg N ha⁻¹) in the form of urea. OSR was sown in the second half of August (20–25 August).

Crop measurements

The measurements were made with a LAI-2000 plant canopy analyzer (Li-Cor Biosciences, Inc., Lincoln, NE). The LAI-2000 is one of the commonly used instruments to measure LAI. The LAI-2000 Canopy Analyzer can be used also to determine the canopy area index (CAI) resulting from radiation measurements made with a made with a “fish-eye” optical sensor (Behrens & Diepenbrock 2006).

The method based on the analysis of optical gap fraction measurements was selected not only due to the ease of its use, but primarily due to the conviction that the stem and pods that contain chlorophyll and shade the canopy are of decisive importance for the results of measurement of the CAI and the final yielding capacity (Fang & Liang 2008).

The evaluation of CAI in the OSR canopy was carried out on five dates: beginning of stem elongation (30–32 scale BBCH – Biologische Bundesanstalt, Bundessortenamt and Chemical Industry); beginning of flower bud development (50–55); beginning of flowering (60–63); end of flowering (67–69) and pod development (76–79).

The fruit of plants belonging to the Brassicaceae family is the “silique,” but in the literature, the term “pod” is used. In order to better understand this paper, the name “pod” is also used in it.

Statistical analysis

The obtained results were statistically analyzed using analysis of variance (ANOVA) and the differences that were proved with an error rate less than or equal to 5% were considered to be significant, while their significance was verified by Tukey’s test. Trying to stress the effect of experimental factors that can be determined by man, the results were presented as 3-year means (2010–2012). Correlation coefficients were calculated using Statistica software on the basis of partial results from the 3-year study (the number of cases (N) for each trait was 81) and proven correlations with an error rate less than or equal to 5% were considered to be significant.

Results

The canopy area index of OSR

Statistically significant differences in the CAI were proved between both OSR varieties and treatments with different row spacings. Compared to the area occupied by plants, the lowest CAI of OSR was found in the treatments where OSR rows were 55 cm apart, while the highest one in the treatments with a row spacing of 33 cm. The differences in CAI between the plots with different row spacing were significant from the beginning of stem elongation (30–32 BBCH) until the end of flowering (67–69), whereas the measurement made at the pod development stage (76–79) showed a trend toward higher CAI at a row spacing of 33 cm compared to the row spacing of 44 and 55 cm, but as a result of large standard deviations, these differences proved to be statistically insignificant (Figure 1).

The differences in canopy area of plants between OSR varieties were significant for the first two dates of CAI measurement. At this time, the hybrid varieties (B and C) were characterized by significantly quicker growth, thus higher CAI, compared to the conventional variety (A). At the initial stages of flowering (60–63), the A variety also showed a quick increase in canopy area that caused the differences in CAI between varieties to be insignificant at this phenological stage and at the next stages (Figure 2). The C variety was characterized by the highest average values of CAI, the B variety showed slightly lower CAI, while the A variety – the lowest one.

Figure 1. Canopy area index (CAI) depending on row spacing (cm) and growth stage Biologische Bundesanstalt, Bundessortenamt and Chemical Industry (BBCH) of oilseed rape (OSR).

Figure 2. Canopy area index (CAI) depending on the variety and growth stage Biologische Bundesanstalt, Bundessortenamt and Chemical Industry (BBCH) of oilseed rape (OSR) (A = conventional variety; B = hybrid variety [F1, father line]); C = low-growth (“semi-dwarf”) hybrid variety (F1, father line).

Yield, yield structure, and content of major nutrients in OSR seeds

The statistical analysis did not show row spacing (33, 44 and 55 cm) and cultivar to have a significant effect on OSR seed yield. In spite of this, a decreasing trend in seed yield with increasing row spacing could be observed. This relationship was determined by the B and C varieties (Table 1). Irrespective of row spacing, the C variety was characterized by the highest seed yield, while the A variety produced the lowest yield. The interaction between the factors also shows clearly that the OSR hybrid varieties produced the highest yield at a row spacing of 33 cm, whereas the conventional variety – at a row spacing of 44 cm (Table 1).

Table 1. Seed yield, fat, and glucosinolates content in oilseeds rape (OSR) depending on row spacing and variety (mean 2010–2012).

	Variety
Row spacing (cm)	A	B	C	Mean
OSR seed yield (t ha^−1)
33	2.79	3.58	3.72	3.36
44	3.26	3.12	3.35	3.24
55	3.10	3.13	3.08	3.10
Mean	3.05	3.27	3.38	3.23
HSD0.05	Variety × row spacing – NS Between varieties – NS Between row spacings – NS
Fat content in OSR seeds (%)
33	44.4	44.0	43.5	44.0
44	44.1	43.8	44.2	44.0
55	44.6	43.9	43.8	44.1
Mean	44.4	43.9	43.8	44.0
	Variety × row spacing – NS Between varieties – NS Between row spacings – NS
Glucosinolates content in OSR seeds (µmol g^−1)
33	13.7	9.9	9.2	10.9
44	11.1	8.8	7.9	9.3
55	11.3	7.8	10.2	9.8
Mean	12.0	8.8	9.1	10.0
HSD0.05	Variety × row spacing – NS Between varieties = 1.9 Between row spacings – NS

A = conventional variety; B = hybrid variety (F1, father line); C = low-growth (“semi-dwarf”) hybrid variety (F1, father line); HSD = the honestly significant difference; NS, not significant.

The experimental factors were not shown to have a significant effect on oil content in OSR seeds (Table 1). Seeds from the plots with higher row spacing contained slightly more oil. Significant differences in seed glucosinolates content were proven between OSR varieties (Table 1). The conventional variety (A) contained significantly more oil than the hybrid varieties (B and C). Higher plant density (row spacing of 33 cm) caused increased synthesis and accumulation of compounds classified in the group of glucosinolates in seeds, compared to lower density of OSR plants (row spacing of 44 and 55 cm). The interaction trends show that the varieties responded differently to a change in crop density in terms of glucosinolates content. The A variety and the B variety had the highest seed glucosinolates content at a spacing of 33 cm, whereas the C variety – at a spacing of 55 cm. The differences between treatments with different row spacings and the interaction differences (variety × row spacing) only had the trend character and were statistically insignificant (Table 1).

The A variety had significantly higher thousand seed weight (TSW) compared to the B and C varieties. The calculated interactions show that the conventional variety produced seeds with the significantly highest TSW at a row spacing of 33 cm, while the lowest TSW was found for the C variety at a row spacing of 44 cm. Irrespective of the variety, the highest TSW was obtained in the plots where OSR was sown at a spacing of 33 cm compared to the spacing of 44 and 55 cm, but these differences were statistically insignificant (Table 2).

Table 2. Height of oilseed rape (OSR) plant, thousand seed weight (TSW), straw, and pods weight depending on row spacing and variety (mean 2010–2012).

	Variety
Row spacing (cm)	A	B	C	Mean
Plant height of OSR (cm)
33	113	130	117	120
44	122	125	113	120
55	119	141	110	123
Mean	118	132	113	121
HSD0.05	Variety × row spacing = 13 Between varieties = 6 Between row spacings – NS
TSW of OSR (g)
33	4.72	4.11	3.86	4.23
44	4.12	4.19	3.69	4.00
55	4.28	3.85	4.09	4.07
Mean	4.37	4.05	3.88	4.10
	Variety × row spacing = 0.59 Between varieties = 0.28 Between row spacings – NS
Straw dry weight of OSR (t ha^−1)
33	3.74	4.14	3.05	3.64
44	3.83	3.47	2.65	3.32
55	3.67	3.58	2.74	3.34
Mean	3.75	3.73	2.81	3.43
HSD0.05	Variety × row spacing – NS Between varieties = 54.6 Between row spacings – NS
Dry weight of OSR pods (without seeds) (t ha^−1)
33	3.98	4.79	4.42	4.39
44	4.02	4.11	3.97	4.03
55	3.81	4.17	4.00	3.99
Mean	3.93	4.35	4.13	4.13
HSD0.05	Variety × row spacing – NS Between varieties – NS Between row spacings – NS

A = conventional variety; B = hybrid variety (F1, father line); C = low-growth (“semi-dwarf”) hybrid variety (F1, father line); HSD = the honestly significant difference; NS, not significant.

The OSR C variety was characterized by significantly lower plant height and lower straw weight compared to the other varieties. The B variety was marked by the highest accumulation of biomass in the form of pod residue (pods without seeds) (Table 2). Irrespective of the OSR variety, higher crop residue (pod and straw weight) was obtained at a row spacing of 33 cm than at a spacing of 44 and 55 cm. The differences in biomass between treatments with different row spacings had the trend character and were statistically insignificant. Also, no significant differences in OSR plant height were found between treatments with different crop densities (Table 2).

Correlation coefficients

The calculated correlations showed significant positive relationships between OSR seed weight per unit area and CAI at all growth stages (Table 3). The highest coefficient of correlation between seed yield and CAI was found at the end of flowering (67–69), at pod development (76–79), and at early growth stages (30–32 BBCH). Very similar correlation relationships occurred between CAI and pod weight. This similarity was confirmed by a high positive correlation between seed yield and pod dry weight.

Table 3. Coefficients of correlation between the examined parameters.

Factors		Seed yield	TSW	Plant height	Straw weight	Pod weight	Weed dry weight	Plant density	Fat content	Glucosinolate content
CAI for growth stage (BBCH )	30–32	0.70*	−0.22	0.47	0.26	0.63*	−0.20	0.49	−0.55*	−0.30
	50–55	0.68*	0.04	0.23	0.25	0.54*	−0.44	0.59*	−0.53	0.03
	60–63	0.59*	−0.10	0.31	0.17	0.44	−0.34	0.63*	−0.35	−0.18
	67–69	0.77*	−0.17	0.23	0.14	0.73*	−0.21	0.67*	−0.44	−0.11
	76–79	0.71*	0.08	0.10	0.26	0.69*	−0.08	0.48	−0.29	0.11
Seed yield		1.00	−0.28	0.19	0.34	0.81*	−0.14	0.51	−0.56*	−0.19
TSW		−0.28	1.00	−0.14	0.39	−0.06	−0.25	−0.32	0.09	0.74*
Plant height		0.19	−0.14	1.00	0.47*	0.34	−0.07	−0.12	−0.16	−0.34
Straw weight		0.34	0.39	0.47	1.00	0.50	−0.23	−0.10	−0.10	0.32
Pod weight		0.81*	−0.06	0.34	0.50	1.00	−0.40	0.37	−0.52	−0.03
Weed dry weight		−0.14	−0.25	−0.07	−0.23	−0.40	1.00	−0.35	−0.01	−0.25
Plant density		0.51	−0.32	−0.12	−0.10	0.37	−0.35	1.00	−0.15	−0.16
Fat content		−0.56*	0.09	−0.16	−0.10	−0.52	−0.01	−0.15	1.00	0.14
Glucosinolate content		−0.19	0.74*	−0.34	0.32	−0.03	−0.25	−0.16	0.14	1.00

CAI, Canopy area index; BBCH, Biologische Bundesanstalt, Bundessortenamt and Chemical Industry; TSW, thousand seed weight.

* The correlation coefficient significant at α = 0.05.

The glucosinolates content in OSR seeds was not correlated with CAI, but it increased with increasing 1,000 seed weight (a high positive correlation between these parameters). On the other hand, the oil content was dependent on CAI and seed yield – the higher CAI, the higher was seed yield and thereby lower seed oil content (Table 3).

There were also significant correlations between the number of OSR plants per unit area and CAI at the following growth stages: beginning of flower bud development (50–55); beginning of flowering (60–63), and end of flowering (67–69).

Discussion

The present experiment showed the highest values of the CAI of OSR (3.96–4.19) at pod development stages (76–79 BBCH), irrespective of the variety and row spacing. On average for the whole measurement period, the hybrid varieties were characterized by a higher value of CAI compared to the conventional variety. Cao et al. (2009) found that OSR reached the highest values of LAI at the early flowering stage. In their study, the hybrid variety showed higher LAI (5.90) than the conventional variety (4.75). The highest LAI at the flowering stage was also found by Momoh and Zhou (2001). In their research, LAI at pod development was more than three times lower than at flowering. The differences in the studies of individual authors were probably caused by different LAI measurement or calculation methods. The present study used the LAI-2000 plant canopy analyzer that measures soil shading by all plant parts (Welles & Cohen 1996). At pod development in OSR and other plants of the Brassicaceae and Fabaceae families, not only the leaves contain chlorophyll but also the stem and pods and they are therefore considered to be the photosynthetic area (Andrews & Svec 1975; Atkins et al. 1977; Singal et al. 1995; Zhou et al. 2009; Jun et al. 2012).

When reducing the number of plants per unit area, one should take into account the risk of decreasing crop yield, in particular in agricultural systems where extensive crop protection and reduced crop rotation are used. The study of Bilgili et al. (2003) showed that seeding rate had no significant effect on seed yield in both years. Seed yield was similar in all seeding rates. In the study of Różyło and Pałys (2011), wider row spacing (33 cm) of OSR plants significantly increased only the number of branches and number of pods per plant, compared to a spacing of 25 cm. But no significant differences were found in yield and in the basic chemical composition of seeds. Studying the response of OSR to a change in row spacing (30, 40, and 50 cm), Mousavi and Bagheri (2011) observed the lowest seed yield and “biological yield” at a row spacing of 30 cm. They obtained the highest seed yield at a spacing of 50 cm and the highest “biological” yield at a spacing of 40 cm. In the study conducted by these authors, the oil content in OSR seeds harvested from the plots with different row spacings was similar.

In the present study, the OSR hybrid varieties (B and C) produced distinctly higher seed yield at the highest plant density, while in the case of the conventional variety when it was sown at the medium plant density (Table 1). However, these differences were statistically insignificant. Also, the study did not prove a significant effect of lower density of OSR plants on pod and straw weight per unit area as well as on oil and glucosinolates content in OSR seeds. The present study is corroborated by the research of Spychaj-Fabisiak et al. (2011), which also did not demonstrate significant differences in seed yield and seed oil content between plots with different row spacings.

In the study of Momoh and Zhou (2001), an increase in OSR plant density resulted in a significant decrease in leaf area per plant and, at the same time, an increase in leaf area per unit area, expressed as the LAP, with increasing plant density at all stages of development.

Özer et al. (1999) observed a positive correlation between the yield of rape seeds and number of pods per plant and weight of 1000 seeds. Similar results were obtained in our experiment.

In the present study, OSR plants showed a high ability to branch out and to produce higher canopy area under the conditions of reduced plant density. This is evidenced by the absence of significant differences in CAI between the treatments with different row spacings at the pod development stage. Jullien et al. (2009) also reported that OSR plants have a high ability to accommodate to habitat conditions. They showed that under nitrogen deficiency conditions OSR significantly increased leaf mass per area (LMA). Increasing row spacing entails the risk of losing a part of yield, in particular in modern low-growth hybrid varieties (“semi-dwarf” varieties). These relationships are confirmed by high positive correlations between seed yield and CAI, in particular at initial growth stages of winter OSR during which the soil surface is not used fully.

On the other hand, wide row spacing in OSR crops creates the possibility of using mechanical weed control methods in extensive or ecological crops in which a small reduction in yield is acceptable.