Response of Winter Oilseed Rape to Sowing Date and Seeding Density

Streszczenie

Rzepak ozimy (Brassica napus L.) należy do najważniejszych gospodarczo roślin uprawnych w Polsce i Unii Europejskiej, a optymalny dobór terminu siewu oraz gęstości siewu uznawany jest w literaturze agrotechnicznej za jeden z kluczowych czynników determinujących jakość wschodów, akumulację biomasy jesiennej, przezimowalność i ostateczną wielkość plonu nasion. Celem niniejszej pracy było określenie wpływu terminu siewu i gęstości siewu — stosowanych samodzielnie oraz w układzie czynnikowym — na wybrane cechy morfologiczne i plonotwórcze rzepaku ozimego odmiany mieszańcowej 'DK Expressive' uprawianego w specyficznych warunkach pedoklimatycznych południowo-wschodniej Polski. Doświadczenie polowe przeprowadzono w stacji doświadczalnej Uniwersytetu Przyrodniczego w Lublinie w trzech sezonach wegetacyjnych (2019/2020–2021/2022), stosując układ split-plot z trzema terminami siewu (D1: około 20 sierpnia, D2: opóźniony o 10 dni, D3: opóźniony o 20 dni) skrzyżowanymi z trzema gęstościami siewu odpowiadającymi dolnemu, środkowemu i górnemu zakresowi zaleceń praktyki agrotechnicznej dla odmian mieszańcowych półkarłowych. Uzyskane wyniki potwierdziły, że termin D1 zapewniał istotnie wyższe tempo wschodów, lepiej rozwiniętą rozetę jesienną oraz najwyższe wartości składowych plonu i plon nasion. Opóźnienie siewu do terminu D3 spowodowało istotne statystycznie obniżenie plonu nasion średnio o 27,9% w stosunku do terminu optymalnego. Wykazano ponadto, że zwiększona gęstość siewu częściowo kompensowała straty plonu wynikające z opóźnienia siewu, a interakcja między terminem siewu a gęstością siewu okazała się statystycznie istotna dla plonu nasion. Wyniki wskazują na konieczność uwzględnienia lokalnych warunków agroklimatycznych Lubelszczyzny przy formułowaniu regionalnych zaleceń agrotechnicznych dla rzepaku ozimego.

Słowa kluczowe: rzepak ozimy; termin siewu; gęstość siewu; plon nasion; przezimowalność; agrotechnika

Abstract

Winter oilseed rape (Brassica napus L.) ranks among the most economically significant arable crops in Poland and the European Union, and the selection of an appropriate sowing date and seeding density is widely recognised in the agronomic literature as one of the primary determinants of seedling establishment quality, autumn biomass accumulation, winter survival, and ultimate seed yield. The aim of the present thesis was to determine the effects of sowing date and seeding density — applied individually and in factorial combination — on selected morphological and yield-forming traits of the semi-dwarf hybrid cultivar 'DK Expressive' grown under the specific pedoclimatic conditions of south-eastern Poland. A field experiment was conducted at the experimental station of the University of Life Sciences in Lublin over three growing seasons (2019/2020–2021/2022), employing a split-plot design with three sowing dates (D1: approximately 20 August, D2: delayed by 10 days, D3: delayed by 20 days) crossed with three seeding densities corresponding to the lower, central, and upper ranges recommended for semi-dwarf hybrid cultivars in regional agronomic practice. The results confirmed that sowing date D1 ensured significantly higher emergence rates, more advanced autumn rosette development, and the highest values of yield components and seed yield. Delayed sowing at D3 resulted in a statistically significant reduction in seed yield averaging 27.9% relative to the optimal sowing date. It was further demonstrated that elevated seeding density partially compensated for yield losses attributable to delayed sowing, and that the interaction between sowing date and seeding density exerted a statistically significant effect on seed yield. The findings indicate the necessity of accounting for the specific agroclimatic conditions of the Lublin region when formulating regional agronomic recommendations for winter oilseed rape production.

Keywords: winter oilseed rape; sowing date; seeding density; seed yield; winter survival; agronomy

List of Abbreviations

ANOVA: analysis of variance
BBCH: Biologische Bundesanstalt, Bundessortenamt und CHemische Industrie phenological scale
CV: coefficient of variation
FAME: fatty acid methyl ester
FLC: FLOWERING LOCUS C gene
FT: FLOWERING LOCUS T gene
GDD: growing degree days
HSD: Honestly Significant Difference
LSD: Least Significant Difference
QTL: quantitative trait locus
RED: Renewable Energy Directive
TKW: thousand kernel weight
TSW: thousand-seed weight

Introduction

Winter oilseed rape (Brassica napus L.) occupies a position of singular importance within the agricultural systems of Poland and the broader European Union, representing one of the most economically significant arable crops cultivated on the continent. As the second most valuable oil crop in global trade by aggregate market value, rapeseed commands a central role in both the food processing industry — as a source of high-quality edible vegetable oil characterised by a favourable fatty acid composition — and in the energy sector, where it serves as the principal feedstock for biodiesel production under the frameworks established by successive iterations of the European Renewable Energy Directive. The sustained expansion of the rapeseed cultivation area observed across Central and Eastern Europe over the past three decades reflects not only the maturation of plant breeding programmes that have progressively elevated yield potential and stress tolerance, but also the growing recognition of the crop's agronomic value as a break crop in cereal-dominated rotations, where its inclusion disrupts pathogen and pest cycles, improves soil structural conditions, and contributes to the overall productive efficiency of the cropping system. In the context of Polish agriculture, winter oilseed rape consistently ranks among the three most economically valuable field crops, and Poland is recognised among the five largest oilseed rape producing member states within the European Union, with the cultivated area and average yields reflecting the technological progress achieved by Polish agronomy over the past two decades.

The agronomic importance of winter oilseed rape is paralleled, however, by a degree of biological and ecological complexity that renders the crop more sensitive than the principal cereal species to suboptimal management practices, particularly those determining the quality of crop establishment during the critical autumn vegetative period. Among the agronomic decisions that exert the most decisive influence on the trajectory of individual plant development and the ultimate expression of yield potential, the choice of sowing date and seeding density have been identified in the research literature as primary determinants of establishment success, autumn biomass accumulation, overwintering capacity, and the expression of yield components at harvest. The sowing date governs the length of the pre-winter vegetative period available to developing seedlings, thereby conditioning the extent to which individual plants attain the morphological and physiological state — characterised by the formation of a sufficiently robust root neck, the accumulation of adequate reserves of soluble sugars and compatible solutes, and the development of a canopy capable of effective light interception during spring regrowth — regarded as necessary for tolerance of the freeze-thaw cycles and ice encasement events that characterise Central European winters. Seeding density, in turn, modulates the competition for resources among individual plants and between the crop stand and the weed flora, influencing the tiller architecture, branching capacity, and per-plant investment in reproductive structures, and interacts with sowing date in ways that are not fully predictable from the separate effects of each factor considered in isolation.

The challenge of optimising sowing date and seeding density is further compounded by the progressive intensification of inter-annual climate variability across Central and Eastern Europe, a trend that has substantially increased the risk of both suboptimal and excessive autumn development in winter oilseed rape stands. The occurrence of extended warm and dry periods during the conventional sowing window — particularly during August and the first half of September — has been associated with delayed seedling emergence, reduced seedling vigour, and elevated mortality during establishment, conditions under which the agronomic logic of late sowing may be undermined by the same unfavourable meteorological circumstances that made early sowing impractical. Conversely, unusually warm autumn seasons following timely sowing may promote excessive vegetative development, elongation of the stem apex, and premature entry into reproductive phases, increasing the vulnerability of well-established stands to late frosts following periods of mid-winter warming. The complexity of these interactions between agro-meteorological conditions and crop physiological responses to management inputs has stimulated sustained experimental inquiry across multiple European countries, generating a substantial body of knowledge regarding the quantitative relationships between sowing date, seeding density, and yield outcomes. The majority of this experimental knowledge, however, has been derived from investigations conducted in the temperate-maritime or sub-oceanic climatic conditions prevailing in Western Europe — particularly in France, Germany, the United Kingdom, and the Nordic countries — and the direct transferability of the agronomic recommendations derived from such experiments to the climatic conditions characterising the continental and transitional zones of Central and Eastern Europe remains a subject of acknowledged uncertainty.

The specific environmental context of south-eastern Poland, and of the Lublin region in particular, presents a combination of climatic and pedological characteristics that justify the conduct of dedicated regional experimentation rather than the uncritical application of recommendations derived from more extensively investigated climatic zones. The Lublin region is situated within a transitional continental-oceanic climatic belt, characterised by pronounced inter-annual variability in the thermal and moisture conditions prevailing during the August–September sowing window that is critical for oilseed rape establishment, by an appreciable frequency of early-autumn drought episodes capable of delaying emergence and limiting seedling development, and by a winter thermal regime that, while generally less severe than that of north-eastern Poland, is nonetheless subject to irregular freeze-thaw cycles and episodic late-winter frost events following periods of mid-season warming. The soils of the region, predominantly Haplic Luvisols of loessial origin characterised by favourable texture for arable cultivation but susceptibility to structural degradation under adverse moisture conditions, constitute an additional site-specific factor influencing seedling establishment dynamics and the response of the developing root system to temperature fluctuations. Against this background, the reliance of agricultural practitioners and advisory services in the region on generic national or European recommendations calibrated to different environmental conditions has represented a recognised source of uncertainty in the practical management of winter oilseed rape, and the generation of locally relevant, multi-year experimental data has been identified as a research priority for the improvement of crop management guidelines applicable to the specific conditions of south-eastern Poland.

The present thesis was undertaken with the objective of addressing this regional knowledge gap through the analysis of data generated by a systematic three-season field experiment conducted at the Experimental Station of the University of Life Sciences in Lublin during the 2019/2020, 2020/2021, and 2021/2022 growing seasons. The experiment employed a factorial design crossing three sowing dates — representing early, optimal, and late sowing scenarios relative to the regional agronomic calendar — with three seeding densities, applied to the hybrid cultivar DK Expressive, which is widely grown in the region and representative of the semi-dwarf hybrid type that has come to dominate commercial rapeseed production in Poland over the past decade. The selection of a multi-year experimental framework, encompassing seasons characterised by distinct agro-meteorological profiles including at least one climatically unfavourable year, was considered essential for the generation of conclusions with sufficient generality to support practical agronomic recommendations, given that the results of single-season experiments are inherently constrained in their applicability across the range of climate scenarios likely to be encountered at a given site over a commercial cultivation period. The investigation was designed to quantify the responses of key crop performance variables — including seedling emergence rate, plant population at harvest, individual plant biomass at the end of the autumn vegetative period, grain yield, and selected yield components — to variation in sowing date and seeding density, both as main effects and in their interaction.

The experimental work presented in this thesis was guided by three principal research hypotheses formulated on the basis of existing knowledge and the specific environmental context of the study site. First, it was hypothesised that delayed sowing, relative to the optimal regional window, would result in statistically significant reductions in autumn plant biomass accumulation, measured as root neck diameter and shoot fresh weight at the end of the vegetative period, and that this effect would be most pronounced in seasons characterised by below-average effective temperature accumulation following sowing. Second, it was hypothesised that increasing seeding density would partially compensate for the adverse effects of late sowing on canopy establishment and would be associated with a significant increase in yield under delayed sowing conditions, while under optimal sowing conditions higher seeding density would not confer a significant yield advantage and might be associated with increased within-canopy competition and reduced individual plant productivity. Third, it was hypothesised that the interaction between sowing date and seeding density would be statistically significant for grain yield and for at least one yield component, reflecting the non-additive nature of the combined effects of these management factors on crop performance, and that the magnitude of this interaction would vary across growing seasons as a function of the contrasting agro-meteorological conditions characterising each experimental year.

The thesis is structured to address the stated research objectives through a progression from theoretical and contextual foundations to empirical analysis and practical synthesis. Chapter 1 provides a comprehensive account of the biological characteristics and agricultural significance of winter oilseed rape, encompassing the botanical classification and morphological description of Brassica napus L., the genetic determinants of vernalisation and photoperiodic responses underlying the transition to reproductive development, the principal components of yield and the physiological mechanisms governing their expression, and the economic importance of the crop in both national and European contexts. Chapter 2 reviews the agronomic determinants of crop establishment and yield formation, with particular emphasis on the documented effects of sowing date on autumn plant development and winter hardiness, the principles governing seeding density optima and their modification under contrasting environmental conditions and hybrid cultivar types, and the practical implications of the reviewed evidence for the formulation of management recommendations applicable to south-eastern Poland. Chapter 3 presents the empirical core of the thesis, comprising a full description of the materials and methods employed in the field experiment, the statistical analysis of the resulting data, and a structured discussion of the findings in relation to the research hypotheses and the broader body of experimental evidence reviewed in the preceding chapters. The thesis concludes with a synthesis of the principal findings, an assessment of the practical implications for the management of winter oilseed rape under the specific conditions of the Lublin region, and a delineation of the directions for future research that have been identified on the basis of the limitations of the present investigation.

Chapter 1. Biological Characteristics and Agricultural Significance of Winter Oilseed Rape

1.1. Botanical Classification and Morphological Description of Brassica napus L.

Winter oilseed rape belongs to the family Brassicaceae (formerly Cruciferae), which encompasses one of the most agriculturally significant plant groups recognised in botanical science, characterised by extensive intra- and inter-specific morphological and phytochemical diversity that has contributed to the global importance of its constituent species in crop production.^{[8, p. 2]} Within the Brassiceae tribe, members of the genus Brassica occupy a position of particular economic relevance, as selection for targeted traits in both seed and vegetative tissues has generated high-value oilseed, vegetable, and condiment crops over centuries of cultivation.^{[8, p. 2]} The species Brassica napus L. encompasses both tuberous forms such as swede (rutabaga) and leafy forms including fodder rape and oilseed types used for animal nutrition and human consumption, and represents perhaps the most economically significant member of this genus in contemporary European agriculture.^{[2, p. 2]}

From a genomic perspective, Brassica napus is an allotetraploid species with a genomic constitution of AACC (2n = 38), originating from a natural interspecific hybridisation event between two diploid progenitor species: Brassica rapa L. (AA genome, 2n = 20) and Brassica oleracea L. (CC genome, 2n = 18).^{[2, p. 2]} This allopolyploid origin is consistently confirmed through genomic analyses of the species' sub-genomes.^{[3, p. 2]} As described within the so-called Triangle of U framework, three allopolyploid species — B. napus (AACC, 2n = 38), Brassica juncea (AABB, 2n = 36), and Brassica carinata (BBCC, 2n = 34) — each arose through hybridisation between pairs drawn from three ancestral diploid species: B. rapa, B. oleracea, and Brassica nigra (BB, 2n = 16).^{[2, p. 2]} It is further noteworthy that these three diploid progenitors are themselves ancient polyploids that underwent a lineage-specific whole-genome triplication, meaning that B. napus contains a particularly complex and redundant genomic architecture with multiple homeologous gene copies.^{[2, p. 2]} The amphidiploid nature of the species confers both opportunities and challenges for breeders, as homeologous gene interactions influence phenotypic expression across traits ranging from flowering time to seed quality.^{[6, p. 1]}

Based on whole-genome resequencing data, the estimated time of B. napus formation has been placed at approximately 1,912 to 7,178 years ago, with subsequent divergence between ecotypes occurring over a considerably shorter timescale: the winter and semi-winter ecotypes are estimated to have diverged roughly 60 years ago, while the winter and spring ecotypes diverged approximately 416 years ago, and oilseed and non-oilseed forms approximately 277 years ago.^{[2, p. 3]} Historical and genetic evidence points to winter oilseed rape as the original form of B. napus, with spring-type rapeseed having been developed around the year 1700; the semi-winter ecotype, which predominates in East Asia, has a shorter history, having arisen from winter-type germplasm introduced from Europe during the 1930s and 1940s.^{[2, p. 9]} These divergence timescales underscore the relatively recent nature of many adaptations distinguishing ecotypes, including those governing photoperiod sensitivity and vernalisation requirement, which are of direct agronomic relevance to crop management.

The distinction between winter and spring forms of B. napus is of fundamental agronomic importance and is reflected in every aspect of crop management strategy. Winter annual-type rapeseed possesses a clear requirement for extended cold temperature exposure — a process known as vernalisation — in order to transition from vegetative to reproductive development, in contrast to summer annual-type rapeseed, which has little or no such requirement.^{[1, p. 2]} This biological distinction underlies the entire framework of autumn sowing, winter rosette establishment, and spring flowering that characterises oilseed rape production under temperate European conditions. Modern breeding programmes have substantially modified the nutritional quality profile of B. napus, achieving dramatic reductions in the concentrations of erucic acid and aliphatic glucosinolates in the seed while simultaneously improving oil content, seed yield, and disease resistance, transitions that have been essential to the crop's acceptance in edible oil markets.^{[2, p. 2]}

Morphologically, winter oilseed rape is characterised by a robust taproot system that penetrates deeply into the soil profile under structurally favourable conditions, providing secure anchorage and access to subsoil moisture and nutrient reserves during the critical autumn establishment phase and the subsequent spring growth period. The stem, which is erect and typically branched, develops a characteristic glaucous appearance resulting from epicuticular waxes that also coat the surface of the leaf laminae. Basal rosette leaves formed during the autumn vegetative phase differ substantially in size, shape, and developmental origin from the upper cauline leaves that emerge following stem elongation in spring; both types display a distinctive blue-green colouration attributable to epicuticular wax deposits, which confer partial resistance to foliar desiccation and contribute to reflectance characteristics relevant to remote sensing assessment. The inflorescence is a raceme-corymb bearing four-petalled yellow flowers, a morphological arrangement characteristic of the Brassicaceae and relevant to the sequential nature of flowering within a single plant. Following pollination — which involves both insect visitors and wind-borne pollen transfer — seed development proceeds within elongated siliquae that typically bear numerous seeds arranged in two parallel rows. The seeds are spherical, dark brown in colour, with an oil content and composition that has been progressively optimised through decades of selective breeding aimed at maximising the proportion of oleic and linolenic acids while minimising erucic acid concentration.

Table 1.1. Key botanical and taxonomic characteristics of *Brassica napus* L. in the context of the genus *Brassica*
Characteristic	Description
Family / Tribe	Brassicaceae (Cruciferae) / Brassiceae
Genomic constitution	AACC, 2n = 38 (allotetraploid)
Progenitor species	B. rapa (AA, 2n = 20) × B. oleracea (CC, 2n = 18)
Estimated species formation	~1,912–7,178 years ago (based on whole-genome resequencing)
Winter–spring divergence	~416 years ago; winter–semi-winter ~60 years ago
Principal ecotypes	Winter annual, spring (summer) annual, semi-winter
Inflorescence type	Raceme-corymb; four-petalled yellow flowers
Leaf surface	Blue-green; glaucous bloom from epicuticular wax deposits
Key breeding improvements	Reduction of erucic acid and glucosinolates; increased oil content, seed yield, disease resistance
Seed calorific value	26.6 MJ/kg; straw 15.8–19.1 MJ/kg

1.2. Developmental Stages and Phenological Growth Phases

The ontogeny of winter oilseed rape is conventionally described through the application of the Biologische Bundesanstalt, Bundessortenamt und CHemische Industrie scale (BBCH scale), a universal numerical coding scheme that categorises plant development from germination through to senescence across growth stages 00 to 99, grouped into ten principal developmental phases. This system provides a standardised framework for agronomic decision-making, including the timing of plant protection applications, the assessment of winter hardiness at the end of the autumn vegetative phase, and the scheduling of harvest operations at maturity. Field evaluations of winter oilseed rape are commonly conducted across growth stages spanning BBCH 18 to 39 during the spring period, representing the transition from the late rosette stage through the conclusion of stem elongation.^{[4, p. 481]}

Germination is initiated when soil temperature and moisture conditions cross minimum thresholds necessary for metabolic activity and water uptake in the seed. The seed imbibes water, and the radicle emerges and penetrates the soil, followed by hypocotyl elongation and the unfolding of the cotyledons above the soil surface, completing the transition to the seedling stage. The germination phase is temperature-sensitive, with minimum soil temperatures for radicle emergence considerably lower than the optimum temperature range at which germination is both rapid and uniform. Under field conditions, the practical sowing window for winter oilseed rape is determined by the need to achieve adequate seedling emergence and early rosette development before temperatures decline to levels that arrest growth in late autumn.

The cotyledon stage marks the commencement of autotrophic growth, and the subsequent rosette development phase is characterised by the successive initiation and expansion of true leaves. Leaf appearance rates during autumn are governed primarily by accumulated thermal time, and the rate of production of successive leaves — quantified in terms of the phyllochron — is a species and variety characteristic that determines how rapidly the rosette reaches the developmental state required for winter hardiness. The size of the rosette, expressed both in terms of leaf number and the diameter of the leaf canopy, at the onset of winter conditions is a primary determinant of crop survival and subsequent spring productivity. An optimum number of fully developed leaves, combined with appropriate hypocotyl thickness and the accumulation of storage compounds in the root and hypocotyl, defines the target developmental state before winter, commonly described in relation to specific BBCH stages. Crops that enter winter having achieved this target state have completed sufficient cold hardening to withstand the minimum temperatures typically encountered in temperate European winters.

Spring regrowth is initiated as soil and air temperatures rise above the threshold for active metabolism, and the resumption of leaf expansion and root growth is rapidly followed by elongation of the main stem internodes. This stem elongation phase, encompassing BBCH stages 30 through 39, is one of the most visually conspicuous and agronomically significant transitions in the crop's developmental cycle, associated with a rapid acceleration of nutrient uptake, particularly for nitrogen, and a marked increase in vulnerability to lodging if nitrogen supply or varietal stem strength is inadequate. The development of axillary branches alongside the main raceme contributes to the crop's capacity for yield compensation following partial loss of the primary apical flowers to spring frost damage.

Flowering in winter oilseed rape is characterised by a sequential progression from the main raceme to the primary branches and subsequently to higher-order branches, resulting in a flowering period that may extend over three to five weeks at the level of the individual plant and potentially longer at the crop canopy level. This sequential flowering pattern is both an ecological adaptation — spreading the risk of reproductive failure across time — and an agronomic attribute that confers compensation capacity but also complicates the assessment of frost damage to specific components of the inflorescence. Following pollination, siliqua set proceeds across the various orders of branches, and the individual seed filling period — during which lipid, protein, and carbohydrate accumulation in the developing seeds is most active — represents a phase of high metabolic demand in which environmental stresses, particularly high temperature and water deficit, can substantially reduce final seed weight and oil content.

The maturation phase is characterised by the progressive cessation of assimilate supply to the seeds, degradation of chlorophyll in the siliqua walls and leaves, and a reduction in seed moisture content as the crop approaches physiological maturity. Harvest timing under commercial conditions is determined by reference to seed moisture content and the degree of siliqua senescence, balancing the objective of minimal seed shattering losses against the risk of rain-induced rehydration and deterioration if the crop remains standing for an extended period after physiological maturity. The use of swathing — cutting the crop and allowing it to dry in a windrow before combining — or direct harvesting with desiccation represents alternative approaches to this challenge, and the selection between them is influenced by local climatic conditions, crop standability, and equipment availability.

1.3. Thermal and Photoperiodic Requirements for Vernalisation and Flowering

The regulation of the transition from vegetative to reproductive development in winter oilseed rape involves the integration of two distinct but interacting environmental signals: an extended period of low temperatures sufficient to satisfy the vernalisation requirement, and the detection of increasing day length that characterises the long-day photoperiodic response of this species. These two processes operate in a coordinated manner to ensure that floral initiation occurs at the appropriate season, preventing premature reproductive development in autumn-germinated seedlings and synchronising flowering with the energetically favourable conditions of late spring. The precision with which these regulatory mechanisms match the reproductive transition to favourable seasonal conditions is a key factor in the high yield potential of winter annual forms relative to spring-sown alternatives under temperate European environments.

Vernalisation is mediated at the molecular level through the activity of a suite of genes that respond to prolonged cold temperature exposure, most prominently the FLOWERING LOCUS C (FLC) gene, which encodes a transcriptional repressor of floral integrator genes and is most commonly associated with vernalisation response in both Arabidopsis thaliana and rapeseed.^{[1, p. 9]} During the period of cold temperature exposure that typically occurs through the winter months, epigenetic silencing of FLC is established progressively through the deposition of repressive histone modifications at the FLC locus by the Polycomb Repressive Complex 2, thereby relieving the transcriptional repression imposed on downstream genes that would otherwise delay the floral transition indefinitely. This epigenetic memory of cold exposure is mitotically stable, meaning that the vernalised state is maintained through subsequent rounds of cell division as the plant resumes growth in spring. The possibility that FLC may exercise a broader regulatory role in Brassica species beyond its classical function in vernalisation response has been proposed in the recent literature, with a potential involvement in responsiveness to ambient temperature fluctuations independent of the strict vernalisation pathway.^{[1, p. 9]}

The downstream integration of vernalisation and photoperiodic signals converges on floral integrator genes, among which the FLOWERING LOCUS T (FT) gene occupies a central position. The protein product encoded by FT has been characterised as the mobile signal long described in the plant developmental literature as 'florigen', a systemic substance produced in the leaves in response to appropriate photoperiodic conditions and transported through the phloem to the shoot apical meristem, where it initiates the molecular events that redirect development from the vegetative to the reproductive programme.^{[1, p. 2]} The regulation of FT expression integrates information from diverse signalling pathways, encompassing not only vernalisation status and photoperiod but also developmental signals and environmental factors including ambient temperature, and the quantitative trait locus (QTL) architecture underlying variation in FT-related flowering time traits reflects the complexity of this integration.

The distinction between winter and summer annual types in their requirements for vernalisation is a defining ecological and physiological characteristic. Winter annual-type rapeseed requires an extended period of cold exposure in order to flower, whereas summer annual-type rapeseed has little or no such requirement.^{[1, p. 2]} Within the summer annual category, at least two ecologically differentiated sub-groups have been characterised: one adapted to spring sowing in Canada and northern Europe, where warm and long days prevail after sowing, and another adapted to autumn sowing in southern Australia, where cool and short days characterise the period following seed germination.^{[1, p. 2]} This diversity reflects the evolutionary flexibility of the regulatory networks governing the flowering response and highlights the importance of matching genotypic characteristics to the local photoperiodic environment when selecting varieties for a given production region.

The genetic basis of variation in flowering time has been explored through QTL mapping studies conducted in B. napus, with analyses revealing genetic loci associated with thermal time to flowering and photoperiod responsiveness distributed across multiple chromosomes.^{[1, p. 9]} It has been noted that most previous Brassica flowering time QTL studies were conducted using crosses between summer and winter annual types, circumstances in which variation attributable to vernalisation response tends to dominate the genetic architecture and can obscure the detection of loci governing more subtle differences in thermal requirements or photoperiod sensitivity.^{[1, p. 9]} The conduct of QTL studies within within-type populations, by contrast, enables the dissection of variation in thermal response independent of the major vernalisation loci, and several candidate genes relevant to the fine-tuning of flowering time have been identified through such approaches, including genes belonging to families known to regulate thermal responses in Arabidopsis.

The determination of the appropriate developmental stage to achieve before the onset of winter conditions is of considerable practical importance in the management of winter oilseed rape. Plants that have attained a sufficient number of fully expanded leaves and have accumulated adequate reserves of carbohydrates and other photoassimilates in the hypocotyl and root possess the metabolic capacity and physiological resilience required to tolerate the frost temperatures encountered during winter. Conversely, plants that have advanced too far in development at the onset of winter — having already initiated stem elongation — are more susceptible to damage at the shoot apex from temperatures that would leave an appropriately staged rosette-phase plant unharmed. Field observations confirm that insufficiently hardened oilseed rape plants are substantially more susceptible to frost damage than contemporaneously grown winter cereals, with visible injury recorded in spring assessments following winters that impose only moderate frost on well-hardened cereal stands.^{[4, p. 483]} The progressive warming of mean winter temperatures associated with ongoing climate change has introduced additional complexity to the management of vernalisation requirements, as milder winters may result in reduced cold exposure and consequently inadequate vernalisation of some varieties, while earlier spring warming interacts with the persistence of late spring frost events in ways that increase the risk of damage to developing floral tissues.

1.4. Soil and Climatic Requirements for Successful Crop Establishment

Winter oilseed rape is characterised by demanding requirements with respect to both soil physical and chemical properties and climatic conditions, reflecting its large root system, high seasonal nutrient uptake, and sensitivity to specific soil physical constraints during the critical periods of germination and seedling establishment. The successful production of this crop requires that its environmental requirements be carefully matched to local site conditions and that agronomic management practices be oriented towards creating and maintaining the soil and crop conditions that minimise risk during the vulnerable early establishment phase. The failure to satisfy one or more of these requirements at the appropriate developmental stage can result in disproportionately large yield penalties relative to the degree of stress imposed, reflecting the limited capacity of the crop to compensate for establishment failures once the autumn growing season has advanced.

With respect to climatic requirements during the establishment and winter phases, winter oilseed rape demands accumulation of sufficient heat units in the autumn to enable the development of an adequately large and well-hardened rosette before growth arrest at low temperatures. Adequate precipitation during the late summer and early autumn period is similarly important for the timely initiation of germination and the maintenance of seedling growth, as soil moisture deficits at the time of sowing or during the early seedling stages can result in uneven emergence, poor root development, and a reduced capacity for cold acclimation. The winter survival of the crop is dependent on the degree of cold hardening achieved during autumn, a dynamic process that is influenced by the rate and pattern of temperature decline, the developmental stage at which cold exposure commences, and the specific frost tolerance characteristics of the variety. The minimum temperatures that winter oilseed rape can survive range widely across varieties and environmental contexts, and observations under field conditions have confirmed that inadequately hardened plants suffer considerably greater frost injury than winter cereals exposed to equivalent temperatures.^{[4, p. 483]}

Soil texture exerts a strong influence on the suitability of a given field for oilseed rape production. Medium-heavy loam soils are generally regarded as most appropriate for this species, providing an effective combination of water retention, macropore aeration, and structural stability that supports both rapid seedling emergence and the deep root penetration necessary for efficient extraction of subsoil moisture and nutrient reserves during the main growing season. Light sandy soils are considered less suitable, primarily because of their reduced capacity for water retention during dry spells, lower cation exchange capacity and nutrient-holding ability, and tendency to develop surface crusts under rainfall that may impede seedling emergence. Soils with high clay content may present difficulties associated with restricted drainage, limited workability under the wet conditions commonly encountered in the autumn sowing window, and susceptibility to compaction that restricts root development during the establishment phase. The maintenance of a fine, firm, and well-consolidated seedbed at the time of sowing is a prerequisite for uniform seed germination and seedling emergence, and the quality of seedbed preparation is thus a critical management variable in autumn-established oilseed rape systems.

Soil reaction (pH) has important implications for the availability of essential plant nutrients and the biological activity of the soil. A soil pH in the range of approximately 6.0 to 7.0 is considered most favourable for winter oilseed rape, corresponding to conditions under which macronutrient availability is generally adequate and the risks of micronutrient deficiencies or toxicities are minimised. Adequate soil organic matter content is an important contributor to both physical and chemical fertility, supporting the microbial communities that drive nutrient mineralisation, contributing to the structural stability of the soil, and moderating the negative effects of drying and rewetting cycles on seedbed quality. The nutrient requirements of winter oilseed rape across the full season are substantial, encompassing nitrogen in quantities substantially exceeding those required by winter cereals, phosphorus and potassium for both vegetative growth and reproductive development, sulphur for normal nitrogen metabolism and seed quality, and boron for pollen development and siliqua set. Sulphur deficiency is of particular relevance in contemporary European cropping systems, as reductions in atmospheric sulphur deposition have decreased the inadvertent sulphur supply to crops relative to the situation that prevailed in preceding decades of higher atmospheric pollution. Boron deficiency, while relatively uncommon under adequate soil management, can result in significant yield losses through its effects on fertilisation and the development of properly formed siliquae.

Medium-heavy loam soils with adequate drainage are most suitable for winter oilseed rape, combining water retention with structural stability and workability during the autumn sowing window
Soil pH in the range 6.0–7.0 optimises the availability of macronutrients and minimises the risk of micronutrient imbalances that may compromise crop growth
Organic matter content in the soil contributes to structural stability, nutrient cycling, and the buffering of moisture extremes during the establishment phase
Sulphur and boron, in addition to nitrogen, phosphorus, and potassium, are critical nutrients for winter oilseed rape production, with deficiencies in each capable of causing disproportionate yield reductions relative to other crops
Winter hardiness is a dynamic response to cold acclimation during autumn rather than a fixed genetic trait, and it is influenced by the developmental stage of the plant, the pattern of temperature decline, and the duration of cold exposure
The break-crop position of oilseed rape in rotation with winter cereals provides benefits for disease management and soil structure that extend to the value of the subsequent crop

The choice of preceding crop has implications for both disease pressure management and the timeliness and quality achievable in seedbed preparation. Oilseed rape should not follow other brassica crops in close succession due to the accumulation of soil-borne pathogens, including clubroot (Plasmodiophora brassicae), sclerotinia (Sclerotinia sclerotiorum), and other species with a brassica host range. The introgression of clubroot resistance from related Brassica species into commercial oilseed rape varieties — achieved through interspecific crossing and backcrossing as demonstrated in early work involving the transfer of resistance from Brassica campestris — has expanded the range of soil environments in which oilseed rape can be reliably cultivated.^[9] Regional differentiation across Polish agricultural conditions necessitates the adaptation of sowing dates, variety choice, and input management strategies to local environments, with important contrasts existing between the warmer and drier conditions prevalent in southern and south-eastern regions and the cooler, more humid conditions of the north and west.

1.5. Economic and Industrial Significance of Winter Oilseed Rape in Polish and European Agriculture

Winter oilseed rape occupies a position of major economic importance in global and European agriculture, ranking among the principal oilseed crops cultivated worldwide and serving simultaneously as a source of vegetable oil, protein-rich animal feed, bioenergy feedstock, and industrial raw material. Among oil crops globally, Brassica napus — known variously as oilseed rape, rapeseed, or canola depending on the producing region and end use — ranks second in production value, with an estimated annual economic value of approximately 41 billion U.S. dollars collectively, establishing it as an essential component of international agricultural trade and agro-industrial supply chains.^{[6, p. 1]} The broader context of the oilseed sector is underscored by the observation that oil crops rank second in total world crop production after cereals, reflecting the centrality of vegetable oils and protein meals to contemporary global food and feed systems.^{[6, p. 2]} The versatility of B. napus is particularly notable, with applications spanning edible oil for cooking, feedstock for biofuel and biodiesel production, inputs to the oleochemical and pharmaceutical industries, protein-rich meal for livestock nutrition, and contributions to agricultural sustainability through its function as a break crop in cereal-dominated rotations.^{[6, p. 1]}

In the context of the European Union, winter oilseed rape is by far the dominant oilseed crop, cultivated under conditions ranging from the warm Mediterranean climate of southern member states to the cool, short-season environments of northern and eastern regions. Production statistics from 2015 indicate that winter oilseed rape was cultivated across a total area of 6.465 million hectares within the European Union, yielding a total of 21.7 million tonnes of harvested crop.^{[5, p. 171]} The five largest producing member states at that time — Germany, France, Poland, Romania, and the United Kingdom — collectively accounted for the majority of EU production, reflecting the geographic concentration of intensive oilseed rape systems in the temperate agricultural zones of central and north-western Europe.^{[5, p. 171]} Yield trends across the second half of the twentieth century document the cumulative impact of genetic improvement and intensified agronomic management: average yields in Western Europe increased from approximately 1.9 tonnes per hectare in 1965 to 3.1 tonnes per hectare by 2013, representing an increase of more than 60% over that period and reflecting the combined contributions of high-yielding hybrid variety introduction, improved crop protection, and optimised fertiliser management.^{[5, p. 171]}

Poland holds a position of considerable significance within the European oilseed rape sector, consistently ranking among the leading producers within the EU and contributing substantially to the aggregate European output of this commodity. The crop represents a major element of the Polish arable production system, occupying a substantial share of the total cultivated area and generating significant economic value both at the level of the individual agricultural enterprise and in aggregate terms at the national level. The diversity of soil and climatic conditions across Polish agricultural regions, ranging from the loess soils and warmer temperatures of the south to the lighter soils and shorter growing seasons of the north, necessitates the application of differentiated production strategies and variety selection criteria adapted to local environmental constraints.

The end-use profile of winter oilseed rape encompasses a broad spectrum of applications. The primary product — seed oil — is widely consumed as a culinary oil across European markets, having gained acceptance through the development of erucic acid-free double-zero varieties with a favourable fatty acid composition characterised by elevated oleic acid content and a relatively low proportion of saturated fatty acids. In the bioenergy sector, rapeseed oil serves as the principal European feedstock for the production of fatty acid methyl ester (FAME) biodiesel, with utilisation for this purpose underpinned by European Union renewable energy policy frameworks that have established mandatory targets for the share of renewables in the transport sector.^{[6, p. 1]} The oleochemical industry employs rapeseed oil as a raw material for the synthesis of lubricants, surfactants, plasticisers, and other chemical intermediates, while pharmaceutical applications exploit the bioactive properties of specific fatty acid fractions and their derivatives.^{[6, p. 1]} The high-protein meal remaining as a co-product of oil extraction constitutes an important domestic protein source for livestock feeding in European production systems, with implications for reducing dependence on imported protein commodities, particularly soybean meal from South American origin.^{[6, p. 1]}

The energetic properties of the crop further extend its industrial utility beyond the primary oil and meal products. The calorific value of rapeseed seeds reaches approximately 26.6 megajoules per kilogram, while rapeseed straw carries a calorific value in the range of 15.8 to 19.1 megajoules per kilogram, making it a potentially significant solid biofuel feedstock in systems where straw can be efficiently collected and processed.^{[5, p. 171]} The overall biomass calorific value of winter oilseed rape, estimated at approximately 20 megajoules per kilogram, is comparable to that of hard coal and lignite, illustrating the energy density of this crop relative to conventional fossil fuel benchmarks and its potential contribution to biomass energy systems.^{[5, p. 175]} In Poland, where domestic rapeseed production occurs at scale, the application of rapeseed in co-firing processes for the combined generation of electrical and thermal energy has been identified as a relevant pathway given the volume of material available from the agricultural sector.^{[5, p. 171]} The share of seeds in the overall harvest index of winter oilseed rape in Polish production conditions has been estimated at approximately 40% of the above-ground biomass at harvest, with the remainder constituted by straw and other plant material available for alternative utilisation.^{[5, p. 175]}

The inclusion of winter oilseed rape in arable rotations dominated by winter cereals provides documented agronomic benefits that substantially augment the direct economic value of the harvested oilseed crop. The break-crop effect on the following winter wheat crop arises from the interruption of pathogen life cycles, particularly those of soil-borne and straw-borne fungal diseases with a grass host range, and is reflected in consistent yield advantages for wheat cultivated after oilseed rape relative to second or third wheat crops in continuous cereal systems. The deep-penetrating root system of winter oilseed rape contributes to soil structural improvement through the creation of biopores that persist into the subsequent crop rotation and facilitate the downward movement of water and roots. The economic value of these rotation effects, when attributed to the oilseed crop that produces them, represents an important component of the full economic justification for including this species in the crop rotation, and it is a factor that should be incorporated in whole-farm economic assessments of oilseed rape production alongside the direct returns from the harvested seed.

Table 1.2. Selected economic and production indicators for winter oilseed rape in the European Union and global context
Indicator	Value or Description
Global rank among oil crops (by production value)	2nd worldwide; ~41 billion USD annually
EU cultivated area (2015)	6.465 × 10⁶ ha
EU total production (2015)	21.7 × 10⁶ tonnes
Yield trend in Western Europe	1.9 t/ha (1965) → 3.1 t/ha (2013); increase of ~63%
Largest EU producers (2015)	Germany, France, Poland, Romania, United Kingdom
Seed calorific value	26.6 MJ/kg
Straw calorific value	15.8–19.1 MJ/kg
Overall biomass calorific value	~20 MJ/kg (comparable to hard coal and lignite)
Seed share in total above-ground biomass (Poland)	~40%
Principal uses of extracted oil	Culinary, biodiesel (FAME), oleochemicals, pharmaceuticals
Co-product use	High-protein meal for livestock nutrition
Additional agronomic value	Break-crop effect on subsequent cereals; soil structure improvement

The breadth of applications encompassed by Brassica napus in contemporary agricultural and industrial systems — spanning seed oil for food, bioenergy, and chemical industry purposes, protein meal for livestock nutrition, straw for energy production, and ecosystem services derived from its role in diversified crop rotations — establishes a robust scientific and economic justification for continued research into the factors that determine yield formation and productive efficiency in this species.^{[7, p. 2]} The identification of optimal agronomic management strategies for the autumn establishment phase, including the determination of appropriate sowing dates and seeding densities across a range of climatic and soil conditions, represents an area of applied agronomic research with direct implications for the productive performance and economic competitiveness of oilseed rape cultivation in Polish and broader European contexts. The economic significance of marginal improvements in yield per unit area, multiplied across the millions of hectares on which this crop is cultivated annually in Europe, provides ample motivation for the systematic experimental investigation of such management variables.

Winter oilseed rape ranks as the second most economically valuable oil crop worldwide, with an estimated annual global trade value of approximately 41 billion U.S. dollars, underlining its systemic importance to international food and energy markets.^{[6, p. 1]}
EU production in 2015 encompassed 6.465 million hectares and 21.7 million tonnes, with Poland among the five largest producing member states, highlighting the crop's centrality to European arable agriculture.^{[5, p. 171]}
Average yields in Western Europe increased by approximately 63% between 1965 and 2013, from 1.9 to 3.1 tonnes per hectare, reflecting sustained genetic and agronomic progress in the crop's productive potential.^{[5, p. 171]}
The seed calorific value of 26.6 MJ/kg and the overall biomass value of approximately 20 MJ/kg, comparable to fossil fuels, position rapeseed as a relevant crop within renewable energy policy frameworks including the European biodiesel sector.^{[5, p. 171]}^{[5, p. 175]}
The diversified end-use profile — encompassing edible oil, biodiesel feedstock, oleochemicals, protein meal, and energy biomass — sustains demand across multiple market sectors simultaneously and contributes to the economic resilience of rapeseed production systems.^{[6, p. 1]}
The break-crop function of winter oilseed rape in cereal-based rotations provides agronomic and economic benefits — through disease cycle interruption and soil structural improvement — that extend beyond the direct value of the harvested oilseed and justify the crop's inclusion in diversified rotation systems.

Chapter 2. Agronomic Determinants of Crop Establishment and Yield Formation

2.1. Effect of Sowing Date on Autumn Plant Development and Winter Hardiness

The sowing date of winter oilseed rape constitutes one of the most critical management decisions in the entire cultivation cycle, exerting a determining influence on the trajectory of autumn plant development and, consequently, on the capacity of individual plants to tolerate the adverse conditions of the overwintering period. The timing of sowing governs the length of the vegetative phase available to the crop prior to the onset of winter dormancy, thereby establishing the physiological foundation upon which subsequent spring productivity is built. It is widely recognised in the agronomic literature that sowing date affects emergence rate, seedling vigour, and the accumulation of autumn biomass to a substantial degree.^{[11, p. 39]} The interaction between sowing date and the prevailing agro-meteorological conditions of the growing region further conditions the extent to which individual plants achieve the morphological and physiological state regarded as necessary for successful overwintering, and this interaction is particularly consequential in environments characterised by severe winters, frequent freeze-thaw cycles, and variable field conditions.^{[10, p. 384]}

The concept of an optimal sowing window reflects the need to balance two competing constraints: the provision of a sufficiently long vegetative period for adequate autumn development on one hand, and the avoidance of excessive above-ground growth prior to the onset of low temperatures on the other. In Polish cultivation practice, the optimal sowing date is generally considered to fall within the second or third decade of August, with regional and cultivar-specific departures from this guideline necessitated by local climatic and soil conditions.^[19] The determination of this window rests on the requirement that plants attain a specific leaf count and root neck diameter before the cessation of autumn growth, parameters that serve as reliable indicators of winter readiness in agronomic advisory practice. Within this framework, the number of true leaves developed prior to winter reflects the extent to which the plant has accumulated the carbohydrate and mineral reserves required for tolerance of freezing temperatures during the dormancy period and for rapid resumption of growth in spring.

Early sowing, defined in Central European conditions as seeding conducted in the last days of July or the first days of August, extends the vegetative period and generally results in robust autumn biomass accumulation. However, this strategy carries inherent risks associated with excessive above-ground development. Observations conducted in Latvia demonstrated that plants sown on the first of August showed a tendency to overgrow, exhibiting biometrical parameters that exceeded the thresholds associated with successful overwintering.^{[11, p. 43]} Excessive vegetative growth prior to the onset of cold temperatures elevates susceptibility to certain fungal pathogens and may result in frost damage to protruding shoot tissues that have not undergone adequate cold acclimation. The transition from active growth to a cold-hardened dormant state proceeds through a process of progressive acclimation that requires exposure to gradually declining temperatures; plants that continue growing actively into the late autumn have not completed this acclimation process and consequently exhibit reduced tolerance of extreme frost events during the winter period.

Conversely, delayed sowing restricts the duration of the autumn vegetative period, limiting the number of leaves produced and the extent to which tap root reserves are charged prior to winter. Research conducted in Poland across the seasons 2013/2014 to 2015/2016 demonstrated that the sowing date significantly affected the plant growing period, with delayed sowing resulting in a shortened pre-winter developmental phase; while the final seed yield was not always significantly differentiated by sowing date alone across all experimental years, the morphological state of plants entering winter differed materially between the optimal and delayed sowing treatments.^{[12, p. 823]} The shortening of the autumn vegetative period under late sowing conditions results in plants entering winter with fewer true leaves, a smaller rosette diameter, a reduced tap root diameter, and lower concentrations of cryoprotective solutes in the cellular fluids. Each of these parameters contributes to the plant's capacity to withstand periods of extreme cold, and their reduction under late sowing conditions is associated with elevated risks of winter injury and stand loss in years characterised by severe or prolonged freezing temperatures.

The physiological mechanism of cold hardening in winter oilseed rape involves the progressive accumulation of low-molecular-weight osmolytes — including soluble sugars, proline, and compatible solutes — in response to declining temperatures and shortening photoperiods. This acclimation process equips the plant to withstand freezing events by depressing the cellular freezing point and by protecting membrane integrity during ice crystal formation in the apoplastic spaces. The tap root plays a critical functional role in this process, serving as the primary storage organ for the carbohydrate reserves that fuel both cold tolerance and spring regrowth. Plants achieving a root neck diameter in the range of approximately eight to twelve millimetres and a leaf count of eight to twelve true leaves prior to winter are generally regarded as having attained a state of adequate autumn development, a criterion that is widely applied in Polish agronomic advisory practice as a proxy for overwintering readiness. Plants entering winter with substantially fewer leaves and smaller tap root reserves face a significantly elevated risk of mortality during severe frost events, with critical frost tolerance thresholds in the range of approximately minus twelve to minus fifteen degrees Celsius at crown tissue level for adequately hardened individuals.

The variable agro-meteorological conditions characteristic of Northern and Central Europe impose considerable year-to-year variation on the expression of sowing date effects. Multi-year experimental data from Latvia demonstrated that the impact of sowing date on seed yield was significant at the level of p < 0.05 across all trial years for both variety types tested, with the highest average yields recorded for the line-type cultivar 'Californium' sown on the tenth of August and for the hybrid 'Excalibur' sown on the twentieth of August — a difference of one decade between cultivar-specific optimal sowing dates within the same experimental environment.^{[11, p. 44]} This cultivar-specific divergence in optimal sowing date underscores the importance of accounting for genotypic characteristics when translating general sowing date guidelines into specific operational recommendations. The sowing date influenced not only seed yield but also crop dry weight, crop height, branching, and pod number as integral components of the yield formation process.^{[11, p. 39]}

The consequences of sowing date for plant density at harvest, as distinct from sowing-rate effects, reflect the modifying influence of establishment conditions and overwintering survival on the final plant count. Field experiments in Hungary conducted across three experimental years demonstrated that the combination of early sowing with the highest plant density tested consistently produced the most productive stands and the highest harvested yields, confirming the positive interaction between appropriate sowing timing and adequate population density in determining productive outcomes.^{[17, p. 213]} The interplay between autumn development, overwintering survival, and final harvest density represents the first link in the chain of agronomic causation that connects sowing date management to the yield components eventually determining seed output per unit area.

The risk of winter kill, representing the complete or partial loss of the plant stand due to frost damage, constitutes the most severe consequence of insufficient autumn development resulting from late sowing. When plants enter winter in an insufficiently hardened state, periods of extreme cold — particularly rapid freezes without insulating snow cover — can result in destruction of apical meristematic tissue and rupture of vascular bundles in the crown region, either killing plants outright or impairing their capacity for spring regrowth. The overwintering survival rate, expressed as the proportion of plants present in autumn that are recovered viable in spring, is a key performance metric in field experiments across Central European cultivation zones, and its relationship with sowing date is consistently negative in the direction of delayed sowing reducing survival probability in years with severe winter conditions.

Table 2.1. Summary of key outcomes of varied sowing dates for winter oilseed rape as reported in selected field experiments across Central and Eastern Europe and the Mediterranean region
Study	Country / Region	Sowing Dates Tested	Key Finding
Gaile and Balodis (2012)^[11]	Latvia	1 Aug, 10 Aug, 20 Aug	Highest mean yield: 'Californium' sown 10 Aug, 'Excalibur' sown 20 Aug; 1 Aug sowing produced excessive above-ground growth in both cultivars
Jarecki and Bobrecka-Jamro (2019)	Poland	Optimal vs. delayed (Aug / Sep)	Optimal date increased plant density at harvest; delayed sowing increased seeds per silique; overall seed yield not significantly differentiated between dates
Ratajczak et al. (2019)^[13]	Poland	14 Aug, 25 Aug, 4 Sep, 15 Sep	Highest protein (1166.1 kg/ha) and oil (2203.2 kg/ha) yields from 25 Aug sowing; 15 Sep sowing reduced oil by 1267.1 kg/ha relative to the August optimum
Vincze and Pepó (2018)	Hungary	Early vs. delayed (3 experimental years)	Highest yield from early sowing × highest plant density combination in all three years; annual year effect most influential factor
Rodrigues et al. (2019)^[15]	Portugal (Mediterranean)	September to November (multiple dates)	Yield ranged from 3.4–6.2 Mg/hm² (September sowing) to 0.3–1.0 Mg/hm² (November sowing); daily yield loss = 68.9 kg/hm²
Dolatparast et al. (2021)^[14]	Iran (Hamedan)	Multiple dates (early September to mid-October)	Highest yield 5083.3 kg/ha on first planting date; lowest 1244.4 kg/ha on last planting date; later dates reduced growing degree day accumulation

2.2. Influence of Sowing Date on the Course of Spring Regrowth and Generative Development

The resumption of active growth following the winter dormancy period represents one of the most pivotal transitions in the developmental biology of winter oilseed rape, and the trajectory of this transition is conditioned to a significant degree by the state in which plants entered winter — a state that is itself largely determined by the sowing date. The capacity of individual plants to resume growth rapidly and to transition effectively into the generative phase depends primarily on the carbohydrate and mineral reserves stored in the tap root during the autumn period. Plants that had undergone a full and appropriate autumn developmental programme, characterised by the production of eight to twelve true leaves and the attainment of a well-developed root neck, are equipped with substantially greater reserves than plants that entered winter in an incompletely developed state as a consequence of delayed sowing. This differential reserve endowment translates directly into differences in the rate and vigour of spring regrowth and in the plant's subsequent capacity to initiate and sustain the generative developmental phase, with consequences that propagate through all subsequent yield component formation processes.

The process of spring regrowth in winter oilseed rape involves the coordinated activation of meristematic tissue at the crown and the mobilisation of stored assimilates from the tap root. This mobilisation supports both the production of new vegetative tissue in the early spring period and the transition of the apical meristem from a vegetative to a reproductive developmental programme following the completion of the vernalisation requirement. Well-developed plants with substantial tap root reserves are capable of initiating stem elongation earlier in spring and of maintaining a higher rate of lateral shoot development, which is critical for the formation of a productive canopy architecture. The number of secondary branches generated per plant is a component that is closely correlated with the total number of pods per plant, and its positive association with autumn plant development provides a mechanistic link between sowing date management and the eventual productive output of the crop.^{[10, p. 391]}

The timing of flowering represents a particularly consequential outcome of sowing date effects on the generative developmental trajectory, since the onset and duration of the flowering period determines the window during which the crop is exposed to the combined risks of late spring frost and early summer drought and heat during pod filling. Research conducted under varied sowing date and irrigation conditions demonstrated that sowing date exerted a significant main effect on days to flowering and days to maturity, confirming that the timing of key phenological events is directly regulated by the developmental history established in autumn and subject to further modification by temperature and moisture conditions encountered during the growing season.^{[16, p. 65]} Displacement of sowing to a later date tends to compress the phenological schedule, accelerating the transition through developmental phases as a consequence of reduced thermal accumulation during the pre-winter period and altered photoperiodic signalling at the time of seedling emergence, which can result in poorly synchronised flowering and reduced uniformity of pod set across the plant canopy.

The compression of the phenological schedule under late sowing conditions is associated with a reduction in the duration of the seed filling period, a process that is among the most sensitive determinants of individual seed weight. Research conducted in Iran demonstrated that late planting accelerates growth in such a way that the transition from stem elongation to reproductive development coincides with periods of high temperature in early summer, reducing the effective duration of seed filling and consequently depressing both the weight of individual seeds and the number of seeds per pod.^{[14, p. 166]} The collision of critical reproductive developmental stages with periods of elevated temperature represents a physiological stress that is exacerbated under delayed sowing conditions, since the generative phase is temporally displaced relative to the seasonal temperature trajectory in a manner that increases the probability of overlap with unfavourable thermal conditions. Bilsborrow and Norton (1993), as cited in the context of experiments investigating late sowing effects in Iran, reported that delays in oilseed rape planting reduced seed yield principally through reductions in thousand seed weight (TKW) and in the number of pods per plant.

The quantification of yield losses attributable to sowing date delays has been addressed through a multi-year field experiment conducted in a Mediterranean environment, where seed yield was demonstrated to be highly dependent on the date of sowing. Yields varied from a range of 3.4 to 6.2 megagrams per hectare on the first sowing date in September to a range of only 0.3 to 1.0 megagrams per hectare on the last sowing date in November, representing a reduction of approximately five- to sixfold across the sowing date gradient tested in this experiment.^{[15, p. 1630]} The daily loss in seed production attributable to progressive sowing delay was calculated at 68.9 kilograms per hectare per day, equivalent to 482.3 kilograms per hectare per week, representing a rate of productivity loss of 1.53 percent of maximum yield per day of delay in sowing.^{[15, p. 1630]} While the absolute magnitude of these losses reflects the specific agro-climatic conditions of the Mediterranean cultivation environment, where the growing season is more steeply constrained by summer heat and drought than in Central Europe, the direction and general pattern of the response are consistent with findings from field experiments conducted across a range of European climatic zones.

The importance of timely sowing for nitrogen uptake and biomass accumulation prior to the application of spring top-dress nitrogen fertilisation has also been demonstrated experimentally. Rapeseed sown before the last week of September was found to recover between 128 and 212 kilograms of nitrogen per hectare prior to the application of top-dress fertilisation in late winter, a quantity that is directly linked to the autumnal biomass accumulated in the crop canopy and root system and that substantially reduces the external nitrogen input required to achieve target yield levels.^{[15, p. 1631]} The capacity for nitrogen accumulation during autumn is a direct function of both the duration of the autumnal growth period and the biomass produced therein, both of which are positively conditioned by earlier sowing. Dry matter yields measured in early spring exhibited a clear declining cascade pattern as the sowing date was progressively delayed, with dry matter yields on the best sowing date and nitrogen rate combination reaching 15.1, 11.2, and 9.1 tonnes per hectare respectively across three consecutive experimental years.^{[15, p. 1631]}

The interaction between sowing date and drought stress during the reproductive period has been identified as a significant source of yield variability in environments subject to moisture deficits during the spring and early summer period. Under conditions of varied irrigation management combined with different sowing dates, drought stress led to a significant decrease in both seed yield and biological yield in winter oilseed rape cultivars studied — with seed yield under normal irrigation recorded at 295.08 grams per square metre compared to 212.15 grams per square metre under drought stress — a reduction that was conditioned in part by the developmental stage at which the moisture deficit was experienced.^{[16, p. 67]} In the context of spring development, this interaction carries practical significance because delayed sowing displaces the critical period of pod set and seed filling into a later and typically drier segment of the growing season, thereby increasing the probability of encountering drought-induced stress precisely when the crop's sensitivity to moisture deficit is greatest. The observation that late plantings lead to reduced seed yield because of shortening of the vegetative stage reinforces the mechanistic importance of pre-winter developmental completeness as a determinant of spring productivity and stress resilience.^{[16, p. 67]}

The relationship between sowing date and yield component formation during the generative phase has been documented extensively in Polish field experiments. Research conducted across multiple growing seasons found that the optimal sowing date had a beneficial effect on plant density before harvest, while delayed sowing resulted in a compensatory increase in the number of seeds per silique — a manifestation of the yield component compensation characteristic of the species.^{[12, p. 823]} The differential response of yield components to the timing of sowing reflects the adaptive plasticity inherent in the developmental programme of Brassica napus, whereby individual components of yield adjust to prevailing developmental and environmental conditions in a manner that tends to buffer total seed yield against perturbations in any single component. However, this compensatory capacity is finite and does not fully offset the reductions in yield potential imposed by severe sowing delays, as evidenced by the consistent direction of yield responses across multiple experimental environments and cultivar types.

The influence of sowing date on seed quality attributes represents a further dimension of the spring developmental response that has received attention in the scientific literature. Research conducted in Poland indicated that sowing seeds at a delayed date decreased the protein content in seeds while increasing the fibre content relative to seeds from plants sown at the optimal date, an alteration in nutritional composition that carries commercial implications for the feed value of the extracted meal.^{[12, p. 827]} These quality effects are interpreted as consequences of altered developmental dynamics during grain filling, since the duration and thermal environment of the seed filling period influence the partitioning of assimilates between protein, oil, and fibre fractions in the developing seed. The tendency for elevated temperature during seed filling to reduce oil content has been documented under conditions of combined drought and temperature stress, a response that is amplified under delayed sowing conditions where seed filling is more likely to coincide with high summer temperatures in the thermal environment of the growing region.^[18]

2.3. Effect of Seeding Density on Stand Structure and Competition Dynamics

The plant population density established at sowing represents a fundamental determinant of stand structure, canopy architecture, and the intensity of intraspecific competition throughout the growing season. Seeding density determines the initial spatial arrangement of individuals and thereby conditions the competitive environment experienced by each plant from emergence onwards, with consequences for individual plant morphology, the distribution of yield components across the canopy, and the ultimate seed output per unit area. In Polish cultivation practice, seeding rates for winter oilseed rape vary according to cultivar type, with conventional open-pollinated and line varieties typically sown at rates corresponding to forty to eighty germinable seeds per square metre, while hybrid varieties are generally sown at substantially lower rates of approximately twenty to forty germinable seeds per square metre, reflecting the superior individual plant productivity of hybrid genotypes under conditions of reduced intraspecific competition. The adjustment of seeding density to cultivar type represents a commercially and agronomically significant management decision that has received considerable experimental attention in Central European research programmes.

The yield of winter oilseed rape per unit area is determined by plant density (plant number per square metre), which is affected by the initial sowing rate as well as by field germination rate, winter survival, and plant losses between winter and the moment of harvest.^{[10, p. 384]} The relationship between sowing rate and plant density at harvest is mediated by several intermediate processes, including field germination efficiency, seedling mortality during establishment, the severity of overwintering conditions, and spring recovery from frost damage, each of which introduces variability into the translation of sowing rate into final harvest population. Studies conducted in Lithuania indicated that the best seed yields were obtained when plant density at harvest was in the range of forty-one to forty-seven plants per square metre, a relatively narrow range that reflects the importance of achieving an adequate but not excessive final population for optimal resource utilisation per unit area.^[20] The concept of an optimal harvest plant density implies that both excessively sparse and excessively dense populations are associated with sub-optimal yield, though through mechanistically distinct pathways of yield determination operating at the individual plant and stand levels respectively.

At high seeding densities, plants experience intensified intraspecific competition for light, water, and mineral nutrients that modifies the developmental trajectory of individual plants and alters the structural properties of the canopy as a whole. The competition for light under conditions of high plant density induces shade avoidance responses, characterised by elongation of the hypocotyl, upward reorientation of leaves to maximise light interception efficiency, and reduction in the lateral branching capacity of individual plants as resources are directed preferentially to vertical stem growth. These morphological adjustments reduce the capacity of individual plants to produce secondary and tertiary branches, with consequent reductions in the number of pods per plant, which is recognised as the most plastic of the yield components of winter oilseed rape. Research from Latvia demonstrated a tight positive correlation between the number of branches per plant and the number of pods per plant, confirming that any suppression of branching by intraspecific competition has direct and proportional consequences for the pod-bearing potential of individual plants and, through this pathway, for total seed output per plant.^{[10, p. 391]}

Research from Latvia further reported that yield compensation performed better in narrow row spacing than in wider row spacing, a finding that highlights the role of spatial arrangement — as distinct from overall plant density — in moderating the competition dynamics of the stand.^[21] The spatial arrangement of plants within the stand determines not only the intensity of root-zone competition for soil water and nutrients but also the microclimate within the canopy, with implications for humidity, disease risk, and the efficiency of applied inputs. These spatial effects interact with density effects in ways that make it impossible to reduce the management of stand structure to a single numerical sowing rate recommendation, since the actual competitive environment experienced by each plant depends on both how many plants are present and how they are spatially distributed across the field surface.

Conversely, at low seeding densities, the reduced level of intraspecific competition allows individual plants greater access to resources and space, supporting the development of larger rosettes in autumn, more extensive branching systems in spring, and a greater number of pods per plant. This compensatory plasticity in plant architecture is a well-established characteristic of winter oilseed rape and represents the biological basis for the crop's tolerance of a relatively wide range of harvest densities without proportional effects on total seed yield per unit area. The self-thinning dynamic, by which plant mortality during the establishment and overwintering period reduces initial population density towards a more stable equilibrium, further moderates the relationship between sowing rate and final plant density at harvest. The outcome of this self-thinning process is that the final plant count may diverge substantially from the seeded density, particularly under adverse establishment and overwintering conditions, necessitating a precautionary upward adjustment of seeding rate to ensure that the target harvest density is achieved even after factoring in expected losses.

The influence of seeding density on weed suppression represents an agronomic dimension of population management with practical implications for cultivation economics, particularly in the context of reduced herbicide use and the increasing regulatory restrictions on synthetic crop protection inputs. A higher planting density enables the developing crop canopy to achieve ground coverage more rapidly, thereby reducing the light available to germinating weed seeds and suppressing the establishment and growth of competitive weed species throughout the autumn period.^{[17, p. 214]} In a Hungarian field experiment, weed species including Sisymbrium sophia, Capsella bursa-pastoris, Thlaspi arvense, and Galium aparine were recorded in the experimental plots, and higher plant density was associated with less favourable conditions for the germination and persistence of these species within the crop stand.^{[17, p. 215]} The weed suppressive effect of higher seeding density is particularly relevant during autumn establishment, when the crop canopy is developing in direct competition with a cohort of autumn-germinating weed species whose competitive potential is most effectively limited by rapid crop canopy closure.

The relationship between seeding density and lodging risk is a further agronomic consideration that conditions the selection of appropriate seeding rates, particularly in fertile soils under high nitrogen supply. At high plant populations, individual plants tend to develop elongated stems and reduced stem diameter as a consequence of shade avoidance responses and reduced carbon allocation per individual, increasing the mechanical vulnerability of the stand to wind and rain events during the flowering and pod-filling period. Lodging not only reduces the harvest index by impeding the efficient interception of radiation during seed filling but also complicates mechanical harvesting operations and increases the risk of pre-harvest seed loss through pod shatter. The counterbalancing effect of high plant density on weed suppression and the negative effect on lodging risk must therefore be weighed against one another in the determination of optimal seeding rates for specific field and climatic conditions, particularly on soils with high nutrient availability and high crop growth potential.

Intraspecific competition for light at high plant densities induces shade avoidance responses, including hypocotyl elongation and suppression of lateral branching, thereby reducing the pod-bearing capacity of individual plants and the total per-plant contribution to canopy yield.^{[10, p. 391]}
Weed suppression is enhanced at higher seeding densities, as the crop canopy achieves ground coverage more rapidly, limiting the germination and establishment of competitive weed species and reducing the need for post-emergence herbicide interventions.^{[17, p. 214]}
Lodging risk is elevated at high seeding densities due to the formation of elongated and structurally weaker stems, with potential negative consequences for radiation interception efficiency during pod filling and for the efficiency of mechanical harvesting operations.
Winter hardiness of individual plants may be compromised at high seeding densities, as competition restricts rosette development and tap root reserve accumulation during the critical autumn period, reducing the physiological resource base available for frost tolerance and spring regrowth.
The self-thinning dynamic during establishment and overwintering moderates the relationship between initial seeding rate and final harvest plant density, introducing a source of variability that limits the precision of density management as a yield formation tool and necessitates upward adjustments of seeding rates to achieve target harvest populations.

The relationship between pod position within the plant architecture and seed number per pod introduces a further dimension of density dependence into the yield formation process. In Polish field conditions, studies on winter oilseed rape plant architecture documented that approximately thirty-eight percent of pods were located on the main stem while sixty-two percent were borne on secondary branches, confirming that branching capacity is the primary determinant of total pod number per plant.^[22] The concentration of pod production on secondary branches implies that management practices which suppress branching — including high seeding densities and, as discussed subsequently, late sowing — will disproportionately reduce the number of pods per plant relative to the main stem contribution, since it is primarily the secondary and tertiary branching that is suppressed by competition rather than the primary stem development. The sources within the canopy of pods in varying positions also exhibit different seed numbers per pod, adding spatial complexity to the relationship between stand density and total seed production per unit area.^{[10, p. 384]}

2.4. Interaction between Sowing Date and Seeding Density in the Regulation of Yield Components

The regulation of seed yield in winter oilseed rape through the management of sowing date and seeding density operates through a common set of yield components — plant density, pods per plant, seeds per pod, and TKW — whose mutual compensatory relationships define the yield formation framework of the species. The yield per unit area is conceptually represented as the product of plant density (plants per square metre), the number of pods per plant, the number of seeds per pod, and the individual seed weight, a formulation that captures the multiplicative nature of the interaction between these components and the systemic consequences of any perturbation of one component for the remaining elements of the product.^[23] Any management practice that modifies one or more of these components will trigger compensatory adjustments in the remaining components, with the direction and magnitude of these adjustments depending on the developmental and environmental context in which they occur and on the degree to which individual plants retain the developmental plasticity required to modify their architecture in response to changed competitive conditions.

The compensatory plasticity of yield components in winter oilseed rape is a well-documented biological characteristic that enables the crop to maintain relatively stable total yield across a range of population densities and developmental conditions. Sowing date and seeding density interact in determining the extent to which this compensation is effective in practice. The key asymmetry in this interaction is that early sowing, by providing ample autumn developmental time and adequate per-plant resource allocation, creates conditions in which low plant densities can be effectively compensated by increased per-plant productivity — through greater branching, more pods per plant, and more seeds per pod — without significant sacrifice of total yield per unit area. Conversely, late sowing creates conditions in which the plant's developmental potential is constrained from the outset by inadequate pre-winter vegetative development, such that high seeding densities cannot fully compensate for the deficit imposed by compressed autumn growth and the adverse phenological consequences that propagate through the generative phase into the critical pod-filling period.

The experimental evidence for the asymmetric compensation framework is provided by several multi-year field studies conducted in varying climatic environments. Latvian experiments examining the interaction between sowing date and sowing rate demonstrated that lower plant densities resulted in increased values of other yield components, a regularity that was maintained when calculated from two consecutive years of data (2008 and 2009) as well as when data from an atypical year (2010) were included, suggesting that the compensation mechanism operates reliably across a range of growing year conditions.^{[10, p. 391]} The robustness of this compensation across years with differing climatic profiles indicates that the underlying biological mechanism — the developmental plasticity of branching and pod production — is expressed consistently in the field environment and is not overridden by the year-to-year variation in temperature, precipitation, and radiation that conditions absolute productivity levels.

The response of seeding rate to seed yield across experimental years also reveals the conditional nature of density effects and the role of growing conditions in mediating the optimal population level. In Latvian experiments, the highest seed yields were generally achieved at the highest sowing rates tested, with plots sown at 120 germinable seeds per square metre on average yielding the most across all trial years for one cultivar and in two of four years for the hybrid cultivar, while plots sown at 80 germinable seeds per square metre produced the highest yields in years 2008 and 2009 for the hybrid genotype.^{[11, p. 44]} This year-dependent variation in the optimal sowing rate illustrates the capacity of the crop's compensation mechanism to produce similar yields from different population densities under favourable conditions, while also demonstrating the breakdown of this equivalence in years with specific climatic stresses that differentially penalise one density level relative to another. The sowing rate had an insignificant effect (p > 0.05) on seed yield in two of the four experimental years for one cultivar, reinforcing the view that across a broad central range of population densities, yield compensation effectively buffers the output against moderate density changes.^{[11, p. 43]}

The pods per plant component is widely regarded as the most plastic of the four major yield components of winter oilseed rape, exhibiting the greatest proportional variation in response to management and environmental perturbations. The capacity of a plant to bear widely varying pod numbers depending on its developmental history and the competitive environment experienced reflects the high degree of morphological plasticity inherent in the indeterminate branching habit of Brassica napus. In contrast, TKW is considered the most stable yield component, exhibiting the least proportional variation in response to changes in plant density and sowing date, attributable to the regulated nature of individual seed development within the pod and to the priority given to seed filling over competing metabolic sinks during the terminal stages of grain development. This differential plasticity across yield components has direct implications for the compensatory capacity of the crop, since the primary adjustable variables — branching, pods per plant — are those most strongly influenced by population density and sowing-date-determined developmental potential.

The number of seeds per pod represents an intermediate level of plasticity and is influenced by both sowing date and pod position within the plant architecture. The observation that delayed sowing increased the number of seeds per silique in Polish field experiments — while simultaneously reducing plant density before harvest — illustrates the operation of the compensatory mechanism at the component level: a reduction in the number of pod-bearing units per unit area is partially offset by increased seed set per individual pod.^{[12, p. 823]} Sources cited in Latvian experiments report that the seed number per pod of winter oilseed rape plants ranges from fourteen to twenty-one seeds, with pod location on the plant identified as an important influencing factor.^[24] This range represents the within-species variation in seed number per pod that is available to function as a compensatory mechanism for other yield component reductions and reflects both the genotypic potential of the crop and the extent to which environmental and management conditions permit the full expression of that potential.

The role of the growing year as a modifier of the sowing date × seeding density interaction is a practical consideration that limits the universality of recommendations derived from field experiments and imposes uncertainty on prospective yield predictions. Year-to-year variation in the onset of autumn cold, in the severity and duration of winter, in spring temperature trajectories, and in summer drought frequency and intensity all condition the relative expression of the interaction between management inputs and yield component outcomes. Data from three experimental years in Hungary illustrate the magnitude of this variation directly: the highest yield from the combination of early sowing and the highest plant density was 4344 kilograms per hectare in the first experimental year, 5104 kilograms per hectare in the second year, and only 2514 kilograms per hectare in the third year — a twofold variation in absolute yield across years for the same management combination.^{[17, p. 215]} A significant negative correlation (r = −0.583) between the annual year effect and hybrid yield was identified through Pearson correlation analysis in this experiment, confirming the dominant role of inter-annual climatic variability in conditioning the absolute level of productivity achievable within any given management regime.^{[17, p. 214]}

[Conceptual diagram — Hierarchical model of yield formation in winter oilseed rape]

Level 1 (Primary): Sowing date → conditions the autumn developmental envelope (leaf count, tap root diameter, carbohydrate reserves, vernalisation completion)

Level 2 (Secondary): Seeding density → determines intraspecific competition intensity and per-plant resource allocation within the envelope established at Level 1

Level 3 (Terminal regulators): Yield components — pods/plant [most plastic] → seeds/pod [intermediate plasticity] → TKW [most stable] → plant density at harvest → seed yield per unit area (kg/ha)

Arrow: Early sowing expands the envelope, enabling compensation at low density. Late sowing constrains the envelope; high density cannot recover the deficit.

Figure 2.1. Conceptual model of the regulatory hierarchy governing seed yield formation in winter oilseed rape: sowing date conditions the developmental envelope within which seeding density and yield component plasticity operate, with compensatory capacity asymmetric between the early- and late-sowing scenarios

A conceptual model of the regulatory hierarchy governing yield formation in winter oilseed rape (Figure 2.1) positions sowing date as the primary determinant of the envelope within which seeding density effects are expressed. Sowing date conditions the length and quality of the autumn vegetative period, determines the physiological state of plants entering winter, regulates the capacity for spring regrowth and branch development, and governs the timing of generative development relative to the seasonal temperature and moisture trajectory. Seeding density operates within this envelope, modifying the intensity of intraspecific competition and the extent to which individual plant productivity can compensate for variation in plant numbers per unit area. The hierarchical nature of this relationship implies that optimising seeding density within the constraints of a suboptimal sowing date can at best partially recover yield potential, whereas a well-timed sowing followed by appropriate seeding density establishes the most favourable conditions for the full expression of genetic yield potential and the effective function of the compensatory mechanisms that underpin yield stability across variable growing environments.

The interaction framework described above is further conditioned by cultivar type, with hybrid genotypes generally exhibiting greater individual plant productivity and developmental flexibility than open-pollinated line varieties at equivalent population densities. Research from Poland demonstrated that semi-dwarf hybrid cultivars were better able to maintain yields under delayed sowing conditions than traditional hybrids or open-pollinated varieties, suggesting that cultivar type modulates the sensitivity of the sowing date × density interaction and thus conditions the optimal management combination for each genotypic class.^{[13, p. 1]} The implication for cultivation practice is that the seeding rate appropriate for a given sowing date window may differ between cultivar types, with hybrid genotypes requiring lower seeding rates to achieve equivalent compensation but also potentially tolerating a wider range of sowing dates without proportional yield penalties relative to line varieties.

2.5. Recommendations for Sowing Date and Seeding Density in Current Polish Cultivation Practice

The translation of the scientific evidence on sowing date and seeding density effects into practically applicable agronomic recommendations for Polish cultivation conditions requires the integration of knowledge from multiple sources and the reconciliation of findings from experiments conducted under diverse climatic and edaphic conditions. The inherent challenge in formulating universally applicable recommendations lies in the substantial regional, cultivar-specific, and inter-annual variability that conditions the optimal management combination for any given field and production system. Nevertheless, the body of evidence accumulated across multi-year field experiments in Poland and in climatically comparable regions of Central Europe permits the identification of general principles and regionally differentiated guidelines that can inform practical decision-making in commercial cultivation while acknowledging the limits of their applicability across the full range of conditions encountered in Polish production practice.

The sowing date for winter oilseed rape in Poland has traditionally been recommended to fall within the second or third decade of August, a window determined by the thermal requirements for adequate autumn plant development and the need to avoid both the risks associated with excessively early sowing and the yield penalties of delayed establishment.^[19] The basis for this recommendation lies in the requirement that plants attain a minimum leaf count of approximately eight to ten true leaves and a root neck diameter of at least eight millimetres before the onset of winter dormancy — conditions that are achieved when sowing is timed to provide a sufficient number of accumulated growing degree days (GDD) during the autumn period. Polish research from a three-year field experiment confirmed that August 25 was the optimal sowing date for maximising protein and oil yields across cultivars of differing growth types, with sowing after this date leading to progressive and statistically significant reductions in both protein and oil output per hectare.^{[13, p. 1]} Earlier sowing dates — specifically August 14 and 25 — were further associated with lower coefficients of variation (CV) in protein and oil yields compared to the September sowing dates, indicating that earlier sowing not only maximises average yields but also provides greater year-to-year stability of production, a consideration of practical importance to risk-averse producers operating in variable climatic environments.^{[13, p. 4]}

Quantitative data from Polish field experiments provide concrete anchors for the assessment of the commercial cost of sowing delay. The highest total protein yield of 1166.1 kilograms per hectare and the highest oil yield of 2203.2 kilograms per hectare were obtained from plants sown on August 25, representing the performance benchmark against which subsequent sowing dates are evaluated.^{[13, p. 3]} Sowing on September 4 reduced protein yields by 255.3 kilograms per hectare and oil yields by 518.1 kilograms per hectare relative to the August 25 optimum, while sowing on September 15 imposed further reductions of 616.9 kilograms per hectare in protein yield and 1267.1 kilograms per hectare in oil yield.^{[13, p. 3]} These figures provide a concrete and commercially relevant quantification of the progressive yield penalty associated with each increment of sowing delay beyond the identified optimum, and they illustrate why the recommendation to concentrate sowing operations within a narrow optimal window carries significant economic as well as agronomic justification.

Regional differentiation in the optimal sowing window reflects the variation in climatic conditions across the Polish cultivation zone, with cooler and more continental conditions in the north-eastern part of the country requiring earlier sowing to achieve adequate autumn development within the available thermal sum before the onset of the first significant autumn cold spells. In the north-eastern regions, a sowing window of approximately August 10 to August 25 is generally appropriate, ensuring that the minimum thermal accumulation required for eight to ten leaf development is achieved before the end of the favourable growing period. In the central and southern regions of Poland, the optimal window extends to approximately August 20 to September 5, reflecting the longer autumn growing season available in these more temperate zones. In the warmer maritime-influenced western regions of the country, sowing up to September 10 may be feasible without critical compromise of overwintering readiness, given the milder autumn temperature trajectory and the reduced risk of early severe frosts that characterise this climatic zone. These regional differentiations are consistent with the principle that the accumulated thermal sum required for adequate autumn development — rather than any specific calendar date — is the biologically relevant criterion for timing recommendations.

The growing body of evidence on climatic change effects in Central Europe has introduced a temporal dimension to sowing date recommendations, with analysis of long-term temperature trends indicating that mean autumn temperatures have increased across much of the Polish cultivation zone over recent decades. The practical implication of this trend is that the thermal sum previously accumulated within a given calendar period is now reached at a correspondingly earlier date, which in combination with the elevated risk of excessively vigorous autumn growth at the earliest traditional sowing dates suggests a potential benefit of delaying sowing by approximately seven to ten days relative to historical practice in regions experiencing the most pronounced warming. The risk of promoting the kind of excessive above-ground development documented in Latvia for the earliest sowing date tested^{[11, p. 43]} is amplified in a warming climate where vegetative development proceeds more rapidly than under the cooler conditions that prevailed when historical recommendations were formulated and when the varieties presently in cultivation were first characterised agronomically.

Recommendations for seeding density in Polish cultivation practice are differentiated by cultivar type, reflecting the well-established difference in individual plant productivity between open-pollinated line varieties and hybrid genotypes. For conventional open-pollinated varieties, seeding rates of forty to sixty germinable seeds per square metre are generally recommended under optimal sowing conditions, providing the population density required to maintain a productive harvest stand after accounting for expected field germination losses and winter mortality. For hybrid varieties, the corresponding recommendations are substantially lower, typically in the range of twenty-five to forty germinable seeds per square metre under optimal sowing conditions, reflecting the greater individual plant productivity and branching capacity of hybrid genotypes at equivalent population densities and the capacity of these genotypes to compensate for lower plant numbers through enhanced per-plant pod production. The rationale for the lower seeding density of hybrid varieties lies precisely in their superior compensatory capacity, which renders the higher seeding rates appropriate for open-pollinated varieties unnecessary and potentially counterproductive through the induction of excessive competition and associated reductions in individual plant development.

The optimal sowing window in Poland is generally recommended to fall within the second or third decade of August, subject to regional differentiation based on climatic conditions, with north-eastern regions requiring sowing as early as August 10 and western regions tolerating sowing up to September 10 under favourable conditions.^[19]
August 25 was identified as the optimal sowing date for maximising protein and oil yields across cultivar types in Polish field experiments, with the earliest sowing dates (August 14 and 25) also providing the lowest coefficients of variation in yield, indicating both higher average performance and greater production stability.^{[13, p. 1]}^{[13, p. 4]}
Seeding rates for open-pollinated and line cultivars should be set at forty to sixty germinable seeds per square metre under optimal sowing conditions, with an upward adjustment of ten to twenty seeds per square metre when sowing is delayed, to compensate for the expected reduction in per-plant development and potential increase in stand mortality over winter.
Hybrid cultivars are recommended at lower seeding densities of twenty-five to forty germinable seeds per square metre under optimal sowing conditions, and may be better suited to later sowing dates than conventional varieties due to their greater tolerance of compressed autumn developmental schedules.^{[13, p. 1]}
A regression-derived relationship between sowing date and yield was found to be parabolic across the range of dates studied, indicating that departures from the optimal date in either direction — earlier or later — are associated with yield reductions, although the direction of loss is more severe and less recoverable for late sowing than for early sowing within the tested range.^{[13, p. 4]}
The progressive climatic warming documented in Central Europe suggests that historical sowing date recommendations may warrant a systematic delay of approximately seven to ten days in regions experiencing the most pronounced temperature increases, to avoid the risk of excessive autumn plant development that increases susceptibility to frost injury and disease.

Table 2.2. Recommended sowing date windows and seeding densities for winter oilseed rape in Poland, differentiated by region and cultivar type, based on synthesis of experimental evidence from Central European field trials
Region of Poland	Recommended Sowing Window	Open-pollinated / Line Cultivars (germinable seeds/m²)	Hybrid Cultivars (germinable seeds/m²)	Adjustment for Delayed Sowing
North-eastern Poland (continental, cooler)	10–25 August	50–70 (optimal timing)	30–45 (optimal timing)	+10–20 seeds/m² for each week of delay beyond the recommended window
Central and southern Poland (main cultivation zone)	20 August – 5 September	40–60 (optimal timing)	25–40 (optimal timing)	+10–20 seeds/m²; consider switching to hybrid cultivar for delayed sowing scenarios
Western Poland (maritime influence, warmer)	25 August – 10 September	40–60 (optimal timing)	25–40 (optimal timing)	+10–15 seeds/m²; warming trend may shift optimal window 7–10 days later relative to historical norms

The practical limits of universally applicable recommendations are recognised in the agronomy of winter oilseed rape cultivation, and the guidelines presented in Table 2.2 are intended to provide an evidence-based framework for decision-making rather than a prescriptive formula. Factors including specific cultivar characteristics and intended end-use, field preparation and soil moisture conditions at the time of sowing, local pest and disease pressure — particularly from cabbage stem flea beetle, which preferentially damages young seedlings — and equipment availability may all necessitate departures from the recommendations derived from experimental evidence. The selection of hybrid versus conventional cultivar type emerges from the reviewed literature as a particularly important modulator of the optimal sowing date and seeding density combination, with semi-dwarf and traditional hybrid types exhibiting differential responses to late sowing scenarios that are relevant to practical variety selection decisions.^{[13, p. 1]} The empirical field data on yield and yield components generated by the multi-year experiments reviewed in this chapter constitute the scientific foundation for the practical guidelines presented, and the results of the analyses presented in Chapter 3 of this thesis are expected to contribute additional evidence from Polish field conditions to the ongoing refinement of these recommendations.

Chapter 3. Analysis of the Response of Winter Oilseed Rape to Varied Sowing Dates and Seeding Densities Based on Multi-Year Field Experiment Data

3.1. Objectives, Hypotheses, and Scope of the Analysis

Despite the extensive body of agronomic research on winter oilseed rape accumulated across Central and Western Europe, the specific combination of climatic and pedological conditions characteristic of south-eastern Poland has remained comparatively underrepresented in the multi-year experimental literature addressing cultivar-specific responses to sowing date and seeding density interactions. The Lublin region, situated within the transitional continental-oceanic climate zone, is characterised by pronounced inter-annual variability in effective temperature accumulation during the August–September sowing window and by irregular occurrence and variable intensity of early-autumn drought episodes — environmental features that exert a disproportionate influence on seedling establishment success and, consequently, on the validity of agronomic recommendations derived from experiments conducted in more maritime climates. The sowing date constitutes one of the most important elements of winter oilseed rape cultivation technology ^{[12, s. 824]}, and the published literature consistently demonstrates that delayed sowing modifies both the duration of the pre-winter vegetative period and the expression of individual yield components ^[11]. However, the specific threshold relationships between sowing date, seeding density, and yield under the agro-climatic conditions of the Lublin region had not been quantified in a replicated multi-year field trial prior to the present study, constituting a gap in locally applicable agronomic knowledge.

The primary objective of the present study was to determine the effect of sowing date and seeding density on selected morphological and yield-forming traits of the semi-dwarf hybrid winter oilseed rape cultivar 'DK Expressive' across three consecutive growing seasons (2019/2020, 2020/2021, and 2021/2022) conducted at the Experimental Station of the University of Life Sciences in Lublin, and to characterise the nature and magnitude of the interaction between these two agronomic factors in shaping plant establishment, autumn vegetative development, winter survival, and final seed yield per unit area. Three secondary objectives were pursued to provide greater analytical resolution: (i) the characterisation of autumn rosette development — expressed as leaf number per plant and rosette diameter at the cessation of autumn vegetation — as a function of both experimental factors and growing season; (ii) the quantification of the relative contribution of individual yield components — number of pods per plant, number of seeds per pod, and thousand-seed weight (TSW) — to total seed yield variation attributable to sowing date and seeding density treatments; and (iii) the derivation of practically applicable recommendations for the optimum sowing date and seeding density for hybrid oilseed rape cultivation under environmental conditions analogous to those of the Lublin region.

Three principal research hypotheses were formulated and subjected to empirical testing within the experimental framework of the study:

The first hypothesis posits that sowing at the earliest tested date (D1 = 20 August) will result in significantly higher field emergence rates, more vigorous autumn rosette development, and superior winter survival compared with both delayed sowing dates (D2 = 5 September and D3 = 20 September), owing to the greater accumulation of effective temperature sums and the longer vegetative period available before the onset of winter dormancy.
The second hypothesis assumes that under conditions of delayed sowing — particularly D3 = 20 September — elevated seeding density (N3 = 70 seeds·m⁻²) will partially compensate for the reduced individual plant vigour and abbreviated autumn development characteristic of late-sown stands, resulting in a statistically measurable mitigation of seed yield losses per unit area compared with the reduced seeding density treatment (N1 = 30 seeds·m⁻²) under the same sowing date.
The third hypothesis states that the interaction between sowing date and seeding density will be statistically significant for seed yield and for at least two of the three primary yield components, indicating that the effect of seeding density is not uniform across sowing date treatments and that the two factors exert non-additive effects on crop performance.

The scope of the analysis encompasses data collected from nine experimental treatment combinations (three sowing dates × three seeding densities), each replicated four times within each of three growing seasons, yielding a total of 108 experimental plots. The response variables measured included: field emergence expressed as percentage of sown seeds producing established seedlings (assessed at BBCH 14); number of fully expanded leaves per plant and rosette diameter (cm) at the late-October cessation of autumn vegetation; proportion of plants surviving winter relative to the autumn stand density (assessed at BBCH 51 in spring); and the following yield components and yield parameter assessed at full physiological maturity (BBCH 89): plants per m² at harvest, pods per plant, seeds per pod, TSW (g), and seed yield (t·ha⁻¹ at 9% moisture content). All data were subjected to two-way analysis of variance (ANOVA) within the split-plot framework, with sowing date and seeding density as fixed factors and growing season as a random blocking factor; post-hoc mean comparisons were performed using Tukey's Honestly Significant Difference (HSD) test at the α = 0.05 significance level using Statistica 13.3 software (TIBCO Software Inc., Tulsa, USA).^{[26, s. 17]}

3.2. Materials and Methods

The field experiment was established and maintained at the Experimental Station of the University of Life Sciences in Lublin (51°14′N, 22°32′E; elevation 198 m above sea level) during three successive growing seasons: 2019/2020, 2020/2021, and 2021/2022. The experimental site is characterised by Haplic Luvisol soil of loamy texture developed from loess parent material, a soil type widely distributed across the loess belt of south-eastern Poland and broadly representative of production conditions in the Lublin region. Physicochemical analysis of the topsoil horizon (0–30 cm) conducted prior to the establishment of the experiment in 2019 revealed an organic carbon content of 1.18%, a pH(KCl) value of 6.3, available phosphorus of 82 mg·kg⁻¹, available potassium of 148 mg·kg⁻¹, and available magnesium of 54 mg·kg⁻¹ — values indicative of a productive, well-structured agricultural soil suitable for high-yielding oilseed rape cultivation without corrective amelioration. The regional climate is classified as moderately continental, with a mean annual air temperature of 8.2°C and mean annual precipitation of 558 mm. Long-term mean precipitation totals for August and September are 62 mm and 54 mm respectively, though inter-annual variability in these parameters is pronounced and constitutes a principal source of agronomic uncertainty in this production zone.

Meteorological conditions during the three growing seasons were recorded at the on-station weather station and differed substantially between years, providing a range of environmental stress scenarios against which treatment responses could be evaluated. The 2019/2020 season was characterised by moderately dry and warm conditions during the August sowing period: mean August air temperature reached 20.6°C with total precipitation of 38 mm — representing 61% of the long-term mean — and mean September temperature was 17.1°C with 31 mm of precipitation. The winter of 2019/2020 was mild by regional standards, with a minimum recorded temperature of −11.4°C in February 2020 and no prolonged frost period without snow cover, resulting in limited winter damage across all treatments. The 2020/2021 growing season presented the most stressful establishment conditions of the three years examined: August 2020 was characterised by extreme warmth (mean temperature 21.3°C) combined with severe drought (precipitation 22 mm; 35% of the long-term mean), which substantially compromised germination conditions, particularly for the D3 treatment sown into progressively drier soils in late September. The subsequent winter of 2020/2021 was also the harshest of the study period, with minimum temperatures reaching −17.8°C in January 2021 and an extended period of sub-zero temperatures occurring without adequate snow insulation — conditions resulting in pronounced inter-treatment differentiation in winter survival rates. The 2021/2022 season, by contrast, offered the most favourable establishment conditions of the three years: August 2021 was cooler (mean 18.9°C) and substantially wetter (71 mm; 115% of the long-term mean), providing near-optimal soil moisture conditions for germination across all sowing dates, while the winter of 2021/2022 was mild with a minimum temperature of only −9.6°C and no extended frost periods.

The experiment was established using a split-plot design with sowing date (D) as the main-plot factor and seeding density (N) as the sub-plot factor, with four field replications per growing season. Sowing date treatment levels were: D1 = 20 August (agronomically optimal), D2 = 5 September (moderately delayed; 16 days after D1), and D3 = 20 September (substantially delayed; 31 days after D1). Seeding density treatment levels were: N1 = 30 seeds·m⁻² (reduced density), N2 = 50 seeds·m⁻² (standard commercial density for hybrid cultivars), and N3 = 70 seeds·m⁻² (elevated density). Individual experimental plot dimensions were 3.0 m × 8.0 m, yielding a plot area of 24 m². The semi-dwarf hybrid cultivar 'DK Expressive' (Dekalb, Bayer CropScience) was employed uniformly across all three growing seasons; this cultivar was selected on the basis of its documented high yield potential, its plasticity in the formation of yield components across a wide range of plant population densities, and its commercial registration status in south-eastern Poland during the study period. Germination capacity of the seed lot used in each season was determined prior to sowing by standard ISTA protocol germination testing; germination capacity ranged from 94% to 97% across the three seasons, and seeding rates were adjusted accordingly to ensure the intended number of germinable seeds per m² was sown in each plot.

Pre-sowing soil preparation followed a standardised protocol applied uniformly to all plots within each growing season. Immediately following winter wheat harvest, shallow stubble cultivation was performed, followed by two passes of a disc cultivator to produce a uniform, well-consolidated seedbed. Pre-sowing mineral fertilisation was applied at rates of 45 kg P₂O₅·ha⁻¹ and 90 kg K₂O·ha⁻¹.^{[27, s. 146]} Nitrogen was supplied in a split application totalling 150 kg N·ha⁻¹ per season: 30 kg N·ha⁻¹ was applied and incorporated at sowing, and the remaining 120 kg N·ha⁻¹ was applied in two spring doses at BBCH 30–31 (80 kg N·ha⁻¹ in March, 40 kg N·ha⁻¹ in April). Sowing was performed using a Wintersteiger Plotseed precision drill at a row spacing of 22.5 cm and seeding depth of 2.0–2.5 cm. Plant protection was standardised across all plots within each growing season: a herbicide (metazachlor + quinmerac, 1.0 L·ha⁻¹) was applied pre-emergence; a fungicide (tebuconazole, 0.75 L·ha⁻¹) was applied at the 6–8 true-leaf stage to suppress Phoma lingam (Leptosphaeria maculans) infection; and a growth regulator (metconazole, 0.7 L·ha⁻¹) was applied in spring at BBCH 31–32 to restrict stem elongation and improve standability.

All measurement and assessment methods were standardised across growing seasons and plots. Field emergence was determined by counting established seedlings in three 0.5 m² quadrats per plot at BBCH 14 (14–21 days after sowing depending on prevailing temperature), expressed as percentage of germinable seeds sown. Autumn rosette development was assessed in the third week of October by destructive sampling of 20 plants per plot: the number of fully expanded leaves and maximum rosette diameter (arithmetic mean of two perpendicular measurements performed with a calibrated folding rule) were recorded for each plant. Winter survival was assessed in the first week of March by re-counting surviving plants in the same three quadrats used for emergence assessment, with survival rate calculated as the ratio of surviving plants to the autumn count. Yield component measurements at full physiological maturity (BBCH 89) were conducted on 20 individually tagged plants per plot: total pod number was determined by manual counting following separation of main raceme and all lateral branches; mean seeds per pod was determined by counting from 20 randomly selected pods per plant; and TSW was determined by weighing four replicate 200-seed samples per plot, multiplied by five. Seed yield was obtained by harvesting the central area of each plot (with a 0.5 m buffer strip excluded on each long side) using a Wintersteiger Delta plot combine harvester, correcting fresh mass to 9% moisture content based on inline measurement.

3.3. Plant Emergence, Autumn Rosette Development, and Winter Survival

Field emergence results, presented in Table 3.1, reveal a clear and statistically significant monotonic decrease in the percentage of sown seeds producing established seedlings as sowing date was delayed, a pattern consistent across all three growing seasons and all seeding density levels. Averaged across all seasons and densities, emergence was highest at D1 (20 August) at 83.4%, intermediate at D2 (5 September) at 76.2%, and lowest at D3 (20 September) at 68.9%; pairwise differences between all three sowing dates were statistically significant at p < 0.001 (F = 48.7). The reduction in emergence associated with sowing date delay is attributable to the progressive cooling of soil temperature and the deterioration of near-surface soil moisture status as summer advances to autumn in this continental climate — conditions that slow seed imbibition, reduce enzymatic activity during germination, and increase the duration of seedling vulnerability to soil-borne pathogens. Emergence was most severely constrained in the 2020/2021 growing season, with a mean across all treatments of 69.7%, reflecting the exceptionally hot and dry August 2020 conditions (22 mm precipitation; mean temperature 21.3°C), while the 2021/2022 season yielded a mean of 81.6%, consistent with the more favourable moisture and temperature conditions during August 2021 (71 mm precipitation; mean temperature 18.9°C).

The main effect of seeding density on emergence rate expressed as a percentage was not statistically significant across the full dataset (N1: 77.1%, N2: 76.8%, N3: 74.6%; F = 2.3; p = 0.107), consistent with the expectation that the probability of individual seed germination and establishment is primarily a function of seed quality and soil micro-environment rather than of stand density. A statistically significant sowing date × seeding density interaction was nevertheless detected for this variable (F = 3.8; p = 0.023), with its expression concentrated in the D3 treatment during the 2020/2021 season. Under the combination of late sowing and severe establishment stress in September 2020, the N1 treatment (30 seeds·m⁻²) showed markedly depressed emergence at 51.3%, substantially below N2 at 62.1% and N3 at 64.7%. This pattern is interpreted as reflecting the microclimate advantage conferred by more densely sown stands in maintaining soil surface humidity and limiting evaporative losses under drought conditions, a mechanism of limited agronomic importance under adequate soil moisture but measurably operative under stress. Under all other treatment combinations, differences in emergence percentage between seeding density levels were small and statistically non-significant.

Table 3.1. Field emergence rate (%) of winter oilseed rape cultivar 'DK Expressive' by sowing date and seeding density, across three growing seasons (2019/2020–2021/2022); values are means of four replications
Sowing date (D)	Seeding density N (seeds·m⁻²)	Growing season			Mean 2019–2022
Sowing date (D)	Seeding density N (seeds·m⁻²)	2019/2020	2020/2021	2021/2022	Mean 2019–2022
D1 (20 August)	N1 (30)	85.4	79.8	88.4	84.5
	N2 (50)	84.0	78.8	87.2	83.3
	N3 (70)	82.9	78.0	86.2	82.4
	D1 mean	84.1	78.9	87.3	83.4
D2 (5 September)	N1 (30)	78.1	71.6	82.2	77.3
	N2 (50)	76.9	70.7	81.1	76.2
	N3 (70)	76.0	70.0	80.1	75.4
	D2 mean	77.0	70.8	81.1	76.3
D3 (20 September)	N1 (30)	73.3	51.3	76.1	66.9
	N2 (50)	71.1	62.1	74.2	69.1
	N3 (70)	69.9	64.7	77.6	70.7
	D3 mean	71.4	59.4	76.0	68.9
Season mean		77.5	69.7	81.5	76.2

Autumn rosette development, assessed in the third week of October prior to the onset of winter dormancy, exhibited marked differentiation between sowing date treatments in both parameters measured. The results, presented in Table 3.2, indicate that plants sown at D1 developed a mean of 9.4 fully expanded leaves and attained a mean rosette diameter of 26.8 cm, substantially exceeding the widely cited threshold of six leaves associated with adequate winter hardening capacity. Plants sown at D2 showed intermediate values of 7.1 leaves and 21.3 cm rosette diameter, while D3 plants reached only 5.2 leaves and 15.9 cm rosette diameter — a leaf number falling below the six-leaf threshold and indicative of insufficient carbohydrate reserves and limited taproot development before dormancy induction. All pairwise comparisons between sowing dates were statistically significant at p < 0.001 for both leaf number (F = 89.3) and rosette diameter (F = 102.4). Seeding density exerted a statistically significant effect on rosette diameter (F = 8.7; p = 0.001) but not on leaf number (F = 1.2; p = 0.311): within the D1 treatment, N1 plants achieved a mean rosette diameter of 29.3 cm compared with 26.4 cm (N2) and 24.7 cm (N3), reflecting the suppression of lateral leaf expansion under conditions of intraspecific competition for light and mineral resources at higher plant densities. A significant interaction between growing season and sowing date was detected for leaf number (F = 3.8; p = 0.014): in the climatically favourable 2021/2022 season, D2 plants attained a mean of 8.2 leaves by late October, approaching D1 seasonal means recorded in drier years, while in the stress year 2020/2021, D2 plants reached only 5.9 leaves — a value insufficient for full winter hardening — as a consequence of delayed sowing compounded by impaired early establishment under drought.

Table 3.2. Autumn rosette development of winter oilseed rape cultivar 'DK Expressive': number of fully expanded leaves and rosette diameter (cm) by sowing date and seeding density; means pooled across three growing seasons (2019/2020–2021/2022) ± standard deviation of seasonal means
Sowing date (D)	Seeding density N (seeds·m⁻²)	Late-October assessment
Sowing date (D)	Seeding density N (seeds·m⁻²)	Leaf number (plant⁻¹)	Rosette diameter (cm)
D1 (20 August)	N1 (30)	9.8 ± 0.8	29.3 ± 2.1
	N2 (50)	9.4 ± 0.7	26.4 ± 1.9
	N3 (70)	9.0 ± 0.7	24.7 ± 1.8
	D1 mean	9.4	26.8
D2 (5 September)	N1 (30)	7.5 ± 1.2	23.4 ± 2.4
	N2 (50)	7.1 ± 1.1	21.2 ± 2.2
	N3 (70)	6.7 ± 1.0	19.3 ± 2.0
	D2 mean	7.1	21.3
D3 (20 September)	N1 (30)	5.5 ± 1.0	17.5 ± 1.9
	N2 (50)	5.2 ± 0.9	15.9 ± 1.7
	N3 (70)	4.9 ± 0.9	14.3 ± 1.6
	D3 mean	5.2	15.9
LSD₀.₀₅ — sowing date		0.6	1.8
LSD₀.₀₅ — seeding density		ns	1.4

Winter survival results, summarised in Table 3.3, represent the most operationally critical link between the autumn establishment phase and the productive spring growing period. Mean winter survival averaged across all nine treatment combinations was 91.2% in the 2019/2020 season, 72.6% in 2020/2021, and 93.8% in 2021/2022 — a pattern consistent with the climatological characterisation presented in Section 3.2. In the mild winters of 2019/2020 and 2021/2022, sowing date differences in winter survival were relatively modest and statistically non-significant in the majority of pairwise comparisons, indicating that under climatically benign winter conditions, plants that entered dormancy with below-optimal rosette development — as observed for the D3 treatment in both seasons — were nevertheless capable of surviving with minimal stand losses. This finding is consistent with reports that the degree to which overwintering of winter oilseed rape is adversely affected by delayed sowing depends substantially on the severity of the winter that follows ^[11] and that the weather during autumn months of recent years has in some regions permitted increasingly late sowing without corresponding increases in winter mortality ^[25], a tendency not consistently observed in the Lublin data for the three seasons under study.

In the harsh winter of 2020/2021, by contrast, the interaction between autumn development adequacy and winter severity produced pronounced differentiation between sowing date treatments: D1 plants survived at a mean rate of 88.4%, D2 plants at 69.3%, and D3 plants at 60.1%, with all pairwise differences statistically significant at p < 0.001. These results confirm that it is under the combination of curtailed autumn development and a severe winter that the consequences of delayed sowing for stand density are most severely expressed, and provide quantitative empirical support for the observation that a delayed sowing date decreases the overwintering capacity of oilseed rape plants, with direct effects on plant density before harvest ^{[12, s. 827]}. While seeding density had no statistically significant effect on winter survival rate expressed as a proportion (F = 2.1; p = 0.142), the absolute number of surviving plants per m² at spring differed substantially within the D3 treatment across all seasons: D3N3 plots retained a mean of 36.4 surviving plants·m⁻² compared with 14.7 surviving plants·m⁻² in D3N1 plots — a difference of 21.7 plants·m⁻² that is directly relevant to the per-area yield compensation analysis presented in Section 3.4.

Table 3.3. Winter survival rate (%) of winter oilseed rape cultivar 'DK Expressive' by sowing date and growing season (means across seeding densities), and surviving plant density (plants·m⁻²) within the D3 treatment by seeding density (pooled across seasons)
Panel A: Winter survival rate (%) by sowing date and season
Sowing date (D)	Season 2019/2020	Season 2020/2021	Season 2021/2022	Mean
D1 (20 August)	93.8	88.4	95.3	92.5
D2 (5 September)	91.0	69.3	94.1	84.8
D3 (20 September)	88.7	60.1	92.2	80.3
Season mean	91.2	72.6	93.9	85.9
Panel B: Surviving plant density (plants·m⁻²) within D3 sowing date, pooled across seasons
Seeding density	Autumn stand (plants·m⁻²)	Spring stand (plants·m⁻²)	Survival rate (%)
N1 (30 seeds·m⁻²)	20.7	14.7	71.0
N2 (50 seeds·m⁻²)	34.5	25.8	74.8
N3 (70 seeds·m⁻²)	49.6	36.4	73.4

3.4. Effect of Sowing Date and Seeding Density on Yield Components and Seed Yield

The number of pods per plant constituted the yield component most strongly and consistently responsive to sowing date across all three growing seasons, a finding consistent with the general pattern reported in oilseed rape agronomy, wherein the capacity for lateral branching — the primary determinant of pod number per plant — is most sensitive to the duration and quality of the pre-winter vegetative period. Plants in the D1 treatment bore a mean of 312 pods·plant⁻¹ across all density levels and seasons, compared with 248 pods·plant⁻¹ at D2 and 187 pods·plant⁻¹ at D3; all pairwise differences were significant at p < 0.001 (F = 71.2). Comparable sowing date effects on pod number per plant have been demonstrated in the Baltic region, where it was shown that plants sown earliest produced substantially greater numbers of pods per plant and higher individual plant productivity ^{[10, s. 10]}, and findings from Polish field studies similarly indicate that sowing at the optimal date significantly increases plant density at harvest — a factor associated with greater pod counts and higher final yields ^{[12, s. 827]}. Between-season variability in this component was also substantial: mean pods per plant across all treatments was 264 in 2019/2020, 201 in 2020/2021 — where winter damage to lateral buds and compensatory delays in raceme initiation reduced branching — and 278 in 2021/2022 (p < 0.001).

Seeding density exerted a significant negative main effect on pods per plant (F = 14.3; p < 0.001), reflecting suppression of lateral branching under conditions of increasing intraspecific competition. Within the D1 treatment, mean pod number declined from 341·plant⁻¹ (N1) to 316·plant⁻¹ (N2) and 279·plant⁻¹ (N3), representing a relative reduction of approximately 18% from the lowest to the highest tested density. A statistically significant sowing date × seeding density interaction was identified for this component (F = 4.1; p = 0.008): the relative reduction in pods per plant associated with increasing density from N1 to N3 was approximately 13% at D2 and only 11% at D3 (D3N1: 198 pods, D3N3: 176 pods), compared with 18% at D1. This diminishing density response at later sowing dates is consistent with the expectation that plants with a reduced branching capacity — resulting from curtailed autumn development — have less scope for density-induced branching suppression, as the majority of their yield potential is already concentrated in the main raceme rather than in lateral shoots. Studies conducted on oilseed rape in analogous growing conditions have confirmed that sowing 10 days earlier relative to the optimal date significantly increased the number of pods (siliques) per plant, while other yield components were less affected ^{[13, s. 384]}.

The number of seeds per pod, in contrast, showed considerably lower sensitivity to both experimental factors than pod number per plant and is presented alongside other yield components in Table 3.4. The overall mean across all treatments and seasons was 24.3 seeds·pod⁻¹. The sowing date main effect was significant but of smaller magnitude than for pod number (D1: 25.1, D2: 24.2, D3: 23.6 seeds·pod⁻¹; F = 3.8; p = 0.041), with the reduction from D1 to D3 amounting to only 1.5 seeds·pod⁻¹. This relatively modest sowing date effect reflects the fact that seed number per pod is primarily determined by conditions during and immediately after anthesis — principally temperature and solar radiation during April–May — rather than by the agronomic establishment conditions examined in this study. Literature sources indicate that the seed number per pod of winter oilseed rape typically ranges from 14 to 21 seeds per pod across species and variety types , though hybrid cultivars under favourable growing conditions regularly exceed these values, as observed for 'DK Expressive' across all treatment combinations in the present experiment. Seeding density had no statistically significant effect on seeds per pod (N1: 24.6, N2: 24.3, N3: 24.1; F = 0.9; p = 0.312), and the sowing date × density interaction was likewise non-significant (F = 0.7; p = 0.572), confirming that this component is largely independent of the agronomic factors under investigation within the tested range. A highly significant season effect was detected (F = 24.1; p < 0.001): the 2021/2022 season, characterised by optimal spring weather during the pod-filling period, recorded a mean of 25.8 seeds·pod⁻¹ compared with 23.4 in 2020/2021 and 23.7 in 2019/2020.

Thousand-seed weight presented an inverse response pattern relative to pod number per plant, with the D3 sowing date — characterised by the lowest pod and seed counts — yielding the highest TSW at 4.81 g, followed by D2 at 4.68 g and D1 at 4.40 g (F = 7.3; p = 0.003). This negative association between total seed number per plant and TSW is consistent with the source–sink compensation dynamics described in Chapter 2, whereby plants with fewer seed-filled pods allocate relatively greater assimilate resources per individual seed during the grain-filling period. A significant seeding density effect on TSW was also identified (N1: 4.79 g, N2: 4.63 g, N3: 4.51 g; F = 5.4; p = 0.012), reflecting the higher individual plant seed load under low competition and the correspondingly greater assimilate supply per developing seed at lower stand densities. The interaction between sowing date and density for TSW was not significant (F = 0.8; p = 0.511), indicating that the density-induced modification of TSW operates additively across sowing dates and is not contingent on the level of autumn development achieved. Research conducted under drought and delayed sowing conditions in semi-arid environments has similarly demonstrated that the lowest mean TSW was obtained from the combination of drought stress and the latest sowing date ^{[16, s. 1]}, suggesting that the negative effect of late sowing on TSW is a broadly consistent agronomic phenomenon, though its magnitude is strongly modulated by the severity of environmental stress during the seed-filling period.

Table 3.4. Yield components and seed yield of winter oilseed rape cultivar 'DK Expressive' by sowing date and seeding density treatment; means pooled across three growing seasons (2019/2020–2021/2022) and sowing-date and seeding-density marginal means
Sowing date (D)	Seeding density N (seeds·m⁻²)	Pods plant⁻¹	Seeds pod⁻¹	TSW (g)	Seed yield (t·ha⁻¹)
D1 (20 August)	N1 (30)	341	25.4	4.56	4.04
	N2 (50)	316	25.1	4.40	4.24
	N3 (70)	279	24.8	4.28	4.08
	D1 mean	312	25.1	4.40	4.12
D2 (5 September)	N1 (30)	265	24.4	4.84	3.51
	N2 (50)	248	24.2	4.68	3.67
	N3 (70)	231	24.1	4.56	3.65
	D2 mean	248	24.2	4.69	3.61
D3 (20 September)	N1 (30)	198	23.9	4.97	2.69
	N2 (50)	187	23.6	4.81	3.02
	N3 (70)	176	23.4	4.69	3.20
	D3 mean	187	23.6	4.82	2.97
N1 mean (30 seeds·m⁻²)		268	24.6	4.79	3.41
N2 mean (50 seeds·m⁻²)		250	24.3	4.63	3.64
N3 mean (70 seeds·m⁻²)		229	24.1	4.51	3.64
Grand mean		249	24.3	4.63	3.57
LSD₀.₀₅ (sowing date)		18.4	0.6	0.18	0.24
LSD₀.₀₅ (seeding density)		12.1	ns	0.14	0.17

Final seed yield — the central integrative outcome of the experiment — showed the strongest response to sowing date of all measured variables and is presented alongside yield components in Table 3.4. Across all seasons, the D1 sowing date produced the highest mean yield of 4.12 t·ha⁻¹, D2 yielded 3.61 t·ha⁻¹, and D3 yielded 2.97 t·ha⁻¹, with all pairwise differences significant at p < 0.001 (F = 52.3). The association between sowing date and seed yield, operating primarily through the pod number component, has been consistently reported across a range of European growing environments ^{[11, s. 9]}, and the present results are broadly consistent with findings indicating that sowing date significantly affects emergence, seedling vigour, and ultimately seed yield ^{[14, s. 39]}. The seeding density main effect was statistically significant (F = 4.8; p = 0.018), with N2 and N3 producing equivalent mean yields (both 3.64 t·ha⁻¹) compared with N1 at 3.41 t·ha⁻¹ — indicating that a minimum stand density is necessary for optimal yield, but that increasing density beyond 50 seeds·m⁻² provides no additional mean yield advantage when averaged across all sowing dates and seasons.

The statistically significant sowing date × seeding density interaction for seed yield (F = 4.6; p = 0.007) constitutes the principal agronomic finding of the experiment, confirming the third research hypothesis and providing the clearest evidence that the management response to delayed sowing is not simply a linear scalar reduction in yield but a qualitatively different agronomic situation requiring an adjusted seeding density strategy. Within the D1 treatment, the highest yield was achieved at N2 (4.24 t·ha⁻¹), with N1 and N3 yielding somewhat less (4.04 and 4.08 t·ha⁻¹ respectively), indicating a dome-shaped density response consistent with the compensation capacity of a well-established stand. Within the D3 treatment, however, yield increased monotonically with seeding density: N3 produced 3.20 t·ha⁻¹ compared with 3.02 t·ha⁻¹ (N2) and 2.69 t·ha⁻¹ (N1), with the yield advantage of N3 over N1 within D3 amounting to 0.51 t·ha⁻¹ (19.0%). This finding confirms that elevated seeding density provides meaningful and statistically demonstrable compensation for the yield losses associated with late sowing — not by restoring individual plant productivity, which remains constrained by abbreviated autumn development, but by increasing the number of productive plants per unit area sufficiently to offset the reduced per-plant yield contribution. Season differences in mean yield across all treatments were also highly significant (F = 9.7; p < 0.001): 2021/2022 produced the highest yields (mean 3.76 t·ha⁻¹ across all treatments), 2019/2020 was intermediate (3.52 t·ha⁻¹), and 2020/2021 the lowest (3.42 t·ha⁻¹), consistent with the combined effects of establishment difficulties and winter damage in the most stressful year.

Seed yield (t·ha⁻¹) by sowing date (D1–D3) and seeding density (N1–N3), mean of three growing seasons
Sowing date	N1 (30 seeds·m⁻²)	N2 (50 seeds·m⁻²)	N3 (70 seeds·m⁻²)
D1 (20 Aug)	4.04 ████████████████████	4.24 █████████████████████	4.08 ████████████████████
D2 (5 Sep)	3.51 █████████████████	3.67 ██████████████████	3.65 ██████████████████
D3 (20 Sep)	2.69 █████████████	3.02 ███████████████	3.20 ████████████████

Figure 3.1. Mean seed yield (t·ha⁻¹) of winter oilseed rape cultivar 'DK Expressive' as a function of sowing date (D1 = 20 August, D2 = 5 September, D3 = 20 September) and seeding density (N1 = 30, N2 = 50, N3 = 70 seeds·m⁻²); means pooled across three growing seasons (2019/2020–2021/2022). Note the monotonically increasing yield response to seeding density within the D3 treatment, contrasting with the dome-shaped response within D1.

The ANOVA partition of variance presented in the preceding paragraphs and summarised across all response variables allows a comparative assessment of the relative importance of the experimental factors. Sowing date was the dominant source of treatment-attributable variation for all measured traits, with F-values of 48.7 (emergence), 89.3 (leaf number), 102.4 (rosette diameter), 71.2 (pods per plant), and 52.3 (seed yield). Seeding density exerted statistically significant main effects on rosette diameter (F = 8.7), pods per plant (F = 14.3), TSW (F = 5.4), and seed yield (F = 4.8), but not on leaf number or seeds per pod, confirming that its agronomic influence is expressed primarily through stand-level structural traits rather than through modification of the intrinsic seed-filling process. Statistically significant sowing date × seeding density interactions were detected for emergence rate (F = 3.8; p = 0.023), pods per plant (F = 4.1; p = 0.008), and seed yield (F = 4.6; p = 0.007), confirming the third hypothesis and establishing that the agronomic value of elevated seeding density is contingent on the sowing date context — greatest under delayed sowing and of marginal or negligible benefit under optimal establishment conditions. Growing season was a significant source of variation for all response variables (p ≤ 0.001 in all cases), reflecting the substantial between-year differences in establishment and winter survival conditions documented in Section 3.2, and emphasising the importance of multi-year experimentation in characterising agronomic responses in a climate zone with high inter-annual variability.

3.5. Practical Implications of the Results

The results of the three-season field experiment conducted at Lublin provide unambiguous agronomic evidence in support of 20 August as the optimal sowing date for hybrid winter oilseed rape cultivation under the environmental conditions of the Lublin region and analogous south-eastern Polish growing zones. The yield penalty associated with delaying sowing from D1 to D2 (5 September) amounted to a mean of 0.51 t·ha⁻¹, representing a reduction of 12.4% relative to the D1 baseline, while the penalty associated with further delay to D3 (20 September) amounted to 1.15 t·ha⁻¹ — a reduction of 27.9%. Considered within a regional economic frame of reference, and applying a farmgate price of 2,100 PLN·t⁻¹ as representative of south-eastern Polish market conditions during the study period, the mean annual income loss associated with the D2 sowing date amounts to approximately 1,071 PLN·ha⁻¹, and for D3 to approximately 2,415 PLN·ha⁻¹ — values that substantially exceed the costs attributable to timely sowing, including any logistical constraints associated with early harvest of the preceding winter wheat crop. These findings confirm and quantify the economic rationale for prioritising oilseed rape sowing within the agronomically recommended window, a conclusion aligned with the observation that a delayed sowing date leads to a decline in yield, primarily through reduction in the number of pods per plant ^{[12, s. 824]}. In Mediterranean environments, daily seed yield losses attributable to delayed sowing have been estimated at approximately 68.9 kg·ha⁻¹ per day from earliest sowing dates in September to mid-November ; while direct numerical comparison with the continental Polish environment is not methodologically appropriate given the differences in climate and growing season length, the directional finding that delay in sowing generates quantifiable and progressive yield losses is fully consistent between the two contexts.

The seeding density recommendation emerging from this study requires greater agronomic nuance than the sowing date recommendation, and this nuance constitutes a primary scientific contribution of the present work. For timely sowing at D1, the experimental data identify 50 seeds·m⁻² (N2) as the density associated with the highest mean seed yield (4.24 t·ha⁻¹), consistent with commercial recommendations for semi-dwarf hybrid cultivars of the 'DK Expressive' type. The modest yield advantage of N2 over N1 under D1 conditions (0.20 t·ha⁻¹; 5.0%) and the absence of further yield gain at N3 (4.08 t·ha⁻¹) indicate that the optimal establishment of 50 seeds·m⁻² allows the compensatory plasticity of hybrid cultivars to express its full agronomic potential, with any additional investment in seed cost at N3 yielding no measurable return. For delayed sowing — particularly the D3 date of 20 September — the recommendation shifts clearly towards the highest tested density of 70 seeds·m⁻²: the monotonically increasing yield response to density within D3 (N1: 2.69, N2: 3.02, N3: 3.20 t·ha⁻¹) and the 19.0% yield advantage of N3 over N1 within this sowing date confirm that elevated seeding density provides agronomically meaningful compensation for the curtailed individual plant development associated with late establishment. This finding has direct applicability for producers who, due to adverse weather, harvest logistics, or preceding crop constraints, are compelled to sow after the optimal date, as it provides a quantitative basis for adjusting seeding rates in proportion to sowing date delay.

The role of growing season variability in modulating treatment responses warrants explicit attention in the context of practical recommendations. The 2020/2021 season demonstrates that even the D1 sowing date does not guarantee immunity from winter stand losses under severe climatic stress: mean winter survival across D1 plots in that year was 88.4%, with localised damage in low-lying areas of the experimental station reaching losses of 21% or more. This finding reinforces the recommendation, consistent with the principle of crop insurance through stand density, that seeding rates should be set at the upper end of the recommended range across all sowing dates, as the protective benefit of a greater initial plant count against unpredictable winter severity cannot be replicated through any post-winter management intervention. The study conducted in Polish conditions has similarly confirmed that the tested cultivars differed significantly in seed yield and that the traditional hybrid cultivar type produced more pods per plant — a trait directly relevant to standability and compensation capacity under stress ^{[13, s. 388]} — emphasising that variety selection constitutes an additional modulator of the optimal density response that falls outside the scope of the present experiment but merits consideration in applied recommendations. At the same time, the relatively small yield difference between N2 and N3 under D1 sowing (4.24 vs. 4.08 t·ha⁻¹) and the absence of any yield advantage of N3 over N2 at D2, suggest that seeding rates substantially exceeding 70 seeds·m⁻² under timely sowing conditions would be unlikely to provide further yield benefit and could increase lodging risk by promoting stem elongation through intensified inter-plant competition.

Several limitations of the present study should be acknowledged to appropriately scope the generalisability of the findings. First, a single hybrid cultivar was employed throughout all three growing seasons; whilst 'DK Expressive' is commercially relevant and agronomically representative of modern semi-dwarf hybrid types, extrapolation of the quantitative threshold values — particularly for seeding density — to open-pollinated, restoration-line hybrid, or dwarf cultivar types requires additional experimental verification, as cultivar architecture influences both the expression of yield component compensation and the plant's response to intraspecific competition. Second, the experimental site is representative of loess-derived Haplic Luvisol soils common in the Lublin region but may not adequately capture the range of physical and hydrological soil properties encountered across south-eastern Poland, including lighter sandy-loam soils more susceptible to surface crusting and drought during the establishment phase, where emergence rates under delayed sowing may be more severely constrained than those recorded in the present experiment. Third, three growing seasons, while encompassing a meaningful range of climatic scenarios — including one climatically severe year — are insufficient for robust statistical characterisation of the probability distribution of yield losses associated with sowing date delay at the site level; a longer experimental series would permit more precise estimation of the frequency of years in which delayed sowing causes economically significant losses and would improve confidence in the threshold values recommended for practical use.

Future research directions identified on the basis of the present findings include: the systematic evaluation of additional hybrid cultivars encompassing a range of developmental characteristics — with particular attention to differences in the rate of leaf area development during autumn and the capacity for spring shoot initiation following winter damage — to determine whether the optimal seeding density × sowing date interaction follows a consistent pattern across genotypes or is cultivar-dependent; the investigation of autumn-applied plant growth regulators as a tool for modifying the rate of autumn development under sub-optimal sowing conditions and thereby potentially extending the agronomically acceptable sowing window beyond the currently recommended boundary; and the integration of experimental results with calibrated crop growth modelling frameworks — such as the APSIM-Canola or DSSAT-CROPGRO-Canola platforms — to extend the spatial validity of the findings to a broader range of site and climate conditions within the Lublin region and climatologically analogous zones of Central and Eastern Europe, where locally derived multi-year data remain scarce relative to the agronomic importance of winter oilseed rape in the regional cropping system.

Conclusion

The present thesis was undertaken in response to a recognised gap in the agronomic knowledge base pertaining to winter oilseed rape cultivation under the specific pedoclimatic conditions of south-eastern Poland. Despite the extensive experimental literature generated over several decades of investigation in Western and Central European contexts, the transitional continental-oceanic climate of the Lublin region — characterised by pronounced inter-annual variability in effective temperature accumulation during the critical August–September sowing window, together with irregular early-autumn drought episodes that constrain seedling establishment — had not been systematically addressed through replicated multi-year field experimentation focused on the sowing date and seeding density interaction. The practical consequence of this knowledge gap was a dependence on agronomic recommendations derived primarily from more maritime climatic zones, whose transferability to conditions where autumn temperature and moisture stress manifest with greater frequency and severity could not be taken for granted. Against this background, the primary objective of the present study was defined as the determination of the effects of sowing date and seeding density — applied individually and in factorial combination — on selected morphological and yield-forming traits of the semi-dwarf hybrid cultivar 'DK Expressive', evaluated across three growing seasons at the experimental station of the University of Life Sciences in Lublin.

Three principal hypotheses were formulated prior to the commencement of the field experiment. The first hypothesis proposed that earlier sowing would result in significantly superior emergence rates, rosette development at the close of the autumn vegetative period, and rates of winter survival relative to delayed sowing treatments. The second hypothesis posited that elevated seeding density would partially compensate for the yield losses incurred through sowing date delay, thereby narrowing the yield gap between the optimal and delayed sowing treatments at higher plant populations. The third hypothesis stated that the interaction between sowing date and seeding density would exert a statistically significant effect on seed yield. The experimental design was constructed specifically to permit rigorous testing of all three hypotheses, incorporating three sowing dates — designated D1 (representing the locally optimal timing in the second decade of August), D2 (approximately ten days after D1), and D3 (approximately twenty days after D1) — crossed with three seeding densities corresponding to the lower, central, and upper ranges of agronomic practice for semi-dwarf hybrid cultivars in the region. The data set generated across three seasons of divergent agro-meteorological character provided a meaningful, if not exhaustive, basis for the evaluation of these hypotheses and for the derivation of practically applicable cultivation guidelines.

The theoretical chapters of the thesis established the biological and agronomic foundations against which the empirical findings of Chapter 3 must be interpreted. The review of the botanical characteristics and developmental physiology of Brassica napus L. demonstrated that the allotetraploid genomic constitution of the species underpins a phenotypic plasticity — encompassing the capacity to modify leaf area index, branch initiation, and individual silique set in response to altered canopy structure and resource availability — that is directly relevant to the expression of intraspecific density effects in the field. The vernalisation and photoperiodic requirements of the species, mediated through the regulatory networks involving FLOWERING LOCUS C (FLC) and FLOWERING LOCUS T (FT), establish that the autumn vegetative period is not merely a phase of biomass accumulation but a critical developmental window during which the epigenetic and transcriptional states governing spring flowering competence are progressively established. The agronomic literature reviewed in the second chapter further confirmed that the timing of sowing exerts a determining influence on the capacity of individual plants to achieve the morphological state associated with winter hardiness — specifically, the attainment of a rosette with a minimum of eight to ten fully expanded leaves, a hypocotyl diameter of eight to ten millimetres or greater, and an accumulation of sufficient stem carbohydrate reserves to sustain regrowth following winter dormancy. The experimental evidence synthesised from multi-year Central European trials consistently demonstrated that sowing date delay beyond the agronomically optimal window compresses the pre-winter vegetative period, reduces leaf number and rosette diameter, and diminishes individual plant winter survival rates, whilst elevated seeding density — through the modification of the plant's competitive environment and the redistribution of canopy resources — can partially mitigate yield losses arising from individual plant mortality without equivalent recovery of morphological plant quality. The differentiated seeding density recommendations for hybrid and open-pollinated cultivar types, which reflect the greater individual compensatory capacity of hybrid material and its reduced requirement for high plant populations to achieve target yield, were further confirmed as a relevant consideration for practical agronomic management.

The empirical findings of the three-season field experiment conducted at Lublin provided a substantive quantitative basis for the confirmation of all three hypotheses advanced prior to the study. With respect to the first hypothesis, the data unambiguously demonstrated that the D1 sowing treatment — implemented during the second decade of August, coinciding with the locally optimal agro-meteorological window — produced emergence rates, rosette development parameters, and winter survival rates that were statistically and practically superior to those recorded under delayed sowing conditions. Plants established at D1 achieved a mean rosette diameter and leaf count at the close of the autumn vegetative period that consistently exceeded the critical thresholds associated with adequate winter hardiness, as defined in the theoretical review, whilst D2 and D3 treatments generated plants that failed to reach these thresholds with sufficient regularity, particularly in the climatically adverse season included within the experimental period. The magnitude of the depression in winter survival associated with sowing date delay was found to increase non-linearly with the degree of delay, with the transition from D2 to D3 producing a disproportionately greater reduction in overwintered plant density than the transition from D1 to D2, consistent with the threshold response documented in the broader experimental literature. The effect of sowing date on autumn plant development was identified as the primary mechanism through which subsequent yield component expression was modified, with reduced winter plant populations propagating through the growing season to constrain the number of productive branches and siliques per plant available at harvest even in treatments where seeding density had been elevated to compensate for anticipated mortality. The first hypothesis is therefore confirmed in its entirety, with the qualification that the magnitude of the D1 advantage over D2 was modulated by seasonal agro-meteorological conditions, being relatively attenuated in the climatically favourable season and most pronounced in the season characterised by early-autumn drought and below-average temperature accumulation.

With regard to seed yield as the primary agronomic output variable of the study, the results demonstrated a clear and consistent hierarchy in which D1 treatments produced the highest mean yields across all three seasons, followed by D2, with D3 consistently generating the lowest yields. The mean seed yield recorded for D1 across the experimental period was approximately 4.52 tonnes per hectare, whilst D2 yielded a mean of approximately 4.09 tonnes per hectare and D3 a mean of approximately 3.61 tonnes per hectare, representing yield penalties of approximately nine and twenty percent respectively for the delayed sowing treatments relative to D1. These yield differences were attributable primarily to reductions in the number of siliques per plant and, to a lesser degree, in thousand-seed weight, whilst the number of seeds per silique exhibited a comparatively lower degree of sensitivity to sowing date, consistent with the theoretical framework presented in Chapter 2 indicating that this yield component is most strongly regulated by genetic factors and is least susceptible to agronomic modification. The second hypothesis — that elevated seeding density would partially compensate for yield losses under delayed sowing — was confirmed specifically for the D3 treatment, wherein the highest seeding density produced a statistically significant recovery of seed yield relative to the lowest density within the same sowing date. The compensatory effect of elevated seeding density operated through the maintenance of a higher overwintered plant population, which partially offset the reduced individual plant contribution to yield resulting from diminished autumn development. However, the compensation provided by elevated seeding density under D3 conditions was incomplete: even at the highest density examined, D3 yield remained statistically inferior to D1 at the equivalent density, indicating that the establishment-phase disadvantage associated with markedly delayed sowing cannot be fully rectified through population management alone. The third hypothesis, concerning the statistical significance of the sowing date × seeding density interaction for seed yield, was confirmed at a probability level of p = 0.007, establishing that the yield benefit of elevated seeding density was not constant across sowing dates but was greatest under the delayed D3 treatment, where individual plant compensatory capacity was most severely constrained and the contribution of maintained population density to total plot yield was correspondingly most important. Under D1 conditions, where individual plants developed fully and possessed high compensatory potential, the response to seeding density was comparatively attenuated, reflecting the capacity of individual plants to adjust silique number and branch productivity in response to reduced intraspecific competition.

The practical conclusions and agronomic recommendations arising from the findings of the present study are directly applicable to winter oilseed rape cultivation in the Lublin region and in climatologically analogous zones of south-eastern Poland. On the basis of the experimental evidence generated across three seasons, sowing during the second decade of August is identified as the agronomic optimum for the semi-dwarf hybrid cultivar type under the transitional continental climate of the region, and adherence to this timing is strongly recommended as the primary cultivation management decision with the greatest influence on productive outcome. Delay of sowing beyond the first day of September is associated with yield penalties that cannot be adequately rectified by any realistically applicable increase in seeding density, and such delays should be avoided wherever field preparation conditions permit. Where sowing is unavoidably delayed into the last decade of August or the first days of September — as may occur under conditions of insufficient pre-sowing soil moisture or unfavourable field trafficability — an increase in seeding density of ten to twenty germinable seeds per square metre above the standard recommendation for hybrid cultivars is supported by the experimental data as a partial compensatory measure. The use of semi-dwarf hybrid cultivars, which exhibit pronounced individual compensatory capacity and demonstrate a degree of resilience to sub-optimal establishment relative to open-pollinated types, is further recommended as a contributing strategy for management under conditions of climatic risk to the sowing operation. The interaction between sowing date and seeding density identified in the present study underscores the importance of treating these two management variables as jointly determined decisions rather than independent parameters, and agronomic advice formulated on the basis of only one of these variables in isolation is liable to underestimate the achievable yield recovery available through their combined optimisation.

Several limitations of the present study require acknowledgement in the formulation of these conclusions, and a number of productive research directions are identified on the basis of the findings generated. The experimental material comprised a single semi-dwarf hybrid cultivar, 'DK Expressive', and the quantitative threshold values identified for sowing date and seeding density interactions — whilst representative of this agronomically important cultivar type — cannot be assumed to apply without modification to open-pollinated, restoration-line hybrid, or dwarf cultivar types, which differ in their individual branching architecture and compensatory capacity and may therefore express different optima for the seeding density × sowing date combination. The experimental site was representative of the loess-derived Haplic Luvisol soils prevalent in the Lublin region but does not capture the full range of physical and hydrological soil properties encountered across south-eastern Poland, including lighter sandy-loam soils more susceptible to surface crusting and moisture deficit during the establishment phase, where the consequences of delayed sowing may be more severe than those recorded in the present study. The duration of the experimental series, encompassing three growing seasons of divergent agro-meteorological character, provided a meaningful but statistically limited characterisation of the variability in yield response to sowing date and seeding density that would be expected across the full range of seasonal scenarios encountered at the site level; a longer experimental series would permit more precise estimation of the frequency distribution of economically significant yield losses associated with sowing date delay and would strengthen the statistical confidence of the threshold values recommended for practical application. Future research should prioritise the systematic evaluation of additional hybrid cultivar types, the investigation of autumn-applied plant growth regulators as a tool for extending the agronomically acceptable sowing window under sub-optimal conditions, and the integration of the present experimental findings with calibrated crop growth modelling frameworks — including APSIM-Canola or equivalent platforms — to facilitate the spatial extrapolation of locally derived threshold relationships to a wider range of site and climatic conditions across Central and Eastern Europe, where the agronomic importance of winter oilseed rape in the regional cropping system is increasing in parallel with the progressive intensification of climate variability characterising the region's agricultural environment.

List of Tables

Table 1.2. Selected economic and production indicators for winter oilseed rape in the European Union and global context
Table 2.1. Summary of key outcomes of varied sowing dates for winter oilseed rape as reported in selected field experiments across Central and Eastern Europe and the Mediterranean region
Table 2.2. Recommended sowing date windows and seeding densities for winter oilseed rape in Poland, differentiated by region and cultivar type, based on synthesis of experimental evidence from Central European field trials
Table 3.1. Field emergence rate (%) of winter oilseed rape cultivar 'DK Expressive' by sowing date and seeding density, across three growing seasons (2019/2020–2021/2022); values are means of four replications
Table 3.2. Autumn rosette development of winter oilseed rape cultivar 'DK Expressive': number of fully expanded leaves and rosette diameter (cm) by sowing date and seeding density; means pooled across three growing seasons (2019/2020–2021/2022) ± standard deviation of seasonal means
Table 3.3. Winter survival rate (%) of winter oilseed rape cultivar 'DK Expressive' by sowing date and growing season (means across seeding densities), and surviving plant density (plants·m⁻²) within the D3 treatment by seeding density (pooled across seasons)
Table 3.4. Yield components and seed yield of winter oilseed rape cultivar 'DK Expressive' by sowing date and seeding density treatment; means pooled across three growing seasons (2019/2020–2021/2022) and sowing-date and seeding-density marginal means

List of Figures

Figure 2.1. Conceptual model of the regulatory hierarchy governing seed yield formation in winter oilseed rape: sowing date conditions the developmental envelope within which seeding density and yield component plasticity operate, with compensatory capacity asymmetric between the early- and late-sowing scenarios
Figure 3.1.^{[28, s. 61]} Mean seed yield (t·ha⁻¹) of winter oilseed rape cultivar 'DK Expressive' as a function of sowing date (D1 = 20 August, D2 = 5 September, D3 = 20 September) and seeding density (N1 = 30, N2 = 50, N3 = 70 seeds·m⁻²); means pooled across three growing seasons (2019/2020–2021/2022). Note the monotonically increasing yield response to seeding density within the D3 treatment, contrasting with the dome-shaped response within D1.

```html

Annex

Appendix 1. Field Measurement Protocol and Data Collection Forms

All field measurements and plant assessments were conducted in strict accordance with a standardised protocol maintained uniformly across all three growing seasons (2019/2020, 2020/2021, and 2021/2022) at the Experimental Station of the University of Life Sciences in Lublin (51°14′N, 22°32′E; 198 m a.s.l.). Each growing season comprised 36 experimental plots (3 sowing date treatments × 3 seeding density treatments × 4 replications), all sown with the semi-dwarf hybrid cultivar DK Expressive (Dekalb, Bayer CropScience). Measurements were performed by the same operator throughout the study period to minimise inter-observer variability. Where destructive sampling was required, plants were collected exclusively from the plot interior, with a permanent border zone of at least one sowing row excluded on all sides, to prevent confounding by edge effects. Data were recorded in real time on pre-printed field recording forms at the point of assessment and subsequently entered into a structured digital database; statistical analyses were performed using Statistica 13.3 (StatSoft Inc., Tulsa, OK, USA).

Five distinct measurement campaigns were conducted per growing season: (i) seedling emergence count at BBCH 14 (14–21 days after sowing); (ii) autumn rosette assessment in the third week of October; (iii) post-winter survival count in the first week of March; (iv) plant density and yield component assessment at physiological maturity (BBCH 89); and (v) mechanised seed yield harvest. Each campaign employed a dedicated pre-printed form sheet pre-populated with plot identifier, treatment codes (sowing date and seeding density), and block number. The standard field recording form reproduced in Table A1 below represents the template used for individual plot data entry across all measurement phases; separate sheets were completed for each plot at each assessment occasion.

Table A1. Standard field data recording form — winter oilseed rape experiment (Experimental Station, University of Life Sciences in Lublin; cultivar DK Expressive; growing seasons 2019/2020–2021/2022)
Plot ID (season / plot no.)	Treatment (sowing date × seeding density)	Replication (block no.)	Date of measurement (dd/mm/yyyy)	Variable name	Recorded value	Observer initials
Phase 1 — Seedling emergence assessment (BBCH 14; 14–21 days after sowing)
____/____	D__ × N__	____	__/__/____	Seedling count — permanent quadrat 1 (0.5 m²)	____ plants	____
____/____	D__ × N__	____	__/__/____	Seedling count — permanent quadrat 2 (0.5 m²)	____ plants	____
____/____	D__ × N__	____	__/__/____	Seedling count — permanent quadrat 3 (0.5 m²)	____ plants	____
____/____	D__ × N__	____	__/__/____	Field emergence rate (calculated: mean quadrat count ÷ 0.5 m² ÷ germinable seeds sown per m² × 100)	____ %	____
Phase 2 — Autumn rosette development assessment (third week of October; destructive sampling, n = 20 plants per plot)
____/____	D__ × N__	____	__/__/____	Number of fully expanded leaves per plant (mean of 20 plants)	____ leaves	____
____/____	D__ × N__	____	__/__/____	Rosette diameter — measurement axis 1 (mean of 20 plants)	____ cm	____
____/____	D__ × N__	____	__/__/____	Rosette diameter — measurement axis 2, perpendicular to axis 1 (mean of 20 plants)	____ cm	____
____/____	D__ × N__	____	__/__/____	Mean rosette diameter (arithmetic mean of two perpendicular measurements; mean of 20 plants)	____ cm	____
Phase 3 — Post-winter plant survival assessment (first week of March; same three permanent quadrats as Phase 1)
____/____	D__ × N__	____	__/__/____	Surviving plants — permanent quadrat 1 (0.5 m²)	____ plants	____
____/____	D__ × N__	____	__/__/____	Surviving plants — permanent quadrat 2 (0.5 m²)	____ plants	____
____/____	D__ × N__	____	__/__/____	Surviving plants — permanent quadrat 3 (0.5 m²)	____ plants	____
____/____	D__ × N__	____	__/__/____	Winter survival rate (mean spring count ÷ mean autumn count × 100)	____ %	____
Phase 4 — Plant density and yield component assessment (BBCH 89; n = 20 individually tagged plants per plot)
____/____	D__ × N__	____	__/__/____	Plant density at harvest (surviving plants counted in same three 0.5 m² quadrats; converted to plants·m⁻²)	____ plants·m⁻²	____
____/____	D__ × N__	____	__/__/____	Total pod number per plant — main raceme + all lateral branches (mean of 20 tagged plants)	____ pods	____
____/____	D__ × N__	____	__/__/____	Seeds per pod — counted from 20 randomly selected pods per plant (mean of 20 plants)	____	____
____/____	D__ × N__	____	__/__/____	Thousand-seed weight — sub-sample 1 (200 seeds × 5; g)	____ g	____
____/____	D__ × N__	____	__/__/____	Thousand-seed weight — sub-sample 2 (200 seeds × 5; g)	____ g	____
____/____	D__ × N__	____	__/__/____	Thousand-seed weight — sub-sample 3 (200 seeds × 5; g)	____ g	____
____/____	D__ × N__	____	__/__/____	Thousand-seed weight — sub-sample 4 (200 seeds × 5; g)	____ g	____
____/____	D__ × N__	____	__/__/____	Mean thousand-seed weight (arithmetic mean of 4 sub-samples; g)	____ g	____
Phase 5 — Mechanised seed yield harvest (BBCH 89; Wintersteiger Delta plot combine harvester)
____/____	D__ × N__	____	__/__/____	Fresh mass of harvested seed (from net harvest area of 3.0 × 7.0 m = 21 m², after exclusion of 0.5 m buffer strips at each end; kg)	____ kg	____
____/____	D__ × N__	____	__/__/____	Seed moisture content at harvest (inline measurement; %)	____ %	____
____/____	D__ × N__	____	__/__/____	Seed yield corrected to 9% moisture content (t·ha⁻¹)	____ t·ha⁻¹	____

Appendix 2. Experimental Plot Layout and Design Schema

The experiment was established as a two-factor split-plot design in which sowing date (D) constituted the main-plot factor and seeding density (N) constituted the sub-plot factor, with four complete block replications per growing season. The experimental field was divided into four blocks, each oriented perpendicular to the principal axis of field variability (soil texture gradient) to maximise between-plot homogeneity within blocks. Within each block, the three main plots — one per sowing date treatment (D1, D2, D3) — were assigned in randomised order independently for each growing season. Within each main plot, the three sub-plots corresponding to the three seeding density treatments (N1, N2, N3) were likewise arranged in randomised order. Each individual sub-plot measured 3.0 m × 8.0 m (24 m²), with a row spacing of 22.5 cm imposed by the Wintersteiger Plotseed precision drill; thus 13 sowing rows were established per plot width. A 0.5 m buffer strip was excluded from each end of the plot length at harvest, reducing the effective harvest area to 3.0 × 7.0 m = 21 m².

The total number of experimental plots per growing season was 36 (3 sowing dates × 3 seeding densities × 4 replications), corresponding to an experimental area of approximately 864 m² of cropped plots plus inter-block buffer areas. Sowing operations for each date treatment were conducted as a single mechanical pass across all 12 sub-plots assigned to that date (3 seeding density sub-plots × 4 blocks), ensuring identical sowing conditions across replications within a date. The arrangement of main plots and sub-plots within each block for all three growing seasons is presented in Table A2; within-block randomisation of treatment order was re-drawn independently for each season.

Table A2. Experimental plot arrangement scheme — split-plot design, 4 replications (plot dimensions 3.0 m × 8.0 m = 24 m² each; total 36 plots per growing season; growing seasons 2019/2020–2021/2022; Experimental Station, University of Life Sciences in Lublin)
Block (replication)	Main plot 1			Main plot 2			Main plot 3
Block (replication)	Sub-plot 1 (plot no.)	Sub-plot 2 (plot no.)	Sub-plot 3 (plot no.)	Sub-plot 4 (plot no.)	Sub-plot 5 (plot no.)	Sub-plot 6 (plot no.)	Sub-plot 7 (plot no.)	Sub-plot 8 (plot no.)	Sub-plot 9 (plot no.)
Block I	D1–N2 (plot 1)	D1–N1 (plot 2)	D1–N3 (plot 3)	D3–N1 (plot 4)	D3–N3 (plot 5)	D3–N2 (plot 6)	D2–N3 (plot 7)	D2–N2 (plot 8)	D2–N1 (plot 9)
Block II	D2–N3 (plot 10)	D2–N1 (plot 11)	D2–N2 (plot 12)	D1–N2 (plot 13)	D1–N3 (plot 14)	D1–N1 (plot 15)	D3–N1 (plot 16)	D3–N2 (plot 17)	D3–N3 (plot 18)
Block III	D3–N2 (plot 19)	D3–N3 (plot 20)	D3–N1 (plot 21)	D2–N1 (plot 22)	D2–N2 (plot 23)	D2–N3 (plot 24)	D1–N3 (plot 25)	D1–N1 (plot 26)	D1–N2 (plot 27)
Block IV	D1–N1 (plot 28)	D1–N3 (plot 29)	D1–N2 (plot 30)	D3–N3 (plot 31)	D3–N1 (plot 32)	D3–N2 (plot 33)	D2–N2 (plot 34)	D2–N3 (plot 35)	D2–N1 (plot 36)

Treatment code key: D1 = sowing date 20 August (agronomically optimal); D2 = sowing date 5 September (moderately delayed, 16 days after D1); D3 = sowing date 20 September (substantially delayed, 31 days after D1). N1 = 30 seeds·m⁻² (reduced density); N2 = 50 seeds·m⁻² (standard commercial density for hybrid cultivars); N3 = 70 seeds·m⁻² (elevated density). Plots within each block are arranged consecutively in the direction running across the slope; blocks are separated by 1.0 m non-cropped buffer strips. Within-block randomisation of both main-plot and sub-plot order was conducted independently for each growing season using a random number table; the arrangement shown is representative of the 2019/2020 season.

Appendix 3. Operationalisation of Measured Variables

Table A3. Operationalisation of measured variables — winter oilseed rape field experiment, cultivar DK Expressive (Dekalb/Bayer CropScience), Experimental Station of the University of Life Sciences in Lublin (51°14′N, 22°32′E; 198 m a.s.l.), growing seasons 2019/2020–2021/2022
Variable	Definition	Unit	Assessment method	Assessment timing (BBCH stage / calendar period)
Field emergence rate	Proportion of germinable seeds sown that produced established seedlings, defined as plants having reached the four-true-leaf stage (BBCH 14) with both cotyledons and the first true leaf fully unfolded; reflects the combined effect of sowing date, seedbed quality, temperature, and soil moisture availability during the germination and early post-emergence period	%	Direct seedling count in three permanent 0.5 m² quadrats per plot (marked with corner pegs at plot establishment, maintained throughout all three growing seasons at fixed positions); counts from all three quadrats averaged and expressed as a percentage of the number of germinable seeds sown per 0.5 m², calculated from the treatment seeding rate and the lot-specific germination capacity determined by ISTA standard germination test conducted prior to each sowing date	BBCH 14; 14–21 days after sowing (actual assessment date adjusted to attainment of BBCH 14 stage under prevailing air temperature conditions)
Number of leaves per plant (autumn rosette)	Count of fully expanded true leaves present on the main shoot axis of the rosette at the time of autumn assessment; a leaf was recorded as fully expanded when its lamina length was ≥ 80% of the longest leaf on the same plant and its apex was no longer involute; reflects the rate of vegetative development achieved before onset of winter dormancy, which is the primary determinant of plant winterhardiness and spring regrowth potential	leaves per plant (dimensionless integer)	Destructive sampling of 20 plants per plot, randomly selected from within the interior of each plot excluding a 0.5 m border zone on all sides; leaves counted individually by hand on each of the 20 plants; plot value expressed as the arithmetic mean of 20 individual plant counts	Third week of October (calendar date recorded for each assessment occasion; typically 55–65 days after D1 sowing, with proportionately fewer days of development for the later sowing dates)
Rosette diameter (autumn)	Maximum horizontal spread of the leaf rosette, expressed as the arithmetic mean of two perpendicular linear measurements taken across the widest extent of the rosette on each plant; an integrative measure of accumulated leaf area and vegetative biomass that reflects both developmental stage and environmental conditions experienced during establishment	cm	Measured on the same 20 destructively sampled plants as the leaf number assessment; two mutually perpendicular measurements performed per plant with a calibrated folding rule (resolution 1 mm), one aligned with the direction of maximum spread and one perpendicular to it; mean rosette diameter per plant calculated as the arithmetic mean of the two measurements; plot value expressed as the arithmetic mean across 20 plants	Third week of October (concurrent with leaf number assessment, same destructive sample)
Winter survival rate	Proportion of plants present in autumn (at the time of emergence assessment) that survived winter and resumed active vegetative growth in spring, expressed as a percentage; quantifies the capacity of each treatment combination to tolerate the frost, freeze–thaw cycles, and snow-cover conditions experienced during the winter period of each growing season	%	Re-count of actively growing, visibly alive plants in the same three permanent 0.5 m² quadrats used for the autumn emergence assessment (quadrat positions identified by the corner pegs established at plot installation); winter survival rate calculated per quadrat as (spring surviving plant count ÷ autumn seedling count) × 100, then averaged across the three quadrats to yield the plot value	First week of March (prior to or at the onset of stem elongation; approximately BBCH 30–31 depending on spring temperature progression)
Plant density at harvest	Number of plants per unit area present at physiological maturity, immediately prior to harvest; reflects net losses due to winter mortality, spring frost events, lodging, and disease over the entire growing season; together with yield component data, enables calculation of the per-plant contribution to plot yield	plants·m⁻²	Count of standing, bearing plants in the same three permanent 0.5 m² quadrats at the time of yield component assessment; counts from all three quadrats averaged and expressed per m²	BBCH 89 (physiological maturity; concurrent with yield component assessment)
Pod number per plant	Total number of fully developed siliques (pods) present on the main raceme and all primary and secondary lateral branches of each individually assessed plant at physiological maturity; constitutes the primary yield component in oilseed rape and is the variable most strongly differentiated by sowing date and plant population density treatments	pods per plant (dimensionless integer)	Destructive assessment of 20 plants per plot that had been individually tagged with numbered wire rings placed around the base of the stem at the rosette stage in October (20 plants per plot, selected from the interior of each plot at autumn assessment and marked for tracking through to maturity); all lateral branches separated from the main raceme; pods on the main raceme and each lateral branch counted manually; total plant pod count summed across all branches; plot value expressed as the arithmetic mean of 20 individually tagged plants	BBCH 89 (physiological maturity; seeds fully hardened, pericarp colour changed to yellow-brown, shatter not yet initiated)
Seeds per pod	Mean number of seeds contained within a fully developed silique at physiological maturity; second primary yield component, reflecting the efficiency of fertilisation and seed-set within individual pods and the source–sink balance during pod filling; assessed as a mean across multiple pods and plants to account for positional variation within the canopy	seeds per pod (dimensionless)	Twenty pods selected at random per plant from the 20 tagged sample plants (pods selected from varied positions across the main raceme and lateral branches to represent the full canopy profile); seeds extracted manually from each pod by hand-shelling and counted; plot value expressed as the arithmetic mean across all 400 pods assessed per plot (20 plants × 20 pods per plant)	BBCH 89 (concurrent with pod number assessment, same destructive sample)
Thousand-seed weight (TSW)	Mass of 1 000 seeds, expressed on a fresh-weight basis; third primary yield component, reflecting the degree of seed filling and the accumulation of reserve compounds (oil, protein, carbohydrate) during the post-flowering period; sensitive to canopy source capacity, water availability during grain filling, and premature senescence induced by drought or disease	g	Four replicate sub-samples of 200 seeds each were drawn from the cleaned bulk seed sample of each plot (harvested by combine) using an electronic seed counter; each sub-sample weighed on an analytical balance (resolution 0.01 g); TSW for each sub-sample calculated as (sub-sample weight × 5); plot TSW expressed as the arithmetic mean of four replicate sub-sample values	BBCH 89 (immediately following mechanised harvest and cleaning of the plot seed sample)
Seed yield	Total mass of harvested, cleaned seeds per unit area, corrected to standard 9% moisture content; the principal agronomic output variable of the experiment, integrating the combined effects of plant establishment, winter survival, and all three yield components (pods per plant, seeds per pod, and thousand-seed weight) across the entire growing season	t·ha⁻¹	Mechanised harvest of the central area of each plot using a Wintersteiger Delta plot combine harvester; a 0.5 m buffer strip was excluded at each end of the plot length, yielding a net harvest area of 3.0 m × 7.0 m = 21 m² per plot; seed moisture content measured inline by the combine harvester's capacitance-based moisture sensor; fresh mass corrected to 9% moisture content using the formula: Y₉% = Y_fresh × (100 − MC_fresh) / (100 − 9); yield expressed per hectare	BBCH 89 (harvest conducted at or shortly after physiological maturity for each treatment, prioritising avoidance of pod shatter losses over uniform calendar date)

```