2. Key Factors Affecting the Performance of Cereal-Grain Legume Mixtures
2.1. Soil Type
The yield of CGLM largely depends on the soil type, soil pH, and water availability, which is important under conditions of high variability of climatic conditions. Mixed crops are characterised by greater yield stability compared to monocropping, especially on poor, sandy soils. According to Księżak and Magnuszewska
[22][18] the largest yield of pea and cereal mixtures was noted on soils with neutral pH, and the smallest, on very acidic. Rudnicki and Kotwica
[23][19] pointed out, that on medium-textured soils, the yield of cereal-lupine mixtures depended mainly on the cereal component. Species of lupine (yellow (
Lupinus luteus L.), blue (
Lupinus angustifolius L.), and its sowing density had no significant impact on the overall yield of the mixture. However, on the heavier textured soils, most suited to wheat (
Triticum aestivum L.) production, CGLM differed very much in height, yield, the share of a legume component, and other characteristics
[23][19].
Another important factor influencing the yield of CGLM is the availability of soil water. At low soil moisture content, the mixture of peas (
Pisum sativum subsp.
arvense (L.) Asch.) with cereals yielded better than pure sowing of these species
[24][20]. According to Dudek and Żarski
[25][21], the use of irrigation allowed for the cultivation of mixtures for green fodder even on very light (sandy) soils, providing higher (by 163%) and more reliable yields.
2.2. Choice of Species
Legume-based mixtures show positive interspecies interactions that justify their association with other species. CGLM perform well under temperate climate conditions and yield better and more stably than the same species grown in pure sowing. When composing a balanced mixture, attention should be paid to the duration of the growing season of the CGLM components and matching the harvesting time, as these are important factors affecting their yields. It is also related to the competitiveness of plants. Most cool-season annual legumes suffer from a competitive disadvantage in association with cereals. This serious disadvantage has usually been reported for peas, white lupine, and vetches
[26][22]. According to Księżak and Borowiecki
[27][23] peas mixed with wheat yielded better than with barley (
Hordeum vulgare L.), while oats (
Avena sativa) as a companion crop for lupine gave smaller yields than triticale (× Triticosecale)
[28][24]. Yields of lupine with oats were small and variable between years, and the share of lupine seeds in the yield of these mixtures generally did not exceed 10%. On poor, sandy soils, yellow lupine in mixtures with spring cereals performed better than blue lupine
[23][19]. Droushiotis
[29][25] showed that the higher share of legume (vetch or pea) seeds in the mixture with triticale or barley, the smaller its total dry matter production. According to this author, the highest proportion of legume was noted in the mixture of pea with triticale, assuming the harvest time for the cereal was the phase of grain milk maturity, and for the pea the final stage of the pod formation.
Herper [quoting after 5] argues that if the components of the mixture compete with each other in terms of height, the overall yield is determined by the weaker component of the mixture. Baxevanos et al.
[30][26] suggest that the ideal pea plant type depends on the extent of competitive stress as determined by the intrinsic plant vigour and plant height of the associated cereal and the presence and amount of nitrogen fertilization. A tall semi-dwarf type is preferable under modest competitive stress, while a tall type is preferable under severe competitive stress. Plant modelling work by Barillot et al.
[31][27] on pea-cereal mixtures suggested that pea competitive ability was positively affected by leaf area index (LAI) in early growth stages, and by plant height during the onset of interspecific competition.
Horse bean (
Vicia faba L.) is one of the more popular grain legumes and appears to be very suitable for intercropping with cereals. It has good spatial and temporal complementarity with cereals in most climates
[21][28] and presents less risk of yield failure than sole crops
[1] standing. Its global production stands at 4 Mt (from 2.6 M ha)
[21][28]. Horse beans after inoculation with Rhizobium strains can fix up to 88% of their nitrogen requirements and leave considerable quantities of nitrogen in the soil after harvest
[32][29].
Little attention has been paid to root system effects. Wilson
[33][30] argues that over-yielding of mixtures (i.e., when the yield of the mixture is higher than the yield of the components treated separately) is associated with diversity in the root system structure of its components. Księżak
[34][31] showed that the substrate in which seeds of four species of cereals germinated (wheat, barley, oats, and triticale) stimulated the growth of rootlets of pea and vetch (
Vicia sativa L.). The corresponding substrate from triticale inhibited the growth of yellow and blue lupine rootlets. Moreover, the substrate from beneath the seeds of legumes had an inhibitory effect on the development of rootlets of cereals, as well as on the coleoptile of barley. Księżak and Staniak
[35][32] reported that root secretions from oat seedlings inhibited the development of the rootlet of legumes after 96 h. The grain extracts from oats stimulated the germination of pea seeds after just 24 h, but no such effects were observed in the case of barley exudates. Secretions from pea seeds soaked for 48 h had an inhibitory effect and after 72 h strongly inhibited the germination of barley
[35][32].
2.3. Choice of Varieties
The positive effect of interspecific diversity with legumes on crop yielding is widely acknowledged. In contrast, the effect of large intraspecific diversity within each associated species is little studied. Greater intraspecific diversity of annual legumes, as provided by genotype or variety mixture, reveals some advantages in terms of yield and stability
[12]. Among different traits, plant height is of particular importance, as it strongly determines the architecture of the canopy, and affects lodging and yields of mixtures. Rudnicki
[36,37][33][34] developed a methodology for assessing the suitability of varieties of pea, yellow and blue lupine for mixtures with spring cereals, at the same time providing formulas for determining the composition of such mixtures. The author took into account such aspects as time of maturity, plant height, resistance to lodging, thousand seeds weight, and the protein content in the seeds. According to this author, the height of plants in the mixture and the habitat type determine the stand architecture. The varieties with a shorter stem usually have less favourable light conditions in the mixture. Particularly unfavourable conditions occur when legumes dominate over cereals, which may lead to the lodging and, consequently, to yield decrease
[38][35]. Well-chosen varieties of pea increase the chance of over-yielding in mixtures with barley, triticale, and oats
[38][35]. Semi-leafless pea varieties have a high-yielding potential and a lower rate of transpiration due to a large number of tendrils. Tolerance of varieties to the neighbourhood of other crops is also important, as is resistance to pod shatter In legumes. Nykänen et al.
[39][36] showed that for tall varieties of forage pea, oats, and wheat were better components than barley. Księżak
[24][20] indicated that determinate varieties of vetch yielded better with spring barley than with oats. Darras et al.
[40][37] showed that pea genotypes cultivated as isolated cultivars in mixtures with barley demonstrated the potential to improve field pea competitive ability, particularly with respect to the least competitive genotypes, but this result emerged mainly in suboptimal growing conditions.
2.4. Seeding Rate
The yield of a CGLM depends on the performance of its weaker component. A low proportion of legumes in the mixture reduces the proportion of their seeds in the yield, but also makes these plants more vulnerable to yield fluctuations. In addition, it reduces biological nitrogen fixation and compromises yield quality. Therefore, it is important to optimize the proportion of components in the seed mixture, especially the legume one, taking into account the species, habitat, and crop management. There are uncertainties regarding the effect of sowing density on the grain yield of CGLM. Particular species in the mixture can be sown at either their monocrop seed rates (an additive design) or at a reduced percentage (a replacement design). Most research has been based on the replacement design of 50:50 legume:cereal
[15[15][38][39],
41,42], but there are designs such as 60:40, 80:20
[13], 63:37, 75:25, 85:15
[43][40], 65:35, 55:45
[44][41], along with additive designs of 100:100, 50:50, and 100:50
[15,41][15][38]. Haymes and Lee
[45][42] have shown up to a 40% grain yield increase for intercrops compared with sole crops (Land Equivalent Ratios (LER) up to 1.4). The 100:100 designs have shown a constraint to the legume due to competition from the cereal, but a general intercrop advantage has been widely reported. The other designs indicated the advantages of the legume:cereal crop combinations on weed suppression, protein yield, and nitrogen uptake. In these studies, two factors were taken into account: MAI (Monetary Advantage Index) and IA (Intercropping Index). In any case, the mixtures were more favourable than objects with plants grown as a sole crop. The proportion of components had a significant effect on the yield of mixtures. The yield of the mixture usually decreases with an increasing share of legumes in the mixture
[23,38,46][19][35][43]. This relationship was observed for mixtures of cereals with peas
[38[35][43],
46], as well as for mixtures of cereals with yellow and blue lupine
[23][19].
The results of the meta-analysis of the crop yields in cereal/legume mixtures done by Yu at al.
[18][44] showed that the higher the sowing density of cereal, the higher its relative yield and the higher the yield of the whole mixture. Intercropping of common vetch with different cereals (wheat, triticale, barley, and oat) at two different seeding ratios (55:45 and 65:35) under Mediterranean climate conditions affected the crop yields, competition between cereal and legume and the economic efficiency of the cropping system. The common vetch–wheat mixture at the 55:45 seeding ratio and the common vetch–oat mixture at the 65:35 seeding ratio yielded the best and were the most profitable compared with other intercropping systems
[44][41].