Search

Mixtures of plant species are often higher yielding due to species complementarities. Such effects may be beneficial to agriculture but are under-used in practice. It is therefore important to demonstrate beneficial effects of mixtures that are easy to implement for farmers. Here, we study yield in mixtures of agricultural cover crops. We hypothesized that the degree to which species express complementarity might be affected by the spatial configuration of the mixture. To test this hypothesis, we seeded mixtures using two configurations: (1) seed mixtures in each row (intimate mixing), and (2) different species in separate (alternate) rows (less intimate mixing). Three years field experiments were conducted to compare growth and N uptake by sole stands and the two pairwise mixture configurations of three cover crop species: vetch (Vicia sativa L.), bristle oat (Avena strigosa) and oilseed radish (Raphanus sativus L.). Shoot biomass production and N uptake were determined. Results were analysed using additive partitioning to determine the net biodiversity effect and its components. Mixtures, overall, accumulated significantly more shoot biomass and shoot N than the corresponding pure stands did on average; hence, the net biodiversity effect was positive. The yield benefit of mixtures (observed minus expected) was 0.66 t biomass ha−1 and 10.7 kg N ha-1 compared to an average sole cover crop of 2.58 t biomass ha−1 and 53.1 kg N ha−1. Contribution to the net biodiversity effect for biomass was, on average, equally divided between the complementarity effect and the selection effect. For N uptake, the net biodiversity effect was mostly realized through complementarity. Experimental year and species combination had a significant effect on biomass and N yield of mixtures. Mixtures containing vetch had the greatest gain in biomass compared to expectation. Whether within or between row mixing gave the highest yield and N uptake depended on species combination and year, but in most cases, the effect of configuration was not significant. Thus, overall, using cover crop mixtures resulted in a positive net biodiversity effect irrespective of seeding configuration.

The performance of three vacuum precision seeders was investigated in a field study. Seeding uniformity was determined in three different within-row distances: 14, 18 and 21 cm. The seeders were operated at 1.8, 3.6, 5.4 and 7.2 km h-1. Successive seed spacing along of 3 m of row was measured in three replications on each row. For evaluating the seeding uniformity of seeders, seed spacings were analyzed using the methods (MISS, MULT, QFI and PREC). There were no differences between seeders. For P < 0.01, operating speed affected MISS and QFI values, and the within-row seed spacing affected MULT and PREC values. The best operating speed was 1.8 km h-1 because of the highest QFI value (88.5%). There was no difference between 1.8 and 3.6 km h-1. The speeds, 1.8 and 3.6 km h-1, were different from 5.4 and 7.2 km h-1. The best within-row distance was 18 cm because the QFI value was higher than those of 14 and 20 cm, 86.9%, 82.0% and 81.8%, respectively. The best PREC value was obtained for 21 cm within-row distance (17.4%). PREC values were acceptable for precision seeding in all trials. O comportamento de três semeadeiras de precisão a vácuo foi investigado em um estudo de campo. A uniformidade de semeadura foi determinada para três distâncias de entrelinha: 14, 18 e 21 cm. As semeadeiras operaram nas velocidades de 1,8; 3,6; 5,4 e 7,2 km h-1. O espaçamento entre sementes sucessivas foi feito ao longo de 3 m de linha, com três repetições em cada linha. Para avaliar a uniformidade, os espaçamentos entre sementes foram analisados usando os métodos (MISS, MULT, QFI e PREC) e nos resultados não foi achada diferença entre semeadeiras. Para P < 0,001 a velocidade de operação afetou MISS e QFI e o espaçamento entre linhas afetou MULT e PREC. A melhor velocidade de operação foi de 1,8 km h-1 devido ao seu mais alto valor de QFI (88,5%). Não houve diferença entre as velocidades de 1,8 e 3,6 km h-1, mas elas foram diferentes das velocidades 5,4 e 7,2 km h-1. A melhor distância entre linhas foi de 18 cm, pois seu QFI foi maior em relação à 14 e 20 cm, 86,9%, 82,0% e 81,8%, respectivamente. O melhor PREC foi obtido para 21 cm de entrelinha (17,4) e os valores de PREC foram aceitáveis para todos experimentos.