This study investigates how cover crop management and soil tillage influence the development and yield of cucumber and cabbage crops. Three cover crop treatments—blue lupin, black oats, and their mixture—were evaluated during the autumn/winter season, while
Stylosanthes capitata (Fabaceae), pearl millet (
Pennisetum
[...] Read more.
This study investigates how cover crop management and soil tillage influence the development and yield of cucumber and cabbage crops. Three cover crop treatments—blue lupin, black oats, and their mixture—were evaluated during the autumn/winter season, while
Stylosanthes capitata (Fabaceae), pearl millet (
Pennisetum glaucum, Poaceae), and their mixture were assessed during the spring/summer season, under both conventional tillage and no-till (direct seeding) systems. Cover crops were established in spring/summer (October–November) and, after their management, cucumber (
Cucumis sativus L.) was cultivated from December to February. Subsequently, winter cover crops were grown from May to July, followed by cabbage (
Brassica oleracea var.
capitata) cultivation from July to September. Drip irrigation was used, and organic practices were employed for weed, pest, and disease management. Germination, seedling survival rate, and plant growth (height, number of leaves, foliage cover, and fruit or cabbage size) were evaluated. Finally, crop yield is considered by comparing harvest weight and quality to determine which combination of soil cover and planting method maximizes crop productivity and quality. Obviously, management differences that influence yield will be associated with soil properties. To better understand the causes of these yield differences, the influence of soil chemical properties was explored using multivariate analysis techniques (discriminant analysis) and neural networks. Multivariate techniques allow for the exploration of complex relationships among multiple variables simultaneously, facilitating the identification of key patterns or factors that influence crop yield. On the other hand, neural networks, using machine learning models, allow for the prediction of outcomes based on the soil’s physicochemical properties, as well as the identification of optimal combinations of factors that maximize crop yield. Discriminant analysis and neural networks showed that soil variables such as pH, organic matter (OM), cation exchange capacity (CEC), phosphorus (P), and potassium (K) were key determinants in differentiating the yield groups. Cabbage yield was most strongly associated with pH and OM, while cucumber yield responded more strongly to potassium and CEC.
Full article
