Cover crops are increasingly being adopted to provide multiple ecosystem services, including weed suppression. Understanding what drives weed biomass in cover crops can help growers make the appropriate management decisions to effectively limit weed pressure. In this paper, we use a unique dataset of 1764 measurements from seven cover crop research experiments in Pennsylvania (USA) to predict, for the first time, weed biomass in winter cover crops in the fall and spring. We assessed the following predictors: cover crop biomass in the fall and spring, fall and spring growing degree days between planting and cover crop termination, cover crop type (grass, brassica, legume monocultures, and mixtures), system management (organic, conventional), and tillage before cover crop seeding (no-till, tillage). We used random forests to develop the predictive models and identify the most important variables explaining weed biomass in cover crops. Growing degree days, cover crop type, and cover crop biomass were the most important predictor variables in both the fall (r2 = 0.65) and spring (r2 = 0.47). In the fall, weed biomass increased as accumulated growing degree days increased, which was mainly related to early planting dates. Fall weed biomass was greater in legume and brassica monocultures compared to grass monocultures and mixtures. Cover crop and weed biomass were positively correlated in the fall, as early planting of cover crops led to high cover crop biomass but also to high weed biomass. In contrast, high spring cover crop biomass suppressed weeds, especially as spring growing degree days increased. Grass and brassica monocultures and mixtures were more weed-suppressive than legumes. This study is the first to be able to predict weed biomass in winter cover crops using a random forest approach. Results show that weed suppression by winter cover crops can be enhanced with optimal cover crop species selection and seeding time.
This is a preview of subscription content, log in to check access.
We are grateful to the diverse research teams from Penn State University and the staff of the Russell E. Larson Agricultural Research Center for planting, managing, and assisting in data collection in our experimental plots, and to many undergraduate assistants for their assistance with data collection.
Compliance with ethical standards
Conflict of interest
This work was supported by the USDA National Institute of Food and Agriculture, the Organic Research and Extension Initiative under Project PENW-2015-07433, Grant No. 2015-51300-24156, Accession No. 1007156, and the National Science Foundation Grant No. DGE1255832. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation. The authors declare that they have no conflict of interest.
Dieleman A, Mortensen DA, Martin AR (1999) Influence of velvetleaf (Abutilon theophrasti) and common sunflower (Helianthus annuus) density variation on weed management outcomes. Weed Sci 47:81Google Scholar
Lawson A, Cogger C, Bary A, Fortuna AM (2015) Influence of seeding ratio, planting date and termination date on rye-hairy vetch cover crop mixture performance under organic management. PLoS One 10:e0129597CrossRefGoogle Scholar
Murrell E, Schipanski ME, Finney DM, Hunter MC, Burgess M, LaChance JC, Baraibar B, White CM, Mortensen DA, Kaye JP (2017) Achieving diverse cover crop mixtures: effects of planting date and seeding rate. Agron J 109(1). https://doi.org/10.2134/agronj2016.03.0174
Wallace JM, Williams A, Liebert JA, Ackroyd VJ, Vann RA, Curran WS, Keene CL, VanGessel MJ, Ryan MR, Mirsky SB (2017) Cover crop-based, organic rotational no-till corn and soybean production systems in the Mid- Atlantic United States. Agriculture 7, 34. https://doi.org/10.3390/agriculture7040034