Estimates indicate that 30% of land surface globally is affected by soil acidity, influencing agricultural production. Application of lime increases soil pH and improves crop growth. We tested the hypothesis that liming will reduce rigid ryegrass (Lolium rigidum Gaudin) growth by improving the competitive ability of the crop. Experiments at Merredin and Wongan Hills in Western Australia indicated that application of lime in previous years reduced L. rigidum density, biomass, and seed production in wheat (Triticum aestivum L.) crops in 2018. At Merredin, L. rigidum seed production in 2018 was reduced from 9,390 to 2,820 seeds m−2, and wheat tiller number and yield was increased, following lime application of 0 to 6,000 kg ha−1 in 2016. At Wongan Hills, lime application of 4,000 kg ha−1 in 1994 reduced seed production in the 2018 wheat crop from 4,708 to 1,610 seeds m−2, and application of 3,000 kg ha−1 of lime in 2014 reduced seed production from 3,959 to 921 seeds m−2 in 2018. Again, lime increased wheat tiller number, but not yield. A screen house experiment (in controlled conditions) indicated that lime application increased the initial growth of both L. rigidum and wheat seedlings. This supports the conclusion that reduced L. rigidum growth and seed production in the field resulted from increased competitive ability of the crop, rather than any direct and detrimental impact of lime on L. rigidum growth. Incorporation of lime reduced initial emergence of L. rigidum in controlled conditions, with L. rigidum seeds at a uniform depth, and in the field experiments in situations of high weed density, with seeds buried by the incorporation process. Nationally, the revenue loss from residual L. rigidum in crop is A$93 million per year. The current research confirms that application of lime will increase the competitive ability of crops growing in regions with acidic soils.

The long-term success of weed management programs requires that all crops in a rotation receive satisfactory weed control. Band sowing with inter-band hoeing has been proposed as an innovative weed management strategy for grain crops. In the band-sowing system, crops are sown in a broadcast pattern within a band of some chosen width (here we selected 12.7 cm); weeds between bands are controlled with inter-band hoeing, with or without so-called “blind cultivation,” for example, tine harrowing. Alteration of the crop spatial arrangement from typical single-line rows to a more evenly distributed pattern aims to enhance interspecific competition while reducing intraspecific competition. Field experiments, conducted in Maine in 2016 and 2017, compared band sowing with inter-band hoeing to the region’s standard practice of planting in 16.5-cm rows and tine harrowing in four test crops: spring wheat (Triticum aestivum L. ‘Glenn’), oat (Avena sativa L. ‘Colt’), field pea (Pisum sativum L. ‘Jetset’), and flax (Linum usitatissimum L. ‘Prairie Thunder’). Band sowing improved weed control relative to the standard practice, especially in crops with greater competitive ability (wheat and oat). Despite improved weed control, in most cases, yields were unaffected by treatment. While band sowing with hoeing provided improved weed control in multiple crops, further study is warranted to optimize seeding rate, band width, and inter-band width to improve crop yields.

Spot spraying POST herbicides is an effective approach to reduce herbicide input and weed control cost. Machine vision detection of grass or grass-like weeds in turfgrass systems is a challenging task due to the similarity in plant morphology. In this work, we explored the feasibility of using image classification with deep convolutional neural networks (DCNN), including AlexNet, GoogLeNet, and VGGNet, for detection of crabgrass species (Digitaria spp.), doveweed [Murdannia nudiflora (L.) Brenan], dallisgrass (Paspalum dilatatum Poir.), and tropical signalgrass [Urochloa distachya (L.) T.Q. Nguyen] in bermudagrass [Cynodon dactylon (L.) Pers.]. VGGNet generally outperformed AlexNet and GoogLeNet in detecting selected grassy weeds. For detection of P. dilatatum, VGGNet achieved high F1 scores (≥0.97) and recall values (≥0.99). A single VGGNet model exhibited high F1 scores (≥0.93) and recall values (1.00) that reliably detected Digitaria spp., M. nudiflora, P. dilatatum, and U. distachya. Low weed density reduced the recall values of AlexNet at detecting all weed species and GoogLeNet at detecting Digitaria spp. In comparison, VGGNet achieved excellent performances (overall accuracy = 1.00) at detecting all weed species in both high and low weed-density scenarios. These results demonstrate the feasibility of using DCNN for detection of grass or grass-like weeds in turfgrass systems.

Cover crop residue can act as a mulch that will suppress weeds, but as the residue degrades, weed suppression diminishes. Biomass of cover crop residue is positively correlated to weed suppression, but little research is available regarding the composition of cover crop residue and its effect on weed suppression. Field experiments were conducted to determine the impact of cover crop residue properties (i.e., total carbon, total nitrogen, lignin, cellulose, and hemicellulose) on summer annual weed suppression and cash crop yield. Cover crop monocultures and mixtures were planted in the fall and designed to provide a range of biomass and residue properties. Cover crops were followed by corn (Zea mays L.) or soybean [Glycine max (L.) Merr.]. At termination, cover crop biomass and residue components were determined. Biomass ranged from 3,640 to 8,750 kg ha−1, and the carbon-to-nitrogen (C:N) ratio ranged from 12:1 to 36:1. As both cover crop biomass and C:N ratio increased, weed suppression and duration of suppression increased. For example, a C:N ratio of 9:1 is needed to suppress redroot pigweed (Amaranthus retroflexus L.) 50% at 4 wk after termination (WAT), and that increases to 16:1 and 20:1 to have 50% suppression at 6 and 8 WAT, respectively. Similarly, with biomass, 2,800 kg ha−1 is needed for 50% A. retroflexus suppression at 4 WAT, which increases to 5,280 kg ha−1 and 6,610 kg ha−1 needed for 50% suppression at 6 and 8 WAT, respectively. In general, similar trends were observed for pitted morningglory (Ipomoea lacunosa L.) and large crabgrass [Digitaria sanguinalis (L.) Scop.]. Corn and soybean yield increased as both cover crop biomass and C:N ratio increased where no weed control measures were implemented beyond cover crop. The same trend was observed with cash crop yield in the weed-free subblocks, with one exception. This research indicates that cover crop residue composition is important for weed control in addition to biomass.