Search

The prevalence of herbicide resistance in ryegrass (Lolium spp.) in the wine-growing regions in New Zealand is poorly understood. Cases of glyphosate, glufosinate and amitrole-resistant ryegrass were documented in a few vineyards in New Zealand in 2013, but there have been no regional surveys for resistance. To address this knowledge gap, 106 vineyards were visited across the important New Zealand wine-growing regions of Marlborough and Waipara in late February 2021, and Hawke’s Bay and Gisborne in late February 2022, and seed samples from individual plants at each site surviving weed-control measures were collected. Ryegrass was found in more South Island (68%) than North Island (20%) vineyards. These seeds, and those from a susceptible ryegrass population were sown in marked rows into trays (10-20 seeds per herbicide) and grown in a glasshouse. When seedlings reached the 3-4 leaf stage, trays were sprayed at the highest recommended label rate of glyphosate. Samples with enough seed were also screened against additional herbicides, amitrole, glufosinate or clethodim. The results indicated 39% of the surveyed vineyards contained glyphosate-resistant ryegrass, with cases detected across all regions, including 58% of vineyards in Marlborough. Eleven of the 27 Marlborough vineyards screened contained amitrole-resistant ryegrass; six samples were also resistant to glyphosate. However, glufosinate and clethodim were still effective against ryegrass at the sites tested. Considering the levels of herbicide resistance to ryegrass observed in this study, growers should explore alternative weed-suppression measures, including tilling, cover-crops, grazing, mowing and the use of herbicides with different modes of action.

Weeds can be major environmental and economic burdens in New Zealand. Traditional methods of weed control including manual and chemical approaches can be time consuming and costly. Some chemical herbicides may have negative environmental and human health impacts. One of the proposed important steps for providing alternatives to these traditional approaches is the automated identification and mapping of weeds. We used hyperspectral imaging data and machine learning to explore the possibility of fast, accurate and automated discrimination of weeds in pastures where ryegrass and clovers are the sown species. Hyperspectral images from two grasses (Setaria pumila [yellow bristle grass] and Stipa arundinacea [wind grass]) and two broad leaf weed species (Ranunculus acris [giant buttercup] and Cirsium arvense [Californian thistle]) were acquired and pre-processed using the standard normal variate method. We trained three classification models, namely partial least squares-discriminant analysis, support vector machine, and Multilayer Perceptron (MLP) using whole plant averaged (Av) spectra and superpixels (Sp) averaged spectra from each weed sample. All three classification models showed repeatable identification of four weeds using both Av and Sp spectra with a range of overall accuracy of 70–100%. However, MLP based on the Sp method produced the most reliable and robust prediction result (89.1% accuracy). Four significant spectral regions were found as highly informative for characterizing the four weed species and could form the basis for a rapid and efficient methodology for identifying weeds in ryegrass/clover pastures.