Abstract
Tussock grasslands represent an important but threatened ecosystem in New Zealand, especially in the South Island high country. Tussock grasslands hold significant ecological and cultural value but, like many of New Zealand’s ecosystems, are threatened by invasive species and changing land use. While many areas previously used for agriculture are converting back to natural vegetation, these heavily disturbed systems remain vulnerable to invasion during this process of vegetation change. This transition, in combination grasslands having no canopy cover and relatively nutrient poor soils makes them susceptible to invasion by wilding conifers. Wilding conifers pose a threat to hundreds of thousands of hectares of both productive and ecologically significant land in New Zealand. Due to the rapid nature of the spread of wildings the need for control is time sensitive. One of the most common methods of control is aerial herbicide application. However, the addition of herbicides into an ecosystem can potentially have several impacts on the system, including changes in composition and abundance of both above and below ground biota. A key component in soil microbiota is the presence of arbuscular mycorrhizal fungi (AMF), which form mutually beneficial associations with native tussock species that facilitate the uptake of nutrients and water in naturally poor soils. The effect that herbicides used for wilding conifer control have on AMF is largely unknown, and the nature of this effect could have long term impacts on the functioning of tussock grassland systems.
To assess the effects that aerial herbicide application could have on invaded tussock grasslands two studies were carried out. A controlled glasshouse experiment assessed the colonisation of AMF on roots of the endemic tussock grass, Chionochloa rigida, after receiving a herbicide treatment. TDPA (triclopyr, dicamba, picloram, aminopyralid) and metsulfuron were used as they are currently the two recommended herbicide treatments for wilding conifer control. Glyphosate was the third herbicide used as a positive control as it has been shown to negatively affect AMF in many agricultural systems. All three herbicide treatments resulted in a significantly lower rate of AMF colonisation than the control treatments, however, there was no significant difference between each of the three herbicide treatments. Despite both herbicide treatments being selective for broadleaf species, TDPA had a negative effect on the health of the experimental plants with many having died at the time of sampling, and all of them dying by the end of the experiment. Metsulfuron did not affect the health of the tussocks with all plants remaining in good to excellent condition for the duration of the experiment.
An observational field study was conducted to assess the effects of aerial herbicide application of metsulfuron in a natural setting. An area was selected in which several different sites had received different conifer control treatments. Two sites had been sprayed with metsulfuron (one in 2018 and one in 2020), conifers at a third site had been felled, a fourth site had conifers that had yet to be sprayed and the fifth site was native tussock grassland that was yet to be invaded or receive any type of treatment. AMF colonisation on roots of the endemic tussock grass, Festuca novae-zelandiae, was assessed at each site.
This study was pseudo-replicated due to COVID19 limitations on field work, so results should be interpreted cautiously. The most recently sprayed site (2020) had a significantly lower level of AMF colonisation than the site sprayed in 2018, which had similar levels to the sites that hadn’t received herbicide treatment. This is consistent with herbicides reducing AMF colonisation in the field, and that levels of AMF colonisation may recover relatively quickly. Vegetation surveys were conducted and sites that receive herbicide application were more likely to be dominated by exotic weeds in the period post spraying. This study highlights that aerial herbicide application may have several negative implications for tussock grasslands communities which should be considered when carrying out wilding conifer control. Further studies are required to fully assess these implications using appropriately replicated sites in different locations and areas where TDPA is used instead of metsulfuron.