Controlling cotoneaster – grub, spray or burn
Published source details
Bond W. (2003) Controlling cotoneaster – grub, spray or burn. Conservation Land Management, 4-7.
Published source details Bond W. (2003) Controlling cotoneaster – grub, spray or burn. Conservation Land Management, 4-7.
Old quarry workings on the Isle of Portland have a high conservation value for limestone grassland species but are being overwhelmed by the introduced invasive, small-leaved cotoneaster Cotoneaster integrifolius. Three different approaches to control (mechanical removal, herbicide, and flame gun) were tested on in February 2002.
The mechanical treatment and its consequences are outlined below.
On 14 February 2002, mechanical removal of small-leaved cotoneaster was undertaken using a 13-tonne tracked excavator, fitted with a tilting mechanism at the end of the dipper arm. This enabled the bucket to be angled with greater flexibility around lumps of stone and to follow the random contours. Using this method, Cotoneaster was removed with as much root as possible, and burnt on nearby open ground. A trial area of 1,000 m² was cleared.
Anticipated disadvantages were a relatively high cost, the possibility of ripe berries being shed onto the ground (thus giving rise to new plants), safety implications and the fact that this technique could only be used on about 50% of the area due to steepness of the terrain.
The excavator quickly and efficiently removed the Cotoneaster ‘canopy’ but many roots were left in the soil. The process disturbed almost the entire ground surface leaving it open and loose. Any residual limestone flora was also removed with the cotoneaster. The treatment was slowed down by the need to burn the grubbed up material. The estimated cost of treatment was £0.66/m².
Initial monitoring of the treated areas was undertaken four months later in June 2002. The grubbed area looked messy, and vegetative regrowth of Cotoneaster was visible and abundant on remaining root fragments. Recently germinated plants were also visible. Additionally, there were many small bramble Rubus fruticosus, thistle Cirsium spp., ragwort Senecio jacobaea and dock Rumex spp. plants appearing but few desired limestone specialists were apparent.
Eight months after treatment the area had not become much worse but showed little sign of native vegetation recovery. Overall, mechanical removal was not considered a successful method of cotoneaster control at this site.
The herbicide treatments and their consequences are outlined below.
On 14 February 2002, three herbicides were sprayed onto patches of small-leaved cotoneaster. As an assessment of the effectiveness of each was the test aim, no wetters were used. All herbicides were applied by a qualified worker using a knapsack sprayer. These, (their active ingredients), application rates and estimated costs (£ stirling/m²) were:
Roundup (glyphosate): 10 l/ha - £0.11/m²
Broadsword (24D + dicamba + triclopyr): 4 l/ha - £0.18/m²
Garlon (triclopyr): 30 l/ha - £0.12/m²
Spraying offered flexibility in terms of accessing difficult terrain and crevices with minimal disturbance of ripe Cotoneaster seed and no ground disturbance. A disadvantage was the lack of selectivity when native grassland flora was intermixed with Cotoneaster.
After four months, healthy-looking green Cotoneaster tips were growing beyond the browned off mid-level parts of the bushes. Canopy cover was still dense as there was no leaf loss. After eight months the sprayed areas appeared better as leaves were mostly dead. However, most, although dead were still in place, therefore light did not penetrate. Sporadic new green growth was restricted to ivy Hedera helix and bramble Rubus fruticosus.
All three tested herbicides achieved acceptable results. Broadsword was the most effective, closely followed by Roundup. The authors considered that performance could be improved by using wetters and refining the timing of application. Spraying, in terms of Cotoneaster kill was considered the most effective and cheapest solution of three techniques tested.
The flame gun treatment and its consequences are outlined below.
On 14 February 2002, a triple-nozzle flame gun fuelled by propane gas, as used for weed control in some urban environments, was tested. The flame was applied directly to small-leaved cotoneaster plants. An attempt was made to target the base of the stems where regrowth was otherwise considered most likely to occur.
The flame gun was too small to treat large stands of Cotoneaster. The process was very slow and took time and patience to get damaging heat into the bottom of the stems.
Four months after treatment there was a flush of limestone plant species and although there was some Cotoneaster regrowth it was much less in comparison to the mechanically grubbed area (for a summary see: www.conservationevidence.com/ViewEntry.asp?ID=17). After eight months, development of a good cover of limestone species continued but the Cotoneaster was also recovering and regeneration unacceptably widespread.
Note: If using or referring to this published study please read and quote the original paper.