Scientific principles of weed invasion, establishment and spread
Some introduced species, like gorse, have been spectacularly successful and now occupy thousands of hectares. Others, like nodding thistle, have increased dramatically in numbers and distribution in the last 50 years39. And yet many introduced species have stayed where they were planted, in gardens or on farms. Why do some species become pestilential weeds, whilst others remain perfectly well behaved?
Weed risk assessment (WRA), determining which plants are likely to naturalise, spread and create problems, is a 'new and developing discipline'43. If we know in advance which plants were going to become weeds, we could stop them entering the country in the first place, eradicate them while they were few in number, or make determined efforts to stop them spreading further.
According to the Department of Conservation11, in addition to the 2000 or so exotic species that have become naturalised in New Zealand, 17,000 exotic species grow in our gardens, plantations and farms. Four thousand of these have been listed as weeds elsewhere in the world44. New plants or seeds are regularly brought into the country either accidentally or deliberately, legally or illegally44. According to Smith et al.45 the reported proportion of introduced organisms that become pests ranges from 0.007% in Britain to 17% for plants introduced to Australia as pasture plants. The authors settle on a 'possible central tendency' of 2%.
Some species introduced to New Zealand have been valuable for their economic importance or the pleasure they bring to people. New species continue to be introduced: some are useful and unlikely to adversely affect the environment, whilst others could degrade New Zealand's productivity, landscapes or biodiversity. MAF maintains a list of prohibited species44. MAF's former responsibility for evaluating proposed introductions on the basis of their weediness elsewhere and on other characteristics has now been passed to the Environmental Risk Management Authority (ERMA). The measures aim to prevent the introduction of potentially environmentally-damaging plants as well as those that pose threats to agriculture.
Panetta et al.14 point out that predicting potential problem weeds is much easier in the case of agricultural weeds than it is for environmental weeds. Weeds that become problems in pastures are likely to be poisonous, prickly or otherwise unpalatable to stock, and invasive. Environmental weeds have a much wider range of characteristics. Although Panetta14 says 'Screening systems designed solely to prevent importation of agricultural weeds are thus unlikely to limit the introduction of many invaders of natural areas', screening systems for detecting invaders of natural areas will probably account for most of the potential agricultural weeds.
Clearly, for any plant species that escape border controls, early eradication is the best option. This has been achieved in isolated instances like skeleton weed (Chondrilla juncea) and water lettuce (Pistia stratiotes) in New Zealand, and is the principle behind
Central Government's involvement in the control of four potentially serious weeds currently of limited distribution.
Newly introduced plants often follow the population growth pattern shown in Fig. 4. Most Regional Council Plant Pest Strategies use similar graphs to explain their strategies. The length of the lag phase is variable and may depend on the plant species becoming acclimatised to local conditions or building up numbers to the point where it can expand rapidly. In some cases the lag phase may end when external conditions change in such a way that the plant population can rapidly expand.
Figure 4. Population growth pattern of a newly introduced species.
Plants that have become naturalised, and regarded as less dangerous than the four controlled by MAF, or that are too widespread for easy eradication, become the responsibility of Regional Councils. Weeds listed as pest plants by Regional Councils may include those that:
- Have been serious weeds for a long time (blackberry, gorse and ragwort) and have probably more or less reached the limits of their distribution, but whose populations wax and wane with the prosperity of farming and with the kind of livestock being farmed,
- Have long been considered noxious and which may or may not still be expanding their ranges (barberry, boxthorn, some thistle species),
- Are of limited distribution at present but could certainly spread or increase in numbers (Chinese pennisetum, nut grass, kiwi fruit, nassella tussock),
- Are established in the wild, and may yet become serious weeds (lag phase, or Groves' 'sleepers'14) (olives9).
Plant species that can clearly be identified as potential weeds should be controlled while still in the 'lag' phase. Guthrie-Smith46 described the difficulties of spotting potential weeds early enough:
'The fact is that all the King's horses and all the King's men cannot catch up with a weed that has obtained a start. No action is ever taken in time; to begin with, the new plant is not noticed in its unit stage; when it numbers hundreds a few of the more observant settlers become interested; when thousands appear it is talked of as a newcomer; only when the hundred thousand phase is past, when the plant has been carried or blown abroad to every corner of every province in New Zealand, is legislation attempted.'
How then should Regional Councils categorise weeds and decide which are important enough to warrant the costs of declaration as plant pests? Panetta14 suggests that, at this level, WRA is needed as a basis for prioritising actions against known weed species. In Australia, Virtue et al.47 have devised a scoring system for determining weed issues of national significance. This proposed National Significance Assessment System treats agricultural, forestry and environmental weeds equally, and considers the five issues of current and potential distribution, value of different land uses, invasiveness, impacts and feasibility of control.
New Zealand's Regional Councils already use WRAs in setting their pest plant strategies. When MAF was responsible for noxious plants, Esler carried out WRAs of all noxious plants48. The same system was then used by Champion on 124 plants for the Auckland Regional Council49, an approach widely adopted by other regional councils in preparing their Pest Management Strategies.
In preparation for the 2001 revision of Pest Plant Strategies, the Biosecurity Officers Working Group developed a report on preparing these. Appendix III, by Simon Harris50, developed a model for cost/benefit assessment of pest plant control and many Regional Councils are using this to support their pest plant strategies. The model includes current and potential distribution, time to reach that potential, landowner efforts at control, value of land use, and potential impact due to lost production and control costs. The model guideline is useful for Regional Councils and provides them with a uniform and partially objective means of comparing the importance of weeds. It certainly goes some of the way towards addressing WRA, and could easily be developed further. Two aspects of 'guesswork' involved in the model are the potential area infested, and the time taken for the weed to achieve its full potential. Both these are difficult or impossible to estimate accurately.
Weed Risk Assessment models are still being developed and improved. MAF and other agencies concerned with pest plants must keep abreast of overseas developments in such models.
Contact for Enquiries
MAF Information Services
Pastoral House
25 The Terrace
PO Box 2526
Wellington, NEW ZEALAND
Fax: +64 4 894 0721
Contact this person
