NSW WRM system - Weed Risk

Important assumptions

Current weed management

Weeds that are not controlled by current management practices are more likely to reach high numbers and have greater impacts than those that are controlled.

Assumptions about the current routine weed management practices for each land use examined need to be clearly stated and recorded, for example, standard herbicide use, cultivation and/or physical control. In practical terms this may mean what herbicides are used and the number and type of cultivation events that are actually performed by farmers in a land use like the dryland cropping rotation. Practices may vary considerably between (and even within) land uses and there may be no routine weed management undertaken for some land uses, for example in natural environments. Stating and recording these assumptions will help all participants to accurately and consistently answer the questions and reduce any possible conflict.

Each question assesses weed potential under current routine weed management practices. As outlined above, this may or may not affect the weed. This also assumes no future government intervention.

Targeted weed management in land use

The cost and effectiveness of specific, targeted control of different weeds is taken into account when assessing the feasibility of coordinated control. For example, participants would list any additional herbicide treatments, physical or mechanical control methods that would be used. Stating and recording these assumptions will help all participants to accurately and consistently answer the questions and reduce any possible conflict.

Density of weed in land use

Another important assumption to record is the density of the weed in the land use, whether low, medium or high. This is used in the Impacts section.

Notes for all assessors

Weed Risk Assessment section

The comparative weed risk questions are divided into three main criteria;

Invasiveness, Impacts and Potential distribution.

In the weed risk assessment section, higher scores indicate a higher weed risk.

Invasiveness looks at the weed's ability to establish, reproduce and spread.

Impacts are the economic, environmental and social affects the weed would have caused when established.

Potential distribution indicates the total area the weed could spread to.

Scores for each of these criteria (each ranging between 0 and 10), are multiplied to give a weed risk score out of 1000.

All questions should be answered assuming only the existing weed control practices in the land use.

Invasiveness

Invasiveness is a measure of the weed's ability to establish, reproduce and spread. Faster spreading weeds are a higher priority for control. Invasiveness will often vary between different land uses due to differences in management practices, disturbance regimes and resources available for weed growth. Direct measurement of spread is difficult without information collected sequentially over time.

This section uses five questions to analyse how quickly a weed can spread within a particular land use. These questions account for how well the weed can establish, reproduce and spread. Answer all questions with the land use in mind, except for question 5(a) because people often deliberately spread plants irrespective of land use.

Invasiveness Question 1

What is the ability of the weed to establish amongst existing plants?

Importance of the question

Weeds that are more able to establish among existing plants are likely to be more of a problem.

Greater significance is given to weed species that can readily establish amongst existing vegetation, having the capacity as seedlings/juveniles (young plants) to tolerate competition for light, moisture and/or nutrients. Establishment may occur from seeds or from vegetative units, for example, bulbs, root fragments, tubers or rhizomes. Larger propagules (seeds or vegetative units) will usually have greater reserves for establishment. Seedlings that have shade tolerance, nitrogen fixation, rapid root growth and/or drought tolerance are also likely to have greater establishment ability amongst competing vegetation. Weed species that have a poor ability to establish amongst existing vegetation mainly establish after significant vegetation disturbance events, such as fire, cultivation, drought or over-grazing.

Explanation and assumptions

Features which can help a plant establish amongst existing plants include:

• the ability to germinate under the canopy of other plants, for example, weeds that have a staggered germination in crops
• large seeds or vegetative propagules, for example, bulbs, root fragments, tubers or rhizomes provide more reserves to help the weed establish in competition with other plants
• the ability to tolerate or avoid competition, for example, by rapid root or shoot growth, nitrogen fixation or drought tolerance.

Weeds that invade well-managed land uses (where a dense cover over the soil is maintained) are assumed to have a higher weed potential. High scoring weeds include Bridal creeper (Asparagus asparagoides), and the Dodders (Cuscuta species) that are parasitic plants, and establish well among other plants.

Assume no weed control practices for this question. Depending on the land use 'vegetation' may be crops, pastures, lawns or native vegetation. The density of existing plants will depend on the land use, for example dense vegetation in the grazing modified pasture land use may be quite sparse when compared to dense vegetation in a natural environment land use such as a forest.

Assume the weed has just arrived.

"Seedlings" includes growth from dispersed vegetative propagules, for example, broken fragments of Alligator weed (Alternanthera philoxeroides) stems or Silver-leaf nightshade (Solanum elaeagnifolium) roots, and spores in addition to seeds. "Seedlings" do not include new vegetative growth attached to the parent plant, for example, by stolons, rhizomes or lateral roots. This feature is covered in Question 3(c).

Invasiveness Question 2

What is the weed's tolerance to average weed management practices in the land use?

Importance of the question

Weeds that are more able to tolerate management will have greater survival and spread.

This question examines whether the new weed is killed by the weed management practices that are commonly used across the land use, for example a new grass weed may be killed by a grass herbicide along with existing grass weeds in a broad-leaf dryland crop. If few individuals of the new weed are killed then changes to weed management practices will eventually be needed.

Explanation and assumptions

Assume the weed is new to the area.

Current weed management practices may include herbicides, cultivation, cutting/slashing, grazing and fire and maintaining the competitiveness of desirable vegetation, for example by fertiliser or pest/disease control. The types and timing of these practices may vary within a land use. For example, weed management may vary between cereal cropping and broad-leaf cropping within the Dryland cropping land use, or between vineyards and citrus production in the Dryland perennial horticulture land use. Average the weeds survival if this is the case.

If a weed grows and sets seeds when there is normally no weed management (for example during a winter fallow in a summer crop) then it is highly tolerant of common weed management practices.

Weeds that have a high tolerance to routine weed management would include Silver-leaf nightshade (Solanum elaeagnifolium) and Alligator weed (Alternanthera philoxeroides) which are difficult to kill or Fireweed (Senecio madagascariensis) which is quick to set seed and may therefore escape routine controls.

In native vegetation there may be no commonly used weed management practices. If this is the case then record this in your assumptions about the land use in the form.

Invasiveness Question 3

What is the reproductive ability of the weed in the land use?

Importance of the question

Weeds that can reproduce quickly produce more seed and those that have vegetative reproduction are more likely to be a problem.

Explanation and assumptions

This question looks at how well the weed can reproduce, to rapidly build up its numbers at a site under current land use weed management practices. Highly invasive weeds often reproduce both by seed and vegetatively. Three factors are important. Answer each part question.

1. Time to seeding. How long does it take from establishment (from seed or vegetative propagules) to the production of viable (live) seed? Annual plants, for example Fireweed (Senecio madagascariensis) have a faster rate of population growth (that is initially build up numbers faster) than slow growing trees such as Camphor laurel (Cinnamomum camphora).

2. Seed production. How much seed is produced? This is best considered as the average number of viable seeds produced per square metre of ground area underneath the canopy per year.

   A high seed production would be >1000 viable seeds per m².

   Seed release is important since a plant that retains its seed in the canopy for years until an infrequent disturbance event is likely to have a slower rate of spread than one that releases its seeds annually. Recording seed release per unit ground area rather than plant numbers gives a standard method of comparison when one or many plants may be present. Average seed production per year needs to consider the frequency of seeding events, for example, continuous throughout the growing season, annual or biennial, and the number of seeds contained per fruit. Answers to question 2 may influence the amount of seed production.

3. Vegetative reproduction. Does the weed have frequent vegetative reproduction? "New plants" are defined as shoots with their own root system. There may still be some connection to the parent plant.

   Frequent vegetative reproduction would be >10 new plants per year from a mature parent plant.

   Consider vegetative reproduction as the average number of new plants produced each year by such means as bulbs, bulbils, corms, tubers, rhizomes, stolons, root suckers, root fragments, shoot fragments, or new shoots (for example in the case of plantlets produced on leaves by species such as Mother-of-Millions, (Bryophyllum species).

   Vegetative reproduction is not shoot regrowth following shoot removal.

   Management in certain land uses may increase vegetative reproduction. For example cultivation of areas containing weeds like Perennial ragweed (Ambrosia psilostachya) or Black knapweed (Centaurea nigra) will increase the number of new plants arising from vegetative reproduction.

   If a weed never gets to reproduce in a land use then it will score 0.

Invasiveness Question 4

How likely is long-distance dispersal (>100 m) by natural means?

Importance of the question

More significant weeds are likely to be those with more means of dispersal or that have propagules that are regularly moved long distances from parent plants. These weeds tend to spread faster.

Explanation and assumptions

This question looks at how well the weed can spread its propagules (seed, vegetative or spores) by natural means to start new weed outbreaks at a long distance (>100 m) away from the original outbreak. Dispersal ability depends on the number of dispersal modes for a plant species, the number of times this occurs and the dispersal distance achieved.

Answer each part question.

Consider if a plant is adapted for long-distance dispersal by any of the natural means mentioned in this question, that is by flying animals, other wild animals, water or wind. Ignore domestic and farm animals as these are covered in Question 5(d).

Features favouring long-distance dispersal by flying animals and other wild animals include:

• whole fruits that are eaten with viable seed defecated or regurgitated
• seeds that have an aril or coating which is attractive to birds or animals with the viable seed discarded

  Examples include Bitou bush and Boneseed (Chrysanthemoides monilifera subsp. rotundata and monilifera respectively), Broad-leaf pepper tree (Schinus terebinthifolius), Camphor laurel (Cinnamomum camphora), Olives (Olea europaea, both subspecies), Privet species (Ligustrum lucidum and L. sinense) and Sweet briar (Rosa rubiginosa).

• propagules that have hooks, barbs or sticky substances that attach to feathers, hairs or skin, for example, Horehound (Marrubium vulgare) and Noogoora and Hunter burr (Xanthium occidentale and X. italicum respectively)
• very small seeds which can lodge within feathers, hairs or feet, for example, nutgrass, (Cyperus species)
• vegetative components which may be picked up and carried by birds or animals, for example, grass stems carried by grazing animals that may take root.

Examples of wild animals include emus, foxes, kangaroos, rabbits, reptiles and feral horses, goats and cattle. Wild animals disperse seeds either through the gut or by external transport on their fur or feet.

Seeds of most species can be dispersed short distances by water runoff after heavy rainfall events. However, aquatic, coastal and riparian species may be pre-adapted for long-distance water dispersal. Long-distance water dispersal is more likely with:

• propagules which float (consider wind-assisted water movement as water dispersal)
• weeds located in or near to moving water
• areas that receive frequent floods.

Examples of weed species commonly dispersed long-distance by water include many floating aquatic species like Water hyacinth (Eichhornia crassipes) and Water lettuce (Pistia stratiotes), and emergent aquatic species such as Sagittaria (Sagittaria platyphylla), Rubber vine (Cryptostegia grandiflora) and Mimosa (Mimosa pigra). Seed-producing Willows (Salix species) are also dispersed by water.

Research has shown that the majority of wind-dispersed propagules land close to the parent plants. Dispersal beyond 100 metres is generally uncommon. Despite this, long-distance wind dispersal is more likely to be occasional or common for:

• tall trees with light seeds
• weeds with light seeds with wings, plumes or hairs, for example, Siam weed (Chromolaena odorata) and seed-producing Willows (Salix species)
• weeds with propagules which can snap off after fruiting and roll across sparsely-vegetated ground, for example, African turnip weed (Sisymbrium thellungii), Kochia (Bassia scoparia) and Serrated tussock (Nassella trichotoma).

Invasiveness Question 5

How likely is long-distance dispersal (>100 m) by human means?

Importance of the question

Weeds that have more means of dispersal tend to spread faster and are potentially more significant.

Explanation and assumptions

This question looks at how well the plant can spread its propagules (seed, vegetative or spores) by human-influenced means, to start new weed outbreaks at a long distance (>100 m) away from the original source.

Consider if a plant is adapted for long-distance dispersal by any of the means below and how regularly these means of dispersal occur.

Answer each part question.

Deliberate human spread includes plants which are currently or have been historically planted for use in agriculture, forestry, horticulture (including the nursery trade), and for medicinal, aquatic, turf, amenity, shelter and/or soil protection purposes. It also includes those planted in research sites for these purposes but then have escaped. Ignore the land use for question 5a only because people often deliberately spread plants irrespective of land use. Those plants which are, or have been widely planted have greater potential for dispersal due to many introduction points, since these plantings provide sources for new infestations. Examples include African lovegrass (Eragrostis curvula), Pines (Pinus species), Olives (Olea europaea, both subspecies), Blackberry (Rubus species), Willows (Salix species), Gorse (Ulex europaeus), Athel pine (Tamarix aphylla) and Bitou bush (Chrysanthemoides monilifera subsp. rotundata), among many other species. The Positive impacts of these species have been examined.

Deliberate human spread also includes plants which have been deliberately planted or kept in aquaria but later dumped as garden waste or in streams. This may include parts of, or whole plants. In addition, it includes plants that are deliberately moved by people in the mistaken belief they are something else. For example when plants with attractive flowers are picked and then discarded such as the Cape tulip (Moraea species). A further example is when the identification of grasses or other species is difficult and when this confusion results in seed or plants of a weed species being collected and moved in the mistaken belief they are a desirable species.

Although a weed may be legally restricted from sale, it may still be propagated and planted illegally.

Features favouring accidental human and vehicle dispersal are:

Propagules of potential weeds can be dispersed via contaminated produce including farm, mining, and landscaping products such as crop seed, pasture seed, hay, grain, soil, sand, gravel, fertilisers, manures and/or mulch. Contaminated produce also includes the by-products or waste of industries such as stockfeed manufacturers and tanneries and may include weeds on or in rolled turf. Examples of weed species commonly dispersed long-distance as produce contaminants include Paterson's curse (Echium plantagineum), Soursob (Oxalis pes-caprae) and Wild radish (Raphanus raphanistrum). Do not consider contaminants in wool when this relates to animals that are sold between properties as this is covered in question 5(d) (of this question).

Features favouring dispersal by domestic/farm animals, for example, sheep, cattle, horses, goats and dogs include:

• when whole fruits are eaten with viable seeds defecated or regurgitated, for example, ryegrass (Lolium species)
• propagules that have hooks, barbs or sticky substances that attach to feathers, hairs, wool or skin, for example, Horehound (Marrubium vulgare) and Noogoora burr (Xanthium occidentale)
• very small seeds which can lodge within feathers, hairs or feet, for example, nutgrass, (Cyperus species)
• weeds that grow in or near pasture, paddocks, stables, cattle yards, watering holes, homesteads, tracks or roads, for example, Serrated tussock and Chilean needle grass, (Nassella trichotoma and N. neesiana respectively).

Impacts

This section helps to determine the potential environmental, social and economic impacts the weed could have. Weeds can cause a loss in the supply, quality or usage of desired products and services, for example, agricultural production, nature conservation, recreation, water supply and urban infrastructure. There are various difficulties in placing dollar amounts on these impacts. For this reason, this section focuses on the types and size of the impacts, as related to weed density and abundance.

Each question is answered with a land use in mind. Assume that the weed has spread across a whole field, orchard, plantation, nature reserve or water body and that the commonly used weed management practices have not changed to specifically target the weed.

Decide if the weed is likely to reach a low, medium or high density in the land use. In most cases the impact of the weed will be related to the weed's density or abundance. Stating and recording this assumption is important to help all participants accurately and consistently answer the questions.

If the weed is well controlled by these common practices then it will occur at a low density and will have minimal impacts. Alternatively, if the weed is poorly controlled by these common practices then it may get to a high density and have substantial impacts. If the weed has an effective biocontrol agent established which substantially reduces its growth then the weed's impacts will be reduced.

Generally only negative impacts should be considered in Questions 1-5 of this section. Some positive impacts of a weed are considered at Question 6. Any other positive impacts should be recorded as text in the Positive impacts section.

It may also be important to record the timeframe you are considering the impact of the weeds over in the assumptions section. This is particularly important if the impact of the weed varies with time. For example, the impact of the weed Lantana (Lantana camara) is often most severe in terms of native species establishment in a natural ecosystem during the years immediately after invasion but this affect seems to be reduced as time progresses.

Impacts Question 1

Does the weed reduce the establishment of desired plants?

Importance of the question

Weeds that greatly reduce the establishment of desirable plants are more likely to have greater impacts.

Explanation and assumptions

This question looks at whether the weed prevents the establishment of desired plants so that the density of these is reduced. Desirable plants can be crops, pastures, trees or native vegetation. A reduction in desirable plants may cause a reduction in the abundance of animals dependant on these plants.

The weed may prevent the germination of desired plants by dense shading, or by forming a physical barrier to water movement or light to the soil. The weed may kill or stunt seedlings by competing for moisture, light and nutrients. Weeds can also reduce the establishment of desirable plants through allelopathy (when one plant produces chemicals that reduce the establishment and/or growth of another plant).

Desired plants may mainly establish after a major disturbance, for example, after cultivation or a fire. Weeds may also establish at these times and impact on desirable plant establishment.

Weeds that are likely to cause greater than 50% reductions in establishment are Broad-leaf privet (Ligustrum lucidum), Gorse (Ulex europaeus), early germinating and unsprayed Paterson's curse (Echium plantagineum) in pastures and Phalaris (Phalaris species) in native vegetation.

Impacts Question 2

Does the weed reduce the yield or amount of desired vegetation?

Importance of the question

Significance is given to weeds that greatly reduce the yield or amount of desired vegetation. Following on from Question 1, this question looks at the growth achieved by plants that did establish despite the weed. It also considers the reduction in yield or growth of plants that were already established before the weed invaded.

Explanation and assumptions

This question looks at the degree of yield loss in crops, pastures or other produce, for example, fruit or forestry, or suppression of mature native vegetation caused by the weed. Weeds reduce the growth of other plants by competing for moisture, light and nutrients and/or via allelopathy.

Weed competition is greater where a weed is larger, (for example, a tall weed with a dense leaf canopy and an extensive root system) and grows at the same time as a desirable plants. Some weeds also compete by forming physical barriers that stop plants growing to reach light, water and nutrients, for example the tuber mat of Bridal creeper (Asparagus asparagoides).

A special case are parasitic weeds which directly attack other plants, for example, Golden dodder (Cuscuta campestris).

Weeds that can cause greater than 50% reductions in the yield or amount of desired vegetation include Serrated tussock (Nassella trichotoma) and Paterson's curse (Echium plantagineum).

The question is answered on a per hectare basis in comparison to similar vegetation that is free of the weed. For native vegetation, it may be useful to consider percentage cover or biomass instead of the amount of mature native vegetation.

Some weeds may increase the amount of useful vegetation in a land use, for example a perennial weed of grazing land may provide nutritious summer feed thus increasing the total amount of pasture available through the year.

Impacts Question 3

Does the weed reduce the quality of products, diversity or services available from the land use?

Importance of the question

Weeds cause reductions in the quality of products, diversity or services available.

Explanation and assumptions

This question looks at whether a weed causes a loss in the supply, quality or usage of desired products, diversity or services. Our focus is on the negative impacts that are caused by weeds and the size of these impacts. In most cases, the magnitude of impacts will be related to the weed's density or abundance. This weed density is the same as that assumed throughout Impacts section.

A weed may affect products by tainting of meat or milk, discolouration, tainting or otherwise reducing the quality of water, weed seed contamination of grain, seed, hay, wool, fruit or timber. Indigenous use of native bush tucker and materials and the quality of products of sustainable harvesting should be considered here if applicable. Impacts on fishing and hunting by all members of the community are also considered here.

In native vegetation, the decline of native plant species diversity and abundance are the main concerns (with flow on effects to animal diversity). This affects ecosystem structure and function and eventually conservation significance, recreational and tourism values. Weeds may threaten biodiversity by negatively impacting threatened plant and animal species or communities. Specific details for weeds impacting biodiversity in NSW can be found elsewhere (Coutts-Smith and Downey 2006).

In residential areas, the weed may cause damage to physical infrastructure such as buildings, roads and footpaths. Damage to human infrastructure such as fences should also be considered here.

Reductions in visibility and aesthetics are considered as low impacts.

Examples of high effects on quality may include Dodder (Cuscuta species) preventing the sale of seed crops, the zero tolerance of certain weeds in grain crops, or floating mats of Hymenachne (Hymenachne amplexicaule) escaping from ponded pastures and significantly reducing biodiversity in high value conservation areas or causing damage to human infrastructure such as fences.

Reductions in livestock condition or weight should not be considered here as these are due either to a reduction in available feed (Question 2), or animal health effects caused by eating the weed covered in Question 5, for example, the red flowering form of Lantana camara. Restriction of physical movement is considered in Question 4.

For this question, ignore the status of the weed with regard to moving contaminated produce in New South Wales but do consider priority weed lists and seed quality standards of other states and countries. This prevents bias against weeds that are currently not a priority when comparing them to existing priority weeds.

Impacts Question 4

What is the weed's potential to restrict the physical movement of people, animals, vehicles, machinery and/or water?

Importance of the question

Weeds that restrict the physical movement of people, animals, vehicles, machinery and/or water are likely to have greater impacts.

Explanation and assumptions

This question looks at the degree to which a dense infestation of the weed physically restricts movement. Weeds may restrict movement by being tall, thorny, tangled and/or dense.

Examples of weeds limiting movement include:

• blocking or slowing access of cars, bikes, quad bikes or other machinery by a physical barrier, by tangling and/or by tyre puncture
• blocking or slowing farm machinery at sowing and/or harvesting
• interference with boat access and/or maneuverability
• blocking or slowing of water flow
• interference with thinning operations in forestry
• preventing livestock access to pasture and/or water by physical barrier or discomfort
• preventing animal access to nesting sites by physical barrier or discomfort
• impeding movement of people on foot by physical barrier or discomfort.

Examples of weeds that score highly include Blackberry (Rubus fruticosus spp. agg.) and Gorse (Ulex europaeus) at high densities, and cacti species at lower densities such as the Opuntia, Harrisia or Cylindropuntia species such as Hudson pear (Cylindropuntia rosea and C. tunicata), all forming impenetrable thickets. An aquatic example is a dense infestation of water hyacinth (Eichhornia crassipes) making aquatic access difficult or blocking drainage so that flooding and damage to areas where humans live. Ignore any human health impacts that result from flooding of areas where humans live as they are considered in Question 5.

For this question, ignore any deliberate restrictions on movement aimed solely at limiting the spread of weed propagules, for example quarantine measures.

Impacts Question 5

What is the weed's potential to negatively affect the health of animals and/or people?

Importance of the question

Weeds adversely affect the health of people, livestock and/or native animals.

Explanation and assumptions

This question looks at how the weed affects the health of animals (domestic livestock and native) and people.

If a weed is toxic but is not palatable then it may not actually be grazed. Ignore any starvation effects from reduced growth of pasture or reduced access to pasture, as these have been covered in Questions 2 and 4.

Many plants will have a negative effect on human health and these effects may vary considerably from person to person. For example an individual may have an allergic reaction to a certain plant but many others will not. Similarly, respiratory problems may also arise in some people and not others.

A weed with high effects on human health would be Poison ivy (Toxicodendron radicans) while a weed with medium effects on health would be Parthenium weed (Parthenium hysterophorus). There are also many weeds that can potentially poison animals if eaten with two common species being St. John's wort (Hypericum perforatum) and all red flowering varieties of Lantana (Lantana camara) in NSW.

Some aquatic weeds have the potential to slow water flows. This may lead to mosquitoes breeding and an increase in mosquito-borne human diseases. Alternatively, aquatic weeds may block drainage resulting in flooding of areas where humans live again resulting in human disease concerns. Infestations of such weeds may result in medium effects on human health. Ignore any impacts caused by aquatic weeds on flooding and damage to areas where humans live as these are considered in Question 3.

Impacts Question 6

Does the weed have major positive or negative effects on environmental health?

Importance of the question

Greatest significance is given to weeds that are 'ecosystem transformers', that is, those that change the character, condition and nature of ecosystems over substantial areas. Often these plants form monocultures.

Plants that have a major negative impact on a range of environmental health measures are likely to have the greatest impacts.

Explanation and assumptions

This question looks at whether the weed has major long-term effects on a land use's environment. These effects may be beneficial or harmful. Effects are more likely where the weed substantially changes the vegetation structure, such as woody weed invasion of grassland. A major effect is one that is well known, that is backed up by scientific studies, or by expert opinion and has significant impacts across the landscape. A minor effect should also be well known but is either limited in the area of its impact or, if its impact is widespread, is not significant.

A long-term effect becomes apparent over a number of years and may even be opposite the initial effect the species gave. For example, Lippia (Phyla canescens) was once planted to stabilise soils but is now recognised as causing significant slumping of soil banks in riparian areas.

A species is considered not to have an effect if it has been well studied, but there is nothing in the literature that refers to an effect on environmental health.

Answer each part question.

Examples of negative effects caused to food/shelter include Blackberry (Rubus fruticosus spp. agg.) harbouring rabbits, or grass weeds such as Johnson grass (Sorghum halepense) hosting cereal crop pests and pathogens.

Examples of positive effects may be Lantana (Lantana camara), African boxthorn (Lycium ferocissimum) and African olive (Olea europaea subsp. cuspidata) providing food and shelter for native animals where none other exists, although this is likely to be a minor positive effect in most cases.

Ignore pasture for livestock as this was considered in Question 2 and the death of native animals as this was considered in Question 5.

For fire regime, consider changes in normal frequency, intensity and/or timing of fires.

Examples of weeds having major effects include introduced grasses invading shrubby native vegetation and increasing fire intensities.

Some plants increase nutrient levels, for example, legumes can increase soil nitrogen. This may make native vegetation more prone to invasion by other weeds, but would be beneficial to agriculture. Other plants alter nutrient cycling and this can result in result in changes to the diversity of native vegetation, for example Lantana (Lantana camara) alters nitrogen distribution in the soil. Ignore competition for nutrients (decreased nutrient levels) as this was covered indirectly in Question 2.

Soil salinity may be altered by plant growth. For example the leaves of some plants are high in salt and leaf decomposition may increase salinity at the soil surface, for example Athel pine (Tamarix aphylla).

Soil stability may be affected because plants may increase the risk of soil erosion or silting of water ways, for example Lippia (Phyla canescens), Willows (Salix species) and Athel pine (Tamarix aphylla).

Other plants may substantially raise or lower the soil water table compared to other plants. This may have positive or negative effects. Ignore competition for water as this was indirectly covered in Question 2.

In the 'Source and comments' response for this question, mention any other negative effects. These effects will be assessed separately. Examples may include:

• decaying infestations of water weeds such as Salvinia (Salvinia molesta) or other weeds swept into waterways reducing oxygen levels and pH in water
• soil acidification caused by decaying Pine (Pinus species) wilding leaf material
• any increase in flooding regime caused by stream bed blockage caused by, for example, Willow (Salix species)
• if the specific negative effect is unknown. For example Kudzu (Pueraria lobata) has the ability to 'transform' ecosystems by forming monocultures resulting in many negative, but as yet, unresearched effects in Australia.

All positive impacts should be recorded as text in the Positive impacts section.

Potential distribution

This section indicates the total area at risk of invasion if the weed is left to spread uncontrolled. Another way to express this is the 'worst case scenario' of where the weed will spread to if it is left unchecked. For terrestrial weeds, this is determined by climatic, soil and land use tolerances. For aquatic weeds, water temperature and other water properties such as salinity, depth, clarity and flow are important (for more details see Champion and Clayton 2001).

Potential distribution Question

Within the geographic area being considered, what is the percentage area of land use that is suitable for the weed?

Importance of the question

The potential area at risk of invasion will affect the overall risk a weed poses. Weeds that are able to invade larger areas are more of a risk.

Explanation and assumptions

This question looks at what percentage of a land use is at risk from the weed in question. This is the worst case scenario of where the weed will spread to if it was not controlled. This will depend on the climate and soil preferences of the weed. For example, some weeds may only be suited to higher rainfall areas of the geographic area being considered, or may only be a problem on certain soil types like red earths or heavy clay alkaline soils.

Differences within land use also need to be considered. For example, if an irrigated perennial horticulture land use is used then a weed may only be a problem in citrus rather than in vineyards, as is the case for Bridal creeper (Asparagus asparagoides).

The percentage area of land use suitable should be based on where the weed will grow at the density you have assumed in scoring the Impacts. This means that if you have assumed a high density when scoring the Impacts then ignore areas where the weed would only persist at a low density when determining Potential distribution.

Potential distribution may be described in various ways including the area of the land use at risk (ha) or the percentage area of the land use at risk (%). This question uses the percentage area of land use at risk.

Predictions should be conservative and only include areas where the weed will have a significant impact. If conditions are marginal for the weed then it will not reach a high enough density to have a significant impact.

It is important to ensure that only one taxon is covered by a name. For example, Blackberry (Rubus fruticosus) is a species aggregate and has a number of different species all covered by the one common and scientific name. If more than one species is covered then it is likely that inaccurate predictions on where the weed will grow will be made.

This question is best answered with topographic, land use and soil information maps for the geographic area being considered. There are several ways to obtain this information:

1. The most useful of these is the National Agricultural Monitoring System. This system allows the mapping of climate (as 'Climate Zones Detailed'), soils (as 'Soil Type') and land use (as 'Dryland land use' or 'Irrigated land use') across a range of scales including Local Government Areas (LGAs). Multiple LGA's can be selected initially, with each of the above factors mapped on individual maps. For example, all dryland land uses are mapped on one map only and this map needs to be compared to separate soil type and detailed climate zones (this can be done visually by saving the files, printing and comparing). Although coarse, the detail in the soil type and detail climatic zone maps is suitable for analysis for this question.
2. Climatic, soil and land use information may also be available via the Australian Natural Resource Atlas and the New South Wales Natural Resource Atlas. A number of information layers are under development on these sites.
3. Data and information may be obtained from NSW DPI. Alternatively NSW DPI can produce draft maps for validation at a local level.

Coarser scale climate matching of introduced species and where these species may potentially spread in Australia is available from Climatch (Crombie et al. 2008).

Mapping can be done either electronically using a Global Information System (GIS) such as ArcView and then matching climatic and soil tolerances using any one of a number of programs, for example CLIMEX or CLIMATE.

If using paper maps the following steps will help in estimating the percentage area of the land use in the geographic area being considered that is suitable for the weed.

1. Map the land use in your geographic area being considered. If a land use map is not available, shade areas on clear plastic laid over topographic maps.
2. Consider the climatic preferences to which the weed is suited, for example median annual rainfall, and mean maximum and minimum temperatures. Lay a sheet of plastic over the land use map and shade only the areas of land use which are suitable for the weed based on the climatic preferences.
3. Consider the soil preferences to which the weed is suited. Lay a sheet of plastic over the land use map and shade only the areas of land use which are suitable for the weed based on the soil preferences.
4. Match the climatic and soil preferences of the weed (overlays from steps 2 and 3) to the land use map (from step 1) and estimate the percentage of the land use that is suitable to the weed. This may be done more easily by dividing the land use map and the climatic and soil preferences up into small grids or boxes, counting the number of boxes that are suitable for the weed and then comparing this to the total number of boxes the land use covers. Answer as follows

Comparative weed risk score

The score for weed risk is calculated by adjusting the Invasiveness, Impacts and Potential distribution scores to range from 0 to 10, and then multiplying these. Weed risk will have a maximum of 1000 and a minimum of 0. The electronic form does this for you.

The higher the score, the greater the risk posed by the weed, that is the greater the invasiveness of the species, the negative impacts caused by the species and the potential area effected.

Splitting up these possible scores into bands of 20% gives cut-offs for categories of weed risk as follows:

The weed risk categories can then be combined with the feasibility of coordinated control categories to determine appropriate weed management priorities and principles.

Do not compare scores between land uses. Land uses differ in their intrinsic values and the relative value of each land use is very difficult to measure. Average weed risk scores may also be lower in some land uses compared to others, for example there may be lower scores in agricultural land uses because of the greater level of weed management undertaken. This would not necessarily mean that agricultural weeds are less important.

Why multiply the Invasiveness, Impacts and Potential distribution scores?

• Multiplying gives a greater spread in the scores than adding (i.e. range from 0-1000 compared to 0-30).
• Multiplying is logical as it recognises the interactions between the criteria. Say the Impacts of a weed can be measured in dollars per hectare per year, the Potential distribution is known in hectares, and the Invasiveness (rate of spread) is measured in terms of the increase in hectares compared to the previous year:

When multiplying, all the hectare units cancel so that weed importance is measured in total dollars per year. In multiplying the Invasiveness, Impacts and Potential distribution criteria scores, we are mimicking the above calculation, without having the actual dollar and hectare figures.

Weed risk uncertainty score

The uncertainty score for weed risk assessment is determined by calculating the percentage of 'do not know' answers that have been recorded in the Invasiveness, Impacts and Potential distribution sections. In the case of part questions, for example Invasiveness Questions 3-5 and Impacts Question 6 record the individual scores from each 'do not know' question in each part to determine the section uncertainty score. Do not combine the scores from each 'do not know' question to calculate a score for that question as was done to calculate the question score. See How to calculate scores in the Background Information book for an example. The electronic form (XLS, 241.5 KB) does this for you.

The resulting percentages will be used to determine the overall uncertainty score.

To calculate manually, adjust the section uncertainty score as follows:

The resulting percentages will be used to determine the overall uncertainty score.

Acknowledgements and References

Author: Dr Stephen Johnson
Industry and Investment New South Wales
ORANGE NSW 2800
September 2009

Other acknowledgements

References cited

Other information sources