module 5

Weed invasions

Objectives

This module is designed to introduce the concept of weed invasions and the factors that are important to the invasion process. At the end of this topic you will:

Module Outline


Introduction

(Notes based on: Scott, JK 2000, ‘Weed invasion, distribution and succession’, in Sindel, BM (ed.) Australian Weed Management Systems.RG and FJ Richardson, Melbourne, Victoria, pp. 19-38.)

Plant communities are subject to continual invasion and change which create opportunities for weed establishment. An understanding of the ecological processes that allow weed invasions is essential for weed management. Successful weeds possess traits which ensure their survival and reproduction in the invaded area.

The origin of weeds

Invasions begin when an introduced species is able to survive and give rise to subsequent generations (i.e. it becomes naturalised). About 30% of introduced weed species in southern Australia originate from Europe and the Mediterranean region. Many northern Australian weeds originate from tropical countries such as the Americas.

The pattern has changed in recent times, with an increasing number of naturalised plant species in Australia now coming from Asia and other parts of Africa.

Naturalisation

It is estimated that 10% of introduced species become naturalised, with a small percentage of these becoming serious weeds. It is also estimated that, on average, Australia is acquiring at least 6 or 7 new species of naturalised plants each year. Ornamental and agricultural species account for many of these weeds.

About 65% of recent naturalisations were originally deliberately introduced as ornamentals and 7% for agricultural purposes.

Native species

Native species can also become invasive. For example, Australia’s floral emblem, golden wattle (Acacia pycnantha), and other Acacia species from eastern Australia have escaped cultivation and become weeds in Western Australia.

There are many other examples: West Australian bluebell creeper (Sollya heterophylla) is a weed in Victoria; sweet pittosporum (Pittosporum undulatum) has extended its range from eastern Australia to other areas; Queensland umbrella tree (Schefflera actinophylla), from northern Queensland, has become a weed in southern Queensland.

The invasion process

Communities at risk

Communities at risk from the invasion of weeds tend to have a low level of plant cover, be subject to frequent disturbance and are close to sources of potential invaders. Examples of communities at risk include agricultural areas, riparian zones, and nature reserves in urban areas.

Agricultural areas are easily invaded by weeds as they are regularly disturbed (by cultivation, herbicides, and harvesting). This disturbance often creates bare areas, and roadsides and neighbouring paddocks provide a ready source of potential invaders.

Riparian areas can be subject to disturbance from flooding, are frequented by animals (particularly in times of drought), and invading species are easily spread downstream by the flow of water.

Nature reserves in urban areas are at greater risk because they are usually surrounded by large numbers of potential immigrants (garden ornamentals), have higher numbers of ‘visitors’, and are prone to dumping of unwanted garden waste. For example: almost 40% of the plants present in Kings Park, a nature reserve in the centre of Perth, are exotic.

Invasiveness

All species must possess the ability to survive in their native habitat otherwise they would become extinct. Therefore all plants have the traits necessary for invasion in at least one environment. However, there are no specific traits that are uniquely associated with invasiveness (i.e. there are no hard and fast rules).

Hence it is difficult to predict with any certainty which species will invade native or agricultural ecosystems. For this reason ‘weed assessment risk’ is not always successful. However, the weed history of the species overseas may be a useful guide (i.e. has it already become naturalised and weedy in other countries?).

Invasion windows

An ‘invasion window’ occurs when the right conditions for invasion are apparent and the weed is present. It may occur because of a combination of factors including climatic suitability, degree of disturbance and biotic influences.

For example: having three consecutive years of above average rainfall following a long drought (which has depleted the seed bank of competitive native species and created gaps in the vegetation).

A weed is more likely to establish in Australia where climatic conditions match the original distribution of the species. Because of this, climate matching computer software (e.g. Climex) is sometimes used to predict whether a species is likely to invade and where it is most likely to be a problem. However, invasions are not necessarily limited to areas with matching climates – because other pressures on the weed may not be as high.

Invasion phases

The area invaded by a weed often appears to increase slowly at first and then seems to expand quite quickly, at which time it is recognised as a problem.

There are two possible explanations for this. Firstly, it may be a two-phased pattern of expansion with an initial period of slow spread called the ‘lag phase’. Species that are thought to be in this phase of spread are sometimes called ‘sleeper weeds’. One of the explanations for this is that long distance dispersal is related to population density (i.e. significant long distance dispersal only occurs when a certain population density is reached).

Secondly, the expansion may be exponential (i.e. with a constant rate of increase) and may only be perceived to be otherwise. The initial slow rate of spread may just be due to perception (i.e. it is not noticed until it has a measurable negative impact) or, where data has been collected, there is bias in the method that the data was collected.

Disturbance promotes invasion

Disturbance is any change caused to an ecosystem due to external factors (e.g. from floods, fire, grazing, or human interference). European settlement of Australia has led to a large increase in the frequency and intensity of disturbance through activities such as clearing, agriculture, mining, urbanisation and increased grazing.

Increased frequency and intensity of disturbance enlarges the invasion window and increases the chance of invasion. Combinations of disturbance can also enhance invasion. Disturbance can facilitate invasion by creating microsites for germination and increasing the chances of weed seeds arriving at the site. For example, a road corridor built through native vegetation disturbs the soil, removes the existing vegetation, increases run-off and provides the mechanism for spreading the weed seeds (i.e. via vehicles and animals).

However, some disturbance is usually part of the natural process of native ecosystems and the natural disturbance regime should be maintained as far as possible. Removal of certain types of natural disturbance may actually facilitate invasion. For example: the removal of natural fires from the vegetation may promote the invasion of fire prone species (e.g. Monterey pine has invaded the ACT after the removal of fires from eucalypt forests and; the native coastal wattle has invaded sub-coastal areas after the removal of fires from the nearby grassy heathlands).

Susceptibility to invasion due to disturbance can vary in different plant communities.

Dispersal in the new area

Kinds of dispersal

A weed’s ability to disperse to new areas influences the rate of invasion. There are two categories of dispersal: diffusion, which is a gradual localised spreading and jump dispersal, when weed populations jump distance to establish in new areas away from the parent.

In the first category spread is by seed or clonal growth while in the second category weed populations establish at a distance from the parent population, aided by human activity.

Both forms of seed dispersal are affected by barriers (i.e. deliberate or natural). Quarantine control on borders is an example of a deliberate barrier. The Nullarbor Plain is an example of a natural barrier preventing dispersal between the western and eastern halves of Australia.

Mode of dispersal

The main method of dispersal in most species is by seed, although plant fragments and specialised plant parts (e.g. bulbils, tubers or corms) are important in some species.

Seed dispersal has two phases: Phase 1, where seed disperses from the parent plant to a surface such as soil or another plant (e.g. parasitic plants such as mistletoes) and Phase 2, where seed redistributes from the soil surface to another location (horizontal movement) or incorporates into the seed bank (vertical movement).

In Phase 1, agents such as animals, wind or water may be involved (for example fruit-eating animals help spread seeds; many invasive species have fleshy fruits to facilitate this).

In Phase 2, agents such as wind, rain, water and gravity may be involved. Seed harvesting animals may also play an important role (e.g. ants).

Weed distribution

Distribution range

Most plant species are found over a small geographic area in their native range. However, a few species, including many weeds, have very large distribution ranges.

In Australia, introduced weeds are often found in a particular climatic region or associated with a particular soil type. For example: yarrow (Achillea millefolium) is usually associated with cooler temperate regions, and is particularly common in alpine areas; gamba grass (Andropogon gayanus) is only naturalised in tropical pastures; boneseed (Chrysanthemoides monilifera subsp. monilifera) is a weed of sandy soils; and alligator weed (Alternanthera philoxeroides) is associated with moist habitats.

Differing land use patterns, and varying management regimes, can lead to very heterogeneous distributions of weed species. For example, a landscape may contain recently cultivated cropping areas, land recently established with pastures, and permanent pastures (areas that may have been intensively managed with herbicides, fertilisers and irrigation) – and next to these areas may be roadsides and native habitats with varying degrees of disturbance and land management. This mosaic of management is also placed on a mosaic of different soil types, geography, and vegetation history.

Distribution patterns may also be highly localised. For example, in furrowed cultivated fields a weed may even be distributed either at the top, side or bottom of the furrow. Very small changes in elevation can affect the distribution of Trifolium legumes within a pasture.

Factors limiting distribution

Potential distribution in a species’ native habitat (i.e. its theoretical maximum spread) is limited by climate (e.g. temperature and moisture). Competing species, diseases, and pests generally prevent a species reaching its potential distribution and also limit its range of habitats. Therefore the result of combined biotic and abiotic factors produces the observed distribution in its native habitat.

Since introduced species have left behind biotic factors, they may be able to occupy a wider range of habitats/climate regions. For example: Bridal Creeper (Asparagus asparagoides) occupies temperate regions of the Southern Cape province in South Africa, but in Australia is found in a wider range of climates including Mediterranean South and Western Australia. In South Africa very closely related species occupy the Mediterranean climates, which may be preventing its establishment in these regions.

Natural enemies (i.e. pests and diseases) may also not have been introduced with the weed into Australia, thereby allowing it to survive in a wider range of habitats. This is one of the theories underpinning classical biological control.

Predicting distribution

Predicting the potential spread of a weed is important if the weed is to be managed effectively. Potential distribution of introduced weed species can be predicted using computer software.

This often involves matching climate types in new areas with the original distribution, taking into account the species’ temperature and moisture requirements. CLIMEX and BIOCLIM are computer models which assume that climate factors limit a species’

distribution, not taking into account other factors. GIS approaches, which take both biotic and abiotic information into account, will be important for future predictions.

Succession

The changing plant community

Succession is the process of change in a plant community. It can be defined as ‘The gradual and orderly process of plant ecosystem development brought about through changes in the composition of the plant community’.

Succession occurs when open sites become available for colonisation, and species are differentially available at a site, or species have different capacities for surviving at a site and for interacting with one another.

Types of succession

Primary succession occurs on areas of bare ground, often newly created, where no plants are already present. Examples of sites that may undergo primary succession include: areas that have been subject to lava flows or large landslides, coastal areas where large amounts of sand have been deposited, etc.

Plants initially colonise the soil (i.e. pioneer species) and are replaced by other plants over time. There is often a recognisable sequence over time, with smaller shorter-lived plants being replaced by larger longer-lived species (e.g. from grasses to shrubs

and then trees). Succession can take place over a long period of time. It can be difficult to recognise when a plant community is at the end of the cycle.

Secondary succession occurs when an established plant community is exposed to some form of disturbance. Cultivation and fire are common disturbances of agricultural and native habitats, respectively, that may cause secondary succession to occur.

The original vegetation is replaced by different species that occupy the disturbed area. For example: disturbance from cultivation may advantage species that have evolved to grow in similar situations of natural soil disturbance, thereby increasing their prominence in agro-ecosystems undergoing regular cultivation.

Models of succession

There are three models that explain the process of succession – facilitation, tolerance and inhibition. Combinations or variations of these may occur in plant communities.

Facilitation is when species present in a habitat early in succession modify the environment, making it more suitable for other species than themselves. In this model particular species, or species with particular attributes, are required for the succession to proceed on to its climax.

Tolerance is when a species begins the succession, but these are replaced by more competitive species or species that are more tolerant as resources become limited. In this model the initial species are not required, and are later replaced by more tolerant species. They may have better dispersal abilities and simply occupy a site until the more competitive and tolerant species arrive.

Inhibition is where the development of species composition as a site depends on which plants get there first, and whether they are able to suppress or exclude later successional species. In this model succession proceeds from short-lived to long-lived species, but particular early species may modify the environment in a way that inhibits late successional species, and may therefore prevent the climax community from developing.

Weeds and succession

In primary succession weeds often become part of the successional process and can be present during any stage, though individual species may be confined to a particular stage. Many weeds have very good dispersal abilities and are therefore often among the first plants to arrive in a new habitat, while others have seeds with long-lived dormancy and may already be present in the soil. Such species (i.e. ruderal species) will usually form a prominent part of the primary succession. However, species with traits such as shade tolerance or resistance to grazing may become prominent during the later stages of succession.

Many weeds are well adapted to disturbance, but they can react differently to particular forms of disturbance. Fire is an important driving force in the succession of native plant communities, and weeds in these communities can react differently to fire. For example: Rubbervine (Cryptostegia grandiflora) plants are usually killed or badly damaged by fire, but Chinee apple (Ziziphus mauritiana) plants usually survive and re-sprout.

Other influences on weed distribution

Biotic influences, such as grazing, can open the way for invasion or decrease the abundance of a weed. An example of this is when increased grazing pressure from rabbits and kangaroos is thought to be the main cause of the invasion of Wilson’s Promontory National Park by coastal tea tree (Leptospermum laevagatum).

Another example is when native herbivores have been shown to reduce the cover of bridal creeper (Asparagus asparagoides) in the nature reserve on Garden Island in WA.

Reserve sizes may influence the extent of invasion as small reserves are at greater risk from edge effects and nearby weed sources such as farms or gardens. For example, it has been shown that where rainforests are fragmented weeds colonise the borders while the native rainforest species are most abundant in the interior parts of the fragment. The border size can increase with the level of disturbance.

icon
Reading
  • "Weed invasion, distribution and succession’ (pages 19 to 35) in Australian Weed Management Systems.

  • Hobbs, RJ & Huenneke, LF 1992, ‘Disturbance, diversity and invasion: implications for conservation’, Conservation Biology, vol. 6, pp. 324-337.

    This review discusses types of disturbance and how they allow for invasion of a native ecosystem. The authors make the point that disturbance may be a natural part of an ecosystem’s dynamics, but it is a change in the disturbance regime, or the type of disturbance that can lead to invasion by weeds. They also point out that the natural disturbance regime is unlikely to persist in small reserves, due to fragmentation of the ecosystem and human intervention.

Summary

Plant communities are subject to continual disturbance, invasion and change. Many weed species are able to rapidly invade when conditions are suitable. Succession, along with its associated disturbance, plays a poorly-understood role in weed invasion. A greater understanding, particularly in Australian pastures, is needed to manage weeds and maintain desired plant communities.

Further reading

Gleadow, RM and Ashton, DH 1981, ‘Invasion by Pittosporum undulatum of the forests of central Victoria. I. Invasion patterns and plant morphology’, Australian Journal of Botany vol. 29, pp. 705-20.

Groves, RH 1986 Plant invasions in Australia: an overview’, in: Groves, RH and Burdon, JJ (eds.) Ecology of Biological Invasions: An Australian Perspective, Australian Academy of Science, Canberra, ACT, pp. 137-149.

Hobbs, RJ and Humphries, S 1994, ‘An integrated approach to the ecology and management of plant invasions’, Conservation Biology vol. 9, pp. 761-70.

Humphries, SE, Groves, RH and Mitchell, DS 1993, ‘Plant invasions: homogenizing Australian ecosystems’, in Moritz, C and Kikawa, J (eds.) Conservation Biology in Australia and Oceania, Surrey Beatty and Sons, Sydney, Australia, pp. 149-170.

Rozefelds, ACF, Cave, L, Morris, DI and Buchanan, AM 1999, ‘The weed invasion in Tasmania since 1970’, Australian Journal of Botany, vol. 47, pp. 23-48.