Definitely in some situations. On the Cape Flats and other parts of the fynbos biome of South Africa a handful of woody shrubs from Australia, in particular Acacia cyclops, are outcompeting and excluding coastal fynbos, the most diverse flora in the world (when measured on scales of 10s of metres). Conversely, a coastal fynbos plant, Bitou, Chrysanthemoides monolifera, is doing the same to Wallum, another diverse coastal sand flora, in Eastern Australia.

Aquatic weeds such as Salvinia molesta and Water Hyacinth, Eichornia crassipes, proliferate over extensive areas of subtropical lakes and rivers in Australia and South Africa.

All of these weeds are transformers, they change the structure and ecological dynamics of the vegetation - including rates of flow, decomposition, seed dispersal, fire frequency and intensity. This affects the entire biota. Invasive species are definitely one of the Key threats to Biodiversity.

See the chapter in this superb free textbook:

http://www.mongabay.com/conservation-biology-for-all.html

Certainly in the Top End of Australia, particularly when you look at the invasive super-grasses like gamba grass (Andropogon gayanus) and annual mission grass (Cenchrus pedicellatus). These are changing the fire regimes of the areas that they invade by creating later, hotter fires that are completely changing the vegetation composition of the landscape, including a significant reduction in biodiversity. These "Key Threatening Processes" are of major concern to us in the tropics, particuarly when they were introduced into agriculture to "improve" pastures, which gave them an already wide distribution so that when they went feral, they were impossible to control.

Rob, would you say the same of Rhododendron ponticum in Britain?

I think it is reasonable to distinguish between weeds that have ecological equivalents and fit in, or have evolved within a particular community and reintroduced in some way, versus those that transform the vegetation dynamics to a new, more or less permanent state. In many cases, however, weeds are prolific because of other changes to ecological dynamics, such as with fire, grazing and bracken fern.

Thank you all for your concern. Some weed species in gradual time loose their identity as a weed like Elodea canadensis, Prunus serotinaas mentioned by Mr Rob. But we should not expect the same for harmful weed species like Parthenium hysterophorus.

In many cases, weeds, whether exotic or indigenous, spread due to imbalances in a system, so that they can quickly take advantage. Overgrazing, which prevents local plants from making seeds, is a common cause in India, so that plants inedible to cattle tend to take over pastures. The same may be said for the Chir Pine (Pinus longifolia) which is indigenous in the Himalaya, but has established a regime of forest fires to spread. The reason it is so successful is that broadleaf forests have been lopped for fodder, so ecosystem functions are at an all time low, enabling the Chir to spread. Parthenium does not grow in the shade. Where canopy has been destroyed, Parthenium moves in. Similarly, Lantana camara was considered a weed. However, it only moves into degraded areas, where it helps hold the soil until trees can germinate and grow a canopy, which is when it dies out.

If there are imbalances, something is going to move in to take advantage of it. perhaps we should be happy that something moves in, for over large parts of India, nothing moved in when the local vegetation was destroyed by cattle. Travel through parts of Gujarat, Karnataka, Tamil Nadu.. even Uttar Pradesh. Even parthenium would have bee welcome there to keep down the dust and to see something alive and not contorted.

Peter Smetacek raises a good point. You have to define what you mean by weeds. Some native species have weedy, ruderal growth forms or life histories but play an important ecological role as pioneers and facilitators of succession. Although weedy, they are generally out-competed over time.

Then there are the weeds so labeled because they are of little value to humans or undesirable in agriculture. These weeds can actually increase alpha diversity in what would otherwise be a crop monoculture. This actually benefits diversity, although perhaps at some economic cost.

Finally, you have the invasive weeds (both exotic and native) that are persistent and competitively dominant. These displace native communities, often in areas subjected other forms of human disturbance (land clearing, nutrient pollution, etc.) These are the ones that profit from "imbalances in the system" and may reduce diversity as a consequence.

Totally agree with Rebecca Rooney. In some cases the loss of biodiversity is apparent because some taxa (e.g. Ailanthus altissima in Catalonia, Spain) spread widely and are very difficult to eradicate. But in many other cases, such as weeds in croplands (e.g. Agrostemma githago), not only increase biodiversity, but also are in danger of extinction due to the abandonment of agricultural activity.

I think every case is different and we can not generalize as well have exposed the experts.

An invasive plant that is thinly spread through a landscape may raise local plant diversity but that in itself is not necessarily a good thing. Something else must be either displaced or in reduced numbers. If the plant is not closely related to a local species then it is likely to be of less benefit in terms of ecosystem function, or may encourage other species that are detrimental to the local ecology. Some plants have profound effects on fertility e.g. legumes and can cause a cascade of change over time. Even the argument (earlier comments) that Prunus serotina is no longer a weed in Europe seems simplistic. Even if it only displaces other cherry species it will have differences that will have flow on effects through the local ecosystems.

Raising local diversity by letting exotic species invade does nothing for global diversity, particularly if it puts pressure on endemic species.

Invasives that are abundant, must be displacing indigenous species and disrupting natural ecosystem processes. Yes there are still processes and it is just a shift but species and processes natural to that area will be negatively impacted. Some rare ecosystems such as frost hollows are particularly vulnerable to invasive species.

Protected areas can be managed for species, for ecosystems or for processes. Some argue that managing for continuance of rare species or keystone species will provide for the other values; some argue for an ecosystem focus, safeguard the ecosystem and all species will be provided for; others focus on ecosystem function which is a very human centred approach and is not particularly concerned with endemic diversity or even indigenous diversity, rather it seems to be focussed on ecosystem services, which puts human values first. Only for the latter is ‘ecological shifting’ or ‘integration’ of exotic invasive species an acceptable approach.

That is a good point Graeme. Activities such as agriculture, forestry, and other horticulture purposes also have resulted in human deliberately introducing plants to new habitat. The majority of Australia’s weeds species, for instance, were deliberately introduced for cultivation as food, horticultural and/or ornamental plants with about 96% still in cultivation.In some cases this increased competition from weeds can threaten native species with extinction. For example, competition from environmental weeds such as Bridal Creeper (Asparagus asparagoides) threatens many populations of Rice Flowers (Pimelea spicata) in the country. Weeds could therefore be threat to diversity in some cases.

Yes Michael

And then even if Asparagus doesn't completely displace Pimelea spicata it may reduce and fragment the population to where it disrupts cross pollination, seed dispersal and recruitment, in the long run the effect will be the same but could take decades. Each exotic species needs to be considered in the context of where it is and where it might spread to from where it is now as well as any changes that may happen to its behaviour with climate change.

Over the last couple of days there has been a discussion about loss of diversity by hybridisation in regard to ngaio, Myoporum laetum and an Australian ngaio on the NZ Plant Conservation Network forum http://www.nzpcn.org.nz/page.aspx?conservation_forum The thing here is that even botanists have trouble recognising the different spp but it looks like our ngaio will in time vanish and no one knows what consequences that may have for other species like the threatened ngaio weevil.

It is true that in some situations some weeds are harmful and some weeds are beneficial for ecosystem. Therefore is it possible to predict those weed species which will become either threat for diversity or beneficial for ecosystem in near future.

From the California perspective, EVERY single exotic plant that can become naturalized in a wild lands situation is a threat to the native ecosystem, and in no cases is a naturalized weed ever beneficial. There are at least two reasons for this, "spatial extinction" of the natives caused by the exotic, and that the exotic did not evolve with the biome to play a particular supporting role.

The "spatial extinction" issue, is the spot where you have an individual exotic plant growing, it has taken the place of a native plant, causing spatial extinction on that spot.; And in our case in California, it only takes a dozen weeds to cause spatial extinction of a whole grassland biome that consisted of over 100 species of native working together.

Each native plant growing on its own spot, is like a little battery, sending energy into the ecosystem, and each native plant is wired together with the surrounding plant and animal community. The exotics rewire the ecosystem for their sole purpose of using that space for themselves, so the animals that were dependent on the native plant growing on that spot usually lose out and in turn become spatially extinct.

The most successful exotic, is one that mimics or has an analog species in the native ecosystem. For example, we have summer blooming members of the sunflower family, and when those tarweeds were eradicated, then Yellow start thistle also a member of the sunflower family, easily moved into that vacant niche.

Any exotic species has the ability to be naturalized, whenever a plant is able to survive in a wild lands area without continual care from humans, that is a huge, huge threat to any ecosystem, and if possible eradicated and replaced with the native species that it displaced.

That process of conversion from exotic plants in wild lands back to original natives is very , very expensive to get started, and most governments, even of the wealthiest countries, have been unwilling so far to fund adequate money to get the process started on any scale.

For example for my project in Palo Alto, California at http://www.ecoseeds.com/arastradero.html, I am estimating that I will need $500,000 to do the first 1/10th hectare, mostly to get the native seed stocks commercially reproduced. Then to do the rest of the 30 hectares, about $2 million dollars, and that is only to get about 35 of the original 100 species back in place.

Fortunately, my test plot is indicating that if I get at least 35 of the 100 natives back in place, then that planting should be resistant to future weed invasions from the sides, because I will have all the major niches filled, for the various plant families.

In the Macaronesian islands (e.g. in the Canaries) we have very good examples of invasive species which invade very easily good preserved and important habitats with a lot of endemic plants and animals (mainly invertebrates). One of the most dangerous species is Pennisetum setaceum, an African grass that modify the habitat and the landscape. Its eradication is very difficult (if not impossible), and at this moment we would need a lot of money to maintain the populations controled, at least in protected areas. The situation in the central islands of the Canary archipelago (Gran Canaria and Tenerife) is simply catastrophic in many areas, mainly in lower and middle zones, as this plant tends to colonize mostly arid and semiarid sites. 

Surely exotic invasive species are threat to biodiversity. For instance Parthenium hysterphorus, Lantana camara, Prosopis juliflora, Eupatorium odoratum, Eupatorium riparium and Eupatorium adenophorum are posing threat to biodiversity in India. Herbaceous  Parthenium hysterphorus is responsible for depletion of biodiversity in Plains of India while shrubby  Lantana camara has become threat to biodiversity in hilly states of Uttarakhand and Himachal Pradesh, and the  forest of the Vindhyan region. Eupatorium riparium and Eupatorium adenophorum are emerging as a threat to biological diversity in North-eastern states of India. Eupatorium odoratum and Prosopis juliflora are responsible for depletion of biodiversity in Kerala and Gujarat states of India, respectively.