Is plant diversity on the skids? The scales of biodiversity loss


Imagine you want to monitor changes in plant diversity, to see whether diversity is increasing or decreasing over time. What spatial scale would you study?

Would you search for changes at the global scale, or at continental, regional or local scales? More importantly, would you expect to see the same trend or different trends if you examined changes at many scales?

At the global scale, biological diversity on Planet Earth is on the skids. Thousands of species face extinction and over a thousand hectares of forest are cleared every hour. Species diversity is falling because species are going extinct faster than new species can evolve. And we’re causing it.

At the continental scale, we see the opposite pattern. In Australia and other continents, total plant diversity is increasing rather than decreasing, because we’ve introduced more species than we’ve made extinct. There are over 19,000 species of vascular plants in Australia. Since European colonization, over 60 species have gone extinct and nearly 1,200 species are now threatened with extinction. In the same period, about 2,000 species have been introduced from elsewhere in the world and are now naturalized. These naturalized species can reproduce in the wild without human assistance, and are now part of the Australian flora.

The same pattern usually occurs at the regional scale. In every region of Australia, the total number of plant species has increased in recent centuries and decades because more species have been introduced than have gone extinct. We may not want these exotic species but we can’t ignore them, as all plants influence how ecosystems function, not just native species.

Small-scale changes

If the diversity of plants is decreasing at the global scale, and increasing at continental and regional scales, then what is happening at the small scale? Is diversity consistently increasing or decreasing at small scales? A decade ago, ecologists Dov Sax and Steven Gaines (2003) asked this question in a great paper:

Within anthropogenic environments, such as parking lots, housing developments and agricultural fields, diversity of species… has clearly decreased dramatically from conditions before human disturbance. Within local systems that are more or less intact, however, net changes in diversity are not as well understood. Although ‘intact’ local systems are perhaps the most studied of all ecological systems, ironically it is at these small scales that we have the poorest conception of how diversity has changed. (Sax & Gaines 2003, p. 563)

It’s important to understand how plant diversity is changing at small scales, as the diversity and composition of plants in small areas affects processes like nutrient cycling and biomass production. If diversity changes greatly at small scales, then ecosystem processes could also change.

Venn alps quadrat
Surveying plant diversity in a quadrat in the alps (Photo source: Susanna Venn).

Plant ecologists often survey vegetation at small scales, in quadrats or plots. Imagine that someone sampled thousands of quadrats 50 or more years ago and the quadrats were all re-surveyed recently. Do you think that species diversity at the quadrat scale would have consistently increased or decreased over the 50 year period? Remember to include exotic species when you think about the question. When averaged across all of the quadrats, would species richness have risen, fallen or stayed constant over the last 50 years or so?

Now broaden your horizons and imagine that thousands of quadrats were surveyed across the planet. When averaged across all of the quadrats, would species richness have risen, fallen or stayed constant over time? Is there a global trend in changes in species diversity when measured at the small, quadrat scale?

A global analysis

Fortunately a new paper answers this question for us. An international team led by Mark Vellend documented changes in plant diversity at the quadrat scale by analyzing studies from across the globe. Using meta-analyses, they studied every available paper that reported changes in plant diversity or richness in re-surveyed plots. The median plot size was 44 m2 (e.g. a rectangular quadrat, 4 m x 11 m in size) and most analyzed plots were less than 1,000 m2 (about 30 x 30 m square). The studies ranged from 5 years to over 100 years in duration, and included areas disturbed by humans, such as urban areas and grazed pastures, and natural areas disturbed by hurricanes, fires and other processes. In total, their analysis was based on 148 papers containing 346 data-sets and over 16,000 plots. That’s a whopping big data-set.

Their first, big question was: has species richness and diversity (they analyzed both attributes) at the quadrat scale increased or decreased over time? They then divided the data into groups to see whether patterns varied across continents and vegetation types, and how disturbance, pollution and other processes influenced changes in diversity. What do you think they found?

Plant diversity has increased at regional scales, but what happens at small scales?
Is plant diversity consistently increasing or decreasing at small spatial scales?

The take-home message may or may not surprise you. Contrary to patterns at global, continental and regional scales, plant diversity has not consistently increased or decreased over time at the quadrat scale. Species richness increased in some studies and declined in others. These changes balanced out, and the net effect was no change in species richness (or diversity and evenness) across the 16,000 plots.

This pattern occurred in all habitat types (forests, grasslands, wetlands, etc.) and across most continents, as can be seen in the chart below. Plant diversity is declining at the global scale, increasing at continental and regional scales, but remaining stable at the quadrat scale.

Vellend 2013 global richness
Change over time in plant species richness at the small, quadrat scale. See text for explanation of symbols. Source: Fig. 3 in Vellend et al. (2013).

This chart from the Vellend paper summarizes their findings. How do you interpret the colored circles and lines? The circles show how much species richness changed over time on average. If circles are close to the black zero line, then little change occurred. The colored horizontal lines (50% and 95% credible intervals) show how variable the changes were among all of the studies. A longer horizontal line means that changes in richness were more variable for a given number of studies. Circles and lines to the right of the black zero line indicate that species richness increased over time, while those to the left indicate that species richness declined. Lines that cross the zero line indicate that species richness increased over time in some studies and decreased in others.

As the chart shows, different patterns emerged when the dataset was divided into groups. The strongest responses were for ‘post-disturbance’ studies, where species richness increased over time, and for ‘invasion’ studies, where species richness usually declined, often greatly. In the authors’ words (Vellend et al. 2013, p. 2):

Consistent with intuition, marked increases in species richness over time were found in studies in which authors attributed vegetation change to succession following major disturbances such as fire, severe storms, or logging… and to a lesser extent, to the cessation of grazing.

By contrast, declines in species richness were often attributed to invasive species and, to a smaller extent, to climate change. Surprisingly, no other processes consistently caused species richness to decline.

The finding about invasive species needs to be interpreted carefully. Some invasive species – the ‘transformative species’ – can reduce plant diversity by altering ecosystem processes or out-competing existing plants. The big declines in species richness due to ‘invasion’ were caused by transformative exotics. Most exotic species don’t have such big impacts, and exotic invasions often increase species richness in quadrats, as they do across regions and continents. There’s obviously no logical reason why an entire species will go extinct when a single exotic plant colonizes a quadrat.

Comparing scales

How do we reconcile the conclusion that plant diversity has remained stable at small scales, with the knowledge that thousands of hectares of forest are cleared for agriculture every day? Surely clearing and cropping reduce plant diversity? In part, this inconsistency reflects biases in the underlying data. Few ecologists study whether plant diversity drops when rainforests are cleared for soybeans as the outcome is blatantly obvious. Consequently, few quadrats from cleared areas were included in the analysis. Vellend’s analysis wasn’t restricted to pristine areas (it includes quadrats in urban areas and pastures) but it doesn’t document changes due to the destruction of natural ecosystems. Instead, it provides the best analysis we have of global trends in small scale diversity, in ecosystems that haven’t been completely transformed by humans.

The authors concluded:

…in the absence of wholesale habitat conversion (e.g., turning a tropical rainforest into a parking lot or a crop monoculture), local-scale plant diversity has not, on average, declined over the last century… nor do the data suggest any reason to predict the future will be fundamentally different from the past. (Vellend et al. 2013, p. 3)

As the ecologist Chris Thomas states in an interesting commentary article, this global study is extremely valuable because it helps us to more precisely understand how and where plant diversity is changing:

The biodiversity crisis has not gone away, but we definitely need to be considerably more precise in identifying which elements of biodiversity are in decline, where, whether and why such declines are concerning, and what we should and can do about it. (Thomas 2013, p. 2)

So what’s happening in your patch of bush? Is plant diversity increasing or decreasing? Is diversity changing at the patch scale or the plot scale, or both? Which of the two bothers you the most? And which of the two is easiest to fix?


Update (April 2014)

A more recent study in the journal Science, by Maria Dornelas and colleagues, found similar findings to the study by Mark Vellend and colleagues. They analyzed long-term datasets for many species groups (not just plants) and found no evidence of consistent declines in species richness (or diversity) at the plot scale. However, they found that species composition changed greatly over time. They concluded:

The absence of systematic change in temporal α [alpha] diversity we report here is not a cause for complacency, but rather highlights the need to address changes in assemblage composition, which have been widespread over at least the past 40 years…. There is a need to expand the focus of research and planning from biodiversity loss to biodiversity change.

Reference: Dornelas, M., Gotelli, N.J., McGill, B., Shimadzu, H., Moyes, F., Sievers, C. & Magurran, A.E. (2014) Assemblage time series reveal biodiversity change but not systematic loss. Science 344(6181), 296-299.


Ecologists use the term ‘species diversity’ and ‘species richness’ to refer to different things at different scales. At small scales, species richness (or species density) refers to the number of species in a defined area regardless of the abundance of each species. If ten species occur in a square meter then species richness is 10 species per square meter, no matter how many plants there are of each species. By contrast, the term species diversity incorporates species richness and the relative abundance (or evenness) of each species. If all ten species are equally abundant then species diversity is greater than if one species is abundant and the other nine are rare. At large scales, the two terms are used inter-changeably to refer to the number of species. Thus, Australia is a ‘mega-diverse’ continent of high plant diversity. I’ve used both terms in this post. The Vellend paper analyzed richness for all plots and diversity and evenness for a subset of plots, and found similar trends for all three.


Many thanks to Susanna Venn for allowing me to copy the photo of quadrat sampling in the Victorian alps from her great blog site.


Sax DF, Gaines SD (2003) Species diversity: from global decreases to local increases. Trends in Ecology & Evolution 18(11), 561-566. [The full paper can be freely downloaded from this link].
Thomas CD (2013) Local diversity stays about the same, regional diversity increases, and global diversity declines. Proceedings of the National Academy of Sciences (early view). [The full paper can be freely downloaded from this link].
Vellend M, Baeten L, Myers-Smith IH, Elmendorf SC, Beauséjour R, Brown CD, De Frenne P, Verheyen K, Wipf S (2013) Global meta-analysis reveals no net change in local-scale plant biodiversity over time. Proceedings of the National Academy of Sciences (early view). [The full paper can be freely downloaded from this link].

22 thoughts

  1. As someone who measured alpine quadrats with CSIRO 50 years ago I found this fascinating. But as someone who has worried about invasives and watch each year, like now, as yellow box grassy woodland understories turn to St Johns Wort, Pattersons Curse, ALG, Serrated Tussock, Fleabane etc etc etc I wish I knew a prescription for preservation against what seems like a tsunami of invasives!

    1. Hello Max, you’re certainly facing a suite of horror-show weeds there. We don’t have many 50 years old data sets in Australia unfortunately. I wonder how your old alpine quadrats have changed over half a century, and whether they’ve been re-surveyed since? Hopefully some of the plots have improved in quality rather than degraded over the period. It would be fascinating to know how they have fared. Best wishes Ian

      1. The data sets were actually those of the wonderful Alec Costin and I was a mere helper, but Alec said to me a few years back “the gene jockeys” got rid of the ecologists at CSIRO P.I a long time ago so who would know? Australia, with a few notable exceptions like Lindenmayer and Dickman have very few long term ecological study sites (ie with a history of around 30+ years) oh for a Rothamstead or two in Australia.

      2. I should have added Ian that walking over those sites over the last 50 years, not measuring them, there does appear to have been a re-thickening of the vegetation and certainly many/most of the species appear to be natives, post the removal of the cattle, only the damn horses to contend with now.

    1. Hello again Max, thanks for replying. I’ve got a vague memory that some of the old Alec Costin plots have been re-surveyed relatively recently. I’ll make a few inquiries and see if I can shed some light on the matter. Thanks again, and best wishes, Ian

  2. Hi Max – you’ll be pleased to know that Alec Costin’s and Dane Wimbush’s plots are still being monitored in one way or another. This has been important given the drought and fire that has occurred in the Australian Alps in the last 20 years. The most substantial piece of work was by Pascal Scherrer from Griffith University who undertook his PhD in the early to mid 2000s, using Alec’s plots to (in part) assess long-term vegetation change. He has produced one scientific paper (Scherrer & Pickering 2005 Arctic, Antarctic and Alpine Research 37, 574-584) and I’ve reproduced his thesis abstract below. I’m guessing you would have been involved in monitoring using point quadrats or line intercepts (both favourite methods of Alec). Pascal used photo-quadrats to interpret vegetation cover changes. It’s worth chasing up the thesis – link provided at the bottom. Best wishes, JOHN

    Monitoring vegetation change in the Kosciuszko alpine zone, Australia
    Pascal Scherrer, Griffith University

    This thesis examined vegetation change over the last 43 years in Australia’s largest contiguous alpine area, the Kosciuszko alpine zone in south-eastern Australia. Using historical and current data about the state of the most common vegetation community, tall alpine herbfield, this thesis addressed the questions: (1) what were the patterns of change at the species/genera and life form levels during this time period; (2) what were the patterns of recovery, if recovery occurred, from anthropogenic disturbances such as livestock grazing or trampling by tourists; (3) what impacts did natural disturbances such as drought have on the vegetation and how does it compare to anthropogenic disturbances; and (4) What are the benefits, limitations and management considerations when using long-term data for assessing vegetation changes at the species/genera, life form and community levels?

    The Kosciuszko alpine zone has important economic, cultural and ecological values. It is of great scientific and biological importance, maintaining an assemblage of vegetation communities found nowhere else in the world. It is one of the few alpine regions in the world with deep loamy soils, and contains endemic flora and fauna and some of the few periglacial and glacial features in Australia. The area also forms the core of the Australian mainland’s most important water catchment and is a popular tourist destination, offering a range of recreational opportunities.

    The vegetation of the Kosciuszko alpine zone is recovering from impacts of livestock grazing and is increasingly exposed to pressures from tourism and anthropogenic climate change. At the same time, natural disturbances such as drought and fire can influence the distribution, composition and diversity of plants. Thus, there is a need for detailed environmental data on this area in order to: (1) better understand ecological relationships; (2) understand existing and potential effects of recreational and management pressures on the region; (3) provide data against which future changes can be assessed; and (4) provide better information on many features of this area, including vegetation, for interpretation, education and management. The research in this thesis utilised three types of ecological information: (1) scientific long-term datasets; (2) photographic records; and (3) a comparison of disturbed and undisturbed vegetation.

    This research analysed data from one of the longest ongoing monitoring programs in the Australian Alps established by Alec Costin and Dane Wimbush in 1959. Permanent plots (6 transects and 30 photoquadrats) were established at two locations that differed in the time since grazing and have been repeatedly surveyed. Plots near Mt Kosciuszko had not been grazed for 15 years and had nearly complete vegetation cover in 1959, while plots near Mt Gungartan showed extensive impacts of grazing and associated activities which only ceased in 1958.

    Some transect data from 1959 to 1978 have been analysed by the original researchers. The research presented in this thesis extends this monitoring program with data from additional surveys in 1990, 1999 and 2002 and applies current methods of statistical evaluation, such as ordination techniques, to the whole data set for the first time. Results indicated that the recovery from livestock grazing and the effects of drought have been the main factors affecting vegetation. Recovery from livestock grazing at the three transects at Gungartan was slow and involved: (1) increasing genera diversity; (2) increasing vegetation cover; (3) decreasing amounts of bare ground; and (4) a directional change over time in species composition. Patterns of colonisation and species succession were also documented. In 2002, 44 years after the cessation of grazing, transects near Mt Gungartan had similar vegetation cover and genera diversity to the transects near Mt Kosciuszko, but cover by exposed rock remained higher. A drought in the 1960s resulted in a temporary increase of litter and a shift in the proportional cover of life forms, as grasses died and herb cover increased at both locations. Proportions of cover for life forms reverted to pre-drought levels within a few years. The results also highlighted the spatial variability of tall alpine herbfield.

    The photoquadrats were surveyed in the years 1959, 1964, 1968, 1978 and 2001 and are analysed for the first time in this thesis. After comparing a range of methods, visual assessment using a 130 point grid was found to be the most suitable technique to measure vegetation cover and genera diversity. At the 18 quadrats near Mt Gungartan, there was a pattern of increasing vegetation cover as bare areas were colonised by native cudweeds and the naturalized herb Acetosellavulgaris. Revegetation from within bare areas largely occurred by herb species, while graminoids and shrub species predominately colonised bare ground by lateral expansion from the edges, eventually replacing the colonising herbs. At the 12 quadrats near Mt Kosciuszko, vegetation cover was almost complete in all years surveyed except 1968, which was at the end of a six year drought. Similar to the results from the transect study, the drought caused an increase in litter at both locations as graminoid cover declined. Initially herb cover increased, potentially as a result of decreased competition from the graminoids and a nutrient spike from decaying litter, but as the drought became more severe, herb cover also declined. Graminoid cover rapidly recovered after the drought, reaching pre-drought levels by 1978, and was at similar levels in 2001. Herb cover continued to decline after peaking in 1964. The photoquadrat study also documented the longevity and growth rates of several species indicating that many taxa may persist for several decades. It further provided insights into replacement patterns amongst life forms.

    In addition to assessing vegetation change following livestock grazing and drought at the long-term plots, recovery from tourism impacts was examined by comparing vegetation and soils on a closed walking track, with that of adjacent undisturbed tall alpine herbfield at a series of 22 paired quadrats. Fifteen years after the track was closed there was limited success in restoration. Over a quarter of the closed track was still bare ground with non-native species the dominant vegetation. Plant species composition differed and vegetation height, soil nutrients and soil moisture were lower on the track which had a higher compaction level than adjacent natural vegetation.

    The results presented in this thesis highlight that tall alpine herbfield is characterised by nearly entire vegetation cover which is dominated by graminoids, followed by herbs and shrubs in the absence of disturbance by livestock grazing, trampling or drought. The studies also showed that under “average” conditions, the relative cover of herbs and graminoids remained fairly stable even though there can be considerable cycling between them. Spatial variability in terms of taxa composition was high. The only common introduced species in unrehabilitated sites was Acetosella vulgaris, which was effective at colonising bare ground but was eventually replaced by other native species. However, in areas actively rehabilitated, such as on the closed track, non-native species introduced during revegetation efforts still persist with high cover 15 years after their introduction.

    Monitoring of vegetation change is also important at the landscape scale. This thesis provides a review of the potential use, the limitations and the benefits of aerial photography to examine vegetation change in the Kosciuszko alpine zone. Numerous aerial photography runs have been flown over the area since the 1930s for government agencies, industry and the military. Some of these records have been used to map vegetation communities and eroding areas at a point in time. Other studies compared different types and scales of photographs, highlighting in particular the benefits and potential of large scale colour aerial photography to map alpine vegetation. However, despite their potential to assess vegetation change over time, a temporal comparison of vegetation in the Kosciuszko alpine zone from aerial photographs has not been completed to this date. Historical photographs may not be easy to locate or access and difficulties with vegetation classification may restrict the practicality of using historical aerial photographs to assess vegetation change. Despite these issues, aerial photography may provide a very useful and efficient tool to assess changes over time when applied appropriately, even in alpine environments. The development of digital classification techniques, the application of statistical measures of error to both methodology and data, and the application of geographic information systems are likely to further improve the practicality of historical aerial photographs for the detection of vegetation change and assist in overcoming some of the limitations.

    The results presented in this thesis highlight the need for limiting disturbance, for ongoing rehabilitation of disturbed areas and for long-term monitoring in the Kosciuszko alpine zone. The results contribute to our understanding of how vegetation may change in the future and may be affected by new land use activities and climate change. This type of information, which otherwise would require the establishment of long-term studies and years of monitoring, can assist land managers of this and other important protected areas. The study highlights how the use and expansion of already existing datasets to gather ecological information can save considerable money and time, providing valuable data for current and emerging issues.

    Thesis available for download from ADTP at

      1. Hi Ian, Max and John,
        You might be interested to know, that several of the transects measured by Costin, Wimbush and later Scherrer are being re-visited regularly by Gen Wright (NSW DECC) and Catherine Pickering (Griffith Uni). Also, many of Alec’s other scientific plots, from measuring ice-shear on rocks are still being monitored (occasionally) by Ken Green (NSW NPWS). Some of the metal pins out behind Mt Twynam are still there (the ones that are too thick to shear off). There really isn’t anything that Costin didn’t investigate, and of what’s left, Ken and others including myself, are trying to understand.

    1. Thanks John that is terrific to read and unfortunately shows that a life outside of academia can lead to missing out on current work. I put in time with Alec, Dane and Max Gray doing casual field work as did, I think, Jim Peacock back in those early collections.

      1. It’s great to see how blogs can help to connect the dots across both space and time – half a century in this case. Thanks again John & Max for your great comments. Best wishes Ian

  3. Thanks for another very interesting post, Ian. A couple of weeks ago I had a bright young thing who works for a local conservation group looking at the remnants of indigenous plants on my property near Castlemaine. He looked at the weed infested paddock across the road and lamented the lack of biodiversity. I didn’t have the heart to tell him that the paddock is undoubtedly much more diverse now than it was prior to European colonisation because of its spectacular diversity of exotic grasses and other weeds. I also lacked the heart to mention how the exotic willows, blackberry and almost shoulder high grasses that line the creek in the paddock are the only things that stop the creek turning into a nasty erosion gully.

    As much as I enjoy growing indigenous plants and playing my own little role in fending off their extinction, Nature is ruthlessly pragmatic and couldn’t care less about such sentiments.

    1. Hi Mel, well, I can certainly sympathize with him. I wonder how people will deal with this issue in 50 years time? Very differently to the way we deal with it I suspect. But to a large extent, future attitudes will depend on the state of the world we leave them, so let’s hope we can save as many native ecosystems as possible, despite the ingress of exotics. Thanks again, best wishes Ian

  4. The notice of your latest post (Is biodiversity on the skids) reached me just after seeing a segment on Landline (ABC 16 Nov 2013) about developments on Flinders Island – in particular a property called Markarna Park and its new owner’s activities:

    “FIONA BREEN: His dreams for the 12,000 hectare property are huge. He has a team of people working on a bush conversion project. Hectares are being transformed into grazing land. There’s another team working on an irrigation project to support those expanding grazing lands.
    GEORGE ADAMS: We’re about 90 per cent through that construction program of drainage and canals. Within the next three years we’ll be able to flood irrigate some 2,500 hectares here to enable us to double the size of our breeding herds.
    FIONA BREEN: A huge drain is being built on the property. When finished, it’ll be 3.5 kilometres long. It will eventually divert water out of the surrounding hills to a new irrigation scheme.” See the full transcript here

    This puff piece and the accompanying vision of giant machines scraping the surface off the land and reshaping landforms sent shivers down my spine and left me feeling quite depressed. I wonder whether anybody knows or cares what native vegetation might have existed there?

    Re Mel’s comment, now I’m well into my second half century I reflect on the immense changes in landcover I’ve seen in my life and I’m sure that most people will only feel the changes or loss noticed within their lifespan – but that’s a game of diminishing returns as political and administrative controls are mostly based on populist sentiments. I’m doing my best to help biodiversity on my plot but what happens after me, when anything goes in Tasmania.

    Meanwhile I am enjoying reading “White Beech” by Germaine Greer who is restoring a South Qld rainforest property so more power to people like her, as well as dedicated rationalist scientists who go on collecting and compiling the data.

    1. Hi Gwenda, thanks for your comment. Sounds great 😦 I haven’t read Germaine Greer’s book yet but have heard a bit about it, I might try to read it over the break. I gather it’s got a good dose of optimism to counter the doom and gloom. Best wishes Ian

  5. Before anybody else gets in, I did just find the Markarna Park referal form under the EPBC Act to DEH, dated 2006. Obviously the works were approved legally – so does that make it all right then?

  6. Another point that might be worth making is that although Australia may have lost few plant species* , we have certainly lost an incredible number of provenances and many of those that are left could drop off the perch at any time. An example of the latter is a very wide leaved form of Dianella perfragrans that I have found in very small numbers on roadsides about my locality, including one population (now extinct) that was just 50 metres away from my front gate.

    * The proviso being that we’ll never know how many species were lost in the massive clearing of land in the megadiverse SW of WA

    1. Hello Mel, good point. The gradual loss of native species from plot and regional scales is what ultimately drives the loss of diversity at global scales. At regional and national scales, we get more new exotics than the number of native species that we lose, so total diversity increases. But the loss of native species from regions and, if it continues to continents, drives the global decline in diversity. Hence, it will always be critically important to Think Global, Act Local. Best wishes Ian

What Do You Think? Please Leave a Reply.

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s