Eutrophic or euphytic?

A paper has just been published that should be required reading for anyone interested in the management of nutrients in in ecology.   It is a follow-up of a 2006 paper with the catchy title “How green is my river” that set out to provide a conceptual framework for how rivers responded to enrichment by nutrients.   That original paper contained several good ideas but, crucially, not all of them were underpinned by evidence.  A decade on, several of the predictions and statements made in that original paper have been tested, and the time has come to re-examine and modify that original conceptual model.

My reaction to the 2006 paper was that it was very interesting but not fully reflective of the rivers in my part of Britain, whose rougher topography produced quite different responses to nutrient enrichment than that proposed in their original model.   That criticism has been addressed in the revised version, which places greater emphasis on the physical habitat template, which means that it is more broadly applicable than the original version.   But that, in turn, got me wondering about the continued relevance of a term such as “eutrophication” to rivers.

People have been using the term “eutrophic” to describe lakes with high concentrations of nutrients since early in the 20th century.   As the century progressed, evidence of a causal relationship between inorganic nutrients and algal biomass, and the consequences for other components of lake ecosystems grew.   With this foundation, it has then become possible to predict the benefits of reducing nutrients and there are plenty of case studies, particularly from deep lakes, that demonstrate real improvements as nutrient concentrations have declined.

Attempts to apply the same rationale to rivers have, however, met with far less success.   Legislation to reduce nutrients in rivers has been in force in Europe since 1991 (the Urban Wastewater Treatment Directive, followed by the Water Framework Directive) and whilst this has led to reductions in concentrations of phosphorus in rivers (see  “The state of things, part 2”), there has, in most cases, not been a corresponding improvement in ecology.   There are a number of reasons for this but, at the heart, there was a failure to understand that the tight coupling between nutrients and biology that was the case in deep lakes did not also occur in running waters.   What was needed was recognition of fundamental differences between lakes and rivers, and “How green is my river?” and, now, this new paper have both contributed to this.

However, one consequence of recognising the importance of the physical habitat template alongside nutrients is to challenge the relevance of the term “eutrophic” when describing rivers.   “Eutrophic” literally means “well-nourished” so is appropriate in situations where high nutrients cause high plant or algal biomass.   This high biomass (strictly speaking, the primary production arising from this biomass) then creates problems for the rest of the ecosystem (night-time anoxia caused by plants consuming oxygen being a good example).   If high biomass can arise due to, let’s say, removal of bankside shade or alteration to the flow regime, perhaps (but not always) in combination with nutrients, then perhaps we need a term that does not imply a naïve cause-effect relationship with a single pressure?

My suggestion is to shift the focus from nutrients to plant growth by using the term “euphytic” (“too many plants”) as this would shift the emphasis from simply driving down nutrient concentrations (expensive and not always successful) towards reducing secondary effects.  It is possible that strategies such as planting more bankside trees, for example, or altering the flow regime or channel morphology (see “An embarrassment of riches …”) may be just as beneficial, in some cases, as reducing nutrient concentrations.   That said, we also have to bear in mind that nutrients may have an effect well downstream, so focus on amelioration of effects within a particular stream segment will never be a complete solution.

I should emphasise that a lot of work has been done in recent years to understand the concentrations of nutrients that should be expected in undisturbed conditions, and also to understand the nutrient concentrations that lead to changes in community structure in both macrophytes and algae.   These show that many rivers around Europe do have elevated concentrations of nutrients and I am not trying to side-step these issues.  I do, however, think it is important that regulators can prioritise those rivers in greatest need of remediation and, in most cases, they do this without considering the risk of secondary effects.

It is, largely, a matter of semantics.   I have been involved in many conversations over the past couple of decades about how to improve the state of our rivers.  Many of those have centred on the importance of reducing nutrient concentrations (which would be, indisputably, a major step towards healthier rivers).  But there is more to it than that.  And Mattie O’Hare and colleagues are helping to open up some new vistas in this paper.

Note: the photograph at the top of this post shows the River Wear at Wolsingham.  This stretch of the river captures many of the challenges facing river ecologists: nutrient concentrations are relatively low and there is good bankside shade.  However, the flow of the river is highly altered due to impoundments upstream and a major water transfer scheme.  How do all these factors interact to create the often prolific algal growths that can be seen here, particularly in winter and spring?

References

Hilton, J., O’Hare, M., Bowes, M.J. & Jones, J.I. (2006).  How green is my river?  A new paradigm of eutrophication in rivers.   Science of the Total Environment 365: 66-83.

O’Hare, M.T., Baattrup-Pedersen, A., Baumgarte, I., Freeman, A., Gunn, I.D.M., Lázár, A.N., Wade, A.J. & Bowes, M.J. (2018).  Responses of aquatic plants to eutrophication in rivers: a revised conceptual model.   Frontiers in Plant Science.   9: 451

Advertisements

That’s funny …

The most exciting phrase to hear in science, the one that heralds new discoveries, is not “Eureka!” but “That’s funny”
Attributed to Issac Asimov

I have visited Croasdale Beck, in western Cumbria, twenty-eight times since 2015 and I thought I was beginning to understand it’s character (see “A tale of two diatoms” and “What a difference a storm makes”).   It is the unruly sibling of the River Ehen which, usually, offers a far less amenable environment for freshwater algae.  Last week, however, as we walked down the track towards the stream, we were confronted with the unexpected sight of a river bed that was bright green.  Our measurements, too, showed that not only was there a lot of algae in absolute terms, but there was far more here than we had measured in the River Ehen.  Usually, the situation is reversed, with the Ehen having more than Croasdale Beck.

Croasdale Beck at NY 087 170 looking upstream in April 2018.   The position of the gravel bar has shifted over the time that we have visited, with the wetted channel originally being at the right hand side, rather than being split into two.

It was hard to capture the extent of the algae growing on the river bed in a photograph, but the macroscopic image below captures the colour of the growths well, and you’ll have to use your imagination to scale this up to cover half of the stream bed.  Under the microscope, these growths turned out to be virtual monocultures of the green alga Draparnaldia glomerata.  This is common in clean rivers in spring time, and I often find it in the nearby River Ehen (see “The River Ehen in February”).  What my images do not show is the mucilage that surrounds the filaments.   In some cases, the growths can be almost jelly-like, so prolific is this mucilage.   One of the roles of this mucilage plays is to serve a matrix within which enzymes released by the fine hairs at the end of the filaments can act to release nutrients bound into tiny organic particles (see “A day out in Weardale …”).

Growths of Draparnaldia glomerata in Croasdale Beck (NY 087 170) in April 2018.  The upper image shows the filaments growing on submerged stones and the lower image shows the bushy side-branches growing from a central filament.  Scale bar: 100 micrometres (= 1/10th of a millimetre).

We also sample a site a couple of kilometres downstream on Croasdale Beck and, here again, the river bed was smothered in green growths.  I assumed that this, too, was Draparnaldia glomerata but, when I examined the filaments under the microscope, it turned out to be a different alga altogether: Ulothrix zonata (see “Bollihope Bhavacakra” and links therein).   There is little difference between the two sites that might explain this: the latter is slightly lower and is surrounded by rough pasture whilst the other is closer to the fells.   However, I have seen both Ulothrix zonata and Draparnaldia glomerata at several other sites in the vicinity, and a simplistic interpretation based on agricultural enrichment does not really work.

There were also a few obvious differences in the diatoms that I saw in the two samples.   In both cases, we sampled stones lacking green algae but, instead, having a thick brown biofilm.  Several taxa were common to both sites – Odontidium mesodon, for example (broadly confirming the hypothesis in “A tale of two diatoms …”) and Meridion circulare was conspicuous in both.   However, the lower site had many more cells of “Ulnaria ulna” than the upper site.   Again, there is no ready explanation but, at the same time, neither green algae or diatoms at either site suggests anything malign.

Filaments of Ulothrix zonata at Croasdale Beck (NY 072 161).   The upper filament is in a healthy vegetative state (although the cell walls are not as thickened as in many populations).  The lower filament is producing zoospores.   Scale bar: 25 micrometres (= 1/40th of a millimetre).

Diatoms in Croasdale Beck, April 2018.   a. upper site: note the abundance of Odontidium mesodon, plus cells of Gomphonema cf exilissimum, Achnanthidium minutissimum and Meridion circulare; b. lower site: note the presence of “Ulnaria ulna” as well as several of the taxa found at the upper site.   Scale bar: 25 micrometres (= 1/40th of a millimetre).  

So where does this take us?  I talked about the benefits of repeat visits to the same site in “A brief history of time wasting …” and I think that these data from Croasdale are making a similar point.  By necessity, most formal assessments of the state of ecology are based on very limited data, from which, at best, we get an estimate of the “average” condition of a water body over a period of time.  Repeat visits might lead to a more precise assessment of the “average” state but also give us a better idea of the whole range of conditions that might be encountered.  Here, I suspect, we chanced upon one of the extremes of the distribution of conditions.   Cold, wet weather in early spring delayed the growth of many plants – aquatic and terrestrial – as well as the invertebrates that graze them.   Then the period of warm, dry conditions that preceded our visit gave the algae an opportunity to thrive whilst their grazers are still playing “catch-up”.  I suspect that next time we visit Croasdale Beck will have its familiar appearance.   It is, nonetheless, sobering to think that this single visit could have formed fifty-percent of the evidence on which a formal assessment might have been made.

 

A brief history of time-wasting …*

Having talked about diversity on a microscale in the previous post, I thought it would be interesting to place this in context by looking at the variations that I have observed in the River Wear at Wolsingham over the past decade or so.   The River Wear has seen some significant improvements in water quality over this period, but those have mainly affected sections of the river downstream from Wolsingham.  Most of the changes at Wolsingham are, therefore, giving us some insights into the range of natural variation that we should expect to see in a river.

I’ve got 31 samples from the River Wear at Wolsingham on my database, collected since 2005.  Over this period, nine different diatom species have dominated my counts: Achnanthidium minutissimum on 21 occasions, Nitzschia dissipata twice and Cocconeis euglypta, Encyonema silesiacum, Gomphonema calcifugum, Navicula lanceolata, Nitzshia archibaldii, N. paleacea and Reimeria sinuata once each.   I also have records for non-diatoms during 2009, during which time the green alga Ulothrix zonata, and two Cyanobacteria, Phormidium retzii and Homeothrix varians were the dominant alga on one occasion each.   In total, I have recorded 131 species of diatom from this one reach, although only I’ve only found 91 of them more than once, and only 59 have ever formed more than one percent of the total.   I’ve also got records of 22 species other than diatoms.

This – along with my comments in “The mystery of the alga that wasn’t there …” raises questions about just how effective a single sample is at capturing the diversity of algae present at a site.  .    In 2009 I collected a sample every month from Wolsingham and the graph below shows how the total number of species recorded increased over that period.   Typically, I find between 20 and 30 species in a single sample, and each subsequent month revealed a few that I had not seen in earlier samples.   Importantly, no single sample contained more than 40 per cent of the total diversity I observed over the course of the year.  Part of this high diversity is because of the greater effort invested but there is also a seasonal element, as I’ve already discussed.   The latter, in particular, means that we need to be very careful about making comments about alpha diversity of microalgae if we only have a single sample from a site.

Increase in the number of diatom taxa recorded in successive samples from the River Wear at Wolsingham.  In 2009 samples were collected monthly between January and December whilst in 2014 samples were collected quarterly. 

This seasonal pattern in the algal community also translates into variation in the Trophic Diatom Index, the measure we use to evaluate the condition of streams and rivers.  The trend is weak, for reasons that I have discussed in earlier posts, but it is there, nonetheless.   Not every river has such a seasonal trend and, in some cases, the community dynamics results in the opposite pattern: higher values in the summer and lower values in the winter.  It is, however, something that we have to keep in mind when evaluating ecological status.

Variation in the Trophic Diatom Index in the River Wear at Wolsingham between 2005 and 2015, with samples organised by month, from January (1) to December (12).   The blue line shows a LOESS regression and the grey band is the 95% confidence limits around this line.

All of these factors translate into uncertainty when evaluating ecological status.   In the case of the River Wear at Wolsingham, this is not particularly serious as most of the samples indicate “high status” and all are to the right of the key regulatory boundary of “good status”.  However, imagine if the histogram of EQRs was slid a little to the left, so that it straddled the good and moderate boundaries, and then put yourself in the position of the people who have to decide whether or not to make a water company invest a million pounds to improve the wastewater coming from one of their sewage treatment plants.

At this point, having a long-term perspective and knowing about the ecology of individual species may allow you to explain why an apparent dip into moderate status may not be a cause for concern.  Having a general sense of the ecology of the river – particularly those aspects not measured during formal status assessments – should help too.  It is quite common for the range of diatom results from a site to encompass an entire status class or more so the interpretative skills of the biologists play an important role in decision-making.   Unfortunately, if anything the trend is in the opposite direction: fewer samples being collected per site due to financial pressures, more automation in sample and data analysis leading to ecologists spending more time peering at spreadsheets than peering at stream beds.

I’ve never been in the invidious position of having to make hard decisions about how scarce public sector resources are used.  However, it does strike me that the time that ecologists used to spend in the field and laboratory, though deemed “inefficient” by middle managers trying to find cost savings, was the time that they learned to understand the rivers for which they were responsible.  The great irony is that, in a time when politicians trumpet the virtues of evidence-led policy, there is often barely enough ecological data being collected, and not enough time spent developing interpretative skills, for sensible decisions to be made.   Gathering ecological information takes time.   But if that leads to better decisions, then that is not time wasted …

Ecological Quality Ratio (EQR: observed TDI / expected TDI) of phytobenthos (diatoms) at the River Wear, Wolsingham) between 2005 and 2015.   Blue, green, orange and red lines show the positions of high, good, moderate and poor status class boundaries respectively.

* the title is borrowed from the late Janet Smith’s BBC Radio 4 comedy series

Our patchwork heritage* …

The problem with the case I set out for a “switch” from a winter / early spring biofilm community to a summer / autumn assemblage is the sample that I was writing about contained elements of both.   This, I think, is another aspect of an issue that I touched upon in “The River Wear in January”: that the scale that we work at is much greater than the scales at which the forces which shape biofilms operate.   There is no intrinsic driver for this switch beyond the physical forces in the river but each stone will have a slightly different history.  A smaller cobble will be more likely to be rolled than a boulder, as will one that is not sheltered from the main current, or not well bedded into the substratum.  The sample I collect is a composite from the upper surface of five separate cobbles so will blend these different histories.   The more stable stone might have more Navicula lanceolata and Gomphonema olivaceum whilst the recently rolled might be dominated by early colonisers such as Achnanthidium minutissimum.

The same processes can even work on a single stone.   Arlette Cazaubon, a French diatomist, now retired, wrote several papers on this topic (see references at the end of this post).  She highlighted how the diatom assemblages differed across the surface of a boulder, depending on the exposure to the current.  However, that is only part of the story.  The picture at the top of the post was taken in January, when I was collecting my first samples of the year.  You can see the streak where I ran my finger through the biofilm and some other marks, perhaps where the heel of my wader had scuffed the stone (I’m trying to keep my balance in the middle of a northern English river in January whilst holding a waterproof camera underwater, remember).   But such damage could have arisen just as easily from twigs or stones that were being washed downstream.   Taken together with Arlette’s work, it shows how a mature Navicula lanceolata / Gomphonema olivaceum assemblage can live alongside a pioneer Achnanthidium minutissimum assemblage.

A schematic view of the biofilm in the River Wear at Wolsingham, March 2018.   a. Navicula lanceolata; b. Gomphonema olivaceum complex; c. Fragilaria gracilis; d. Achnanthidium minutissimum.   Scale bar: 10 micrometres (= 1/100th of a millimetre).

I’ve tried to depict that in the schematic diagram above.   On the left-hand side there is a mature biofilm, with long-stalked Gomphonema species creating a matrix within which motile diatoms such as Navicula lanceolata live whilst, on the right, there is a pioneer community dominated by Achnanthidium minutissimum.   However, whilst this patchiness is a natural phenomenon, it can contribute to the variability we see in ecological data and, indirectly, to an impression that ecological data are not precise.   If I were to divide the diagram above into two halves, the left-hand side would return a higher TDI than the right.  This is because the diatoms on that side have broader ecological tolerances than those on the other (the sample size, by the way, is far too small to do this seriously but I just want to make a point).   In practice, however, the entire diagram represents little more than the width of a single bristle of the toothbrush that I use to collect samples so a sample is, inevitably, an amalgam of many different microhabitats on a stone.  Our assessment of the condition of the river represents the average of all the patches across the five stones that form a typical sample on that day.

The importance of patchiness in determining the structure and composition of stream communities has been known for some time (see review by Alan Hildrew and Paul Giller in the reference list).   What we have to remember when trying to understand phytobenthos is that patchiness is, to some extent, embedded in the samples we collect, rather than being something that our present sampling strategies might reveal.

* “… for we know our patchwork heritage is a strength not a weakness ..” Barack Obama: inaugural address, 2009

Reference

A useful review on patchiness in stream ecosytems (several other papers in this volume also discuss patchiness in freshwater and marine environments):

Hildrew, A.G. & Giller, P.S. (1994).  Patchiness, species interactions and disturbance in the stream benthos.  pp. 21-62.  In: Aquatic Ecology: Scale, Pattern and Process (edited by P.S. Giller, A.G. Hildrew & D.G. Rafaelli).   Blackwell Scientific Publications, Oxford.

Some of Arlette Cazaubon’s papers on variability in diatom assemblages across the surfaces of single stones:

Rolland, T., Fayolle, S., Cazaubon, A. & Pagnetti, S. (1997). Methodological approach to distribution of epilithic and drifting algae communities in a French subalpine river: inferences on water quality assessment. Aquatic Science 59: 57-73.

Cazaubon, A. & Loudiki, M. (1986). Microrépartition des algues épilithiques sur les cailloux d’un torrent Corse, le Rizzanese. Annals de Limnologie 22: 3-16.

Cazaubon, A. (1986). Role du courant sur la microdistribution des diatomées epilithiques dans une Riviere Méditerranéenne, L’Argens (Var, Provence). pp. 93-107.   Proceedings of the 9th Diatom Symposium.   Bristol.

Cazaubon, A. (1988). The significance of a sample in a natural lotic ecosystem: microdistribution of diatoms in the karstic Argens Spring, south-east France.  pp. 513-519.   In: Proceedings of the 10th Diatom Symposium, Joensuu, Finland.

The mystery of the alga that wasn’t there…

I was back at the River Wear at Wolsingham a few days ago for my second visit of the year (see “The River Wear in January” and “The curious life of biofilms” for accounts of the first visit).   I had wanted to go out earlier in the month but we’ve had a month of terrible weather that has translated into high river flows.  Even this trip was touch and go: the river was about 30 cm higher than usual and the gravel berm that usually stretches out under the bridge on the left bank was largely submerged.

Compare the image of the substratum with the one I took in January: that one had a thick film with a chocolate-brown surface whilst the March substratum had a much thinner film lacking any differentiation into two layers.  When I put a small sample of the biofilm under my microscope, I could see that it was dominated by diatoms with only a few strands of green algae.   Many of the diatoms that I saw in January were still here in March but Navicula lanceolata, which comprised over half the algal cells I saw in January was now just 15 per cent of the total whilst Achnanthidium minutissimum was up from about 15 per cent to about 40%.    However, as A. minutissimum is a much smaller cell, N. lanceolata still formed more of the total biovolume.   One other difference that I noticed as I peered down my microscope was that there was much less amorphous organic matter in the March sample compared with the one from January.

The substratum at the River Wear, Wolsingham on 24 March 2018.   The photograph at the top shows the view from the road bridge looking downstream.

When I looked back at notes I had taken after my visit in March 2009, I saw that the riverbed then had been covered with lush growths of the green alga Ulothrix zonata (you can see a photograph of this in “BollihopeBurn in close-up”).   I did not see this on my visit last week.  That might be because the high water level means that I could not explore as much of the river as I wanted, but it was more likely a consequence of the preceding conditions.   The graph below shows at least three separate high flow events during March, the first of which associated with the melting of the snow that fell during the “Beast from the East”.   I suspect that these high flow events would have both moved the smaller substrata (the ones I usually pick up to sample!) scouring away the biofilms in the process.

A view of the biofilm from the River Wear, Wolsingham in March 2018.

River levels at Stanhope, 20 km upstream from Wolsingham across March 2018 showing three separate high flow events.  A screenshot from www.gaugemap.co.uk.

The final graph shows the trend in the three algae that I’ve been talking about over the course of 2009, which is similar to what I am seeing in 2018 except that that the timing of the decline in Navicula lanceolata and Ulothrix zonata along with the increase in Achnanthidium minutissimum is slightly different.   In very broad terms N. lanceolata is typical of winter / early spring conditions, favoured by thick biofilms partly created by the matrix of stalks that Gomphonema olivaceum and relatives creates.   Achnanthidium minutissimum, on the other hand, is the most abundant alga through the summer and early autumn.  It is a species that thrives in disturbed conditions, such as we would expect after the weather we’ve experienced this March.   However, we must not forget that the grazing invertebrates that thrive

during the summer months also represent a type of disturbance.  Ulothrix zonata thrives in the late winter / early spring window (see “The intricate ecology of green slime”).   I would have expected it to have persisted beyond March but, as I said earlier in the post, I may have missed some as it was difficult to get a good impression of the whole reach due to high flows.

This moveable switch between a “winter” and “summer” state creates a problem when we are sampling for ecological status assessments.   The Environment Agency has, for as long as I have worked with them, had a “spring” sampling window that starts on 1 March and runs to the end of May.  As you can see, this straddles the period when there is a considerable shift in the composition of the flora.   I’ve always suggested that they wait as long as possible within this window to collect diatom samples to increase the chance of being past the switch.  However, with a huge network to cover in a short period, along with other logistical considerations, this was always easier said than done.   I’ve worked closely with the Environment Agency to manage as much of the variation in their diatom analyses as is possible (see “Reaching a half century …”); one of the mild ironies is that simply being a huge Behemoth of an organisation can, itself, be the source of some of the variation that we are trying to manage.

Trends in approximate biovolume of three common taxa discussed in this post in the River Wear at Wolsingham during 2009.  

Baffling biodiversity …

Few of the participants in the UK / Ireland diatom ring-test that I described in my previous post felt any need to thank me for my choice of slide for our 50th test.  The slide came from a spring in County Mayo, Ireland, which is part of the Agricultural Catchments Programme, a large study into the effect of farming on water quality. The sample itself came from the stems and leaves of the submerged water cress (Nasturtium officinale*) plants which fill the entire channel.  It was a real stinker, with a mess of Gomphonema forms, several of which did not neatly fit any species description that we could find.   A conservative reckoning is that there were at least eight different Gomphonema “species” and that raises a further question about what it was about this habitat that led to so much diversity within a single genus within a single sample.

First, a quick tour around some of the Gomphonema forms that we found.   There was general agreement that the most common type was close to G. micropus Kützing 1844 but not a perfect match to published descriptions (the stria density, in particular, was too low).   The situation was further complicated because the status of G. micropus was questioned at times, with it being treated as a variety of G. parvulum and placed in the G. angustatum complex by different authorities during the 20th century.  Then there were a number of valves with more rounded ends and a higher striae density than G. micropus but which, if you look closely, are not symmetrical around the long axis.   We thought that these were close to G. cymbelliclinum Reichardt & Lange-Bertalot 1999.   Unfortunately, there were also quite a lot of valves that had intermediate properties, making it hard, in many cases, to say whether it was one species or the other.

Gomphonema cf micropus from Cregduff spring, Co. Mayo, Ireland, September 2017.  Photographs: Bryan Kennedy.  Scale bar: 10 micrometres ( = 100th of a millimetre).  The image at the top of the post shows Cregduff spring (photo by Lauren Williams)

Gomphonema cf cymbelliclinum from Cregduff spring, Co. Mayo, Ireland, September 2017.  Photographs: Bryan Kennedy.  Scale bar: 10 micrometres ( = 100th of a millimetre).

We also found some valves that were close to descriptions of Gomphonema utae Lange-Bertalot & Reichardt 1999 and some that were close to G. parallelistriatum Lange-Bertalot & Reichardt 1991.  We also found representatives of the G. parvulum complex, G. tergestinum and G. subclavatum (more about this one in the next post).

Gomphonema cf utae from Cregduff spring, Co. Mayo, Ireland, September 2017.  Photographs: Bryan Kennedy.  Scale bar: 10 micrometres ( = 100th of a millimetre).

Gomphonema cf parallelistriatum from Cregduff spring, Co. Mayo, Ireland, September 2017.  Photographs: Bryan Kennedy.  Scale bar: 10 micrometres ( = 100th of a millimetre).

So what is going on here?   There are, I suspect, two key elements to the story that we need to explain.  The first is the limits of species within Gomphonema.  I’ve touched on this before (see “Diatoms and dinosaurs”) and some recent studies that combine morphological and molecular biological evidence also cast doubt on our ability to differentiate within this genus using classical approaches.   Whilst I was struggling to disentangle the species in this sample, I had a conversation with an eminent taxonomist and she hinted darkly that Gomphonema was “over-described”.  There is a readiness to “split” established taxa and describe new species that, in her opinion, runs ahead of the evidence.

The limitations of taxonomy cannot explain all of the variation that we observed in this sample, so the second question to ask is what it is about the conditions here that allow so many representatives of one genus to thrive.   I’ve touched on this subject before (see “Baffled by the benthos (1)” and “Baffled by the benthos(2)”).  In these posts I introduced G. Evelyn Hutchinson’s “paradox of the plankton” in which he suggested that environments that look uniform, to mortals six orders of magnitude larger than algae are, in fact, considerably more heterogeneous  and, so offer more opportunities for “variations on a theme” to thrive.   In the second post I went on to suggest that this type of diversity imparts resilience to an ecosystem and so should be looked upon as a positive feature of the ecosystem when doing ecological status assessments.

There is, however, one final possibility that, to my knowledge, has not yet been explored.  The presence of transitional forms in the diatom assemblage at Cregduff may be an artefact of our inability to differentiate biological species based on a limited range of morphological criteria on offer. However, it is also possible that we are looking at a situation where the Linnaean species are not reproductively isolated from one another, allowing hybridisation.   The concept of a “hybrid swarm” is well known in some other groups (e.g. orchids) but has never been formally demonstrated in diatoms.  However, the wide morphological diversity within a single genus in one sample alongo with, in some cases, apparent continua of variation, does raise questions about speciation within thi genus.

The final twist to this story is that, from the point of view of current ecological status assessments, all this diversity has little effect.  Though everyone grumbled about the difficulties in naming the Gomphonema species, the results, as the box-and-whisker plot in the previous post show – were less variable than in many of our other ring tests.  What I suspect happened is that the underlying taxonomic confusion means that many of these taxa have “mid-range” scores for the TDI (and other indices), so the final calculation cancels out the identification issues.  Bear in mind that this may not always be the case!

* I understand that this is the correct name now, rather than Rorippa nasturtium-aquaticum.  See Al-Shehbaz, A. & Price, R.A. (1998).  Delimitation of the genus Nasturtium (Brassicaceae).  Novon 8: 124-126.

References

The two papers that deal with variation within Gomphonema to which I refer are:

Abarca, N., Jahn, R., Zimmermann, J. & Enke, N. (2014).  Does the cosmopolitan diatom Gomphonema parvulum (Kützing) Kützing have a biogeography? PLOS One 9: 1-18.

Kermarrec, L., Bouchez, A., Rimet, F. & Humbert, J.-F. (2013).  First evidence of the existence of semi-cryptic species and of a phylogeographic structure in the Gomphonema parvulum (Kützing) Kützing complex (Bacillariophyta).  Protist 164: 686-705.

Reaching a half century …

Last week saw a small career achievement as I sent out the result of the 50th diatom ring-test that I organise for the diatom analysts in the UK and Ireland.  “Ring-test” is the informal term for an inter-laboratory comparison, when two or more laboratories analyse the same sample and compare their results.  We started out doing regular ring-tests in 2007 for all the people who were analysing diatom samples for assessments associated with the Water Framework Directive, sending out five slides each year to staff in the UK and Irish environment agencies and contractors who worked with them. Now, a decade later, the scheme is still going strong, with participants from Germany, Sweden and Estonia joining the British and Irish contingents.

There are a number of similar schemes around Europe with the same basic model: the organiser sends out copies of a slide made from the same sample, all participants then analyse the slide and send in their results, which the organiser collates.   There is usually one or more designated “expert” against whose results everyone else is judged.  Most of the other schemes then organise a workshop at which participants gather to discuss the finer points of diatom taxonomy.   We have had workshops in the past, but these are not directly linked to the ring-tests.  Instead, we send out a report that summarises results and provides notes on the identification of difficult or unusual taxa.   The money we save on workshops means that we can circulate more slides.  I’m a great believer in “little and often” for this type of quality control.

A second feature of our scheme (which some of the other European schemes have also now adopted) is to use a panel of experienced analysts to provide the benchmark that other participants should achieve.   This means that we have an idea of both the average result and the scale of the variation associated with this.  We learned early on that some samples gave much less variable results than others, even when the analyses were performed by experienced analysts.  We use this knowledge to adjust the size of the “target” that participants must achieve.   The graphs below show the results for our most recent test.  The horizontal blue lines on the left hand graph show two standard deviations around the mean of the “expert” analyses (expressed as TDI).  This is the “warning limit”; if an analyst exceeds this then he or she should be looking at their results to see if they have made any mistakes.  The red line is the “action limit”, seven TDI units either side of the expert mean.   We know from other studies (see lower graph, left) that it is very unlikely that two replicate analyses have a greater difference than this, so analysts who exceed this should definitely be checking their results.

The results of the 50th UK / Ireland diatom ring test showing (left) difference in TDI and (right) number of taxa (N taxa) between experts and other participants.  Blue lines: mean TDI ± two standard deviations of expert panel’s mean; red lines: mean TDI ± 7.   Note that it is unusual for the between-analyst variability to be quite as narrow as it was for this slide.

The reason why we need flexible “warning limits” is illustrated in the right hand graph below.   This shows the similarity between two counts as a function of the diversity of the samples.   The relationship has a wedge-shape (illustrated by the blue line – the regression line through the 90th percentile of the data).   There are a number of reasons why two analysts are unlikely to get identical results, one of which is that they disagree on the identities of the taxa that they encounter (the reason why we are doing the audits in the first place).  But what a wedge-shaped relationship is also telling us is that there seems to be an upper limit to the similarity that can be achieved at any given diversity.   This is an inherent stochastic quality of the data and has nothing to do with the competence of the analysts.

Left: some of the data from which the “action limit” for the ring-tests was established.   These are the results of audits of 67 samples from Northern Ireland in which the original (“primary”) analysis was checked against the result of an independent (“audit”) analysis.   Right: The effect of diversity on the similarity between primary and audit analyses for the same dataset.

A further way in which our scheme differs from others is that no-one “passes” or “fails”.  That might seem counter-intuitive as this is supposed to be a test of competency.   A regular reader of this blog, however, should understand that there absolute truth is often elusive when it comes to identifying diatoms and other algae.  The hard objectivity needed for a real test of competency always has to be moderated by the recognition of the limitations of our craft.   Moreover, turning this exercise into a calibration exercise runs the risk of turning the analysts into machines.  Rather, we use the term “reflective learning”, encouraging participants to use the reports to judge their own performance relative to the experts, and to take their own corrective action.

Some of the organisations whose analysts participate use the ring-test as part of their own quality control systems, and will take corrective action if results stray across the action limit.  That seems to be a sensible compromise: quality control should be the responsibility of individual laboratories, rather than delegated out to third parties.   At the same time, organisations need to understand that the people who perform ecological analyses are professionals, not treated as if they are one more machine in a laboratory that needs to be calibrated.

 

If you are interested in joining the UK / Diatom ring test scheme, or just want to learn a little more about it, get in touch with me and I’ll do my best to answer your questions.

Reference

Kelly, M.G. (2013).  Building capacity for ecological assessment using diatoms in UK rivers.  Journal of Ecology and the Environment 36: 89-94.