In the footsteps of Sherlock Holmes …

There is a long standing tradition that those of us with an interest – professional or amateur – in diatoms gather for a weekend at the end of October to share our enthusiasm.   The tradition includes free time on the Saturday afternoon to collect and examine diatoms which, itself, necessitates a rural location. For some inexplicable reason, this weekend has to coincide with the autumn half-term break so that, just as the privations of Ramadan make the feast of Eid more enjoyable, so sitting in a traffic jam on a motorway en route to this rural location intensifies the pleasure of a weekend looking at and talking about diatoms.

And so it came to pass that the British diatomists gathered at Baskerville Hall Hotel in the Wye Valley, each bringing fresh tales of deadlock on the M6, M5, M4, M25 …   If the name “Baskerville Hall” sounds familiar, then, yes, it is “Baskerville” as in “Hound of the …”.   The legend is that Arthur Conon Doyle was a family friend of the Baskerville family and had the idea for the story on the nearby Hergest Ridge (also the inspiration for an album by Mike Oldfield).   He changed the location to Dartmoor, apparently, to keep the tourists away.   Judging by the rather faded décor of the hotel, that seems to have been a spectacularly successful decision on his part.

I’m not really complaining. British Diatom Meetings are usually located in field study centres with accommodation in draughty dormitories, with lowest common denominator catering and appalling coffee.   This year, at Baskerville Hall, I was able to lie in bed and watch Match of the Day. Twice.   This was necessary because, as a scientist, I believe in replication, so I needed to watch the Sunday morning repeat to make sure that West Ham beating Manchester City was not a dream.


Baskerville Hall Hotel, near Hay-on-Wye, Powys, location of the 2014 British Diatom Meeting.

Thinking about ecology in a location with strong links with Sherlock Holmes was itself instructive. Maria, my co-worker on the studies on the River Ehen, which I have written about here many times (most recently in “Pearl mussels with some unexpected bedfellows…”) gave a talk on our work and it did make me think about the art of deduction in the context of environmental studies. Like Sherlock Holmes, we don’t get involved in problems until someone has realised that there is a problem, which means that we rarely have as much baseline data as we would like. In the case of the River Ehen, we are trying to unpick the reasons for the large quantities of algae that are present in the river, yet our only evidence that the quantities here are greater than they were in the past is the observations of a colleague who visited in the 1980s, though he made no actual measurements at this time. And so it goes on. We have chemical records from nearby locations, for example, though the types of measurements, their detection limits and frequency aren’t ideal for our purposes (nonetheless, we still have a more extensive network of monitoring stations in the UK than in many parts of Europe).   Slowly, as we trawl through the data, patterns emerge but they are rarely as crystal clear as we would like. Our conclusions are always hedged with caveats and there never seems to be enough evidence to unmask a villain with a dramatic flourish in the final paragraph. Were Sherlock Holmes an environmental scientist, I doubt he would ever have felt confident enough to say “elementary, my dear Watson”. Unfortunately.

Why are ecologists so obsessed with “monitoring”?

This is my 200th post on my blog and I am going to celebrate with a rant about a subject that continues to needle me.

A few years ago I searched the Newcastle University library catalogue for all books that had “monitor” or “monitoring” in the title. I then grouped the 433 books that fulfilled this criterion into their disciplines and made a bar chart of the results.   35 per cent of these books, the largest proportion in any discipline by far, were from the environmental sciences, followed by engineering and health (both just under 20%), then disciplines such as economics, sociology and education, each contributing less than ten per cent to the total. Then I sat back and wondered why the environmental sciences are so obsessed with “monitoring” compared to other disciplines.


Books with “monitoring” in the title in Newcastle University library by discipline. Based on a search of the catalogue performed in 2008.

Over the past month I have watched as my youngest son has come to terms with a diagnosis of diabetes. When it was first diagnosed he spent a few days in hospital while his condition was stabilised and so that the specialist nursing staff could teach him how to monitor his blood sugar and adjust his insulin levels.   He has continued monitoring his blood sugar subsequently and the consultant and nurses have used these data to adjust the parameters that he uses to calculate the quantity of insulin that he requires.   “Monitoring”, in other words, is an integral part of the management of diabetes.   It is so integral that when I search Web of Science for papers on diabetes, I find that there are about 2000 with both “diabetes” and “monitoring” in the title compared with almost 11000 with “diabetes” and “management”. My theory is that many of these papers on management of diabetes will, by necessity, discuss monitoring and that, conversely, there is little need to write papers on “monitoring” alone as this is so much embedded within the management of the condition.

By contrast, environmental science is a much younger discipline, and academic and practical aspects are still not as tightly dovetailed as they need to be. “Monitoring” offers academic scientists an opportunity to convert conceptual ideas into potentially useful data but we still often lack strong enforcement regimes and budgets to turn theory into practice. So the “monitoring” has become uncoupled from the overall management regime.

The second part of my gripe is to ask why the term “biomonitoring” has crept into usage in the academic literature in recent years.   Why bolt a Greek prefix onto a word derived from Latin? What does “biomonitoring” tell us about the process that is not already obvious from the context?  Very little.   Personally, I prefer the term “ecological assessment” as it is a truer statement of what is actually taking place.   “Monitoring”, derived from a Latin word meaning “warning” implies a temporal dimension whereas many of the studies in the academic literature that describe themselves as “biomonitoring” are, essentially spatial studies often with limited temporal replication.  The word does seem to have been around for some time: the oldest reference that I found on Web of Science was a 1974 paper by A.S. Pronin in Meditsinskaia tekhnika volume 3, pages 57-60. Interestingly, the paper is in a Slavic language (I don’t know which) and “biomonitoring” appears in the translation of the title. I had suspected that “biomonitoring” might be derived from a language that makes extensive use of compound words, the process perhaps accelerated in recent years by the large number of non-native speakers who now edit scientific journals and this discovery appears to support this hypothesis. “Biomonitoring” is, in my opinion, a completely unnecessary word in English. Actually, most of the papers with “biomonitoring” in the title are also completely unnecessary, but I’ll save that rant for my 300th post.

The Clear Mirror


My microscopical investigations of Pangong Tsu a couple of weeks ago (see “Diatoms from the Roof of the World”) set me of on a hunt for an intriguing book.   In this post I mentioned that the eminent British-born limnologist G. Evelyn Hutchinson had visited the lake in the 1930s as part of the Yale North India Expedition. In addition to the scientific outputs from this expedition, he also wrote a travel book, The Clear Mirror. It is now out of print but I managed to track down a secondhand copy from a German book dealer via Abebooks.   It is an intriguing book, divided into three parts: a preliminary account of the art and architecture of catholic churches and institutions in the then Portuguese enclave of Goa, followed by a description of Buddhist monuments and ceremonies in and around Leh. Finally, he moves on from Leh to Pangong Tsu and surrounding lakes, and gives an interesting description of the ecology of the lake and of the vegetation of the surrounding areas.

The book captures Hutchinson’s extraordinary breadth of interest. The prose style is, to be frank, somewhat dated and my version had no illustrations to liven the experience, but I could not but be impressed by a writer who is prepared to tackle the art and symbolism of two different religions and the ecology of terrestrial and aquatic habitats in a single book. It is not the first time I have stumbled on Hutchinson during my travels: whilst working on the vegetation history of peninsula Italy during the late 1980s, I stumbled across a volume he had edited on the history of Lago di Monterosi , a small lake just north of Rome. Once again, the work managed to juxtapose ecology with disciplines as diverse as archaeology and classics.

Hutchinson’s breadth of interest did not come at the expense of depth in his own specialist research, because he was a very notable limnologist in his own right, the author of a very significant textbook, A Treatise on Limnology, and responsible for many significant advances, not least both proposing and partially resolving the paradox of the plankton (explaining the great diversity of plankton species in the face of the competitive exclusion principle).   They don’t make them like him any more.


Hutchinson, G.E. (1970). Ianula: an account of the history and development of Lago di Monterosi, Latium, Italy. Transactions of the American Philosophical Society 60: 1-178.

The state of things, part 2

If the last post presented a fairly optimistic picture of the quality of Britain’s rivers over the past 30 years, this one is more of a reality check, highlighting some areas where our rivers are still suffering from pollution which means that they are unlikely to meet EU targets.

The first pollutant I will consider in this post is phosphorus.   The data for this and other inorganic nutrients is not summarised by regions but by the predominant type of agriculture in the catchment, reflecting the importance of run-off from land as a source of these pollutants.   However, much phosphorus enters our rivers via sewage works and, as we saw for BOD and other pollutants associated with organic pollution, the trend over the past twenty years or so is generally downwards. Note that the start of the downward trend does not start until the mid 1990s.   This is because reducing phosphorus in effluents only became a legal requirement when the EU’s Urban Wastewater Treatment Directive came into force. The horizontal lines, once again, represent the UK standards for dissolved phosphorus in rivers.   There is a story behind these standards, in which I played a minor role.


Trends in average inorganic phosphorus (PO4-P) concentrations in rivers draining three different land types in Great Britain. Note that the vertical axis is on a logarithmic scale.   The horizontal lines show recent UK standards required to support different classes of ecological status (see text for more details).

You’ll see that the trend lines for the two lowland land types are approaching but have not decisively crossed the line which indicates “good status”.   Presence of a certain phosphorus concentration does not, itself, determine status but these indicate guide values that ecologists believe should be attained if the river is to support a healthy ecosystem. I was involved in setting the original standards and our team originally set values that were much lower than those that were eventually adopted.   We had worked out concentrations that were associated with locations where we had found healthy ecosystems.   However, when these figures made their way through the bureaucracies of our regulatory organisations, they encountered strong resistance from those charged with actually achieving the reductions.   You’ll see that average concentrations have more than halved during the first decade of the 21st century – a considerable achievement. To be told that these concentrations were not low enough was a bitter pill to swallow. To cut a long story short, statistical black arts were performed behind the scenes (not by us, I hasten to add) to justify a higher standard, eventually set at 0.12 milligrams per Litre.

The cavalier fashion in which this original phosphorus standard was set meant that, before long had passed, weaknesses started to become apparent.   Not least of which was that rivers that should, on paper, have had conditions suited to good ecological status continued to have plant and algal communities that were not characteristic of good status.   Eventually, the phosphorus standards were revisited and, in most cases, made more stringent, though not without opposition from the water companies who would bear much of the burden of meeting these.   These new standards would pull each of the horizontal lines on my graph lower, and so make the overall position look yet more pessimistic.

The trend for one other important plant nutrient in GB rivers is equally pessimistic. Concentrations of nitrogen as nitrate have shown virtually no reductions at all over the period that the HMS has been operating.   There are, I think, two reasons for this. The first is that ecologists generally stress the importance of phosphorus over nitrogen as the key nutrient in freshwaters, which means that nitrogen concentrations have received less attention than, perhaps, they deserve. The second reason is more political: the biggest source of nitrates in our rivers is agriculture and DEFRA has to balance the interests of the environment with that of the vocal and politically-powerful farming lobby.   That there is no ecologically-based standard for nitrates in UK rivers tells its own story.   We could easily have produced a standard for nitrates at the same time as we revised the phosphorus standards (the information we needed was in the same spreadsheet) but were warned off. There is, in theory, a standard for nitrates in freshwaters, produced to meet the requirements of the EU’s Nitrates Directive but this was designed to protect human health, not ecology, and is set at a very high concentration. Rather than use this, I have plotted the standards used in the Republic of Ireland on my charts and, assuming that an equivalent UK standard would be of a similar magnitude, these show that the nitrate concentration of lowland rivers is generally much higher than these values.   This is, of course, a very broad-brush picture but it gives us a rough idea of what is going on.


Trends in average inorganic nitrogen (as nitrate) concentrations in rivers draining three different land types in Great Britain. There are no current UK standards for nitrate-N concentrations required to support different classes of ecological status; those plotted here are for the Republic of Ireland.

Looking back at the two posts, the picture that emerges is of reasonably good regulation of those types of pollutants that have been the traditional focus of regulation.   Both the water companies and the regulators understand the processes need to reduce organic pollution levels in sewage effluents, and the necessity for doing this.   The last twenty years or so have seen a change in focus towards inorganic nutrients and here we run into problems. First of all, the benefits for the public are often not immediately clear which, in turn, makes increasing water bills to pay for the expensive process of phosphorus removal more difficult. I have heard staff from water companies raise this argument several times when arguing against tighter regulation of phosphorus.   But the second problem is that regulation of nutrients needs to embrace diffuse inputs from agriculture as well as point sources.   This presents a huge problem as there are many more farms than sewage works to be visited, many of which are struggling to survive in a tough economic climate.   Consequently, there is little political will to drive the process from the top.   These are, indeed, challenging times for freshwater ecologists.

The state of things, part 1

I found a very useful dataset online that gives a good overview of long-term trends in the condition of British rivers. This is the fruits of the Harmonised Monitoring Scheme (HMS), a programme that has ensured the collection of comparable data from locations all over Great Britain (but not, it seems, from Northern Ireland).   The earliest data are from 1980, before the formation of the National Rivers Authority (predecessor to the Environment Agency), when water quality monitoring was the responsibility of ten separate Regional Water Authorities in England and Wales, so collection of comparable data would have required an enormous amount of effort, and this dataset represents a major achievement.

I am going to divide my discussion about this dataset across two posts: the first dealing with the good news and the second considering some aspects where the regulators have made some progress but perhaps not enough.   I am also going to relate the results from the HMS to the current UK environmental standards for each variable, as these give us a snapshot of what ecologists regard as acceptable concentrations for each pollutant.   The precise value of each standard varies depending on the type of river under consideration and, to make life easier, I have used the most lenient of the values for any particular standard.

The first variable I looked at was Biochemical Oxygen Demand (BOD) which is a good indicator of the level of organic (i.e. sewage) pollution in a river.   This measures the amount of oxygen needed by the various bugs involved in breaking the organic matter down to its constituent parts.   As I live in north-east England, I thought that it would be interesting to compare the average values for Great Britain as a whole with the averages for this region. These statistics are, however, not especially illuminating: although there is a downward trend, all values are below the threshold required to support high ecological status.   In this case, the “average” is pulled down by the large number of streams in rural areas with low population densities.   The maximum values recorded in NE England are, on the other hand, very interesting. Note the way that values first increase (suggesting a gradual deterioration in water quality), then decrease. And note the point at which this change starts: it is in the late 1980s.   The significant date here is 1989, when the then Conservative government split the regulatory functions of the old Water Authorities away from their duties to treat sewage.   Previously, the Water Authorities were both “poacher” and “gamekeeper”; from 1989 the National Rivers Authority had no conflicting interests and the effect is visible on this graph. In 1995, the NRA became the Environment Agency, which continues to be responsible in England. The overall impression here is that even the most polluted streams in NE England generally have BOD levels that should be of little concern to ecologists.


Trends in average Biochemical Oxygen Demand (BOD) in Great Britain (closed circles) and north-east England (open circles) between 1980 and 2012. The open squares show the maximum values of BOD recorded in NE England during the same period and the horizontal lines show the current UK standards required to support different classes of ecological status.

If the bugs responsible for breaking down organic matter in our rivers are not using so much oxygen then there should, in theory, be more in our rivers for fish and other wildlife to use.   This is borne out by the evidence, though the trends are not as strong as for BOD. Note that in this case we are most interested in the minimum concentrations available. There are UK standards for dissolved oxygen but these are expressed as percent saturation, rather than concentrations, so we cannot compare the HMS data with these.   One reason for the weak trend shown here is that the Environment Agency takes monthly “spot” measurements in the rivers it monitors and several factors including time of day can also affect the amount of oxygen present.   A river that is visited in on a summer morning will have much lower concentrations of oxygen than one that is scheduled for a visit late in the afternoon of the same day.


Trends in average dissolved oxygen concentrations in Great Britain and north-east England and minimum values in north-east England, between 1980 and 2012.   Legend as for previous graph. Note that UK standards for dissolved oxygen are not expressed as milligrams per litre.

Another measurement that gives us a very good indication of the scale of organic pollution in a river is the concentration of ammonium.   Proteins are broken down first to urea, which organisms excrete, and then, via the action of microbes, to ammonium.   If there is a general trend of better regulation and improved sewage treatment then we would expect a gradual fall in the concentrations of ammonium over the past thirty years and this is, indeed, what we see.

As was the case for BOD, ammonia concentrations show a steady downward trend. In this case, the decline only starts after the establishment of the NRA in 1989. Once again, the average state of our rivers looks promising and even the maximum values recorded are now mostly within the standard for “good ecological status”, which is the EU’s threshold for acceptable river conditions.

So at the end of the first part of my survey of river conditions everything seems to look promising, with concentrations of some of the main pollutants associated with organic pollution within the limits that should enable the UK to meet targets set by the EU.   The Environment Agency has a great deal of experience in managing sewage discharges and the technology for reducing concentrations of pollutants such as ammonium is well understood.   In the next post, I’ll move on to consider two other pollutants where the situation is not quite so optimistic and several challenges still remain.


Trends in average ammonia-N concentrations in Great Britain (closed circles) and north-east England (open circles), and maximum values in north-east England, between 1980 and 2012. The horizontal lines show the current UK standards required to support different classes of ecological status.

Fieldwork in the rain

Fortune dealt a bad hand for the annual GEO2042 fieldtrip to the Ouseburn.   For the first time in six years, it rained before and during our visit to collect water and invertebrate samples.   By lunchtime, the water levels had gone up so much that we were worried that the afternoon’s session may have to be abandoned, for safety reasons. Fortunately, the rain eased at about 1300 and the river levels started to drop again.


Fieldwork on the Ouseburn, Jesmond Dene, October 2014. Left: kick sampling for invertebrates in the river; right: investigating the contents of a pond net.

The progress of the day’s storm are neatly demonstrated on the Environment Agency’s excellent realtime water level monitor, situated about a kilometre upstream from where we were working.   The two groups of students were out between 1100 and 1200 and between 1300 to 1400 – either side of the highest level recorded at about midday.   Look how quickly the water level rose from the baseline.   This particular rainfall followed a long period of warm, dry weather, which has kept water levels down all over the region. The Ouseburn flows through built up areas of Newcastle for most of its short length, which means that a lot of the water will run straight off hard surfaces, into drains and into the river.   Hence the rapid rise of the river levels, followed by the gradual drop as the water that had soaked into the ground gradually found its way to the river. Compare this brief storm event to the hydrograph for the River Coquet that I showed earlier in the year (see “Fieldwork in Northumberland”).    In this instance, the river level went up more gradually, reflecting the much lower proportion of hard surfaces in the upstream catchment, before gradually declining.   To the trained eye, these graphs show the effect of man’s alteration of rivers just as clearly as any measurement of “pollution”.


River Levels in the Ouseburn, 5th – 7th October 2014, from the Environment Agency’s monitoring station at Crag Hall, about a kilometre upstream from Jesmond Dene (

One of the legacies of less-enlightened times that we have inherited is a system of combined sewers that carry both foul waste (don’t ask) and storm runoff.   One effect of prolonged rainfall is to fill these sewers with water from drains and, for this reason, there are overflows built into the system which let the excess flow straight from the sewers to the rivers.   Unfortunately, this overflow includes untreated sewage as well as storm runoff and, by Monday afternoon, the river had a distinctly unsavoury odour. The long-term plan is to replace these combined sewers with separate networks of storm drains and foul sewers. That, however, will take a long time, a lot of money (an awful lot of money) and, as most sewers run under our roads, serious disruption, to implement. So we will probably have to live with these combined sewer overflows for some time to come.

A hint for any GEO2042 students who have read this far: link the words “Ouseburn” and “combined sewer overflows” in your minds now. This might come in useful when you write up your project later this term.   Enough said

Diatoms from the roof of the world

Whilst I was enjoying myself in Venice and the Dolomites, most of my family were hard at work earning an honest crust or studying. One, however, was looking at mountains considerably higher than anything that Europe has to offer.   As she was, technically, reconnoitring locations for a study visit to Kashmir and Ladakh next summer she will tell us that it was work rather than holiday, but we should treat this claim with a generous pinch of salt.


Pangong Tsu, looking east towards the distant mountains of Tibet.   September 2014 (photo: H. Kelly)

One souvenir of her visit was a small sample of algae from the littoral zone of the remote Pangong Tsu at an altitude of 4266 metres on the Indian-Tibetan border.   Problem #1 for anyone working in such locations is how to preserve samples for the journey home. There is a simple solution: buy a bottle of cheap local vodka.   My limited experience of Indian spirits is that using them to preserve algae is a better option than drinking them and, in any case, the high altitude of Ladakh will give the partially-acclimated traveller a cracking headache without any need for chemical assistance. The downside is that the cell contents are not preserved very well but as diatomists are mostly interested in the silica cell wall this is not a problem.


A view of the periphyton from the littoral zone of Pangong Tsu (approx.. 33° 45’ N, 78° 38’ E), photographed at 400x magnification.

A first look at the sample down my microscope revealed a number of diatoms that looked broadly familiar though which differ in detail from species with which I am familiar with in Europe.   There were, for example, a lot of cells of Gomphonema, each at the end of a long stalk, though these were not a species that I recognised.   Within this tangle of stalks, I could see a number of zig-zag chains of a species of Diatoma but, again, it was not a form I had seen in Europe.   There were other taxa, too, but I will wait until I have cleaned the material and prepared a permanent slide before commenting further.


Gomphonema cells from the littoral zone of Pangong Tsu, September 2014. The four cells on the left are in valve view; the two on the right are in girdle view.   The stalk is visible on the right hand image. Scale bar: 10 micrometres (= 1/100th of a millimetre).


Cells of Diatoma from the littoral zone of Pangong Tsu, September 2014.   The lefthand image is in valve view; the central and right hand images are in girdle view.   The central image shows a cell that has recently divided.   The cells are joined at the corners to form zig-zag colonies. Scale bar: 10 micrometres (= 1/100th of a millimetre).

There seems to be very little published on Pangong Tsu, as befits its remote location.   One early visitor was the famous limnologist G. Evelyn Hutchinson, as part of the Yale North India Expedition in the early 1930s.   The papers that I did find told me that the water of Pangong Tsu is slightly brackish, as the lake has no outlets and water is lost by evaporation. However, I usually associate Gomphonema with freshwater habitats, suggesting that the brackish influence is very weak. There is also evidence of very high concentrations of phosphorus in the lake, although nitrogen concentrations are very low. This suggests that growth of the organisms in the lake may be limited by nitrogen rather than phosphorus, as is usually the case in lakes.   I would be interested to know why a lake that is so remote contains phosphorus concentrations that are usually associated with human activity.

The next step is to make some permanent slides from the material and have another look, then see if any of the diatoms in this sample correspond to species found elsewhere in the region or beyond. There is a chance, based on the lake’s isolation and recent discoveries that the diversity of diatoms was much greater than hitherto thought, that there may be some previously undescribed species living in this unusual habitat.


Bhat, F.A., Yousuf, A.R., Aftab, A., Arshid, J., Mahdi, M.D. & Balkhi, M.H. (2011). Ecology and biodiversity in Pangong Tsu (lake) and its inlet stream in Ladakh, India. International Journal of Biodiversity and Conservation 3: 501-511.

Hutchinson, G.E. (1933). Limnological studies at high altitudes in Ladakh. Nature (London) ##: 132-136.

Hutchinson, G.E. (1937). Limnological studies in Indian Tibet.   Int. Hydrobiol. 35: 134-177.