A question of scale …

It has taken some time to convert the observations from my last visit to the River Wear (see “Spring comes slowly up this way …”) into a picture.  Then, if you remember, the river was balanced between its “spring” and “summer” guises, the cool, wet weather that we experienced in March seems to have held the plants and animals that I usually see at this time of year back.   The result was a patchiness that was easy to see with the naked eye, but harder to visualise at the microscopic level.

First there were quite a few diatoms, Achnanthidium minutissimum in particular, – that suggested a thin biofilm subject to grazing by invertebrates (and I could see some chironomid larvae moving amongst the biofilm as I was sampling).   However, there were also diatoms such as Ulnaria ulna and Gomphonema olivaceum that suggested a thicker biofilm.    And finally there were filaments of the green alga Stigeoclonium tenue, mostly in discrete patches.   I never see healthy filaments of Stigeoclonium tenue smothered in epiphytes, which I have always assumed to be due to the copious mucilage that surrounds the plants.  However, I wondered if, nonetheless, Stigeoclonium contributes to overall habitat patchiness for the diatoms, as they subtly alter the way that water flows across the stone, reducing drag and shear stress in a way that favours Gomphonema and Ulnaria.   This is just speculation, of course…

That brings me back to a familiar theme: the problems of understanding the structure of the microscopic world (see “The River Wear in January” and “Baffled by the benthos (1)”) and, tangentially, to a paper on organisms’ responses to climate that was quoted at a scientific meeting I attended recently.   In this, Kristen Potter and colleagues demonstrated that there was typically a 1000 to 10,000 fold difference between the scale at which the distribution of organisms is studied and the size of those organisms.   That might be enough to draw out some coarse-scale patterns in distribution of species, but organisms actually live in microclimates, which may be patchy and which can often be quite different to the prevailing macroclimate (the difference between being exposed to full sun in open grassland and in the shade of a forest being a good example).   They suggested that the ideal spatial resolution is between one and ten times the organism’s length/height.

I see no reason why the same challenge should not also apply to the pressures faced by organisms in rivers where, again, we can get a certain amount of useful information from a coarse analysis of distribution in relation to (let’s say) average nutrient concentrations in a reach, but cannot really understand the reasons behind the spatial and temporal variation that we see in our data.  This mismatch between the scale at which organisms respond and the scale at which we study them is, I suspect, an even bigger problem for those of us who study the microscopic world.

A second illustration came at the same meeting in a talk by Honor Prentice from the University of Lund in Sweden.  She was dabbling in molecular biology years before this became a fashionable pastime for ecologists and has, over her career, developed some fascinating insights into how the structure of both plant communities and populations of individuals vary over short distances.  Her work has focussed on the island of Öland in Sweden which has the largest extent of alvar (limestone pavement) in Europe.   The system of grikes (the slabs) and clints (the fissures which separate the grikes) create quite different microclimates – the cool, moist conditions in the latter can create bog-like conditions with much lower pH than the limestone clints.   These differences influence not just the composition of the community but the genetic structure of species within these communities too.  I left thinking that if she could detect such differences at a scale barely more than one order of magnitude greater than the organisms, then how much more variation am I missing, with perhaps a five order of magnitude difference between organism size and sampling scale?

Based on these two studies, we would need to sample biofilms at a scale of about 1 mm2 in order to get a meaningful understanding of habitat patchiness in stream benthic algae.  That might just be possible with Next Generation Sequencing technologies, though I am not sure how one would go about collecting environmental data at that scale needed to explain what is going on.  Meanwhile, I am left with the coarse approach to sampling that is inevitable when you are five orders of magnitude bigger than the organism that you want to collect, and my imagination.


Potter, K.A., Woods, H.A. & Pincebourde, S. (2013).  Microclimatic changes in global change biology.   Global Change Biology 19: 2932-2939.

Prentice, H.C., Lonn, M., Lefkovitch, L.P. & Runyeon, H. (1995).   Associations between allele frequencies in Festuca ovina and habitat variation in the alvar grasslands on the Baltic island of OlandJournal of Ecology 83: 391-402.



Leave a Reply

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

WordPress.com Logo

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

Google+ photo

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

Twitter picture

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

Facebook photo

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


Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.