This is a slightly shortened version of a talk I gave last week at the WWF chalk-stream conference: I was asked to address what river restoration should mean for chalk-streams.
The definition of restoration is the act of returning something to its former condition. Most of the habitat and ecological enhancement works undertaken on chalk-streams are described as “river-restoration” and yet – if we are honest – only vaguely do most “restoration” projects actually aim at a specific idea of former condition.
I’m not convinced we have a commonly agreed understanding. The danger is that what we do to chalk streams – even acting with the best of intentions – proves to be not so much restoration, as just another form of modification.
With chalk-streams this danger has a keener edge to it than on any other type of river. Chalk-streams are globally rare, confined to south east England and northern France. Lately we’ve made great play of this, arguing for enhanced levels of protection. But their rarity demands care on our part too.
Chalk-streams were created by forces and processes that are no longer at work on the landscape. They are relic imprints of a vanished past. But they are such gentle rivers that, once changed, they cannot easily or quickly undo that change. We ought to think twice, in other words, about what we do to them.
More than once I’ve been told that the idea of restoring any given state, let alone the first state, is a nonsense because the palaeolithic chalk-stream was never a stream. It was more of an ooze: braided threads of spring water seeping across a marshy, tree-covered valley floor. The distinct channels that exist now, so this argument goes, are the work of man, not nature.
Usually in tandem with the first objection is the idea that whatever they were once like, chalk-streams have been so modified that the original streams are beyond recovery. There is nothing left of them. The question is not how do we restore them, but what do we want from them?
I don’t agree with either of these anthropocentric arguments. Sure, the palaeolithic chalk-stream was braided here and there, was impounded by fallen trees, and flowed through a seeping, wet-woodland marsh: but it also flowed along distinct, meandering channels. Mazy, sinuously indecisive and utterly natural in pattern, these channels were not built by men – though they have certainly been erased by men.
It might well be a pipe-dream to think of returning great lengths of our chalk rivers to a pre-historic state. But that is not to say that we can’t take lessons from what we deduce about this former state and apply them to conceptions of future states.
This is something the Norfolk Rivers Trust has been trying to do recently on a chalk-stream in Norfolk.
It began when WWF commissioned me to write a Water Framework Catchment Restoration Plan for the River Nar. The idea was to come up with an easily digestible description of the catchment, the river, the problems it faced, and what could be done to improve its ecological status. It needed to express complex ideas accessibly and engagingly so that we could involve all the stakeholders in the restoration vision for the river, and of course, so that we could raise funds.
The plan certainly helped a successful bid to the Catchment Restoration Fund and also the Drainage Board’s own bids from other sources. At the same time WWF and Coca Cola continued with their sponsorship. All this allowed the Nar Restoration Group and the Norfolk Rivers Trust to think big.
So, what did “restoration” mean, when the time came to turn grand ideas and money into physical manipulation of the river? What could we deduce about the ‘former states’ of a river that carries so many historic imprints and that has been so changed over the centuries?
There is hardly a yard of the River Nar that doesn’t bear the impact of historic canalisation, of dredging, and widening, the construction of levées, as well as the impounding effects of bridges and fords, and mills.
It seemed to me that the best way to approach the “restoration” of the river, was to begin with a forensic interrogation of its history, reach by reach, digging down into why things were as they were, looking at tithe maps and early Ordnance Survey maps and aerial photos, tracking the progression of changes.
I’m a writer. I love this kind of thing.
But even so, I’d argue that restoration has to begin with historical detective work. How else can one discern which of the former states must be preserved, or should be recreated, or more specifically what would actually constitute “restoration”?
One manifestation of all those modifications on the River Nar is the amount of burr-reed and eel grass relative to ranunculus and starwort. It’s cited by Natural England as a reason why parts of the river are in “unfavourable” condition.
This is where the river flows through Castle Acre Village. You’ll see that in places the natural planform of the river is still intact. In other places the river has been diverted.
The red lines show where burr reed and eel grass are the dominant in-stream plants. They coincide with where the river has been dredged and no coincidence, where levées of dredged spoil have been piled up on the banks.
The blue lines mark step changes in gradient as a result of a culvert, a ford and diversions that relate either to the water meadows or the Norman castle and priory.
Each of these structures, all old, some very old, each representing a phase in the river’s history, formed a gentle impoundment to the river’s flow: the culvert because the river must squeeze to get through it, the ford because fords are built up with gravel, and the diversions because diversions inevitably form impoundments as the river runs flat from the centre of the valley to the contour line at the side.
Here’s another satellite image a mile downstream where the river has been diverted to the northern and then the southern edge of the valley to create what is effectively part of the leat that served West Acre Mill downstream. The blue line marks the lip of the gradient change.
All these impoundments slowed the river and caused a gentle ponding upstream and – over centuries – an accumulation of sediment, creating conditions that favoured the plants which prefer impounded flows: eel grass and burr-reed.
But what came next is important too. In the post-war era of agricultural intensification, these reaches will have been the most problematic. They became the focus of “maintenance” weed-dredging, which will have compounded the problem as the stream was made wider and deeper, year on year.
I had an instinct that long sections of the gradient of the stream would show a sagging belly upstream of the culvert, the ford, and the diversions. The sagging belly being the over deepened stream bed. My instinct was right. They did.
Another more recent discovery has been the extent to which the river has migrated laterally because of the dredging, sometimes by its full channel width.
I’m summarising findings that actually took a fair amount of detective work. But restoration proposals were drawn up for these reaches before all this sleuthing was done and were even installed in places. A technique that works where there is gradient – digging even deeper holes and piling the gravel in shoulders and miniature staircase riffles above them – is just another modification where there isn’t.
Modification not restoration
So, how do we approach the restoration of a reach like this, over-deepened and widened stream?
First, a trip to Buckinghamshire:
This picture is of the suburban river I recently wrote a book about: the River Wye.
It was taken about a hundred years ago beside the London Road upstream of a working paper mill, Rye Mill. You can see the mill-leat on the right of the picture is broad and deep and very much in use for the running of the mill machinery.
The milling era is over now. The impoundment that held flows above Rye Mill has been lowered and as you can see in this photograph which I took in 2008 – the same reach looking upstream – the leat has been left alone for quite a few years.
This photograph is fascinating: with the mill impoundment removed the river has started to dictate its own restoration – and in this sense I am using the word correctly – the river is restoring a former state.
It has started to meander and to construct its own flood-plain – albeit within the confines of the old leat. The meander pattern has a wavelength that correlates to the channel dimensions which in turn are shaped by the flow regime of this particular river.
This river is becoming what it wants to be.
So, we have history and we have the river each telling us things: there are clues in the landscape. A river needs gradient and a river needs to meander. Pools and riffles are a function of these two things. Yet how often do restoration schemes try to address the latter without looking at the former?
The only way properly to restore the dredged and ponded reaches I have described on that part of the Nar is to first restore gradient and bed level. Well, we haven’t the budget or the permission to do that yet. So let’s look at the parts of the Nar which we have begun to tackle.
Here, as with that self restoring mill-leat in High Wycombe, we had to work within certain limits. But what would constitute restoration working within the confines of old leats and water-meadows alongside moats and ancient priories?
That mill-leat in High Wycombe gives a strong steer doesn’t it?
In essence we* tried to recreate, or rather we tried to catalyse the recreation of, what the chalk-stream channel might once have been like and would be like again in a thousand years if left alone – but working within within the context of those tiers of history and modification many of which cannot be eradicated and some of which should definitely be preserved.
On another river the context might be different. But the principle of the former state would be the same: it should be what the river would re-create given half a chance … or a thousand years: a channel shape that – in terms of sinuosity and gradient, and heterogeneity, width and depth and flood-plain, matches its flow regime.
(I spoke about other phases of the work but they are detailed on previous blog entries)
(*Simon Cain, Cain BioEngineering, Acorn Trees Services and myself)