“the assumption that, shaved of their built environment elements, floodplains and river channels at the time had much their present form has to be questioned”
Lewin, 2010
Why examine fluvial history?
Tavistock is a beautiful town. If asked to imagine its best bits then Bedford Square, the bustling market and shopping streets, and the statement grandeur of Plymouth Road probably come to mind. Alongside this built environment will be conjured a picture of the river by the weir at Abbey Bridge; the river which gives Tavistock its name. In its universal way, the water draws us to it. It is a sensory experience. We watch the light, broken and bounced on the surface. We are soothed by the soundscape – the plash, yatter and susurration of the water. When it is running high, we can even smell it, the air imbued with the mineral and humic solutes dispersed into the air by the water’s tumble. It is in these times of spate that it really draws us. I, like many others, pay pilgrimage. We stand together on the footpath, watching; excited by the power and roar of the flood-water, safe in its dangerous presence.
In thinking of all these images of Tavistock, I suspect we view the Tavy as natural. And of course, to a large extent it is. But do we really appreciate how changed it is from the river that it might have been if we had not come along to tinker with the catchment and the channel? Do we appreciate how our prehistoric and historic ancestors might have encountered a different river to that which we see today? Do we appreciate the legacy of the intended and unintended actions of people on the river and floodplains? All of these questions are important for framing our understanding of the Tavy today and for helping us interpret some aspects of our archaeological and historical past.
It is for this reason that I have decided to write about the River Tavy, to give a physical geographer’s perspective to its history. This is because, as Lewin (2010) says, in a paper about Medieval floodplains, “the assumption that, shaved of their built environment elements, floodplains and river channels at the time had much their present form has to be questioned”. I find that, even with my palaeoenvironmental background, the landscape of the present dominates how I perceive the places that I explore. I have to keep reminding myself, with sketches from my knowledge, that things were not always thus. This is particularly true for rivers which have a great capacity for mobility and therefore alteration even without human agents getting involved. By looking at the research literature of fluvial geomorphologists like Lewin, I hope to be able to give an outline of its probable geomorphic history.
I can’t do this in one blog post – there is too much to say – so at least one more will follow. In this first one I will examine how the Tavy might have looked at the end of the last ice-age and how it might have developed in the millennia afterwards, once the climate had warmed. In the research literature, the medieval marks the point at which human impact on rivers significantly accelerated and so I will take the story in this blog up to this point.
Before I get going, I need to add some qualifiers. I will be focusing my attention on the Tavy in and around the Tavistock area. It isn’t that the upper and lower catchment aren’t important, but rivers behave differently from the steep incipient streams to estuarine mouth, so trying to cover all aspects would add even more complexity to an already involved and obscure story. There hasn’t been any specific research (that I am aware of) on the fluvial history of the Tavy, and so the text that I am pulling together is based upon generalised/meta-analysis of the history of British rivers. I will supplement this with local evidence from other Dartmoor rivers where possible.
So, with these qualifiers and boundaries in mind, and with an acceptance that the history is hidden and hazy, let’s begin.
The Tavy Before Us
The place to start is the last glaciation of the Pleistocene because, for our rivers, this determines the legacy on which human impacts have taken place. Whilst there may have been some small ice-cap and glacier development on Dartmoor (Evans et al, 2012), for the Tavy, it is cold dry periglacial conditions that predominate. Imagine a tundra type environment. Here the repeated freezing and thawing of water is an aggressive geomorphological process and one which produces abundant loose rocks.
The extreme cold means that few plants can be sustained and soil development is limited. Without a binding and protective blanket of soil and vegetation, the coarse sediments on the hill-slopes are prone to movement. These colluvial materials head, under gravity and aided by water, into valley bottoms. Once in the river network the sediments become alluvium – the sediments that we see in channels and forming floodplains. Late-glacial and post-glacial English rivers are therefore systems filled with the thick deposits of these cold-condition environments (Lewin, 2014).
I am placing focus on the sediment regime because it is of great significance in determining river morphology. For high energy rivers choked with coarse sediment, a braided river type is typical (Lewin, 2010). In terms of the river form, this means that the river bed is wide and shallow. There are multiple channels; a few bigger, some smaller and some dormant. Water flows between sediment islands and bars. Flood flows can cause reorganisations of the channel patterns. The floodplain surface of such a river is complex, with various active and inactive components and a rough relief. At ordinary flow, braided rivers are easily fordable because the water is quite shallow and is unconfined by steep banks; the bars and islands providing hopping points for passage.
From c.12k yr BP the climate began to warm up. This warm epoch is termed the ‘Holocene’ and we are still living in it now. The small tundra plants of the Late-glacial were replaced by a succession of warmer-loving ones, quickly resulting in forest cover. Soils developed and the warmer Holocene climate was wetter and so river flows increased. In their erosive upper and middle reaches, rivers began to cut down through their Pleistocene gravels (Macklin et al, 2013), entrenching the channel and in some places working down to the underlying bedrock. With the supply of coarse sediment reduced, there would have been a gradual flushing out of coarse sediments from the system, although the hard to transport larger cobbles and boulders would remain. However, there would be much within-channel re-working of floodplain deposits so coarse sediments would continue to be mobilised within the river. Fine sediments originating from the developing soils of the hill-slopes would get into the river through natural erosion processes like river headwater incision and gullying (Macklin et al, 2014). These would have a limited impact on fluvial morphology, with modest fines deposited on floodplains. The process of floodplain accretion had started, but the rate of sedimentation occurring was very low.
In the early millennia of the Holocene the Tavy probably retained braided stretches, but the flushing of the sediment through the system, the entrenching of the river to form river terraces, and the very gradual deposition of sediments on the floodplains would slowly change the river morphology. How long this process would have taken I am not sure. In Scotland, high inputs of coarse sediments even to this day mean that there are frequent examples of braided sections of rivers. Lewin (2010) suggests that such morphology may have been persisted in some English rivers until at least the Medieval. In the middle reaches of the Tavy, it is therefore entirely possible that braided forms existed. However, as the coarse sediments were flushed out of the system and fine sediments were gradually deposited on floodplains, rivers probably moved into an anabranching form. Similar to braided rivers, anabranching ones have multiple channels but there is greater stability and persistence of the channels. The stability of the system allows vegetation to develop on islands and bars of sediment, and this acts as a positive feedback, further helping to bind these features in place (Nanason and Knighton, 1996). Even today the Tavy has anabranching pockets so it very likely these were much more widespread in the not too distant past.
The Early Influence of People on the Tavy
Whilst fine sediment inputs to the Tavy would have been low in the early Holocene, as people began to settle the land, disturbance would have begun releasing material into the river. Deforestation in the Bronze Age would have created the first pulses. Farming, particularly arable, where soils are bare for parts of the year, provides another main input. Bronze Age metal extraction may have had some effect but this was limited. Early tinning would have utilised the superficial and easily accessible deposits. It was only later when the buried tin had to be dug out that significant quantities of sediment entered river systems (Thondycraft et al, 2004) due to metal working. The low populations of pre-history meant that, although not without consequence, these changes caused only small increases in floodplain sedimentation (Macklin et al, 2010). Nonetheless, the period in which humans were impacting the floodplains was beginning.
In response to both the sedimentation building up floodplains, and erosion cutting more entrenched channels in the Pleistocene deposits, some English rivers began to evolve into a morphology called ‘active meandering’. Active meandering channels are ones with a single rather than multiple channels. Meandering rivers are more contained by steeper banks than braided or anabranching rivers. Bars of sediments may form on bends adjacent to pools, and shallow stretches called riffles develop between river bends. Active meandering channels are able to migrate and so gradually (but sometimes more suddenly) shift position within the floodplain (Lewin, 2010). Much of the Tavy’s floodplain in Tavistock has been built over or used for recreational purposes and so it is hard to look at floodplain surfaces for clue. It is not possible to know for definite, but looking at LiDAR images, I can’t see any evidence of relic meander features. However, I can see what looks like relic anabranching so I suspect this is the form that was more common for the Tavy of the past.
Once we get to the Medieval, human activities were to lead to large increases in floodplain sedimentation. Active management of riparian areas through drainage, channelisation, and reclamation, to name but a few, would constrain and immobilise the river. However, up to this point in history, the Tavy would have been largely natural. It would have been subject to evolutionary development as it ‘recovered’ from the last glaciation and it would have been subject to variations imposed by climate (particularly periods of greater and lesser flooding). The style of the river in its middle reaches probably started as wide, shallow, unstable and braided. Through time the river will have become gradually more stable, but still with multiple, anabranched channels. Bars of sediment and stabilised islands would have been frequent. However, there is no simple succession of forms for a river. Different reaches would have developed at different rates and been subject to local conditions (for example, where constrained by local topography).
The Implications of a ‘Different’ Tavy
So what is important about the early Tavy when thinking about landscape and history? For me, it is this. The Tavy of the past would have been far less entrenched than the river we see today. Within Tavistock the river banks have been built up by walls. However, along all its length, banks have been heightened by the glut of Medieval and post Medieval sedimentation on the floodplains. These combined impacts – the intended and unintended channeling and entrenching – prevent the present river of today from migrating. The Tavy of the past was mobile, wider and shallower. There would have been more extensive reaches with multiple channels, bars and islands. Of course, those readers familiar with Tavistock history will be aware that there used to be an island in the Tavy where the Medieval bridge was located and which is shown in maps and illustrations of the town such as on Wynne’s Map of 1755 (see below).
These characteristics described impact on people in a number of ways. For a start they influence how and where people find crossing points. Neolithic and Bronze age people may have found the river more fordable at standard flows than we experience today and may have used locations with split channels and islands as accessible points for river crossing (Lewin, 2010). In terms of bridging, the earliest record in Tavistock was the Medieval ‘Great Bridge’ of the 1270s (Woodcock, 1985). However, it seems likely to me that Tavistock would have needed a bridge in the 300 years between the founding of the abbey in the 10th C and this first recorded stone bridge. We won’t ever know what earlier bridges existed in Tavistock before the ‘Great Bridge’ but imagining the river landscape of the past can help us think more accurately about the logistic and engineering possibilities. A wider, shallower, anabranched channel would have perhaps allowed an initial ‘built’ crossing in the form of stepping stones like at Tarr Steps on Exmoor. Saxons were known to build wooden bridges so perhaps we can imagine a longer wooden bridge, spanning multiple shallow branches of channel, secured between banks and islands. What is important is to remember that, whatever bridging solutions the people of the past may have developed, they would not have been doing so to cross a river that looks like the Tavy of today.
In writing this blog I also found myself struck by the idea, not just of issues around fording and bridging rivers but of how potentially useful they might have been as corridors for movement. Accounts of the past tend to focus on rivers as being useful for communication when navigable. The idea of using rivers to move through the landscape on foot seems foolish. Why walk along a river when you can take a footpath or walk over open ground? But of course, before deforestation gathered pace, moving through forest would have been arduous. Wading through deep rivers isn’t very practical either but of course, as we have been exploring, the rivers of the past weren’t the same then as now. Long reaches of braided and anabranching channels would have created open corridors through which a person could walk across gravel bars and islands and wade in the shallow water of standard and low flows. Compared to the forest, the going, I imagine, would have been easy. Of course, not all reaches of rivers would be easy to walk, but there would certainly have been far greater scope for using rivers for pedestrian accessibility than there is today (see photo above to illustrate how a river might present an access opportunity not a constraint).
Finally, I think we ought to re-imagine the floodplains of a pre-Medieval Tavy. These floodplains would not have been blanketed in the thick rich sediments we see today. The fine sedimentation on them would have been thinner and the floodplain surface probably more rough and stoney. Because the river was less entrenched between banks, it would likely be more prone to flooding. The branched channels of the river would be more mobile, and so floodplains would be subject to changes as channels migrated. I therefore suspect the early floodplain of the Tavy offered greater risk and fewer benefits than was to develop from the Medieval onwards. This would have had consequences for how the floodplains were used. For example, did floodplains provide fertile water meadows to the same extent as they did from the Medieval onwards? With a wider, more mobile and more flood prone river, how did this effect the siting of the earliest buildings in the town? These are themes I intend to explore more in my next blog on this theme.
I hope that this blog has opened the possibility of seeing the Tavy in a new way – one which peels away layers of sediment and infrastructure to a different type of river of the past. When I come back to this topic in a future blog post I am going to look at the river from the early Medieval period through to the present day, exploring the intentional and unintentional ways people have altered the river. Far more is known about about this recent past, be it the management of monastic estates through to enlightenment and industrial engineering, so I reckon there will be plenty more to discover and tell.
References
Evans, D.J., Harrison, S., Vieli, A. and Anderson, E., 2013. Dartmoor glaciation–fact or fiction?–a comment on Allan Straw’s article. Quaternary Research Association, 130, p.11.
Lewin, J., 2010. Medieval environmental impacts and feedbacks: the lowland floodplains of England and Wales. Geoarchaeology, 25(3), pp.267-311.
Lewin, J., 2014. The English floodplain. The Geographical Journal, 180(4), pp.317-325.
Macklin, M.G., Jones, A.F. and Lewin, J., 2010. River response to rapid Holocene environmental change: evidence and explanation in British catchments. Quaternary Science Reviews, 29(13-14), pp.1555-1576.
Macklin, M.G., Lewin, J. and Jones, A.F., 2013. River entrenchment and terrace formation in the UK Holocene. Quaternary Science Reviews, 76, pp.194-206.
Macklin, M.G., Lewin, J. and Jones, A.F., 2014. Anthropogenic alluvium: an evidence-based meta-analysis for the UK Holocene. Anthropocene, 6, pp.26-38.
Nanson, G.C. and Knighton, A.D., 1996. Anabranching rivers: their cause, character and classification. Earth surface processes and landforms, 21(3), pp.217-239.
Rhoads, B.L., 2020. River dynamics: Geomorphology to support management. Cambridge University Press.
Thorndycraft, V.R., Pirrie, D. and Brown, A.G., 2004. Alluvial records of medieval and prehistoric tin mining on Dartmoor, southwest England. Geoarchaeology: An International Journal, 19(3), pp.219-236.
Woodcock, G. (1985). Tavistock’s Yesteryears: episodes from her history – 1. Penwell Ltd: Callington.
Be First to Comment