Between 2014 and 2016 an archaeological aerial survey centred on the catchments of the Rivers Exe, Culm, and Clyst, north of Exeter across Mid and East Devon, was completed by a project team of staff from AC archaeology and Devon County Council, funded by Historic England as a National Mapping Programme (NMP) project and hosted by Devon County Council within the Historic Environment Team.
Read more about other aerial survey NMP projects in Devon.
This project included parts of a rich archaeological landscape in which the Historic Environment is under pressure from threats as diverse as initiatives to reduce diffuse water pollution to housing and commercial development growth. It also includes the western edge of the Blackdown Hills AONB, a protected landscape that has seen little systematic archaeological survey and where the archaeological resource is relatively poorly understood.
The project consulted over nine thousand hard copy aerial photographs loaned from the Historic England Archive, Cambridge University Collection of Aerial Photography and those held by Devon County Council, as well as lidar data and modern digital datasets of vertical aerial photographs.
Some of the most useful aerial photographs for identifying earthworks were taken in November 1946, when the ground was least obscured by vegetation cover and before agricultural intensification seems to have levelled many of the sites. Specialist oblique aerial photographs, targeting archaeological sites when cropmarks are most visible during the height of the summer months, were invaluable for recording sites surviving only or mainly as below-ground features.
A total of 2371 archaeological or historic sites were identified from the aerial imagery and recorded on the Devon Historic Environment Record (HER), available through Heritage Gateway and mapped on the Devon Environment Viewer. More than three quarters of these had not previously been recorded on the HER.
A number of different topics became apparent as dominant themes within the project area and are summarised below, but much more detail is available in the highlight report via the Historic England project webpage.
Devon is known for its cider, and cider apples were crucial to the agricultural economy of this area especially in the 19th and 20th century; there are even apple varieties called the Whimple Wonder and Whimple Queen! The importance of the industry is clearly reflected by the very high proportion of records – nearly a fifth of the survey total – that are of the earthwork remains of tree planting banks. These aided drainage and provided adequate soil depth for the fruit trees.
Whiteways Cyder Company came to dominate the local trade, and this is obvious from the aerial images, which include numerous large Whiteways Cyder Orchard signs within very neatly kept orchards.
Eventually Whiteways expanded into supplying nationally, and then internationally, various cider (and associated) products. Much more information is available in the Whimple Heritage Centre.
The density of historic orcharding is demonstrated by various sources; the banks visible on aerial photographs and lidar often correspond to orchards depicted on the Ordnance Survey First Edition 25 inch maps.
However a significant proportion do not, and these are likely to be earlier phases, although their use may have overlapped. Most of the fruit trees visible on aerial photographs have been removed, but approximately two thirds of the recorded earthwork banks could survive to some degree.
Relict or removed field boundaries were the second most numerous monument type recorded during the survey. Nearly all of those recorded are interpreted as having pre-dated the mid-19th century, and the vast majority have a curvilinear form, enclosing small and irregularly shaped fields, that suggests a medieval origin.
Although Devon’s Historic Landscape Characterisation project recorded field boundary loss that occurred after the late-19th century, only field boundaries not depicted on the earlier (mid-19th century) Tithe Map were routinely recorded during NMP. The scale of loss as fields have been amalgamated into larger land parcels has been very high in some of the lowland parts of the project area.
Over half of the removed boundaries were recorded as earthworks, many of them very wide shallow ditches suggesting that there may have been substantial drainage banks associated with them in some areas. A third were visible as cropmarks only, indicating complete levelling of remains probably through plough action.
In the lower lying areas along the river catchments networks of straight narrow drainage ditches were common, covering large areas of land (for example at Exminster MDV72510). In some places it is difficult to distinguish between drainage ditches and watermeadow systems, which comprised a tenth of all monuments recorded during the survey.
The regionally distinctive combe or hillside catchmeadow was most commonly recorded, but a more unusual form, interpreted as a possible local ‘hybrid’ style, was also identified during the survey.
These small-scale systems on hill slopes were defined by a series of parallel contour gutters bisected by one or more linear channels aligned down-slope, cross-contour.
In many cases, these channels appear to tap a water source located further uphill, or appear to originate at a farmstead, as at Durshayes (MDV108300) perhaps as a means of more directly integrating fertiliser in the form of liquid manure into the system.
Valley bottom water meadows, known as bedworks, are rare in Devon, but numerous examples were recorded, considerably extending the supposed distribution of these within the county. Water meadows of all types were more common in the north of the project area, in the central Exe Valley around Tiverton, in keeping with Turner’s (2007) findings from HLC.
Extracting a Living
Scattered clay, sand, gravel and marl pits were recorded across the project area, mainly on the mudstone, siltstone and sandstone soils, but the aerial images clearly reveal the remains of two extensive and well established former industries on the western part of the Blackdown Hills AONB.
The spoil, trackways and collapsed galleries of whetstone mines exploiting the greensand geology of Black Down and North Hill can be seen around the scarp on a range of images, although they are obscured by trees on later 20th century aerial photographs, and increased vegetation growth means it is difficult to see these from ground level now (MDV110227).
In use from the mid-18th to early-20th century, these workings dominated the landscape and local economy; the white spoil was said to be visible from Cullompton in the early 19th century (Stanes 1993). Whole families were involved in the trade, with the complete absence of local children from school in 1878 attributed to the Scythestone fair in Exeter!
A much earlier industry was recorded in detail from earthworks across the plateau above the whetstone mines. Extensive open workings from extraction of iron ore were visible as earthwork pits and, where land had been levelled, as cropmarks (MDV110229).
Radiocarbon analysis has dated partially levelled pits on North Hill to the Roman or post-Roman/early medieval period (Griffith & Weddell 1996), and they are similar in size to Roman iron ore pits from the industrial landscape of the High Weald. It is likely that associated features such as charcoal burning platforms and smelting sites were present in the wider area but have yet to be identified.
A little-known aspect of Devon’s industrial past is the mining of manganese ore for glassmaking, with the Upton Pyne area the centre of national production between about 1770 and 1800 (Russell 1968), before the focus shifted to Newton Abbot and the Teign Valley then declined in the early-20th century. Prior to the survey the only evidence recorded on the HER was a D-shaped pit but two nearby earthworks identified from lidar data may well have been further extraction pits associated with this regionally important industry (e.g. MDV113729).
Standing on Ceremony
Many of the prehistoric features visible on aerial images are interpreted as monument types with a ceremonial or funerary function. Four in five of these were recorded as cropmarks and these were strongly focused on the sandstone areas; cropmarks do not form as readily over mudstone and siltstone.
Specialist aerial photographs taken as part of the aerial reconnaissance programme in Devon during the 1980s and 1990s were the main source for many of these, but a considerable number (43%) were newly recorded on the HER. The earlier of the monuments, dating to the Neolithic, are unsurprisingly the rarest of these.
Two possible mortuary enclosures (MDV111017 & MDV112719) and a hengiform feature (MDV108465), were identified during the survey as cropmarks and an embanked earthwork respectively, and neither had been previously recorded.
More commonly identified were burial mounds, and surrounding ring ditches, dating to the Bronze Age. Previously unrecorded mounds within the dispersed Upton Pyne cemetery site, possibly additional barrows, could be seen only after the revisualisation and adjustment of lidar data MDV113839.
Levelled barrows visible only as cropmarks include those with evidence of more than one ring ditch, possibly indicating several phases of barrow construction and reuse.
For example, cropmarks visible on military RAF vertical photographs taken in 1946 south-west of Uplowman (MDV110019) might be an outlying element of a nearby small nucleated barrow cemetery visible as a group of four ring ditches.
Although most evidence for prehistoric settlement is Bronze Age or Iron Age in date, the cropmarks of a very large oval enclosure (MDV1279), of possible Neolithic date, were visible. Small features within it have been newly identified and recorded during the survey (MDV110718, MDV111089), although it is not clear whether they were contemporary with it.
Ditched enclosures, typical of later-prehistoric or Romano-British settlements, are relatively numerous at 236 examples. Over a third of these are new monument records, and almost all were recorded from cropmarks that had formed above the buried remains of infilled enclosure ditches and levelled banks.
The enclosures with a partially surviving curving earthen bank tended to be larger and may have been more substantial defended sites, rather than the smaller and perhaps less robust straight-sided enclosures. One example (MDV108251) is sited on the intriguingly named Castle Hill!
Cropmark evidence for enclosures was strongly focused on the more freely draining land, and this type of monument was widespread across the south and western parts of the project area.
In part their distribution coincides with the areas of greatest growth in road infrastructure around Exeter, and several have been destroyed by road cuttings, although roadworks also offered opportunities for excavation. Some of the enclosures near Pond Farm were found to have a 2nd century AD date during construction works in 1975 (MDV10043).
Evidence of much later abandoned settlements was also seen on the aerial imagery. Around Poltimore House for instance the earthwork remains of Bargain Farm and Pitt Farm (MDV113064) have been identified from lidar data.
These settlements were depicted on the Tithe Map but not the First Edition OS mapping, by which time the area had been incorporated into Poltimore Park and is depicted as wood pasture with tree clumps.
Fortification and Control
Sites with military or defensive purposes were less frequently encountered than in many NMP projects, but were often highly significant. They range from highly recognisable Iron Age hillfort ramparts, and the buried remains of fortifications that enabled Roman control of the region, to Second World War civil defence, military training or prisoner of war internment centres and Cold War observation posts.
Additional parts of the defensive circuits and interior features were spotted on aerial photographs and lidar data as cropmarks or earthworks at several hillforts (e.g. MDV1360) and Roman camps or forts (e.g. MDV29189). In some cases the imagery allowed a reassessment of their survival (e.g. MDV10188).
The aerial resource, particularly the wartime and immediate post-war coverage, provides very good potential for identifying sites of the two World Wars, and more than half of the monuments of this date that were recorded were new to the HER.
Only a few potential First World War sites were recorded; typical zig-zag and ‘dog-tooth’ shaped training trenches on rough ground at Uffculme Down (MDV107765) and Gaddon Down (MDV108079) and in Tidcombe Plantation (MDV108323).
Training of definite Second World War date could be seen from shell craters in training areas such as Woodbury Common (MDV112541).
Genuine aerial attack was demonstrated by 15 incidences of bomb craters (e.g. MDV107637). Measures to mitigate enemy fire included air raid shelters (MDV108391), anti-aircraft gun emplacements and batteries (e.g. MDV79572, MDV78517) and decoy targets.
Bombing decoys take several forms, and their temporary nature means that often their layout is only known from aerial photographic evidence. One such is the Special Fire or ‘starfish’ decoy site at Ide, established by the end of May 1942 to draw enemy aircraft away from Exeter during the Second World War. This was in response to, but unfortunately too late to mitigate the impact of, the deadly and culturally devastating ‘Baedeker’ raids of late April and early May 1942. These operations targeted historically significant cities and were intended to destroy civilian morale; they were undertaken in retribution following the tactical British attack on Lübeck, which was itself designed to demonstrate the prowess of Bomber Command as much as (or more than) to disable the most critical military targets (Dobinson 2000,164).
Shallow irregular firebreak ditches at Ide (MDV72100) probably surrounded firebaskets and ‘crib’ or oil fires over steel troughs, to simulate the light effects that would be expected in bombed cities. Numerous smaller ring-shaped possible earthworks in the same field may be rare examples of light positions for different types of light display to simulate movements such as doors opening (Roger Thomas, pers. comm.). Slight remains were still visible in 1945.
One major target with its own decoy site (MDV72067, MDV72068) was RAF Exeter (MDV48842). Requisitioned at the outbreak of the war, the airfield played an important role including the Battle of Britain, experimental research, bombing raids and convoy patrols involving Polish-manned squadrons such as the renowned ‘Lwow Eagle Owls‘. In 1944 it became a USAAF station for a time as preparations for D-Day progressed and subsequently casualties were evacuated.
Photographs taken in 1942 show camouflaging of black lines painted over the airfield to imitate field boundaries, with patches along the runway edges designed to break up the straight lines.
The newly extended runway visible at this time testifies to the changing airfield design; by the end of the war the photographs show the standard looped pattern of dispersal which was widely implemented across the country during the conflict and allowed for much more rapid and efficient deployment of aircraft.
Infrastructure associated with military activity seldom survives well, but this was also visible on the aerial photographs. The layout of camps for US servicemen (MDV108461), prisoners of war (MDV80418, MDV57281), RAF Exeter airfield personnel (MDV56271, MDV56268) and unknown purposes (MDV107878, MDV109200) were all recorded.
Several large-scale designed landscapes within the project area have been designated as Registered Parks and Gardens: Bridwell, Killerton House, Knightshayes and Rockbeare Manor. However the survey has added most to the smaller, less well-known parklands.
Earthwork mounds for highly visible tree clumps (MDV113346) at Winslade Park, geometric tree planting enclosure banks at Poltimore, and a possible causeway across an ornamental watercourse on Lord Rolle’s estate (MDV112325) are all examples of additional features identified through the survey.
A particularly interesting and locally distinctive example of estate features with a coherent landscape presence, but not as part of a typical formal parkland, was visible on Woodbury Common.
Groups of paired mounds (e.g. MDV10473) along Yettington Road have been variously interpreted as Bronze Age burial mounds and Napoleonic defences as well as post-medieval ‘follies’.
Given the presence of some even more elaborate star shaped examples just outside the project area south-east of Ottery St Mary (MDV10750, MDV20256 & MDV10757), and Lord Rolle’s instigation of other large landscape projects such as the Rolle Canal, a plausible explanation is that these groups of mounds flanking the roadways are most likely to have been created under the direction of Lord Rolle as foci demarcating routes through his estate, and perhaps incorporating pre-existing archaeological features.
The aerial photographs also illustrate elements of more widespread landscape change, such as the decline in the number of mature hedgerow trees, and changing attitudes towards woodland.
Interesting comparisons can be drawn: the replacement of a line of mature trees with new saplings between 1946 and 1955 along the drive at Winslade Park broadly compares in timeframe to the replacement of a lime avenue with poplars at Poltimore House in 1956. This desire to retain and enhance the historic landscape as depicted on the late-19th century mapping also seems to have waned with subsequent shifts in ownership and management; the avenue at Winslade is now subsumed or completely replaced by a plantation.
Dobinson, C. 2000. Fields of Deception: Britain’s Bombing Decoys of World War II. English Heritage.
Griffith, F. & Weddell, P. 1996. ‘Ironworking in the Blackdown Hills: Results of a recent survey’ in Mining History 13 2: 27-34.
Historic England. 2015. Using Aerial Photographs
Russell, P. M. G. 1968-1970. ‘Manganese Mining in Devon’. Devon & Cornwall Notes & Queries 31: 205-213.
Stanes, R. G. F. 1993. ‘Devonshire Batts: The Whetstone Mining industry and the community of Blackborough, in the Blackdown Hills’ in Report and Transactions of the Devonshire Association 125: 71-112.
Turner, S. 2007. Ancient County: The Historic Character of Rural Devon. Exeter: Devon Archaeological Society Occasional Paper 20.
Wilson, D. 1982. Air Photo Interpretation for Archaeologists. London: Batsford.
Would you like to share additional information on any of the sites we recorded from aerial photographs, or do you have a different interpretation? Let us know! We read all submissions but unfortunately due to the volume of responses we are unable to respond personally to every comment.