Survey Commissioned by Samantha William, Heather and Hillforts Landscape Partnership Scheme Surveyed by I.P. Brooks and K. Laws Engineering Archaeological Services Ltd. registered in England No 2869678 Moel y Gaer, Llantysilio Geophysical Survey July 2009 EAS Client Report 2009/10 CONTENTS List of Illustrations Figure 1: Location Figure 2: Location of Fluxgate Gradiometer Survey Figure 3: Location of the Resistivity Survey Figure 4: Fluxgate Gradiometer, Grey Scale and X-Y Plots Figure 5: Fluxgate Gradiometer Survey, Interpretation Figure 6: Fluxgate Gradiometer Survey Summary Figure 7: Resistivity Survey, Grey Scale and X- Y Plots Figure 8: Resistivity Survey, Interpretation Figure 9: Resistivity Survey, Summary Figure 10: Summary Introduction: NGR Location and Topography Archaeological Background Aims of Survey SUMMARY Methodology: Complicating Factors Survey Results: Area Display Results Fluxgate Gradiometer Magnetic Susceptibility Resistivity Plate 1: Fluxgate Gradiometer Survey Conditions Conclusions Technical Information: Techniques of Geophysical Survey Instrumentation Methodology Copyright Acknowledgements 1 NGR Centred on SJ 16695 46376 Location and Topography (Figure 1) Moel y Gaer, Llantysilio (SAM De 126), is the smallest of the hillforts within the Heather and Hillfort study area (Jones, 2004, 7). It occupies one of the lower summits of the Llantysilio Mountain, between Moel y Gamelin and Moel Morfydd (Figure 1). These two peaks block the view from Moel y Gaer to the east and west leaving the best views to the south towards the River Dee and north over the valley of the Afon Alyn. It also overlooks a relatively low pass between the two valleys at the foot of Moel y Gamelin, now occupied by the route of the county road. The underlying geology consists of fissile mudstones, thinly bedded sandstones and indurated cleaved siltstones and shales of early Ludlow, late Silurian age (Dorning pers. comm.) The single rampart at Moel y Gaer, Llantysilio encloses an area of approximately 0.95 ha. It is most developed on the northern and eastern sides of the hillfort where an external ditch is also evident. Elsewhere the defenses would appear to consist of a single rampart which enhances the natural slope of the hillside. Quarry hollows occur only in the north eastern sector. Only one gateway is evident, facing east. This has an inturned corridor, approximately 22.25 m long and 2.25 m wide Within the hillfort there appear to be the remains of eleven possible hut platforms. These tend to concentrate on the more sheltered eastern side of the hillfort, however the level of heather and bilberry growth within the hillfort would allow for other, more subtle features to exist. Archaeological Background No substantial archaeological work has been documented from the site (Williams, 2004), although the Clwyd-Powys Archaeological Trust carried out an Archaeological Condition Survey (Jones, 2004, 7 - 8) as part of the initial phase of the Heather and Hillfort Project. Also, in 2006 Engineering Archaeological Services Ltd were commissioned to carry-out a topographic survey of the hillfort (Brooks and Laws 2007). Aims of Survey To gather sufficient information to establish the location and extent of any archaeological features within the hillfort and, if possible, to characterise the archaeology located. SUMMARY OF RESULTS Both Fluxgate Gradiometer and Resistivity survey techniques were attempted within the hillfort, of which the Fluxgate Gradiometer survey proved to be most successful. A number of magnetic anomalies were recorded suggesting a level of occupation not previously recorded including the potential of twenty circular anomalies which may represent the positions of roundhouses. Two potential rectilinear structures were also recorded. Linear areas of relatively slight magnetic disturbance may represent the position of track ways within the hillfort. One of these would appear to have a series of anomalies along its edge which may be the positions of pits or large post-holes. The gateway and ramparts were sampled showing the ramparts have a distinct magnetic signature. The Resistivity Survey appears to largely be a response to the underlying geology, although one possible archaeological anomaly was recognised. 7). Interpretation plots are shown as Figures 5 and 8 and the data is summarised in Figures 6, 9 and 10. Methods The Fluxgate Gradiometer survey was undertaken using parts of twenty-two 20 x 20m grid squares laid out as in Figure 2. Readings were taken at 0.5 m intervals along transects 0.5 m apart. These transects were walked in a parallel pattern. Results: Fluxgate Gradiometer Survey (Figure 5) The grey scale plot of the Fluxgate Gradiometer survey shows large areas of magnetic disturbance which are assumed to be geological in origins and are shown in magenta on Figure 5. Despite this level of natural variability within the data the magnetic regime is relatively quite with one standard deviation of the readings only varying between -2.83 and 3.49 nT. This allows for the discrimination of a number of rather feint magnetic anomalies. The survey was carried out using a Geoscan FM 36 Fluxgate Gradiometer with a hand trigger. Grey Scale plots were produced using Geoscan Research “Geoplot” v.3.00e and X Y Plots using Golden Software “Surfer” v. 5.01. The Resistivity Survey used three of the same grid squares as the Fluxgate Gradiometer survey (Figure 3) using a Geoscan RM15 resistance meter with a MXP15 multiplexer. A twin probe setting was used with a separation between the probes of 0.5 m. Readings were taken at 0.5m intervals along transects 1 m apart. Grey Scale Plots were produced using Geoscan Research “Geoplot” v. 3.00e and X Y Plots using Golden Software “Surfer” v. 5.01. Crossing the plot is a linear anomaly (Figure 5, Anomaly A), approximately 2.5 m wide which is the magnetic response to the modern track crossing the site. At the eastern end this anomaly passes between two areas of magnetic disturbance (Anomalies B and C). These anomalies tend to have a line of high readings (up to 20 nT above the background) adjacent to the modern track and zones of lesser magnetic disturbance spreading away from the track. These anomalies correspond with the inturned rampart forming the gateway of the hillfort. A similar level of magnetic disturbance was also recorded where the survey area crosses the rampart elsewhere (Anomalies D and E). There is also an area of slight magnetic disturbance which can be related to the area of quarry hollows in the north eastern sector of the hillfort (Anomaly F). Complicating Factors The use of a hand trigger was partly dictated by the presence of deep heather and moderately steep slopes within the survey area. The heather depth varied, but reached knee height in places (Plate 1) making the use of a sample trigger impossible. The location of the Resistivity Survey was determined by the distribution of the area where the heather was shorter. Survey Results: Crossing the plot are three linear anomalies which are marked only by the lack of variability in the grey scale image. Anomaly G forms a loop south of the existing track across the site up to 5.5 m wide. Anomaly H is north of the track running in a SSW – NNE direction and is approximately 3 m wide. Anomaly I, loops from the northern side of the track to join the northern end of Anomaly H and is up to 4 m wide. All of these anomalies are shown Area The internal area of the hillfort is approximately 0.95 Ha of which 0.84 Ha was subjected to Fluxgate Gradiometer survey and 0.12 Ha to Resistivity Survey. Display The results are displayed as Grey Scale Images and as X-Y Trace Plots (Figures 4 and 2 AE – AL) varying in size from approximately 1 to 2.5 m in diameter which appear to represent a series of features aligned along the possible trackway (Anomaly I). It is not certain whether these anomalies represent possible post holes or pits. in green on Figure 5. The interpretation of these anomalies is uncertain, however, their distribution and form may suggest that they may be tracks or pathways within the hillfort. Within the plot are a series of seven, feint, circular anomalies (Anomalies J – P) of a size and form to suggest that they may represent the site of round houses within the hillfort. There are also fourteen further possible circular anomalies (Anomalies Q – AD) which may also be the sites of possible round houses; however, the anomalies are even feinter on the grey scale plots. Some, but not all of the circular anomalies can be related to topographical features assumed to be house platforms in the previous topographical survey (Brooks and Laws 207). The diameter of these anomalies is detailed below. Anomaly Diameter (m) J K L M N O P Q R S T U V W X Y Z AA AB AC AD 9.57 7.08 7.46 7.46 10.61 7.25 8.15 10.85 9.50 9.65 10.25 8.78 8.36 10.05 10.76 8.31 7.72 8.22 8.61 8.69 7.62 Almost centrally placed within the hillfort are two possible rectilinear anomalies. The clearest of these is Anomaly AM which is marked by a series of discrete anomalies forming a right angle at its south eastern end and an area of low variability within the grey scale plot. This forms an area of approximately 18.5 m x 8.6 m in size. Anomaly AN is much feinter, but appears to define a rectilinear area approximately 12.1 x 5.8 m in size. It is assumed that both of these anomalies may represent the positions of rectangular buildings within the hillfort. There are also four feint linear anomalies (Anomalies AO – AR) which generally do not appear to form any coherent pattern. The only exception is AR which appears to be related to Anomalies AK and AL, joining these two anomalies. Magnetic Susceptibility Given the nature of the site it was considered inappropriate to take soil samples in order to assess the magnetic susceptibility of the soils. The relative quiet nature of the magnetic regime and the possibility to discriminate feint anomalies would suggest that the conditions were reasonably good for magnetic survey. Resistivity Survey (Figure 8) The Resistivity survey was restricted to the areas of the hillfort where the heather coverage was at its lowest. The majority of the anomalies located would appear to relate to the underlying geology, although some of the anomalies may relate to possible archaeological features. The distribution of these circular anomalies would suggest that they cannot all be contemporary as they intersect with each other and other anomalies on the plot. Anomalies 1 and 2 (Figure 8) are areas of high resistance within an area of slightly enhanced resistance (Anomaly 3) which are assumed to be geological in origins. Anomalies 4 and 5 Apparently aligned along Anomaly I are a series of eight discrete anomalies (Anomalies 3 Conclusions may also be geological in origins, however, they also roughly correspond with two of the house platforms recorded in the topographic survey and may therefore relate to a spread of material from the destruction of these features. It is a fundamental axiom of archaeological geophysics that the absence of features in the survey data does not mean that there is no archaeology present in the survey area only that the techniques used have not detected it. Anomalies 6 and 7 are adjacent to or on the rampart of the hillfort and therefore relate to this feature. They sit, however, within a zone of slightly enhance resistance (Anomaly 8) which is assumed to be geological in origins. The Fluxgate Gradiometer survey of Moel y Gaer Llantysilio proved to be particularly successful in locating a series of previously unknown possible archaeological features and suggesting a level of occupation which was previously unexpected. The success of this survey is partly because of the level of resolution with which the survey was undertaken. It is suggested that a minimum resolution of 0.5 m samples along transects 0.5 m apart is required in order to map the possible feint magnetic anomalies within the hillforts of the project. Anomaly 9 is a zone of enhanced resistance which appears to have a distinct shape including a circular anomaly at its eastern end. This anomaly partly corresponds with a possible house platform recorded in the topographic survey to which it may relate. Only two low resistance anomalies have been located. Anomaly 10 is a slightly irregular area which appears to approximately relate to a possible round house recorded in both the topographic and Fluxgate Gradiometer surveys. If so it may be that the floor of the possible house is acting as a barrier to the local drainage. Anomaly 11 has no obvious origins and may therefore be geological in origins. In all twenty possible roundhouses have been located of which only eight appear to have any physical remains visible as topographic feature. Three other possible house platforms could not be related to magnetic anomalies. It is clear that not all these anomalies are contemporary as they intersect with each other and other anomalies within the plot. It is possible that there were internal divisions within the hillfort marked by possible tracks. The date of these features is uncertain as they are cut by possible round houses and one of the possible rectilinear structures. It is possible that the routes relate to one of the Iron Age phases within this developing site, although the correspondence between the alignment of the possible rectangular building (Anomaly AM and the possible track (Anomaly I) may suggest a later date for the tracks. It is clear that one of these possible tracks (Anomaly I) has a series of features aligned along it which may be a series of pits or large post holes. Of particular interest are the two rectilinear anomalies which are assumed to be the sites of rectilinear buildings. Buildings of this size and shape are unlikely to be Iron Age in date and may represent the re-occupation of the site in the post-Roman Early Medieval period. A few 4 hillforts in Wales show evidence of occupation in the Early Medieval period such as Dinas Emrys in Caernarfonsire in the fifth to seventh centuries AD and Degannwy Denbighshire) and Hen Gastell (Glamorganshire) in the ninth and tenth centuries AD (Arnold and Davies 2000, 159 – 162) and it is possible that Moel y Gaer Llantysilio is comparable with one of these. Acknowledgements The survey was commissioned by Samantha William for the Heather and Hillforts Landscape Partnership Scheme. She also organized transport up to the base of the hill and indeed drove the four by four for many of the days. Thanks are also due to the other members of Countryside Services who also drove the vehicle. References Arnold, C.J and Davies, J.L 2000. Roman and Early Medieval Wales. Sutton Publishing Ltd. Thrupp, Gloucestershire. Brooks, I.P. and Laws, K. 2007. Heather and hillforts survey report: Moel y Gaer Llantysilio. EAS Client Report 2007/01 Burnham, H. 1995. A guide to ancient and historic Wales. Clwyd and Powys. HMSO, London Jones, N. 2004. Heather and hillforts archaeological conditions survey. Clwyd-Powys Archaeological Trust Report 658. Williams, S, 2004 Heather and Hillforts Landscape Partnership. Desktop study of the hillforts. Unpublished report by Denbighshire Countryside Services 5 Techniques of Geophysical Survey: Instrumentation: Magnetometry: 1. Fluxgate Gradiometer - Geoscan FM36 This relies on variations in soil magnetic susceptibility and magnetic remenance which often result from past human activities. Using a Fluxgate Gradiometer these variations can be mapped, or a rapid evaluation of archaeological potential can be made by scanning. 2. Resistance Meter - Geoscan RM4/DL10 3. Magnetic Susceptibility Meter - Bartington MS2 4. Geopulse Imager 25 - Campus Methodology: Resistivity: For Gradiometer and Resistivity Survey 20m x 20m or 30m x 30m grids are laid out over the survey area. Gradiometer readings are logged at either 0.5m or 1m intervals along traverses 1m apart. Resistance meter readings are logged at 1m intervals. Data is down-loaded to a laptop computer in the field for initial configuration and analysis. Final analysis is carried out back at base. This relies on variations in the electrical conductivity of the soil and subsoil which in general is related to soil moisture levels. As such, results can be seasonally dependant. Slower than Magnetometry this technique is best suited to locating positive features such as buried walls that give rise to high resistance anomalies. For scanning transects are laid out at 10m intervals. Any anomalies noticed are where possible traced and recorded on the location plan. Resistance Tomography Builds up a vertical profile or pseudosection through deposits by taking resistivity readings along a transect using a range of different probe spacings For Magnetic Susceptibility survey a large grid is laid out and readings logged at 20m intervals along traverses 20m apart, data is again configured and analysed on a laptop computer. Magnetic Susceptibility: Variations in soil magnetic susceptibility occur naturally but can be greatly enhanced by human activity. Information on the enhancement of magnetic susceptibility can be used to ascertain the suitability of a site for magnetic survey and for targeting areas of potential archaeological activity when extensive sites need to be investigated. Very large areas can be rapidly evaluated and specific areas identified for detailed survey by gradiometer. Copyright: EAS Ltd shall retain full copyright of any commissioned reports, tender documents or other project documentation, under the Copyrights, Designs and Patents Act 1988 with all rights reserved: excepting that it hereby provides an exclusive licence to the client for the use of such documents by the client in all matters directly relating to the project as described in the Project Specification 6 Plate 1: Fluxgate Gradiometer Survey Conditions 7