Magnetic Gradient Survey at the M. S. Roberts (41HE8) Site in Henderson County, Texas

Repository Citation McKinnon, Duncan P.; Perttula, Timothy K.; and McKee, Arlo (2017) "Magnetic Gradient Survey at the M. S. Roberts (41HE8) Site in Henderson County, Texas," Index of Texas Archaeology: Open Access Gray Literature from the Lone Star State: Vol. 2017 , Article 15. https://doi.org/10.21112/ita.2017.1.15 ISSN: 2475-9333 Available at: https://scholarworks.sfasu.edu/ita/vol2017/iss1/15


Introduction
The M. S. Roberts site is located in Henderson County, Texas and it represents one of the few known Caddo mound sites in the upper Neches River Basin in northeast Texas (Figure 1). The site is recorded, the single mound at the site was approximately 24 m long and 20 m wide and roughly 1.7 m in height (Pearce and Jackson 1931). Directly west of the mound was a large depression, which has southern end of an elevated alluvial landform. October of the same year, archaeologists from the University of Texas began investigating the mound a surface collection from the site and excavated an unknown number of trenches in the mound where document that the mound began as a 25 cm deposit of yellow sand constructed on the undisturbed brown depth. Materials collected from the surface as part of the 1931 investigations indicate the presence of a Caddo habitation area surrounding the mound and suggest the site was occupied from the fourteenth to the landscape is part of a residential ranch development where landowners are stewards of the site with a focus on preservation and research.
In January 2015, with the permission of the landowners, renewed interested in the site began with a surface collection and the examination of the artifact collections from the 1931 work held by the Texas series of shovel tests and auger holes were then dug in the mound and surrounding habitation area in mid-2015. Shovel tests and auger holes documented organically-stained and charcoal-rich areas within the mound that were thought to represent the remains of several burned Caddo structures, and also the landform topography, estimate the extent of the current mound dimensions and borrow pit, and to erty surrounding the site at a 2 cm per pixel resolution. The aerial survey of the mound and surrounding landscape and the creation of a high-resolution digital elevation model reveal that the mound dimensions For example, aerial data document both the mound and borrow pit features and show that the mound spatial extent of archaeological deposits in the non-mounded habitation area and investigate the stratigraphy of mound deposits, identify cultural features in the mound, and hopefully obtain charred plant remains or unburned animal bones from these deposits for AMS dating.
To help evaluate and identify the distribution of cultural features in the mound and the surrounding - The magnetic gradient results document the subsurface location of at least two interpreted structures within the mound, the possible locations of three 1931 UT trenches, and several possible pit features proximate to the mound. The combination of aerial and geophysical data and the excavation results are revising our understanding of the archaeological remains and preservation conditions of the site.

Magnetic Gradient
The use of magnetic gradient at Caddo sites located throughout the Caddo people's ancestral lands within the current areas of East Texas, Southwest Arkansas, Northwest Louisiana, and eastern Oklahoma Magnetic gradient interpretations discussed herein are developed using a combination of inductive and deductive approaches. An inductive approach has roots in satellite and aerial image interpretation with form interpretable patterns (Wilson 2000). When anomalies in a geophysical dataset resemble patterns of regular geometric shapes, it can be induced that they are of probable cultural origin.
A deductive approach utilizes known physical properties of the subsurface matrix (artifacts, features, sediments, and soils recorded during excavation) to deduce how instrument sensors might respond and thus certain interpretative deductions can be made about the nature of anomalies revealed in the data. For example, thermoremanent magnetism is the result of burning events, which can produce an anomaly netic value may be deduced as being generated by a hearth, kiln, or a quickly burned house. A soil matrix that has been magnetically enriched through pedogenesis (induced magnetism and magnetic susceptibility) can also produce anomalies containing stronger magnetic values than those in the surrounding matrix structure may be deduced as being pits. Highest magnetic values are typically related to ferrous metal debris buried close to the surface, which can generate anomalies of extreme magnitude. Magnetic values 9 tesla).
gradiometer containing two cylindrical sensor assemblies. Each cylindrical sensor assembly contains two mounted sensors with a 1 meter vertical spatial separation that measure the vertical component of 3 ), the lower sensor is more sensitive to subsurface readings whereas the opposite upper sensor is more sensitive to Earth's Given that the Bartington instrument offers a vertical sensor separation of 1 meter, the sensitivity of the instrument is greatly increased and subsurface magnetic features more pronounced when compared to gradiometers with a shorter sensor separation (Bartington and Chapman 2004).
Data at the site was collected within established 20 x 20 m grids where a survey tape was pulled taut order to maintain 0.5 m spacing along the each grid baseline (X). Collection spacing along each transect (Y) was set to 0.125 m spacing (8 samples per m) and regulated using a focused and practiced walking pace of 1.3 m/second. Data were collected using a zigzag pattern.
of anomalies detected and the need for an organized way to describe and interpret dimensions, possible area are grouped and discussed as (a) those of historic origin (i.e., scattered metallic debris on the sur-preted to be of Caddo culture origin (i.e., circular structures and associated pits and hearths).

Anomalies of Historic Origin
Landscapes that have a long history of farming and agricultural activities are likely to contain numerbolts, barbs from wire fences, and other metal debris. With a magnetic gradient survey, ferrous metallic debris is recorded as a dipolar anomaly of both extreme high and low values in opposition (see Kvamme given that the site has a history of cotton cultivation and other forms of farming activities (Figure 4). Three extreme values represent upright t-posts that were unable to be moved prior to the survey. A long single linear magnetic anomaly to the north likely represents a buried utility conduit or irrigation pipe.

Anomalies of Archaeological Origin
As mentioned, a University of Texas crew excavated several trenches into the mound in the fall of 1931 (Pearce and Jackson 1931). The number, location, size, or extent of the trenches was not recorded during the work. However, magnetic gradient results suggest the location of three possible east-west trenches ( Figure 5). The 1931 Pearce and Jackson report suggests that, at a minimum, there was a trench to the north (herein referred to as Trench A) and one to the south (herein referred to as Trench B). A third possible trench is also suggested related to the "east and west portions" of the mound (herein referred to as Trench C). A 1931 photograph of excavations taken by A.T. Jackson indicates digging did occur on the mound summit in the area of the proposed Trench C.
When the proposed trench anomalies are evaluated using an estimated size of the mound, the suggested Trenches B and C coincide with the 1931 mound extent. This arrangement would have been and shovel… the only [method] possible with present funds" (Pearce and Jackson 1931). Metallic debris associated with proposed Trenches B and C suggest the presence of trash (broken shovels, lost trowels, proposed Trench A is north of the 1931 recorded mound dimensions and suggests that either the anomaly based on estimations of basal extent or earlier recording technologies.

Anomalies of Caddo Origin
As recorded in the 1931 excavations and during recent testing at the site, at least one burned structure is present. Shovel tests and auger holes also suggested other structures buried deeper in the mound The concentration of low magnetic anomalies (rather than high magnetic anomalies associated with high heat burning) is likely the result of the inadvertent removal of magnetically enriched topsoil through foot compaction, regular cleaning, and overall use within the house (see also McKinnon 2013). The removal of higher magnetic topsoil as a result of these activities would produce a concentration of the remaining low magnetic soil matrix recorded in the survey. Interestingly, the proposed extent of the north structure overlaps the 1931 mound dimensions, yet is within the recently recorded mound dimensions. This arrangement offers additional consideration regarding the accuracy of the 1931 mound measurements, or it could represent a structure in which the mound was constructed over only a portion.
within each structure and a third (in the north structure) is situated to the northwest. Additional small high magnetic anomalies within each proposed structure may represent former hearths, internal storage pits, or possible architectural pits. Numerous high magnetic anomalies proximate to each of the three proposed trenches are suggested as the result of soil movement while digging the 1931 trenches.
mound, (b) interpreted mound structures and associated anomalies.
Recent testing documents the remains of a shallowly buried burned structure on the north portion of the mound (see Perttula et al. article in this volume). The location of the burned structure (Unit 1) is -AMS dating of samples of unburned deer teeth from Unit 1 and nutshell from Unit 2 have been obtained. The median 2 sigma calibrated age range of the two samples is A.D. 1294-1405. While the mottled remains in Unit 1 certainly represent a burned structure, it is worthwhile to note indicative of wattle and daub architecture. The disposal of a wattle and daub structure would have burned smolder and create a high magnetic signature.
The low magnetic signatures, lack of large chunks of daub, and mottled charred remains associated with the north structure in Unit 1, suggest the structure was a thatch style structure with little or no wattle and daub. Such a structure would burn very rapidly within a short time span of intensive heat and leave onstrate that the thatch structure was burned, but the lack of wattle and daub reduces the potential for

Conclusions
In closing, recent research at the M. S. Roberts site has provided new insights into one of the few known Caddo mound sites in the upper Neches River Basin in northeast Texas. The combination of provided a more informed set of interpretations of mound construction, historic landscape change, and density of off-mound occupations.
mound with the interpretation of at least two structures in or under the mound. Additionally, the possible location of the UT trenches allows for spatiality to be established related to the 1931 excavations and a better analysis of where the UT crew focused their efforts. Magnetic gradient has also provided insight into the distribution of off-mound anomalies and their arrangement relative to the mound. Shovel and auger testing and excavations at the site have corroborated magnetic gradient interpretations associated with the presence and type of mound structures, suggesting thatch-style construction rather than wattle and daub. AMS dates from material collected during excavations demonstrate the site was in use from at least A.D. 1294 to 1405. While still more work is necessary to fully evaluate the relationship of the M. S. Roberts site to contemporaneous sites situated in the upper Neches River Basin, the recent revisit and