RemoteSensingSurveyPreliminaryReportDillardArchaeologicalSite,CrowCanyon,CO

Figure 1 Dillard archaeological site and geophysical survey interpretations, June 2012.

Submitted by: Meg Watters, PhDCo-PI, Remote Sensing & Visualization CoordinatorOregon Public Broadcasting

Table of Contents

Overview .......................................................................................................................................................5

Introduction ..................................................................................................................................................7

Site Control Survey........................................................................................................................................ 7

Geophysical Methods, Principles, and Equipment .......................................................................................9

Magentometry ..............................................................................................................................................9

Conductivity / Magnetic Susceptibility .......................................................................................................11

Resistance ...................................................................................................................................................13

Geophysical Data Interpretations ...............................................................................................................16

Magentometry ............................................................................................................................................18

Resistance ...................................................................................................................................................28

Conductivity / Magnetic Susceptibility .......................................................................................................33

Conductivity ........................................................................................................................................33

Magnetic Susceptibility .......................................................................................................................38

Discussion of geophysical survey results ....................................................................................................43

Airborne LiDAR Principles and Results........................................................................................................51

Conclusions and Recommendations...........................................................................................................52

Acknowledgements and Credits .................................................................................................................53

Time Team America ....................................................................................................................................53

References ..................................................................................................................................................54

Table of Figures

Figure 1 Dillard archaeological site and geophysical survey interpretations, June 2012. ............................1Figure 2 New site features identified through geophysical surveys.............................................................6Figure 3 Geophysical survey areas covered during Time Team America project. ....................................... 8Figure 4 The magnetic anomaly produced by a kiln is aligned to the dip and direction of the Earth’smagnetic field (From Clark 1996)..................................................................................................................9Figure 5 Duncan McKinnon with the Bartington 601 dual array fluxgate gradiometer. ............................11Figure 6 Electromagnetic induction diagram.............................................................................................11Figure 7 Bryan Haley with the EM38 conductivity meter. ..........................................................................13Figure 8 The flow of current from a single current source and resulting potential distribution...............14Figure 9 A general four electrode array......................................................................................................14Figure 10 The Twin-electrode array commonly used in archaeology........................................................15Figure 11 Duncan McKinnon with the RM 15 resistivity meter and electrode array. ...............................16Figure 12 Geophysical survey areas at Dillard site. ....................................................................................17Figure 13 Magnetic gradient survey results...............................................................................................19Figure 14 Magnetic anomalies caused by iron stakes and/or nails are seen as mono-poles (A) or di-poles(B) with an orientation to magnetic north..................................................................................................20Figure 15 Magnetic survey results with site surface features. ..................................................................22Figure 16 Interpreted magnetic survey results (the red arc identifies part of the Great Kiva berm)........23Figure 17 Interpreted magnetic survey results with site surface features. ...............................................24Figure 18 Auger positions for ground-truthing geophysical survey anomalies. ........................................26Figure 19 New site features mapped through geophysical surveys and ground truthed. Pit house and pithouse like features are purple polygons and blue points identify pits associated with a potential fence.27Figure 20 Resistance survey, Dillard. .........................................................................................................29Figure 21 Interpreted resistance survey. ...................................................................................................30Figure 22 Interpreted resistance survey with overlain site features. ........................................................31Figure 23 Interpreted resistance survey with overlain ground-truthed archaeological features identifiedduring geophysical surveys. ........................................................................................................................32Figure 24 Conductivity survey results. .......................................................................................................34Figure 25 Interpreted conductivity survey results. ....................................................................................35Figure 26 Interpreted conductivity survey results with site features........................................................36Figure 27 Interpreted conductivity survey with ground-truthed archaeological features........................37Figure 28 Magnetic susceptibility survey results. .......................................................................................39Figure 29 Interpreted magnetic susceptibility survey results....................................................................40Figure 30 Interpreted magnetic susceptibility survey results with site features. .....................................41Figure 31 Interpreted magnetic susceptibility survey results with overlain ground-truthed archaeologicalfeatures. ......................................................................................................................................................42Figure 32 Comparison of magnetic gradient (A & C) and resistance (B &D) big pit house (red) andpossible ritual pit house (yellow) anomalies. .............................................................................................43Figure 33 Contrasting soils (red and dark brown) can be seen that define the edge of the big pit housefeature. (Image courtesy of Crow Canyon Archaeological Center.) ..........................................................44

Figure 34 The floor of a section of the possible ritual pit house feature with a sipapu, ritual fire circle,and ash pit. (Image courtesy of Crow Canyon Archaeological Center.)....................................................45Figure 35 Interpretations of all the geophysical survey methods at Dillard..............................................46Figure 36 Core samples over geophysical anomalies help identify cultural (red points) and non-cultural(yellow points) site features. ......................................................................................................................47Figure 37 Ground-truthed archaeological features (purple are pit houses or pit house like structures andnavy blue are individual pits) overlain on site interpretations and surface features. ................................48Figure 38 Mapped site features overlain on interpretations of geophysical surveys. ..............................49Figure 39 Pit house anomaly comparison. The pit house features as ground-truthed are positioned inthe center of each sample above. ..............................................................................................................51Figure 40 LiDAR DTM and the broader landscape with BM III site distribution in reference to the Dillardsite (red)......................................................................................................................................................52

OverviewTime Team America joined the Crow Canyon Archaeological Center (CCAC) in June 2013 to investigatethe Dillard site, a community center during the Basketmaker III (BM III) period, A.D. 500-725. The TimeTeam America challenge at the Dillard site was to (1) try to determine the site population, (2) to betterunderstand why a Great Kiva was built here, (3) to gain insight into the organization of the site, (5) itscontext to the broader landscape, and (6) what this meant for the development of community.

BM III settlements cannot be identified or analyzed from the ground surface, they simply are not visible.Over the past three years CCAC investigations have identified eight pit structures to the south of thegreat kiva through systematic auguring, excavation, and a small amount of resistivity. In the three daysof geophysical surveys conducted by Time Team America that included magnetic gradient, conductivity,and resistivity surveys (Charles 2012), an additional eight to nine pit structures and a variety of otherpit-like anomalies were identified in the area to the north of the Great Kiva (Figure 2). Ground-truthing(excavation and auguring) geophysical anomalies helped provide material that was able to provide datesthe site, from A. D. 610 to 670. One pit structure identified through geophysical surveys was partiallyinvestigated and revealed a suite of ritual features, often found in BM III structures, in this instance theorientation, formality of construction and closing the of the features suggests a relationship to the greatkiva. Auguring of selected single pit anomalies suggests a fence line feature that runs between the greatkiva and cluster of pit structures to the north.

Ground penetrating radar was tested on different parts of the site and ruled out as an effectivetechnique due to the soil properties and failure of GPR to record any useful information. (Part of thistesting was conducted over the balk of the excavation trench at the Great Kiva where the stonefoundation of the kiva should easily have been recorded if ground conditions were amenable.)On May 9, 2012 an airborne LiDAR survey was conducted by Paul Kinder and Adam Riley from theNatural Resource Analysis Center, West Virginia University. As part of the flight, Meg Watters (TimeTeam America, Oregon Public Broadcasting, OPB) and Shanna Diedrerichs (site director, CCAC) werefilmed in the air discussing the Dillard site, LiDAR technology, and what might be captured in the LiDARdata. The goal of the airborne LiDAR survey was to attempt to identify possible site features and tocontribute to a broader landscape perspective for interpreting the Dillard site.

Figure 2 New site features identified through geophysical surveys.

IntroductionGeophysical survey methods of the sub-surface and 3D laser scanning of existing environments providecost-effective means for capturing archaeological information for site recording, investigation, andmanagement. Using non-invasive sub-surface and surface mapping methods can document the basicstructure and layout of site. In instances where historic properties are active sites with maintenanceand potential development impact demands, these methods can guide placement of expensiveexcavations and contribute to site impact strategies when dealing with upgrade of site infrastructure(such as utilities, and landscape management); thus providing large cost savings while reducingdestructive impact upon important archaeological remains.

Geophysical survey methods can provide primary information on site settlement patterns. Thecontinued application and development of broad area coverage for archaeological assessment hasbegun to introduce an alternative perspective into regional, or landscape archaeology (David and Payne1997; Kvamme 2003). Because geophysical surveys are able to cover large areas in comparison to thelimited extent of archaeological excavations, the information they provide introduces a new componentto the concept of the archaeological landscape. Broad area geophysical surveys provide information onthe structure and organization of a site enabling the study of spatial patterns and relationships relevantto research questions. In addition to the large-scale perspective of the site, geophysical survey resultsalso provide a high-resolution focus on individual site features.

Geophysical surveys measure different subsurface properties at regular intervals across broad areas.Contrasting properties in a relatively homogeneous soil can identify buried objects or features such asfoundations, compacted earthen surfaces, pits, stone walls, middens, hearths and any number ofarchaeological features. The different physical properties of the features, measured either in contrast totheir surrounding matrix, or as recorded at the surface are referred to as ‘anomalies’ until they are ableto be ground-truthed through excavation or other methods such as soil coring.

Different geophysical methods are sensitive to specific properties, such as magnetic fields, or the flow ofan electrical current in the earth. Employing a combination of methods over a survey area can helpprovide information as to the nature, or material, of an anomaly, thus providing insight for siteinterpretation. Mapping the distribution of anomalies over a large area can help in the recognition ofanomalies generated through cultural activities revealing the spatial distribution and association withsite features (Kvamme 2003).

Geophysical surveys can provide important information for help in site planning and preservation.These non-invasive methods can help establish priorities and identify areas for further invasiveinvestigations, or for preservation and management. They are a fast and cost-effective method forgaining insight to what is buried beneath the ground. Geophysical survey results can be spatiallyintegrated with other data relevant to archaeological investigations to provide a comprehensive recordof the site environment, both below, and above ground.

Site Control SurveyThe Time Team America geophysical survey grids were established by CCAC surveyors and tied into theirsurvey control. Due to the ground cover and inherent data collection rate for different geophysicalsurvey methods, coverage was different for each method. Magnetic gradient survey covered the entireresearch area targeted (north of the Great Kiva to site fence boundary), while conductivity / magnetic

susceptibility and resistance surveys covered smaller areas (due to method and rate of data collection).The resistance survey was conducted by a group from the Center of Southwest Studies at Fort LewisCollege, directed by Mona Charles. The survey results from their work during the 3 day period of theTime Team America project are presented here. Additional site coverage of this and other related sitescan be read in Charles, M.C., 2012. Electrical Resistance Survey of Three Sites in the Indian Camp Ranch,Montezuma County, Colorado; Report prepared for Crow Canyon Archaeological Center Cortez,Colorado.

Figure 3 Geophysical survey areas covered during Time Team America project.

Geophysical Methods, Principles, and Equipment

MagentometryMagnetometers are passive instruments that measure the magnetic field strength a specific location onthe surface of the Earth. The Earth’s magnetic field varies depending on location relative to the earth’sequator and can be visualized as a large bar magnet that is tilted 11 degrees from the axis of rotation(Heimmer and Devore 1995). Over a small area and in homogeneous soils, the magnetic field is expectedto be uniform (Weymouth 1986). A subsurface target can be detected with magnetic survey as adeviation from this background field reading. The resultant anomaly often has a dipolar form alignedwith the dip and direction of the Earth’s field (Figure 4). The most common unit of measure is thenanoTesla (nT).

The magnetic signal of a target is composed of two parameters: induced and remnant magnetism(Reynolds 1997). A magnetometer measures the remnant magnetism of a target, which is permanentand may be caused by the presence of highly magnetic rock compounds or thermal alterations to soilswhich have high iron content (Heimmer and Devore 1995). Magnetization caused by thermal alterationis called thermoremanence and it occurs at maximum expression at temperatures above about 600degrees Celsius, but there is some effect at any elevated temperature (Aitken 1964).

Figure 4 The magnetic anomaly produced by a kiln is aligned to the dip and direction of the Earth’s magnetic field (FromClark 1996).

Induced magnetism is only visible in the presence of magnetizing field. However, the Earth serves as aconstant magnetizing agent and, therefore, it can be sensed by a magnetometer. The inducedmagnetism is generally referred to as magnetic susceptibility. Magnetic susceptibility is greater in thetopsoil and soils that are organically rich, but often produces relatively subtle anomalies (Clark 1996).Therefore, excavations that rearrange the topsoil are sometimes evident in magnetic surveys, but theseare rather weak in strength. The Geonics EM38B conductivity meter can better measure the inducedmagnetism of the ground.

Magnetic anomalies produced by archaeological targets are often much weaker than signals producedby other sources, usually between 1 nT and 100 nT (Aitken 1961). However, anomalies produced byhistoric period targets are usually much greater than this range. Archaeological objects that may

produce magnetic anomalies include fireplaces, furnaces, burnt clay floors, hearths, kilns, daub, bricks,and walls composed of magnetically anomalous rocks such as basalt (Aitken 1964; Hasek 1999).

Another type of target visible magnetically is ferrous, or iron containing materials (Aitken 1964).Archaeological targets such as historic nails can sometimes be mapped using magnetometers. However,more recent ferrous objects, such as power lines, cars, buried pipes, and surface trash, can easilyobscure archaeological targets (Heimmer and De Vore 1995). Some advantages to the use of fluxgateinstruments are their relative insensitivity to steep magnetic gradients and their speed of acquisition isbetter (Reynolds 1997). Fluxgate instruments have become the workhorse for archaeologicalgeophysical survey in Britain and the United States (Clark 1996).

The magnetic gradiometer was developed in the 1990s and uses two sensor heads. The primaryadvantage of a gradiometer system is that no correction for diurnal drift is necessary (Reynolds 1997,Bevan 1998). In addition, they are much less affected by nearby objects with steep magnetic gradients,such as large masses iron (Bevan 1998). Also, gradiometers tend to emphasize shallow anomalies, abenefit for archaeological survey. One disadvantage is that the accuracy is dependent on a consistentorientation of the sensors (Bevan 1998, Hasek 1999).

Interpretation of magnetic imagery begins by identifying anomalies, which may have strong high andlow amplitude values (Bevan 1998). Next, metal objects can be identified from the shape and amplitude.Anomalies with strong, narrowly spaced dipoles or strong monopoles are usually produced by ferrousmetal objects. If targets are relatively large and the amplitude is not extreme, the shape may beapproximated in the magnetic imagery (Bevan 1998).

Little information about the depth of a target is obtained with magnetic survey. In some cases, the half-width rule can be used to estimate target depth. The half- width rule depends on the amplitude drop offfor readings over a target and assumes a simple and regular target shape (Bevan 1998). However, exceptfor buried iron targets, this technique is often not useful for archaeological targets.

The Bartington 601 fluxgate gradiometer was used for the magnetic survey (Figure 5).Magnetometry survey parameters were:0.125 m sample rate0.5 meter transect spacingZig-zag data collection method (survey grid SW corner to grid NE corner)

The magnetic survey data were processed using Geoplot 3.0. Processing techniques included de-spiking,grid/transect mean zeroing, 3 x 3 low pass and 10 x 10 high pass filtering. Once processed, data wereinterpolated along the x axis.

Figure 5 Duncan McKinnon with the Bartington 601 dual array fluxgate gradiometer.

Conductivity / Magnetic SusceptibilityElectromagnetic (EM) induction instrumentation uses a near surface transmitter coil to emit radiofrequency electromagnetic waves into the subsurface. Objects in the subsurface respond by generatingeddy currents, producing a secondary electromagnetic field (Figure 6). This secondary electromagneticfield is proportional to conductivity and detected by a receiver coil on the instrument and recorded byan attached data-logger (Bevan 1983; Clay 2006).

Figure 6 Electromagnetic induction diagram.

The Geonics Limited EM38B was used in the survey (Figure 7) and allows for simultaneous collection ofboth quadrature-phase (electromagnetic conductivity) and in-phase (magnetic susceptibility)components. Electromagnetic conductivity measures the “ability of the soil to conduct an electriccurrent” (Clay 2006) and is recorded in siemens (mS/m). Theoretically, electromagnetic conductivity isthe inverse of resistivity although methods for recording each are completely different (voltage, samplespacing, soil, volume, sensitivity to metals) and results may not match entirely. The transmission of thequadrature-phase component of the induced electromagnetic field signal is related to the mineral andchemical composition of the soil. Soils high in clay and/or saline composition will produce higherconductivity measurements; whereas soils composed of sand and/or silt will produce a lowerconductivity measurement. Levels of soil moisture also have a dramatic impact on conductivitymeasurements where increased moisture will cause higher conductivity readings (Clay 2006).

Magnetic susceptibility measures “a material’s ability to be magnetized” (Dalan 2006). It isdifferent from magnetic gradiometry in that susceptibility is an active measurement recorded in thepresence of an induced magnetic field. The transmission of the in-phase component of the inducedelectromagnetic field is based on the presence of a magnetic topsoil matrix being greater in magnetismthan proximate soil matrix or materials. The increase in magnetism in topsoil is the result of pedogenesisenhancement from hematite, magnetite and maghematite minerals. Additionally, changes to themagnetic composition of the soil can be caused by human activity, such as fire or the movement ofmagnetically rich topsoil (Dalan 2006).

Both quadrature phase and in phase readings were simultaneously collected for each station, relating toconductivity and magnetic susceptibility properties respectively. This specification results in a maximumdepth sensitivity of about 1 m for the conductivity. For the magnetic susceptibility, the penetration issignificantly shallower.

Conductivity survey data sampling:2 samples per meter0.5 meter transect spacingParallel data collection method (all transects travel grid south to north)

The EM data were processed using Geoplot 3.0. Null values were added in a text editor so that gridlengths and widths were in multiples of 10 meters and these were used to create a single compositedata set.

Data processing methods include a despike operation and a 3X3 low pass, as well as the addition of a 10X 10 high pass filter to a second version. The magnetic susceptibility data was processed in a similarfashion, without the creation of the second high pass filtered version.

Figure 7 Bryan Haley with the EM38 conductivity meter.

ResistanceResistance survey was conducted on site by a group from Fort Lewis College led by Mona Charles.Results of the survey conducted during the Time Team America filming are integrated into this report.After filming, additional areas of the site were surveyed with resistivity, results are presented in Charles,M. 2012. Electrical Resistance Survey of Three Sites in the Indian Camp Ranch, Montezuma County,Colorado. Report prepared for: Crow Canyon Archaeological Center, Cortez, CO.

Resistance survey is designed to measure the electrical resistance of the earth in order to provideinformation on the subsurface structure. The electrical properties of the earth are recorded as afunction of depth and / or horizontal distance. An electrical current is introduced into the earth throughelectrodes and the resulting potential distribution is sampled at the ground surface. The measuredapparent resistivity provides information on the magnitude and distribution of the electrical resistivitiesin the volume of the sampled subsurface (Griffiths and King 1981).

An electric current is caused by the flow of charged particles and is measured in amperes (amps).Amperage expresses the amount of charge that passes any point in a circuit in one second. Ameasurement of the ground resistivity is made by passing an electrical current into the ground throughan electrode acting as the current source (Figure 8). A second electrode, or current sink, enables theelectrical current to exit from the ground completing the circuit. The current flows into the earth in alldirections from the source electrode.

Figure 8 The flow of current from a single current source and resulting potential distribution.

The most common electrode configurations are linear arrays that contain two current electrodes (A andB) that are the current source and sink of equal strength, and two potential electrodes (M and N) thatmeasure the difference in potential between two points (Figure 9).

Figure 9 A general four electrode array.

If the ground is inhomogeneous and a fixed electrode array is moved or the electrode spacing is variedduring survey, the calculated resistivity will vary for each measurement. The resistivity of the earth canvary greatly depending on the composition and structure of the material and ground water saturation.Not only does resistivity vary with rock formations, it also varies from deposit to deposit and on a macroscale within individual deposits depending upon their structure. Resistivity values can vary greatly due tothe unconsolidated nature of near-surface materials. The principles provided for basic rock formationscan be followed when considering the structure of the near surface and resistivity mapping forarchaeological applications (Griffiths and King 1981).

The nature of the archaeological features, the mineral content and compaction of soils in which they areburied, and the saturation levels of the subsurface all affect earth resistivity. The saturation of thesubsurface is dependent on rainfall, soil composition and compaction and subsequent percolation rates,evaporation rates, and water take-up through the roots of vegetation. Weather and geologicalconditions impact on the effectiveness of resistance surveys in archaeological applications and dictatecareful consideration of resulting data (Clark 1996).

A number of electrode arrays are used in resistance surveys. The array, or configuration, refers to thearrangement of electrodes. Linear arrays, which are used more commonly, consist of two currentelectrodes (A and B) and two potential electrodes (M and N). The twin-electrode array is the mostpopular for archaeological surveys. Due to the relative speed of data collection, the benefits of theresulting survey include a high lateral resolution and depth of investigation relative to the spacing of themobile electrodes (Apparao et al. 1969; Apparao and Roy 1971). The basic twin electrode array used inarchaeological applications can be seen in Figure 8 where single current (A) and potential (M) electrodesare set with a fixed distance (a) with the second pair of electrodes (B and N) are placed at a distance 30times the spacing (a) of the primary electrodes (A and M) and fixed separation distance (a) the same asthe mobile probe spacing (Figure 10).

Figure 10 The Twin-electrode array commonly used in archaeology.

The depth of investigation can be defined as the depth at which a thin horizontal layer makes themaximum contribution to the total measured signal at the surface (Barker 1989; Evjen 1938; Roy andApparao 1971; Roy 1972). The separation distance and positions of the current and potential electrodesfundamentally contribute to calculating the most accurate depth estimation. The depth of investigationof electrode arrays should be the depth with which a measurement of apparent resistivity is bestassociated. Although there is no single depth of investigation, a single value is more useful to have as areference. The most practically useful value is the median depth (Edwards 1966; Barker 1989). Themedian depth is defined as the depth from below which and from above which 50% of the signaloriginates.

The RM15 resistivity meter was used for survey at the Dillard site (Figure 11).

Figure 11 Duncan McKinnon with the RM 15 resistivity meter and electrode array.

Geophysical Data InterpretationsThe geophysical survey area was focused to cover the Dillard site from the Great Kiva to the northernperimeter of the area bounded by a fence. The fence was removed to enable effective survey withinstruments sensitive to iron. Figure 12 shows the area surveyed by Time Team America. Each dataimage in this report has a key that defines map layers. EM survey area refers to ElectromagneticInduction survey, or conductivity and magnetic susceptibility. Other abbreviations in the key willinclude: mag – magnetic gradient survey; res – resistance survey; cond – conductivity survey; and magsus – magnetic susceptibility survey.

Site features included in some of the data maps were provided by Crow Canyon Archaeological Center.

Figure 12 Geophysical survey areas at Dillard site.

All of the geophysical survey results were imported to a project GIS using ArcMap 10.1. Data arerectified into the GIS project and polygon files are created to identify and map interpreted anomalies.Data results are presented below with and without interpretations. This is done so that the client maylook at the data and consider what they may see based upon their viewpoint and expertise.

MagentometryThe magnetic gradient survey mapped 8 to 9 new pit house like structures, a number of individual pitsand numerous anomalies that may represent additional buried archaeological features. Figure 13 showsthe results of the magnetic gradient survey, magnetic gradient values range from – 3.56 (white) to 3.39(black) nT.

Figure 13 Magnetic gradient survey results.

When interpreting data it is vital to consider surface features and the signatures that they mightintroduce to the final data maps. A number of things can contribute to the data maps including surfacebrush or trees, these may knock sensors out of alignment and introduce data spikes. While attempt tofilter out data spikes, this site had significant obstacles to survey around and through, some area surfacefeatures having more impact on the final data than others. Piles of surface rock, midden, or backfill canalso appear in the final data plans depending on their geophysical values. Having the surface featureinformation for the Dillard site complements the data interpretation process but all interpretations mustbe assessed visually in the field to see if there may be additional surface features that could be thecause of potential anomalies.

Figure 15 (and subsequent surface feature overlays on other survey methods) shows the site featuresoverlain on the geophysical map of the magnetic gradient survey. The windrows stand out clearly in themagnetic data as strong, linear magnetic anomalies. Iron stakes or nails also stand out clearly as veryhighly contrasting black and white anomalies, either as monopoles (Figure 14, A), or as dipoles orientedto magnetic north (Figure 14, B).

Figure 14 Magnetic anomalies caused by iron stakes and/or nails are seen as mono-poles (A) or di-poles (B) with anorientation to magnetic north.

Figure 16 shows the final interpretation for the magnetic gradient survey (Figure 17 showsinterpretations with overlain surface features). Interpretations were divided into three categories:individual anomalies, mag_AOI (general areas of interest), and pit house – AOI (specific areassurrounding the pit houses). The individual anomalies are interpreted over many positive (black) andnegative (white) anomalies, as well as areas of varying magnetic field strength. Because we ground-truthed a sample of individual magnetic anomalies that turned out to be single pits, interpretations inthis report include many, if not most, of the point anomalies in the survey data. I felt that I could notskip over any, as pit fill may vary, thus the magnetic field strength may vary from pit to pit. There maybe a few instances where some changes in the data are not annotated. The line had to be drawnsomewhere with the interpretations, thus what is included in these maps are to the best of my abilitytaking all information into consideration (surface features, ground-truthed features, possible rodentburrows, etc…).

Areas of interest are identified through what appears as a ‘clustering’ of anomalies, these areas shouldbe closely considered on the ground surface and may represent activity areas. Areas of interestassociated with the pit houses suggest possible activity that may be related to the pit house andactivities that may be associated with them. There are many nuances within this data in particular and

A B

an attempt to identify different categories for further consideration was made. Continued ground-truthing will greatly assist in a more intuitive interpretation of these data and hopefully, this can be acase study for investigation of similar sites in the region.

(Though not interpreted in the GIS, the Great Kiva berm is clearly visible in the magnetic gradient data,Figure 16, red line.)

Figure 15 Magnetic survey results with site surface features.

Figure 16 Interpreted magnetic survey results (the red arc identifies part of the Great Kiva berm).

Figure 17 Interpreted magnetic survey results with site surface features.

As part of Time Team America, geophysical data are used in part to select areas to excavate in the 3 dayformat. Because of the excellent results of the magnetic survey, a series of excavation units, coringtransects, and individual core locations were identified to help characterize what the magnetic survey(and additional surveys as their results were investigated) data was mapping. Figure 18 shows most ofthe cores / augur locations that were conducted. Yellow points identify non-cultural samples and redpoints identify cultural samples that were retrieved through coring. Figure 19 shows the finalinterpretation of the archaeological features mapped through the geophysical surveys and ground-truthed through coring and excavation.

The identification of pit house (and pit house like) structures through the geophysical surveys (Figure 19,dark purple) begins to reveal the distribution of structures and helps estimate the population of the site.As part of the ground-truthing, anomalies that appeared as a double ‘ring’ of magnetic points encirclingseveral pit structures to the north of the Great Kiva were sampled; preliminary results identify four ofthese point anomalies (navy blue) as pits and thus, are interpreted by site archaeologists as analignment of postholes that would have been associated with a fence. This reveals not only informationon the organization of space but also begins to provide insight to social and community organization.Following on the research by Kvamme at Huff Village (2003), further investigation of individual magneticanomalies may provide insight to much more information related to site population, the season that itwas abandoned and more.

Figure 18 Auger positions for ground-truthing geophysical survey anomalies.

Figure 19 New site features mapped through geophysical surveys and ground truthed. Pit house and pit house like featuresare purple polygons and blue points identify pits associated with a potential fence.

ResistanceThe resistance survey covered (Figure 20) a limited part of the entire survey area but clearly mapped(Figure 21) pit house and pit house like structures, as well as provides a number of anomalies for furtherinvestigation. Figure 22 presents the surface features overlain on resistance data and interpretations.Resistance data values range from 8.42 (white) - 29.63 ohms (black).

Resistance survey mapped 4 of the 6 pit house / pit house like features (in the survey area it covered)identified through magnetic survey and coring (Figure 23, purple polygons). Resistance survey did notidentify any of the single pit anomalies that were mapped through magnetic gradient survey, ground-truthed and interpreted as pit features representing a fence (Figure 23, navy blue points).

Figure 20 Resistance survey, Dillard.

Figure 21 Interpreted resistance survey.

Figure 22 Interpreted resistance survey with overlain site features.

Figure 23 Interpreted resistance survey with overlain ground-truthed archaeological features identified during geophysicalsurveys.

Conductivity / Magnetic SusceptibilityConductivity (Figure 25) and magnetic susceptibility (Figure 28) surveys map a number of anomalies thatmay be related to archaeological features. The nature of these surveys may reveal more subtle featuresthat relate to areas associated with archaeological structures (as revealed through resistivity andmagnetic surveys). It is important to remember the issues of ground surface obstacles as discussedearlier in the report, the EM38B conductivity meter is sensitive to instrument orientation and heightabove the ground surface. If knocked off of axis or lifted to avoid surface obstacles, artifacts may beintroduced into the final data, visual inspection of conductivity and magnetic susceptibility plan mapsand the survey surface are recommended when reviewing data.

ConductivityConductivity survey maps 4 of the 7 pit house (pit house like) features as ground-truthed in the fieldwith Time Team America. It is interesting to note that the conductivity survey (individual anomalies andArea of Interest) are located adjacent to the pit house (pit house like) features. These areas ofcontrasting conductivity values may represent deposition and / or compaction as a result of activitiesthat might (because of vicinity) be related to the pit house (pit house like) structures. Conductivityvalues range from 10.2 (white) to 12.33 (black) mS/m, a very fine threshold.

Figure 26 shows conductivity data interpretations. Note the purple polygons that are labeledcond_BPH_InteriorFeature. In this instance, I was interpreting the conductivity data in isolation,attempting not to be influenced by the anomalies interpreted (and ground-truthed) in the othergeophysical survey methods. However, knowing that magnetic gradient and resistance surveys bothmapped a ‘big’ pit house (BPH) feature, I wanted to see how this appeared in the conductivity data. Isaw a higher conductivity anomaly in the location of the big pit house and within this anomaly, smallerindividual areas of higher contrasting conductivity values. The thought was to try to interpret featureswithin the actual pit house (as Kvamme did at Whistling Elk, interpreting the central fire hearth andother pit house features in the magnetic gradient data, 2006). But, when compared to the magnetic andresistance survey data, it appears that the conductivity polygon that ‘defined’ the area of the big pithouse structure overlapped with the footprint, but also continued to the northwest of the pit house. Itis in this area adjacent to the pit house that the individual high-conductivity point anomalies areidentified (cond_BPH_InteriorFeature). It will be interesting to investigate this area more to see what, ifany, features might be discovered and how they might relate to the big pit house.

Figure 27 shows that conductivity data did not map the individual pits that were mapped throughmagnetic gradient survey and are interpreted to indicate a fence line.

Figure 24 Conductivity survey results.

Figure 25 Interpreted conductivity survey results.

Figure 26 Interpreted conductivity survey results with site features.

Figure 27 Interpreted conductivity survey with ground-truthed archaeological features.

Magnetic SusceptibilityMagnetic susceptibility survey identified anomalies at 5 of the 7 pit house (pit house like) features in itssurvey area and a number of additional areas of interest for further investigation. Some surfacefeatures such as the windrow and foot path are clearly visible in this data. The pits identified throughthe magnetic gradient survey that define a possible fence line are not mapped through magneticsusceptibility survey. The data values for the magnetic susceptibility survey range from 0.13 (white) to0.50 SI (black).

Figure 28 shows the results of the magnetic susceptibility survey; interpretations are shown in Figure 29.Figure 30 shows the site surface features overlain on the interpreted magnetic susceptibility results andFigure 31 overlays the ground-truthed archaeological features identified through ground truthing of thegeophysical anomalies.

Figure 28 Magnetic susceptibility survey results.

Figure 29 Interpreted magnetic susceptibility survey results.

Figure 30 Interpreted magnetic susceptibility survey results with site features.

Figure 31 Interpreted magnetic susceptibility survey results with overlain ground-truthed archaeological features.

Discussion of geophysical survey resultsTwo features that appeared as strong anomalies in the resistance and magnetic gradient data wereselected for further investigation on the first day of Time Team America’s project; the big pit house anda second pit house or pit house like structure. Figure 32 shows the survey results (A magnetometry, Bresistance) with the outlined pit house anomaly (big pit house in red, possible ritual pit house in yellow)in each method (C magnetometry, D resistance.)

Figure 32 Comparison of magnetic gradient (A & C) and resistance (B &D) big pit house (red) and possible ritual pit house(yellow) anomalies.

Ground-truthing of both of these pit house anomalies feeds back to the interpretation of thegeophysical data. The site surface area of the big pit house was cleared down to the top of the pit housefeature, defining the edges of the pit house structure (Figure 33). A sample trench was excavatedthrough the second pit house feature to the feature floor; initial interpretations suggest this is a typicalpit house that has an alignment of features that include a ritual sipapu, fire hearth, and ash pit, (Figure34). It is interesting to note the complementary feature anomalies of the magnetic gradient andresistance surveys. Magnetic gradient appears to have relatively the same strength over the central partof both pit house features; resistance survey appears to give a more accurate outline of the extent ofthe big pit house feature.

Figure 33 Contrasting soils (red and dark brown) can be seen that define the edge of the big pit house feature. (Imagecourtesy of Crow Canyon Archaeological Center.)

Figure 34 The floor of a section of the second pit house feature with a ritual sipapu, fire hearth, and ash pit. (Imagecourtesy of Crow Canyon Archaeological Center.)

Viewing all of the geophysical survey interpretations together presents a colorful picture (Figure 35).Obviously, there is a lot going on in this area. To help better understand the geophysical surveys, andthus more intuitively interpret the data and archaeological nature of the site, initial ground-truthingthrough coring (Figure 36) helped identify areas that contained cultural materials, enablinginterpretation of possibly 9 pit house or pit house like features and a series of aligned pits that areinterpreted as a fence (Figure 37). Combining the geophysical interpretations with the ground-truthingand overlaying the site surface features (Figure 38) enables a new approach to considering the buriedarchaeological resource at the Dillard site.

Figure 35 Interpretations of all the geophysical survey methods at Dillard.

Figure 36 Core samples over geophysical anomalies help identify cultural (red points) and non-cultural (yellow points) sitefeatures.

Figure 37 Ground-truthed archaeological features (purple are pit houses or pit house like structures and navy blue areindividual pits) overlain on site interpretations and surface features.

Figure 38 Mapped site features overlain on interpretations of geophysical surveys.

One of the primary pieces of information we are gaining from these surveys and ground-truthinginvestigations is an understanding that pit house features, though not visible on the ground surface,area visible through geophysical surveys. However, it is important to note that no two pit housefeatures appear the same in any of the geophysical survey methods. To best comprehend thisstatement, Figure 39 presents each of the pit houses as mapped, or not, through geophysical surveys.

Figure 39 Pit house anomaly comparison. The pit house features as ground-truthed are positioned in the center of eachsample above.

The point of comparing the geophysical signatures over each of the pit houses shows that thus far, thereis no specific ‘guideline’ to data interpretation nor a specific ‘signature’ in any of the geophysical surveymethods for pit house features. The general conclusions from this comparison of geophysical surveymethods are:

1. Magnetic gradient was both the fastest and most effective tool for mapping buriedarchaeological features at the Dillard site.

2. Resistance was a much slower method, but effective in mapping some of the pit house features.3. Conductivity and magnetic susceptibility were slightly faster survey methods than the

resistance, but were not consistent at discerning the actual location of pit house features with ahigh level of confidence.

It is interesting to note, is that these two latter methods may provide valuable information to activitiesrelated to the pit house (pit house like) features.

Airborne LiDAR Principles and ResultsAirborne LiDAR, or light detection and ranging, measures the height of the ground surface and anyfeatures (i.e. trees, buildings) that may be on it and provides high definition and accurate models of thelandscape to a resolution of 1 m to 0.5 m in archaeological applications. LiDAR uses a pulsed laser beamthat scans from side to side as a plane flies at a low altitude over the survey area. 20,000 to 100,000points per second build the ground model. In post-processing the first returns can be removed from thedata providing a ‘bare earth’ model (or Digital Terrain Model, DTM) that accurately represents theground surface.

The airborne LiDAR data were acquired by the NRAC, West Virginia University. NRAC operates anOPTECH ALTM-3100C airborne laser (small-footprint) mapping system. The system integrates a laseraltimeter, a high-end Applanix Pos/AV Inertial Measurement Unit (IMU), also called an InertialNavigation System (INS), and a dual frequency NovAtel GPS receiver. This integrated system is capableof 100 kHz operation at an operating height of 1,100 meters (3,609 feet). LiDAR technology offers fast,real-time collection of three-dimensional points that are employed in the creation of Digital ElevationModels (DEMs), Digital Terrain Models (DTM), landscape feature extraction, forest stand structureanalysis, as well as many other research applications.

Data were collected in multiple, low altitude acquisition passes over the core area of the Dillard site(Figure 40) to yield ground LiDAR point densities of 15-20 per square meter (vertical accuracy of 15 cmor better). Integrated data have a vertical error of 15cm or less at the 95% confidence level for areas ofopen terrain and moderate slopes of 10 degrees or less (based off manufacturer’s specifications). Dataare recorded in the applicable Universal Transverse Mercator (UTM) zone, NAD83 datum (CORS96)while heights are orthometric, referenced to the North American Vertical Datum of 1988 (NAVD88)using GEOID09.

The resulting ‘bare earth’ model from the LiDAR data provides an excellent model of the landscape,

Viewing the site and interpreted features draped on a LiDAR digital terrain model (DTM) shows itslocation within the broader natural and cultural landscape. The site’s orientation to local landmarkssuch as the San Juan Mountains to the east, the Mesa Verde questa to the south, and lone UteMountain to the west confirm the site’s expansive view shed of the prehistoric Mesa Verde Region.Despite this emphasis on view shed, the LiDAR DEM demonstrates that the site sits on one of many lowlying ridges, making it easily accessible to the 107 known BMIII habitation sites in the surroundingsettlement providing insight on its role in the larger community.

Figure 40 LiDAR DTM and the broader landscape with BM III site distribution in reference to the Dillard site (red).

Conclusions and Recommendations

The geophysical methods used for the Dillard site survey were very effective at mapping archaeologicalfeatures as well as providing potential contextual information related to individual pit house features.We are able to make a preliminary statement about the distribution of structures and an estimate of thepopulation of the site. The ground-truthed features and remaining geophysical interpretations provideimportant information to consider not only the use and organization of space at the Dillard site but alsoprovide insight to social and community organization.

This work proves that magnetic gradient survey is the most effective tool for mapping Basket Maker IIIsites in the Crow Canyon region (i.e. similar archaeological features, ground matrix, and environment).It is recommended that for future site exploration magnetic gradient would be the most effective tool.Depending on further site investigations at the Dillard site, conductivity / magnetic susceptibility(electromagnetic induction with the EM36B) survey may provide additional information related to thespace around pit house or pit house like features. While a bit on the slow side for data collection,resistivity is an excellent survey method if conducted during a season when the earth is not entirely dry,as it depends on some ground moisture for successful survey.

Additional regional landscape investigation through traditional GIS and LiDAR 3D model analyses mayprovide interesting insight to the location of the Great Kiva at the Dillard site.

Acknowledgements and CreditsSincere thanks are given to Shanna Diederichs for her continued involvement with the geophysical dataand site interpretation long after the Time Team America crew departed the field. Thanks also go toShirley Powell, Steve Copeland, Scott Ortman, Grant Coffey, Caitlin Sommer, Mark Varien, DylanSchwindt and the rest of the Crow Canyon Archaeology Center archaeologists and staff. Our work couldnot have been completed without the support and assistance from Jane Dillard, for this we are sincerelythankful.

Time Team AmericaThis work was undertaken as part of the filming of Season 2 of the PBS prime-time program Time TeamAmerica. The program is co-produced by Oregon Public Broadcasting and Videotext LLC and fundedentirely by a National Science Foundation Informal Science Education grant. Meg Watters is a co-PI1 onthe grant and the Remote Sensing and Visualization Coordinator for the television program. Membersof the Time Team America geophysical survey team include Bryan Haley, Tulane University, and DuncanMcKinnon, University of Arkansas2. Adam Riley and Paul Kinder, West Virginia University, NaturalResource Analysis Center, performed the LiDAR survey and delivered processed data to Watters.The material in this report is based upon the work supported by the National Science Foundation underGrant number 1114113. Any opinions, findings, and conclusions or recommendations expressed in thismaterial are those of the author and do not necessarily reflect the views of the National ScienceFoundation.

1 Noel Broadbent, Smithsonian Institute is the other co-PI and Dave Davis from OPB is the project PI.2 Thanks to Duncan and Bryan for all of their hard work in and out of the field with data collection, processing,interpretation and contributing to this report.

ReferencesAitken, M. J. 1961. Physics and Archaeology. Interscience Publishers Inc., New York.

Apparao, A., A. Roy & K. Mallik. 1969. Resistivity model experiments. Geoexploration 7:45-54.

Apparao, A. & A. ROY. 1971. Resistivity model experiments. Geoexploration 9:195-205.

Barker, R. 1989. Depth of investigation of a generalized collinear 4-electrode array. Geophysics54:1031-1037.

Bevan, B. 1983. Electromagnetics for Mapping Buried Earth Features. Journal of Field Archaeology10:47-54.

Bevan, B. 1998. Geophysical Exploration for Archaeology: An Introduction to Geophysical Exploration.Midwest Archaeological Center, Special Report No. 1. National Park Service, Lincoln, Nebraska.

Charles, M. 2012. Electrical Resistance Survey of Three Sites in the Indian Camp Ranch, MontezumaCounty, Colorado. Report prepared for: Crow Canyon Archaeological Center, Cortez, CO.

Clark, A. 1996. Seeing Beneath the Soil, Revised Edition. London: B. T. Batsford Ltd.

Clay, B. 2006. Conductivity (EM) Survey: A Survival Manual. Cultural Resource Analysts, Inc.

Crutchley, S. and Crow, P. 2009. The Light Fantastic: Using airborne LiDAR in archaeological survey,Swindon: English Heritage.

Dalan, R. 2006. A geophysical approach to buried site detection using down-hole susceptibitliy and soilmagnetic techniques, Archaeological Prospection 13:182-206.

David, A. and Payne, A. 1997. Geophysical Surveys within the Stonehenge Landscape: A Review of PastEndeavour and Future Potential. Proceedings of the British Academy 92:73-111.

Evjen, H. 1938. Depth factor and resolving power of electrical measurements. Geophysics 3:78-95.Griffiths, D. & R. King. 1981. Applied Geophysics for Geologists & Engineers. London: Pergamon Press.

Hasek, V. 1999. Methodology of Geophysical Research in Archaeology. BAR International Series 769.ArchaeoPress, Oxford, England.

Heimmer, D. and S. DeVore 1995. Near-Surface, High Resolution Geophysical Methods for CulturalResource Management and Archaeological Investigations. U.S. Department of the Interior, NationalPark Service.

Kvamme, K. 2003. Geophysical Surveys as Landscape Archaeology. American Antiquity 68 (3):435-457.

Kvamme, K. 2006. Magnetometry: Nature’s Gift to Archaeology, (ed.) J. Johnson, An Explicitly NorthAmerican Perspective, University of Alabama Press.

Roy, A. 1972. Depth of investigation in Wenner, three-electrode and dipole-dipole DC resistivitymethods. Geophysical Prospection 20:329-340.

Roy, A. & A. Apparo. 1971. Depth of investigation in direct current methods. Geophysics 36:943-959.

Weymouth, J. 1986. Magnetic Survey of Archaeological Sites in the Knife River Indian Village NationalHistoric Site: Small and Non-Village Sites. Manuscript, Midwest Archaeological Center, Lincoln.

OK, looks like:Cond: 10.2 - 12.33 mS/mMagSus: .13 - .50 SIMag: -3.56 - 3.39 nTRes: 8.42 - 29.63 ohms