MAGNETOMETER SURVEYS

Meade F. Kemrer, PhD Archaeological Consulting performed a series of magnetometric surveys in the Cañada Alamosa, New Mexico during the April 23-25, 2001 period at the request of Dennis O’Toole, Cañada Alamosa Institute. The purpose of the study was designed to assess the effectiveness of magnetometry for buried prehistoric feature discovery and characterization, and to identify areas for testing in the upcoming archaeological field session.

Karl Laumbach, Alamosa Project Archaeologist, described the general testing areas within the Montoya Ruin (LA 88891). The site contains a series of Classic Mimbres noncontiguous roomblocks manifested as partially exposed masonry alignments (Figure 1). Buried architecture and extramural features undoubtedly exist associated with the roomblocks. Magnetometer surveys could potentially find such places, enhancing testing efficiency.

Figure 1. Montoya Site Map Showing the Magnetometer Survey Blocks.

METHODS

The instrument used for this project is a Geometrics G-856AX magnetometer configured as a gradiometer with two sensors separated by 0.9 m on a vertical aluminum staff. The instrument and survey methods were designed to reduce error and maximum sensitivity to the archaeological domain based on a series of tests and instrument modifications (Kemrer 1999).

The Alamosa survey is a baseline study. Field strategies and methods matched that intent. The rhyolitic iron-poor basement bedrock survey area could produce low contrast, making discerning features difficult. A high-resolution data collection was warranted and employed. Maximizing target detection meant using block surveys with high data collection densities. Throughout the project, the blocks measured 4 x 8 m with data collecting at 0.25 m in all rows and columns. Mr. John Fitch, volunteer to the Alamosa Project assisted aligning survey blocks and transect lines. One rhyolite cobble and a bag of derivative soil samples were collected for computing site magnetic susceptibility and for future magnetic surveys and modeling.

Karl Laumbach identified five areas planned for testing. Because of time available and survey rigor constraints, three of these areas within four blocks were completed in one full day and two half-days (Figure 1). The geomagnetic properties of the project area were collected from the GEOMAG2000 software package. The IGRF2000 model computed the total earth magnetic field as 49,939 nT based on the longitude and latitude and the elevation of the Montoya Ruin. Prior to traveling to Cañada Alamosa, the solar weather predicted by NOAA indicated that the solar flux would be ideal for magnetic survey. During the April 23-25 survey period, actual field values ranged within a 49,389.7 - 49,633.1 nT interval.

SURVEY RESULTS

A total of four block-type surveys were performed. Three blocks were situated within and adjacent to pueblo wall segments for two purposes: 1) to test whether known walls can be magnetically detected, and 2) to increase the odds that buried walls would fall within the survey blocks and detected. The fourth block was placed within an area devoid of surficial indicators for features – a purely prospect sample.

Analyses of the magnetic information met conservative data treatment methods. Tested magnetic models for archaeological materials and features applied to prehistoric Mimbres sites are available and used in these analyses (e.g., Bevan 1996a,b; Breiner 1973).

Survey Block Unit 1: Unit 1 is located in the southwestern portion of the Montoya site (Figure 1) directly north of a pueblo wall segment (Figure 2). Current field information indicates that wall alignment differs from the original survey sketch map shown in Figure 1.

Figure 2. Sketch Map of Magnetic Survey Unit 1.

The raw magnetic data shows a large anomaly in the southeastern corner of the survey block (Figure 3). The nT range is relatively high. The fact that a block-wide gradient ranging from low to high in the north to the south indicates that local conditions are probably caused by topography and bedrock geological properties.

Figure 3. Magnetic Map, Raw Data from Unit 1.

A second-order derivative of the data removed local geological and topographic noise from the raw magnetic information. The result is shown in Figure 4. The nT range width is less from a range from +25 to –40 nT to +20 to –25 nT.

Subsequent the removal of magnetic noise, a horizontal series of anomalies is visible along the bottom of Figure 4, probably a masonry wall that connects to the vertical wall shown in Figure 2. Two less conspicuous alignments are also visible in the Figure 4 map, characterized by thin black lines. The first is a straight north-south line of anomalies. The second is a zigzag pattern. Both are likely artificial patterning produced by random magnetic changes within the linear row and column data collection points.

Figure 4. Magnetic Map, Local Gradient Removed, Unit 1.

Survey Block Unit 2: This block is south of Unit 1 and placed on a roomblock designed to characterize partially buried architecture (Figure 5). A wall containing several stone courses high is exposed in the erosion drainage shown in Figure 3. Scattered masonry occurs within Unit 2.

The visible architectural remains indicate that this roomblock is substantially more complex than noted in the survey sketch map shown in Figure 1. South of Unit 2 are several contiguous intact room outlines (Figure 5).

Survey of Unit 2 produced large anomalies, initially believed to have been caused by steel fencing. Probing reinforced by findings in another unit demonstrated that masonry was the source. The high range of these anomalies is shown in Figure 6. It is likely that these high readings indicate multi-coursed masonry. Despite of the high nT values, that normally mask feature details, a distinct room outline is visible in the eastern portion of Unit 2.

Figure 5. Sketch Map of Magnetic Survey Unit 2.

Figure 6. Magnetic Map with Large Anomalies and Room Outline, Unit 2.

Reducing the size of the anomalies in the eastern portion of Unit 2 and removing the local gradient failed to find any features in the opposite side of the block (Figure 7). Magnetic value reduction did enhance the detailed shape of the room.

Figure 7. Map with Reduced Magnetic Values and Room Outline, Unit 2.

Survey Block Unit 3: Located north of Unit 1 (Figure 1), Unit 3 was designed to explore a featureless area. The first two units clearly demonstrated that at least masonry features could be detected. Therefore unaided prospecting was feasible.

The raw magnetic data show the effects of the local gradient (Figure 8). All of the nT values are negative. The second-order derivative from these data removed this source of noise (Figure 9).

Three potential feature areas occur within Unit 3, outlined in Figure 9. A set of anomalies occurs in the southwestern corner of the block. They may represent an extramural activity area, although intensive subterranean rodent tunneling can also produce similar magnetic disturbances. An intense single dipole anomaly in the eastern edge circled in Figure 9 may represent a burned feature or perhaps a small piece of metal. A relatively large magnetically “quiet” in the north central area may have been by created by a compacted living surface such as a jacal structure, a filled pitstructure, or noncultural random chance alone.

As shown in this block, the identification of small and subtle cultural features often falls into the realm of conjecture. Nonetheless, this block survey was completed at the high-resolution level in six person-hours and found those areas that warrant further inspection. Strip excavation of this block would probably expend six to eight person-days.

Figure 8. Magnetic Map Showing the Local Gradient, Unit 3.

Figure 9. Magnetic Map Anomaly Areas, Unit 3.

Survey Block Unit 4: Similar to Units 1 and 2, Unit 4 was designed to detect buried architecture. Location of this roomblock is on the eastern edge of the Montoya site (Figure 1). The magnetic survey block was placed in the southwestern portion of the roomblock where several wall alignments may have buried walls that would produce complete rooms (Figure 10).

Figure 10. Sketch Map of Roomblock and Unit 4.

Magnetic survey encountered a series of intensive anomalies in the northeastern portion of the unit. The magnetic map and the nT scale show that these were the largest anomalies found in the project area (Figure 11). Note that that the northwest values completely masked all magnetic variation within the remainder portion of Unit 4.

Analysis of magnetic content necessitated reducing the values of the high readings in the northwestern portion of the unit. The result was a unit-wide array of low values produced by the local gradient noted in the previous three survey blocks. A second-order derivative of the data removed this noise (Figure 12).

Based on previous experience in Unit 2 magnetic and probe, the northwestern large anomalies in Unit 4 probably represent multiple-coursed masonry. The shape of the anomalies suggests that they represent a corner or a collapsed wall (Figure 12). Another room was discovered directly east of the large anomalies (Figure 12). The presence of this room confirmed magnetic readings noted during the Unit 4 survey. (Note: Excavations later in 2001 in Unit 4 exposed the intense anomalies in the northwest corner. It proved to be a hearth surrounded with burned stones.) The visible wall stub in the southwestern portion of the unit (Figure 10), manifested in the magnetic map (Figure 12), does not extend further.

Figure 11. Raw Magnetic Data, Unit 4.

Figure 12. Magnetic Noise Removed, Unit 4.

SURVEY RESOLUTION ANALYSIS

Additional analysis indicates that relatively high-resolution magnetic surveys are most appropriate to detect the range of buried features encountered in the Montoya Ruin site. Because the data was collected at 0.25 m in columns and rows, lower survey resolution can be accurately simulated by structured data removal. All surveys were set at the 0.5 m resolution by removing half of the row and column data.

The 0.25 data in Unit 1 (Figure 4) identified a horizontal series of relative large magnetic anomalies that is probably a buried masonry wall that connects to the visible wall shown in Figure 2. The survey simulation of Unit 1 with data collection at 0.5 m (Figure 13) clearly shows three linearly arrayed anomalies on the bottom of the map corresponding to the same area as those in Figure 4. In this instance both surveys would have detected the wall.

Figure 13. Unit 1 Magnetic Map, 0.5 m Simulated Survey.

Survey at the 0.25 m level in Unit 2 identified a masonry room on the eastern side of the block (Figure 7). The feature was outlined by large anomalies produced by the masonry. The simulated 0.5 m survey also identified the room (Figure 14).

Figure 14. Unit 2 Magnetic Map, 0.5 m Simulated Survey.

The original survey of Unit 3 identified three possible cultural feature areas (Figure 9): a set of several cluster of large anomalies that could represent a buried prehistoric activity area, a single dipole anomaly consistent with a buried hearth, and a large area surrounded by anomalies, possibly a living surface or a pitstructure. The 0.5 m simulation survey identified the multiple anomaly possible activity area in the bottom of the block (Figure 15). The possible hearth and the living surface/pitstructure areas do not appear. It this case, single anomaly and subtle magnetic signature areas were not detected at the 0.5 m survey level.

Figure 15. Unit 3 Magnetic Map, 0.5 m Simulated Survey.

Unit 4 survey at the 0.25 m resolution level identified three cultural phenomena (Figure 12): 1) a wall segment, possibly a corner or a collapsed wall, manifested as several large anomalies, 2) a small room identifies by three intersecting small anomaly alignments, and 3) a visible wall segment produced by several large anomalies. The 0.5 m simulated survey (Figure 16) identified the possible collapsed wall/corner at the top of the block, and the wall segment in the southwestern portion of the unit. The features produced by large anomalies are detected by the 0.5 m survey. Features manifested by small anomalies fail to be magnetically visible in the 0.5 m interval data. (Note: Excavations later in 2001 in Unit 4 exposed the intense anomalies in the northwest corner. It proved to be a hearth surrounded with burned stones.)

Figure 16. Unit 4 Magnetic Map, 0.5 m Simulated Survey.

Simulated surveys of the four block units provided methodological lessons applicable to the Montoya site and the surrounding area. Coarse-grained magnetometer surveys collecting data at the 0.5 m interval consistently detected features that produce large anomalies such as masonry comprised of large stone blocks or multiple coursed walls. These studies indicate that this level of resolution is sufficient for mapping buried masonry wall alignments and rooms made of large stones or multiple-coursed. Surveys at the 0.5 m level is also appropriate for detecting nonstructural features that produce large anomalies such as larger-sized roasting features and other extensive burned areas.

High-resolution 0.25 m surveys can detect all features described for 0.5 m detection plus single-anomaly features such as small hearths, magnetically subtle features such as unburned living surfaces and pitstructures, and small-anomaly patterns such as architecture made with small cobble foundations.

REFERENCES CITED

Bevan, Bruce

1996a Geophysical Exploration for Archaeology Volume B: Introduction to Geophysical Exploration. Geosight Technical Report Number 4. Pitman.

1996b Geophysical Exploration for Archaeology Volume B: Detailed Survey Procedures. Geosight Technical Report Number 4. Pitman.

Breiner, Sheldon

1973 Applications Manual for Portable Magnetometers. Geometrics, Sunnyvale.

Kemrer, Meade F.

1999 Controlled Near-Surface Magnetometer Studies in Doña Ana County, New Mexico. Report to Bureau of Land Management, Las Cruces Office. Las Cruces.

MAGNETOMETER SURVEYS

IN THE

By

TABLE OF CONTENTS

INTRODUCTION

METHODS

REFERENCES CITED