Geophysical prospecting formed a central part of the 2010 field season for a variety of reseaons. The most prominent is the potential of geophysical prospecting to survey large areas for archaeological features quickly, thus helping focus other more time intensive archaeological investigations in areas of promise or on known archaeologuical features and deposits. In addition, geophysical prospecting is a non-destructive method (unlike excavations), an important element in the research design for this project as a result of the site being located on a working coffee plantation. The presence of coffee plants at the site requires that the archaeological research be carried out in such a way as to mitigate the disturbance or destruction of the coffee plants (an important facet of the negotiations with the landowner and community that led to permission to carry out this research project). The geophysical prospecting was carried out in collaboration with the Stanford Shared Measurement Facility and in consultation the director of the facility Nigel Crook, a member of the Stanford Geophysics department. Multiple geophysical methods were used as a preliminary step for identifying the efficacy of the various methods in the environmental conditions specific to the site and thus restricting future geophysical survey to the most appropriate methods. Moreover, geophysical methods that allow for rapid survey were used in order to cover as much of the site as possible and determine areas of interest for the 2010 season’s test excavations. The results of the electromagnetic induction, magnetometer and limited ground penetrating radar survey will be summarized below.
Electromagnetic Induction
Electromagentic induction (EMI) is a geophysical prospecting technique that measures variations in the electrical conductivity of the substrate. Higher levels of electrical conductivity detected across the site may be linked to increased moisture, the presence of metals, and other such materials that are more conductive. Lower levels of conductivity may be linked to the presence of rock, more arid substrate etc. Plotting out the conductivity readings across the site may in turn help identify cultural deposits and the remains of structures when they are analyzed and patterns in the conductivity variations are highlighted (i.e. linear features that may result from stone walls) or when areas of very low or very high conductivity are identified (i.e. isolated features filled with organic rich soil, such as garbage pits that would display a high conductivity measurement).
The EMI survey at San Pedro de Aguacatepeque was carried out over the month of July with a Dualem 1S system rented from the Stanford Shared Measurement Facility. The Dualem system consists of two sensors set at 90 degrees from each other which allows sensor 1 to detect anomalies 1.5 meters below the surface, while sensor 2 detects anomalies 0.5 meters below surface at greater resolution. Over 2 hectares of the site were prospected, the results of which are shown in Figure 1 and 2.
While some ambiguous anomalies and areas of interest can be seen, the survey yielded very little in the way of clearly defined linear features and/or isolated anomalies that could be groundtruthed through excavation. However, there are some areas of interest, that correspond to surface concentrations of colonial material culture and topographic features that may be indicative of buried archaeological deposits (see figure 3 and 4) that shows the presence of large conductive anomalies in the terraced area of the site, a more linear and defined one appears at 0.5 meters in depth and a much larger diffuse one that shows up deeper). The EM survey data is currently undergoing further processing and analysis and with the aim of identifying other anomalies and to compare the results of EM survey against the results of the other geophysical methods.
Magnetometry
Magnetometry is a passive geophysical prospecting technique that measures variations in the Earth’s magnetic field. Metal objects, burned material, and soil affected by heat and/ bacterial actions can all contribute to creating subtle variations in the Earth’s magnetic field. Magnetometers can detect these variations if the magnetic properties of the object or soil contrast with the background magnetic field and can lead to the identification of archaeological deposits and features. Stone walls with magnetic properties that are strikingly different that the background magnetism of the soil they are buried in will show up as linear anomalies in a magnetometer survey (hearths, burials and trash pits can be similarly identified). Magnetometer surveys can be quickly carried out, and allow for a large areas to be surveyed, especially when the instrument is used in conjunction with a GPS system that allows for simultaneously logging linked GPS coordinates and magnetic readings. In magnetometer surveys, it is necessary to account for natural diurnal variations in the Earth’s magnetic field. If unaccounted for, natural peaks or valleys in the Earth’s magnetic field may appear as magnetic anomalies in the survey data as false positives. Such issues can be avoided with the use of a second stationary magnetometer utilized as a base station that only takes readings of the Earth’s magnetic field from a single location near the site. This base station will produce a set of readings that will detail the diurnal variations in the Earth’s magnetic field that can then be used to calibrate the survey readings and thus mute the impact of natural magnetic variations on the magnetometer survey.
The magnetometer survey at Aguacatepeque was undertaken over the course of a few days in early August with the aid of Nigel Crook and equipment rented from Geometrics in San Jose, CA. A Cesium Vapor magnetometer linked to a submeter accuracy GPS System was used for the survey, and a less sensitive proton procession magnetometer was used as a base station in order to correct the survey data for diurnal magnetic variations. In total, an area of 3.2 hectares was surveyed with magnetometer, and several magnetic anomalies were detected (Figure 5). However, despite the success of the survey in identifying areas of interest, in general the magnetometer survey was not as successful as hoped due to the elevated magnetic properties of the volcanic surroundings and the lack of contract between archaeological features and the background volcanic substrate they were located in (the large red areas, indicative of highly magnetic anomalies are likely caused by deposits of volcanic material that are highly magnetic). This being said, anomalies detected in the southern portion of the site and in the flattened terraced area that contains concentrations of colonial material culture have shown great promise, specifically because similar types of anomalies were detected in these areas in the EM survey and because of the concentrations of material culture on the surface associated with these magnetic and electromagnetic anomalies. In particular, the terraced section of the site shows a large area with a diffuse negative magnetic reading and some high magnetic anomalies that hints at potentially buried archaeological remains (Figure 6). This area also contains anomalies detected by the EM survey and for this reason was targeted in the auger and test excavations portions of the project. The anomalies in the southern portion of the site range outside of the surface collection limits, however a test excavation was carried out on the location of one of the largest and strongest magnetic anomalies (see Figure ) and yielded architectural remains, as well as intact stratigraphy that includes the dark sand layer that may be the remnants of the 1582 volcanic activity of Fuego. Processing of the magnetometer survey data is on-going and future work will continue to explore and groundtruth the numerous magnetic anomalies detected.
Ground Penetrating Radar
Ground penetrating radar (GPR) is an active geophysical prospection method that operates by introducing an electromagnetic pulse into the ground, and analyzing the resulting reflections of the wave against buried strata and features. GPR can provide detailed profiles of the stratigraphy of an area, as well as identify the location and relative depth of archaeological objects and features. However, while acquiring GPR data is itself a much slower process than acquiring magnetometry data, the processing and interpretation of the data is itself even more time consuming. As a result, GPR surveys are usually undertaken after initial small scale tests have shown the efficacy of the method in the conditions of the site and region, in order to avoid investing substantial time into the processing of data that may indeed show nothing (if the conditions on the ground are not conducive to the GPR method). For this reason, the other prospection methods were carried out at the site this summer at a greater scale, and the GPR work was undertaken mainly as a small scale preliminary test to evaluate the potential of the method at the site of Aguacatepeque.
The ground penetrating radar work at the site was limited to two 50meter long transects through parts of the site that with large concentrations of colonial material culture on the surface (see Figure 13 for location of GPR transects). Nigel Crook, of the Stanford Shared Measurement Facility performed the GPR prospection using a Sensors & Software Pulse-Ekko Pro system with 200Mhz antenna. More GPR transects were planned for the 2010 field season, however issues related to the failure of the GPR batteries as a result of rain and moisture at the site limited the GPR work to two transects. Despite these technical difficulties, the GPR work was quite successful and it was determined that this methods may be the best one for Aguacatepeque. The GPR work provided a detailed rendering of stratigraphy to a depth of between 3-4 meters and helped identify certain anomalies in the stratigraphy (i.e. possible archaeological features dug into strata) as well as possible archaeological materials within each strata (see Figure; note strata and parabolic anomalies). An excavation unit (Test Unit 2) was placed on GPR transect 2 in an area that showed a depression in a strata possibly related to cultural activity. The excavation confirmed the presence of many artifacts as well as confirming that the GPR’s representation of the stratigraphy in the area was indeed accurate (see below for Test Unit 2 profile drawing ). While no clear archaeological features were detected, GPR survey has great potential to do so when used on a larger scale and may facilitate the rapid evaluation of magnetic and electromagnetic anomalies detected across the site.
These conclusions led to the large-scale GPR survey carried out in July of 2011. Check the front page blog for info on the results of the most recent geophysical work at SPA.