Marine magnetic surveys are a commonplace tool for archaeologists to discover, document, and characterize ferromagnetic submerged cultural resources (SCR). It is difficult, however, to quantify the efficacy of a given magnetic survey in terms of actual detection thresholds and, therefore, accurately assess the presence or absence of archaeological remains throughout a given area. Similarly, survey planning methods and data visualization techniques are likewise challenging to approach quantitatively. To address these issues, the Bureau of Ocean Energy Management's Office of Renewable Energy Programs partnered with the National Park Service's Submerged Resources Center to conduct a field research program whereby known ferromagnetic archaeological sites were magnetically sampled to better understand their respective detection thresholds. Incorporating the results of these tests, the team developed a series of custom geospatial processing tools in ArcGIS to assist in quantifying the process of planning, processing, and describing marine magnetic surveys.
Field testing operations, which took place in Biscayne National Park, involved executing pre-determined magnetic survey sampling patterns around known ferromagnetic archaeological objects of various vintage, size, and materials. Acquired data was then processed to yield specific values for the object's magnetic moment, the primary variable needed to quantify induced magnetic field strength and, therefore, a given object's spatial threshold of detection. These were, in turn, used to refine induced magnetic field models subsequently incorporated into magnetic survey planning tools, as well as geospatial processing methods scripted in Python to automate magnetic survey data integration, visualization, filtering, and post-acquisition assessment.
Sampled SCR included modern period steel-hulled vessels, diffused debris fields containing numerous scatters of iron artifacts, iron cannon and shot, historic anchors, and wooden sailing vessels with iron components. This diversity of test sites encompassed an array of archaeological materials typically found in a marine environment. Information yielded insights into the relative magnetic field strength of each these materials and site types, allowing models of induced magnetic field strength to be further refined in terms of a targeted object's anticipated detectability during a given survey.
Four Python scripts were developed, including an Input tool, Generate Survey Boundary tool, Visualization tool, and Confidence Modeling tool. Collectively these scripts comprise the Magnetometer Survey V.1.0 toolbox, which integrates into ArcGIS via ArcToolbox. Once marine magnetic survey data is output from a data acquisition program, these Python scripts automate the remaining data processing and facilitate a quantitative QA/QC assessment based on user-defined parameters. As a result, marine magnetic surveys for archaeological resources can planned, executed, processed, and assessed according to a repeatable and consistent procedure.