GPR Acquisition: A total of 65 GPR lines, representing a linear distance of approximately 40.23 km, were acquired during field activity 13BIM01. Data were collected, over various terrains, either on foot or towed at slow speeds (>10 mph) behind a vehicle. The GPR data were collected with a Geophysical Survey Systems, Inc. (GSSI) TerraSIRch SIR System-3000. This system collects single channel GPR and contains a Digital Control Unit (model DC-3000), 10.8 V Lithium-Ion rechargeable battery (which allows up to 3 hours of survey time), a 200 MHz antenna and cables, and a 16-inch survey wheel (model 620) that acquired 417 ticks/meter. Additional equipment used in conjunction with the GSSI included: an external [puck] GPS Acumen data logger, support poles and hardware (for the GPS system), and harness and tow strap. The GPR system's internal memory capacity is roughly 1 GB of data; however, to ensure there would be no data loss, files were recorded to both the internal memory and a SanDisk CompactFlash (CF) memory card. Acquisition settings for the subsurface profiles were set to 64 scans/s, 20 scans/unit (m), 1024 samples/sec 16 bits/sample, dielectric constant = 15, range = 200 ns, and auto gain. Due to the homogeneous nature of Dauphin Island's near surface deposits, the preceding settings were applicable for most lines; however, adjustments were made when changes to the terrain or subsurface geology warranted such modifications. An Infinite Impulse Response (IIR) filter was used (Lowpass = 600 MHz and Highpass = 50 MHz) to downplay external interference, thus cleaning up the signal. Each day, prior to collecting any data, the distance setting was used to manually calibrate the survey wheel for the terrain by laying out a 10 m long measured line on the survey surface, which varied between sandy dune slopes, asphalt and grass. Calibration steps consisted of (1) entering the calibration distance, (2) positioning the antenna at the start of the calibration line, ensuring that the same part of the antenna positioned at line start also ended at the terminus of survey line and, (3) lastly pulling the wheel the pre-determined distance. These steps were repeated until a satisfactory average value was obtained, at which time it was saved into the system. Data was acquired using RADAN 7 proprietary software [in TerraSIRch mode] and saved in .dzt format.
GPR processing: The GPR data were processed using Sandmeier Scientific Software's Reflexw Version 7.2.2 geophysical near surface processing and interpretation software (http://www.sandmeier-geo.de/reflexw.html)
. Each GPR data file was imported into Reflex, where it was converted from Radan's DZT (.dzt) format to a .dat file. Next, a variety of processing steps were used including applying static correction, subtracting mean (dewow), removing header gain, and applying manual AGC gain. After these initial processing steps, the GPS (either post-processed DGPS or Lidar-derived GPS) and elevation data were integrated into the trace headers. Lastly, all profiles were inspected for data quality to ensure no navigation or trace data gaps were present before being output in American Standard Code for Information Interchange (ASCII) format for the trace data and Joint Photographic Experts Group (JPEG) format for the profile images. If any issues were discovered during the QA/QC process, this information was noted for each affected line and recorded in the metadata included in this report. A velocity of 0.06 m/ns was used to calculate and display depth on the elevation-corrected profiles. Lines with post-processed GPS and elevation data issues are noted in this metadata file.