Tuesday, June 5, 2012

I recently received an e mail asking for public-friendly resources addressing why ground penetrating radar (GPR) is useful for documenting cemeteries.  It struck me as odd, because we've spent the last nine months conducting Cemetery Resource Protection Training (CRPT) workshops that, in part, discuss just that.  Despite all of that effort, and the fact that we've carried out GPR efforts in several cemeteries (even offered a symposium), we haven't ever really blogged about how it can be used to understand these sensitive sites.

St. Augustine Archaeology volunteer Nick McAuliffe helps conduct GPR survey at the Colonial Spanish Quarter.

To get to that, I'll need to back up a step.  In the CRPT workshop, we emphasize that one of the best ways to protect a cemetery is to record it--in the state of Florida, registering any site with the Florida Master Site File (FMSF) offers a layer of awareness and protection.  In order to add a cemetery to the FMSF, the cemetery name and site boundaries are required; other information is optional, including history, details about interments, and the condition in which it was found.

Florida Master Site File Historical Cemetery Form (front side)

Recording boundaries can be a little tricky though, particularly when dealing with an abandoned or historic cemetery.  Many are missing markers, not fenced in, and have little documentation.  In order to determine boundaries, we encourage people to use all of the resources at their disposal--oral histories, historic images including photographs, any existing documentation, and visual survey/observation.

One oral account claims that Civil War spy Francis Kirby was buried in St. Augustine's Huguenot Cemetery (above), where her headstone is; another asserts that she was interred in Palatka.  Using GPR, we found that someone is buried beneath her grave marker.  We cannot, however, say for certain that it is her.

This is where ground penetrating radar can be useful.  Working in conjunction with these other resources, GPR affords the opportunity to take a "look" into the ground and detect cemetery features, including possible burials.

Since 2009 we have worked with local cemetery stewards and archaeologists on public-friendly projects.  Some of those projects led us to investigate some issues that may arise in documentation and locating boundaries.  I've highlighted just a few examples to demonstrate a range of how GPR can be useful to cemetery stewards.

National Cemetery

Dade Monuments at St. Augustine's National Cemetery.  Photo courtesy of Paul Ramey.

At the National Cemetery we investigated the Dade Monuments, three coquina pyramids under which soldiers who fell in the Dade Massacre (Second Seminole War) were buried en masse.  We wanted to understand the nature of the mass interment: was this one large deposit denoted by the pyramids, or three separate vaults?

This GPR readout shows a space between the monuments.

We attempted to address this question by running the GPR along the southern edges of all three pyramids.  The result was fairly clear--we could see the surface-level footprints of the features (note the dark disturbances along the top of the image to the left).  They were separated by an area with no real disturbance to be seen.  Thus they appeared to be separate interments--a theory confirmed when our partner at the cemetery, Greg Smith, discovered a historic photo of the day the burials occurred.  That photo clearly depicts three separate vaults.  That supplementary evidence both confirms our interpretation and suggests the effectiveness of thorough documentary research.

Tolomato Cemetery

An exhaustive survey of the Tolomato Cemetery.  Image courtesy of Matthew Kear.

The Tolomato Cemetery had been thoroughly mapped prior to our involvement by Matthew Kear as part of his M.A. in Historic Preservation from Cornell University.  However, many questions remain.  Why are there so many open spaces there?  Was no one buried in these areas, or have their grave markers simply been lost over the years?   

Field results of GPR in the Tolomato Cemetery.  Trowels (circled in white) indicate GPR anomalies identified as potential burials.  Volunteers standing along the row indicate the same, as we eventually ran out of trowels!

Hyperbolas indicate anomalies beneath the ground surface. Some may be burials.

Again using 2D GPR, we detected a series of hyperbolas, upside-down U-shaped anomalies located at the depth that burials occur.  Some of the hyperbolas may indicate burials.

We attempt to document a fallen headstone using GPR.

Another of the questions we sought to answer at Tolomato followed that research quite naturally: if headstones have fallen and become covered over, can we locate them using GPR?  We tested this using a known entity: a headstone that had fallen but was visible on the ground surface.

Fallen headstone clearly indicated at ground surface.

 It took only one quick pass of the GPR to document a headstone lying flat on the ground.  Given that headstones obscured by overgrowth lie just beneath the ground surface, we can infer that they would exhibit a similar signature.

Huguenot Cemetery

The Huguenot Cemetery once featured a crushed-shell pathway.

At the Huguenot Cemetery we ran a few tests that were similar to those already detailed.  One new effort involved locating an historic pathway that used to run through the cemetery, complete with a roundabout-like feature near the front gates.  This pathway has long since been abandoned; passersby today see only with grassy terrain.  Using an historic aerial photograph depicting both the cemetery and the City Gates, we estimated where the "roundabout" part of the pathway would have been for our GPR testing.

Pathway marker clearly detected just below ground surface.

Like the Dade Monuments and the fallen headstone from our other cemeteries, this feature appeared clearly through a flat disturbance near the ground surface. 

 These are a few of the many types of cemetery investigations that can be addressed through use of thorough research supported by GPR.  It can be used on a variety of surfaces and can detect a variety of different features, particularly when in the hands of a skilled technician.

GPR is a fantastic tool, and offers some unique benefits:

1. It's non-destructive.  Even if you mess up, all you've done is create bad data--that can be a pain to fix, but you haven't disturbed any sensitive deposits below the ground surface.

2. It creates a permanent record as you investigate.  It stores data as you carry out the investigation--it can be seen and interpreted as you go (in the field) and put through software for more concrete results later.

3. As demonstrated above, it can locate a variety of features--liners, vaults, crypts--it can even be used to map root systems!

We use GPR as a tool for engaging the public--all ages can be involved!
4. Community involvement--this one's a biggie for us.  All of our projects have been designed to engage cemetery stewards and the interested public.  When we can engage new people in investigating sites, we have new partners to protect and care for them as well.  GPR is an interesting and easy way to get others involved.  Once the equipment is set up, it just requires the user to push a button, push, and steer.  And even if mistakes are made, it's non-destructive!

In addition to these affordances, there are some constraints to take into consideration when deciding whether to employ GPR:

1. It's expensive.  GPR equipment, along with software can cost tens of thousands of dollars.  Companies exist that do GPR surveys professionally, and they can be expensive as well, depending on the extent of planned investigation.  Some cemeteries, such as St. Michael's in Pensacola, have secured grant funding and worked with the University of West Florida to conduct GPR research.

2.  It's only as good as its interpretation.  The work we did in the field, while conducted under some training, needed a highly trained eye for solid interpretation.  Anyone operating the equipment can see hyperbolas and other disturbances, but accurate interpretation of the aggregate data requires fairly extensive training.

3.  It can create overly confident results.  Despite the need for extensive training, it's really easy to get excited in the field and decide that we're seeing what we hope is there.  This is particularly true when partners and members of the public are involved--we all want it to be clear, easy and definite. We have to be cautious about what we think we see, particularly in the preliminary phases.  Moreover:

4. We can't be positive of anything without ground-truthing.  GPR isn't perfect, and neither are its users.  I once excavated a 1x2 meter unit with nearly nothing in it because the GPR was thrown off by a modern aluminum can just beneath the ground surface.  According to the data, it looked like a burial.  It wasn't.  It was a beer can and a LOT of dirt.

One way to counteract uncertainty in GPR is to start with the known--for instance marked graves--to learn the "signature" of a burial.  Then you can move on to unmarked graves, which are likely to have the same kind of hyperbola.

GPR provides information about burials and other sensitive subsurface features without disturbing them, and for that reason can be uniquely useful for investigating cemeteries.  However, it's best to remember that the better the preliminary research, the better the GPR data.  Cemetery stewards are wise to first conduct exhaustive research, and use that to guide any investigation using this technology.

- Copyright © Going Public - Skyblue - Powered by Blogger - Designed by Johanes Djogan -