Docstoc

Gettysburg's Codori Farm Lane Pr

Document Sample
Gettysburg's Codori Farm Lane Pr Powered By Docstoc
					                                                          Curt Musselman

      Using Global Positioning System and Geographic
              Information System Tools in the
           Rehabilitation of Cultural Landscapes:
 Gettysburg’s Codori Farm Lane Project




G
               ettysburg National Military Park has made use of geographic
               information system (GIS) tools on a regular basis since 1996,
               when full-time staff and dedicated equipment were made a part
               of the park’s resource planning division. The GIS equipment
               consists of Pentium-class computers running Microsoft Win-
dows and ArcInfo/ArcView software from ESRI, Inc. Global positioning
system (GPS) data collection is done using a Trimble ProXR receiver with a
TDC1 data collector. GPS data is post-processed with Trimble’s Pathfinder
Office software.
   Parkwide GIS data creation and of 1993 (“existing conditions”),
integration efforts at Gettysburg were 1927 (“commemorative era”), 1895
undertaken primarily to assist in the (“memorial association era”), and
development of planning and ar- 1863 (“battle era”). Battlefield land-
chaeological studies. These included scape changes were then analyzed
the National Park Service’s (NPS’s) within the GIS for both large- and
archaeological inventory program, small-scale features. The large-scale
the white-tailed deer impact study, features consisted of woods, fields,
and the cultural landscape inventory. orchards, and roads, while the small-
But completion of the park’s new scale features included fences, lanes,
general management plan (GMP) and individual trees. The signifi-
and environmental impact statement cance of key landscape features was
in 1999 provided the greatest also analyzed based upon both the
amount of support for collecting and level of battle action that actually
integrating all of the existing geo- took place and the importance of the
graphic data for the park area and feature from a military point of view.
consistently organizing it in a GIS Finally, the GIS was used to help
database.                                 develop and analyze six alternatives
   The GIS was used extensively in for the GMP. In addition to calcu-
the preparation of the GMP. First it lating areas and lengths, and pro-
was used to document conditions for viding map graphics, the GIS was
the entire 6,000 acres of the park as used to identify the key viewsheds in

40                                                 The George Wright FORUM
the vicinity of the park. The GIS also    a small cultural landscape report that
provided information on the level of      was to include a detailed treatment
resource impacts for each of the          plan. Eventually, the study area grew
GMP alternatives. The use of the          enough to warrant a change in the
GIS in the preparation of the GMP is      name of the project to the “Emmits-
discussed more fully elsewhere            burg Road Ridge Cultural Land-
(Musselman 1998).                         scape Report” (CLR). The northern
                                          boundary of this area was defined by
                                          the historic Codori Farm Lane,
    The demolition of the National        which was abandoned in the early
Tower overlooking the Gettysburg          twentieth century and largely for-
Battlefield on July 3, 2000, was the      gotten since (Figure 1).
inaugural event in the restoration of
the battlefield. But even before that
explosive event, Gettysburg’s newly
approved GMP had called for the               The GIS was used to support the
rehabilitation of much of the battle-     production of the Emmitsburg Road
field to its condition as of July 1863.   Ridge CLR as much as it was used
    NPS regularly rehabilitates de-       for the GMP. The GIS’s mapping
signed and natural landscapes, but        capabilities were put to use creating
rehabilitation of a battle landscape is   period plans for the battle, com-
not an easy task (Latschar 2001).         memorative, and current periods. A
The overall philosophy guiding the        series of battle action maps for the
rehabilitation work at Gettysburg is      Emmitsburg Road Ridge area was
included in the GMP. It also in-          automated and georeferenced and
cludes management prescriptions           key battle landscape features were
that support appropriate parkwide         identified in great detail. But in addi-
preservation treatments and actions.      tion to documenting conditions,
    The Codori–Trostle Thicket area       analyzing landscape changes, and
was the first part of the park where      calculating viewsheds, GPS tools
landscape rehabilitation efforts were     were combined with the GIS to pro-
planned for following the approval of     vide control points for map registra-
the GMP in 1999. It was anticipated       tion and to navigate to feature loca-
that this area would serve as the         tions in the field. Although the larg-
prototype for developing GMP im-          est landscape feature being rehabili-
plementation procedures that could        tated in this area is the Codori-Tros-
then be followed elsewhere in the         tle Thicket, the first feature to be re-
park. Working in cooperation with         habilitated was the Codori Farm
the Olmsted Center for Landscape          Lane, and it is upon the lane that the
Preservation, the park began work on      following discussion will focus.

Volume 19 • Number 1                 2002                                     41
   Numerous maps and photos               maps. The sources that were used for
document the existence of the             documentation included 19th-cen-
Codori Farm Lane. But recon-              tury photographs, the G.K. Warren
structing the fences along its edges in   map of 1868-1869, the 1895 Na-
the proper place required us to take      tional Park Commission map, the
great care in collecting and combin-      Adams County 1996 orthophoto-
ing information from the historic         graph, and a new set of existing-con-
maps with the existing-conditions         ditions maps. The existing-condi-

42                                                  The George Wright FORUM
tions maps contained one-foot con-      Corps of Engineers and is now in the
tours and were created at a scale of    Library of Congress map collection.
1:600 by a local engineering firm.      This map was scanned at 300 dots
Feature locations were based upon       per inch by the Library of Congress
field measurements and photogram-       as a part of the American Memory
metric compilation from 1998 black-     project and provided to the park
and-white aerial photographs.           (Figure 2).
    To register the historic maps to       The scanned historic-map images
the existing-conditions maps, we        were first registered to the existing-
used building corners and fence in-     conditions base maps using the
tersections as control points. Since    ArcInfo command REGISTER.
the study area boundary had grown       This command performs an affine
after the existing-conditions maps      transformation so that if more than
were contracted for, we only had        three control points are used (we
smaller-scale (1:7200) base maps for    used nine) the image locations can
some of the areas that we wanted to     not be exactly matched to the corre-
register the historic maps to. There-   sponding map locations (ESRI
fore, we took GPS measurements at a     1994). To get a more exact match of
number of the fence intersections       the historic data to the existing-con-
and used those values for the exist-    ditions base map, we next digitized
ing-conditions control points. The      the historic features into a vector
positional accuracy of our existing-    ArcInfo coverage. These features
conditions control points was within    were then rubbersheeted while the
two feet on the ground, regardless of   control points were linked exactly
whether the point had come from the     using the ADJUST command. The
1:600 base maps or the GPS meas-        advantage of using a vector coverage
urements.                               is that one can also use the HOL-
    We used the G.K. Warren maps        DADJUST command if there are
of 1868-1869 to determine the loca-     arcs that one does not want to move
tion of the Codori Farm Lane at the     when rubbersheeting.
time of the battle. Although the full
set of original Warren maps is in the
National Archives, the archives staff
and their vendors could only provide       Once the historic and existing
a digital copy at a resolution of 72    features were registered to one an-
dots per inch, which was too fuzzy      other, we used the GIS to make cal-
for our purposes. Luckily, an exact     culations of lengths for fences and
tracing of the map of the part of the   lanes and to determine the acreages
battlefield that we were studying had   of orchards, woods, thickets, and
been made in 1886 by the U.S. Army      stream buffers. The measures were

Volume 19 • Number 1               2002                                    43
then used to help figure the costs for    Produced from true color photos
rehabilitating these historic features.   taken in the spring of 1996 to sup-
A number of viewshed and line-of-         port mapping at a scale of 1:4800,
sight analyses were also run to show      these images have a pixel resolution
the impact on the views of rehabili-      of 2 feet on the ground. A linear,
tating the historic Codori-Trostle        light-colored crop mark is visible on
Thicket at different heights.             the orthophoto in the same location
    Determining the location of 1863      where the registered historic maps
features in relationship to the exist-    show the Codori Farm Lane.
ing conditions was accomplished as a
part of the map registration process,
but it was a very important part of the
analysis provided for this project by        Having georeferenced the historic
the GIS. The location of the Codori       base maps, the next step was to cre-
Farm Lane in 1863 was also overlaid       ate waypoints along the Codori Farm
with the digital orthophotos created      Lane so that we could navigate to
by the Adams County GIS office.           them using the GPS data collector.

44                                                  The George Wright FORUM
Only the fence location along the        it is possible for us to use real-time
northern side of the lane was to be      differential GPS at Gettysburg.
flagged. The fence on the southern       Although not as important when we
side of the lane was simply built par-   are collecting data (because we
allel to, and 20 feet to the south of    routinely post-process all the GPS
the northern fence. We created way-      files that we collect), DGPS is
points at each end of the lane and at    essential for navigating to and
every point along the lane where it      locating features when one wants to
changed direction, where the type of     be within a meter or two.
fence changed, or where another              There are a number of modes that
fence intersected it. The waypoints      can be used to navigate with the
were created as ArcView shapefiles       Trimble ProXR GPS unit, but we
and then imported to the Pathfinder      had the greatest success using the
Office software where they were up-      bearing and distance mode. Since a
loaded to the GPS data collector         differentially corrected location is
(Figure 3).                              only calculated every five seconds, it
    Because we are within range of the   is important to slow down one’s
Cape Henlopen DGPS radio beacon,         walking pace when the GPS indi-




Volume 19 • Number 1                2002                                    45
cates that one is within about 20 feet
of the waypoint. We knew we were at          Building the fences that defined
the waypoint when the distance to go     the Codori Farm Lane was a coop-
remained in the one- to two-foot         erative project between the Friends
range even though the bearing to the     of the National Parks at Gettysburg
waypoint kept changing. We put           (FNPG) and the resource planning
flags at all the waypoints and in a      and maintenance divisions of the
straight line every fifty feet between   park. After the locations of the future
the waypoints.                           post-and-rail and Virginia worm
    The year before this project we      fences had been laid out using GPS,
had done some navigation tests with      the park’s landscape preservation
our ProXR unit to determine how          team prepared the work site. Fence
confident we could be in our ability     rails, posts, and cross-ties were de-
to navigate to a given coordinate lo-    livered in bundles spread out along
cation. Earlier, a scenic easement       the line the lane was to follow. For
boundary on Gettysburg College           the sections of the fence that were to
land had been marked by buried sur-      be Virginia worm, flat stones to sup-
veyor’s pins whose coordinate loca-      port the bottom rail were placed in a
tions had been determined to within      zigzag pattern every ten feet centered
one-half meter when the pins were        on the flagged line. Post-holes were
put in. Over a year later, when there    drilled every ten feet along the part of
were no longer any visual marks on       the line that marked the post-and-rail
the ground to tell us where the pins     fence. Actual construction of the
would be found, we used the GPS          fences was accomplished by a large
unit to navigate to the pins. A metal    number of FNPG volunteers during
detector was then used to find the       their spring workday. One observer
buried pins. One hundred percent of      reported that the scene resembled an
the pins were found within six feet of   ant colony as the volunteers, working
the location we had navigated to us-     in pairs, carefully placed hundreds of
ing the GPS.                             fence rails one by one along the now-
                                         reestablished Codori Farm Lane
                                         (Figure 4).




46                                                  The George Wright FORUM
    Since Codori Farm Lane has been           Using GPS and GIS tools to assist
rehabilitated, we have begun land-        in the rehabilitation of the Codori
scape preservation treatments for the     Farm Lane and other historic fea-
other historic features in the Em-        tures has proven to be very useful. At
mitsburg Road Ridge study area fol-       the beginning of the project, the GPS
lowing the recommendations in the         was used to collect control-point
CLR. A number of other fence lines        measurements that were then used in
have been rebuilt using the process       the rectification of historic maps and
that was developed for Codori Farm        images. The ability to use GPS to
Lane, and in the coming year we ex-       navigate to waypoints for laying out
pect to rebuild the part of the Trostle   the location of historic features was
Farm Lane that was east of the Tros-      invaluable. Along with specialized
tle Farm and south of United States       map creation, the GIS provided inte-
Avenue. Rehabilitation of the             gration of data collected in the field
Codori-Trostle Thicket as well as         with the park’s existing base maps.
the Neinstedt Field has also begun.       Image rectification tools made it pos-
    On a parkwide basis, GIS tools        sible to fit scanned historic maps to
have been used to help start work on      these base maps, too. Since all of this
a five-year implementation plan for       mapping was integrated using a con-
rehabilitation of the major large-scale   sistent map projection, coordinates
landscape features that were identi-      for waypoints could be derived and
fied in the GMP. Additional treat-        used in the field with the GPS data
ment principles are also being devel-     logger. GIS calculations for the
oped that will be appropriate for use     lengths and areas of features helped
throughout the park.                      to define the scope of work and costs
    One recent equipment upgrade          that would be associated with various
that has been particularly valuable       rehabilitation alternatives. Finally,
when laying out the location of his-      viewshed and line-of-sight analyses
toric features was the replacement of     were used to help confirm narrative
the TDC1 data collector by a TSC1         and photographic evidence related to
data collector. The TSC1 provides a       the height of the Codori-Trostle
map display, so that instead of load-     Thicket in 1863.
ing waypoints, the entire georefer-           Having in-house GIS and GPS
enced historic map can be loaded as       capabilities has made possible a
a background. In order to navigate to     quick turn-around on many of the
locations of interest on the map, one     tasks associated with carrying out a
just needs to observe the track of the    cooperative project such as the reha-
GPS and keep walking until the track      bilitation of the Codori Farm Lane.
intersects the point of interest.         In addition, we were able to deter-

Volume 19 • Number 1                 2002                                     47
mine—with an appropriate level of             historic feature, guided primarily by
confidence—the location of a missing          maps and photos.


ESRI [Environmental Systems Research Institute]. 1994. ARC Commands, ArcInfo Version 7.
   Redlands, Calif.: Environmental Systems Research Institute.
Latschar, John. 2001. Remarks from the superintendent at the quarterly Gettysburg National
   Military Park staff meeting. 18 September. Unpublished.
Musselman, Curt. 1998. Using GIS for general management planning at the Gettysburg
   National Military Park. In Proceedings of the 1998 Annual Meeting of the American
   Congress on Surveying and Mapping. Bethesda, Md.: ACSM.

Curt Musselman, Gettysburg National Military Park, 97 Taneytown Road,
  Gettysburg, Pennsylvania 17325; curt_musselman@nps.gov


                                         y




48                                                        The George Wright FORUM