GSA Annual Meeting in Denver, Colorado, USA - 2016

Paper No. 126-5
Presentation Time: 2:50 PM

STORIES FROM THE ARCHIVES: NEW DIGITAL RESOURCES ENABLE USE OF OLD CASE FILES FOR DEVELOPING A LANDSLIDE DATABASE


DELANO, Helen, Pennsylvania Geological Survey, DCNR, 3240 Schoolhouse Road, Middletown, PA 17057, hdelano@pa.gov

Landslide susceptibility studies require an accurate inventory of past landslide occurrences. The Pennsylvania Geological Survey (PaGS) has started a digital landslide study of Allegheny Co. (Greater Pittsburgh), Pa. Publicly available digital tools now make it possible to confidently locate a number of landslide features from old case files that would have been difficult to locate only a few years ago.

Case files at PaGS date from the 1960s to present. Although type and amount of information varies, each documents a known landslide event that led to a request for investigation or information. The case files were written without considering that in the future we would want to render them to an accurate point in a GIS dataset. Similarly, media (newspaper, and more recently TV and internet) reports are evidence that a landslide occurred and someone noticed. They may have photos but only a vague location. Files may include a circle on a map, or only a street name, neighborhood, or owner name. Most Pittsburgh landslides are small, so precise location is essential, and mapping, even with good orthophotos and lidar, is not always straightforward.

Internet map services rapidly locate road names but may yield only approximate house numbers. Street or bird’s-eye views allow comparison with field photos to identify structures and features on slopes. Google Earth’s collection of historical imagery can help locate and confirm buildings now demolished. Allegheny County’s excellent Real Estate Portal links accurate addresses with parcel maps on aerial photos, with some information about recent sales and owner names, which is useful in confirming addresses, owners and abandonments. Digital orthophotos and lidar-derived elevation data provide a check on location and extent of a feature, especially where repair or debris removal is not complete.

Identification of landslides by lidar alone in an urban area can miss many historical features masked by repair or later development. The ability to locate historical events accurately adds to the inventory and improves conclusions based on its analysis. The same approach has been useful in mapping water wells from old records, and may be applicable to other uses of archival data.