Two furnaces and a forge (Lehigh and East Penn Furnaces and East Penn Forge; Figure 1) were built in the Lehigh Valley in the 1820s and 1830s [2, 3, 4, 5] because of the local availability of iron ore, limestone (needed as a flux during smelting) and water power (needed to power the bellows). However, the availability of charcoal was clearly important as well. The iron works were built on the northern and southern sides of the Blue Mountain, which, unlike much of the area, was uncultivated and contained large stands of forest. Along the mountain and deep in the forest, the timber was converted into charcoal at charcoal hearths, which are 10–15 meter in diameter circular areas that are flat and level [6, 7, 8, 9]. Because much of the mountain is sloped, these regular, flat, level areas are relatively easy to see in data derived from LiDAR (Light Detection and Ranging or Airborne Laser Scanning) [10, 11]. LiDAR data and derivatives are openly available from the state of Pennsylvania [12, 13]. However, much of the streaming data provided by the state is subsampled . Herein, I report on the methods and results of using openly available raw LiDAR data and free-and-open source software (FOSS) to produce high resolution digital elevation models as well as hillshade and slope derivatives. This allowed the identification of 758 potential charcoal hearths within the research area.
The data herein are from a 20.5 km stretch of the Blue Mountain (also known as the Kittatiny Mountain), in eastern Pennsylvania from the Lehigh Gap in the northeast to route 309 in the southwest and includes approximately 56 km2. The data straddles the counties of Lehigh, Carbon and Schuylkill and encompasses much of Pennsylvania State Game Lands #217. Although the northeastern boundary of the Lehigh Gap represents a historical limit to charcoal production, the southern boundary does not. Evidence of charcoal production is present south of Route 309 and extends for some distance. A GeoJSON polygon of the research area is included herein (Figure 1).
Geographic Coordinate System: World Geodetic System (WGS) 1984 (EPSG: 4326).
Datum: World Geodetic System (WGS) 1984.
Northern boundary: 40.79576 (decimal degrees)
Southern boundary: 40.70750
Eastern boundary: –75.60758
Western boundary: –75.828852
The purpose of the data discussed herein was to enable the identification of charcoal hearths from LiDAR data. While streaming hillshade data provided by Pennsylvania was originally employed, resolution was low. Therefore, the original data were acquired and a DEM, a hillshade model and a slope analysis were constructed. From these, particularly the latter, 758 potential charcoal hearths were recognized. The methods are described in detail in the “Description of Methods for Identifying Charcoal Hearths along the Blue Mountain of Pennsylvania”  which is part of the data presented herein.
Originally, a streaming hillshade provided by PAMAP, a collaborative project of local, state and federal agencies was used to identify charcoal hearths. Data is available though the Pennsylvania Spatial Data Access program as well as via the user-friendly Pennsylvania Imagery Navigator (Pennsylvania Spatial Data Access “Pennsylvania Imagery Navigator and Download Tool.” [Online]. Available: https://maps.psiee.psu.edu/ImageryNavigator/. [Accessed: 05-Mar-2018]) and was viewed via WMS in QGIS, an open source GIS package.
While I was able to identify numerous 298 charcoal hearths in the streaming data, it soon became clear that resolution was lower than the original data. The streaming data is based upon “Class 8,” or Model Key, which are “thinned-out ground points used to generate digital elevation models and contours” . Therefore, I downloaded the original post-processed LiDAR data in .las format .
The free-and-open source software (FOSS) packages known as LASTools ((version downloaded March 3, 2018) and QGIS (version 2.18.17) were employed to convert the original LiDAR data into a digital elevation model (DEM), a hillshade and a slope analysis (Figure 2). Details of this analysis are presented in the accompanying documentation.
The slope analysis was reviewed by the author and Heather Lash, a student at Muhlenberg College. Charcoal hearths are particularly distinctive in the hillshade analysis, when they are on slopes. Using default visualization in QGIS, flat areas appear dark and sloped areas appear light. Therefore, on the lighted slopes, flat, level areas that are c. 10–15 meters in diameter are particularly distinctive. Often, hearths were dug into the side of the mountain and the fill was placed on the lower side of the hearth. Therefore, the slope immediately above and below the hearth is steeper (lighter) than the surrounding terrain. A total of 758 possible charcoal hearths were identified (Figure 3).
The entire area was reviewed by Carter and the northern half was reviewed by Lash.
Lash and Carter reviewed the area separately (Lash only reviewed the northern half of the research area) at a magnification between 1:1250 and 1:10,000. Any hearths identified by both were considered highly likely to be hearths. Locations that one identified as a hearth, but the other did not (n = 174) were reassessed by Carter. Each of these locations were examined using streaming satellite photos (via Google Hybrid in the QuickMapServices plugin; version 0.19.6) and the Profile Tool plugin (version 4.0.2) in QGIS. When based upon the DEM, the Profile Tool shows the profile of a line drawn across the potential hearth. Many of these questionable features, though somewhat similar to hearths, were recognized as false positives (n = 61). Others were confirmed as potential charcoal hearths (n = 113). After the recognition of false positives through the comparison of the work by the two analysts, Carter reviewed all identified hearths and was able to determine that an additional 13 (eleven in the southern half that Lash did not review) were unlikely to be hearths.
This method works particularly well on slopes (over c. 10%). Charcoal hearths on relatively flat areas (<10%) are particularly difficult to distinguish. We are developing methods to fill in this gap.
(3) Dataset description
This dataset includes six objects that each have their own DOI. These include a description, three raster files (DEM, hillshade and slope) and two vector files (research area and potential hearths).
- Description of Methods for Identifying Charcoal Hearths along the Blue Mountain of Pennsylvania.
- Digital Elevation Model for Blue Mountain Charcoal Research Project.
- Slope Analysis of “Digital Elevation Model for Blue Mountain Charcoal Research Project”.
- Hillshade Analysis of “Digital Elevation Model for Blue Mountain Charcoal Research Project”.
- Blue Mountain Charcoal Project Research Area.
- Identified Charcoal Hearths from “Slope Analysis of ‘Digital Elevation Model for Blue Mountain Charcoal Research Project’”.
- Description of methods.
- Processed data from original LiDAR data as described in (1).
- Processed data from (2).
- Processed data from (2).
- Primary data.
- Interpretation of data from (3).
Format names and versions
- Portable Document format (PDF)
Data was created between 01/09/2017 and 29/05/2018.
Benjamin Carter, Data Collector, Muhlenberg College (ORCID iD: 0000-0002-7464-0989).
Heather Lash, Data Collector, Muhlenberg College (ORCID iD: 0000-0002-6490-811X).
CC BY 4.0
- DOI: 10.5281/zenodo.1255101 and <http://opencontext.org/media/f70b5ca7-af1d-472e-8f57-32733652adca>
- DOI: 10.5281/zenodo.1252441 and <http://opencontext.org/media/e9939449-f5f1-4dd4-9bff-5ce41953f138>
- DOI: 10.5281/zenodo.1252977 and <http://opencontext.org/media/b229cc1b-ed10-4939-b872-2f7349290cf5>
- DOI: 10.5281/zenodo.1252520 and <http://opencontext.org/media/52bf4905-d03d-4997-b54b-eb287f0c91c9>
- DOI: 10.5281/zenodo.1252418 and <http://opencontext.org/media/7fd5f761-3391-4ca3-bb60-85aa578fb40d>
- DOI: 10.5281/zenodo.1252985 and <http://opencontext.org/media/6131e427-b6ec-4ee1-84fb-32fa670d3c47>
(4) Reuse potential
The reuse potential of this work lies within the methods. This analysis employs openly available LiDAR data of which there is a significant and rapidly growing amount available in individual countries and states, as well as in aggregators (OpenTopography and USGS). Similarly, because this technique employs openly available tools (LASTools and QGIS), the methods are replicable by anyone with a computer and internet access. However, these methods will need to be adapted for other LiDAR datasets with different characteristics. Anyone wishing to identify charcoal hearths (or other historic features) in Pennsylvania can use these methods as is. Additionally, the location of specific charcoal hearths within the research area can be used to further understand historic use of this area as well as a comparative data set for other areas in the US and beyond [e.g, 10, 11]. Similarly, this could easily be used as a pedagogical tool within GIS, archaeology or history courses to demonstrate the relevance of LiDAR data to small-scale local situations (rather than, say Maya settlements).