The White Springs Site: Applying Tobler's Hiking Function
Finger Lakes Region, New York State
Using Tobler's Hiking Function is a natural next step for my research, as it would refine the way we define regional territories. My thesis model uses an amorphous but relative cost unit to measure distance in order to find the most likely and efficient routes. However, Tobler's approach uses travel time, which can be quantified more concretely.
Eucledian distances from the White Springs Site to contemporaneous Haudenosaunee settlements and European trading posts
Archaeologists often describe territories, political economies, ranges, and catchments in Euclidean distances. For example, Kurt Jordan's book The Seneca Restoration (2008) examines the 18th century Seneca settlement of White Springs. He proposes classifying satellite communities based on their distance from major towns, using categories like local (within 20 km), regional (20-80 km), and extra-regional (beyond 80 km). However, this does not account for terrain or water routes.
Tobler’s Hiking Function:
Tobler’s hiking function is an exponential function that determines hiking speed taking the gradient into account.
Implementing Tobler in GIS: Dr. Nicholas Tripsevitch (2007)
Using a custom Vertical Factor table with the steps outlined as follows:
1. Calculate slope in degrees from your DEM using the Slope tool.
2. Build a lookup table relating slope in degrees to the vertical factor, calculated as:
2. Build a lookup table relating slope in degrees to the vertical factor, calculated as:
VF = 1/(R/1000) / (6 * e^(-3.5 * |tan(π*D/180)| + 0.05))
Where R is the DEM resolution in meters.
3. Use Path Distance, inputting your slope raster and the vertical factor lookup table.
4. Specify your starting location(s). The output shows the hiking time based on Tobler.
4. Specify your starting location(s). The output shows the hiking time based on Tobler.
The custom vertical factor table approach allows implementing Tobler's equation to account for varying walking speeds based on slope. Since the VF is the reciprocal of the original function, it penalizes steeper slopes properly. This provides a more realistic model than using slope alone.
Unfortunately the vertical factor table option now longer works in ArcMap or ArcGIS Pro versions (as of 2023).
Dr. Devin Alan White's (2015) Tobler Equation
I searched for other sources and solutions implementing different adjustments to Tobler's hiking function for use in ArcGIS. However, nothing worked properly until I found a 2015 piece by Devin White that guided me in the right direction. His approach produced a surface where each cell's value represented the time needed to traverse that cell, using this equation:
T = (R/1000) / (6 * e^(-3.5 * |tan(π*D/180)| + 0.05))
Where:
T = time in hours
R = spatial resolution of DEM in meters
D = slope in degrees
R = spatial resolution of DEM in meters
D = slope in degrees
I still ran into an issue with the resulting values of the least cost path (LCP) polylines. When running the path distance tool, the cost of moving from one cell to the next increases by the value of the next cell. So the LCP adds all these values together, resulting in incorrect travel times. My workaround is to rerun the Path Distance tool as a raster and record the last raster cell's value as the total time for the path. This produces more realistic values.
So far this is my preliminary implementation of Tobler's equation. More research and testing are needed.
Approximate hiking distance and speed from White Springs using Tobler's hiking function. This takes slope and water in to account
By implementing Tobler's Hiking Function and incorporating slope (with water as a barrier in this case), I could refine these regional boundaries to reflect "real world" travel times and define regional territories in a more meaningful way. This could help answer questions about intragroup dynamics and accessibility to trade networks. For instance, how long would it take to travel from White Springs to Albany or Fort Niagara? Incorporating travel time could reveal new insights into Seneca political economies and ability to respond to outside threats. It could also strengthen my current model and allow for richer analysis of Seneca settlement patterns and relationships.
Creating a shovel test grid for Cultural Resource Management in GIS
Astoria Cove Development Project in Queens, New York
The red polygon represents our area of interest. The layers of concrete and asphalt that still cover the area may have left it relatively undisturbed. Additionally, the red diamond represents a fictional datum point for the purposes of this example.
Dr. Celia Bergoffen tested this site. She originally did all measurements with a tape and compass. However, this is an example of how the site could have been digitally mapped using QGIS. First, a polygon is created that outlines the study area. Then, the Vector Creation > Create grid tool in QGIS can be used. Different kinds of grid shapes can be created, but here points are used for the centroids of Shovel Test Pits. Finally, the grid is created using the study area polygon as its extent.
It is usually preferable to align the grid with reference points and site features, like the datum point and parking lot orientation. This grid shows 5 meter spacing.
First, the grid points were moved here so they originate from the datum. Then, they were rotated to match the linear orientation of the parking lot, while maintaining their original distribution. Finally, the unneeded grid points were removed using the intersect tool.
The coordinate attributes are from the original grid locations. Moving the points made these coordinates incorrect. However, we can update the northing and easting coordinates using the field calculator. Then we can remake the attribute table.
As a result, this process produces a regular grid of geographic coordinates based on a defined datum and site orientation. You can then export the coordinates to high precision GPS units, such as the Emlid Real-time kinematic positioning (RTK) Global navigation satellite system (GNSS) Receivers. Their ReachView software can import the .CSV values, thereby providing approximately 1-2 centimeter accuracy in the field.
Haudenosaunee Place Names Project
Faithkeeper Steve Henhawk, Dr. Kurt Jordan, Nicholas Perez
The map Map of Ho-De-No-Sau-Nee-Ga or the people of the long house representing the state as it was in 1851 (but produced later) was compiled by Lewis H. Morgan and Ely S. Parker, a Seneca sachem.
It is loosely based on French maps of 1720 or earlier, showing trails, lakes, rivers, settlements, territories, and principal locations with their (transcribed) Indigenous names.
We compiled all of the transcribed Indigneous names, their English signification, and original meaning from various historical sources including Morgan who wrote extensively about the Haudenosaunee, as well as the journals of Cammerhoff and Zeisberger, and Beauchamp who wrote about their journeys
By georeferencing Morgan's map and comparing it to topographic maps and modern maps, I identified locations and trails either mentioned or illustrated in these historic texts and maps.
Other sources like the English-Cayuga Dictionary contain lists of places that I also mapped. Steve Henhawk, an L1 speaker of Cayuga is working to decipher place names recorded phonetically in historical sources. As these phonetic spellings can be challenging to identify, he has been translating the names and pinpointing other locations in the Finger Lakes area of New York, the territory of the Haudenosaunee Confederacy.
The ultimate goal of this project is to preserve and reintroduce these place names as part of a broader effort to revitalize the Cayuga language. Going forward, we plan to travel through the region with recording equipment and GPS to identify additional locations and document the stories, as narrated by Steve Henhawk, associated with and contained within these places.
