Can you Dig it?: Identifying and Protecting Ancient Chacoan Ruins

Category: Land Cover Change & Disturbances
Project Team: Chaco Canyon Cross-Cutting
Team Location: NASA Marshall Space Flight Center at National Space Science Technology Center – Huntsville, Alabama

This map illustrates results from the Chaco sites habitat suitability model. Inputs included normalized difference vegetation index (NDVI), slope, aspect, elevation, land cover, pre-European land cover, and Terra ASTER Global Emissivity Dataset. Image Credit: Chaco Canyon Cross-Cutting Team

This map illustrates results from the Chaco sites habitat suitability model. Inputs included normalized difference vegetation index (NDVI), slope, aspect, elevation, land cover, pre-European land cover, and Terra ASTER Global Emissivity Dataset. Image Credit: Chaco Canyon Cross-Cutting Team

Authors:
Kelsey Herndon
Dashiell Cruz
Sydney Neeley
Ryan Schick

Mentors/Advisors:
Dr. Jeffrey Luvall (NASA at National Space Science Technology Center)
Dr. Robert Griffin (University of Alabama in Huntsville)
Dr. Tom Server (University of Alabama in Huntsville)

Abstract:

The Chacoan people flourished in northwest New Mexico between A.D. 850 and 1150. Today, remnants of their monumental architecture draw more than 40,000 visitors a year to Chaco Canyon National Park to experience the natural grandeur of the area and to learn about Native American history throughout the San Juan Basin. However, many Chacoan roads and communities are located outside the boundaries of the National Park. These unprotected sites are threatened by encroaching infrastructure associated with resource extraction, such as drill pads with requisite access roads and pipelines. Currently, the National Park Service (NPS), Binghamton University, the University of Nebraska-Lincoln, and the University of Colorado Boulder rely on expensive and time-consuming ground surveys, imagery from Google Earth, and the Landsat series to identify the extent of Chacoan roads and houses. This project used Landsat 8 Operational Land Imager (OLI) surface reflectance, Shuttle Radar Topography Mission Version 2 (SRTM-v2) digital elevation models (DEMs), Terra Advanced Spaceborne Thermal Emission and Reflectance Radiometer (ASTER) emissivity data, Thermal Infrared Multispectral Scanner (TIMS) data, Hyperspectral Thermal Emission Spectrometer (HyTES) data, Sentinel-1 Synthetic Aperture Radar (SAR) data, and Sentinel-2 MultiSpectral Instrument (MSI) data to identify highly probable locations of unknown Chacoan sites. The goal was to determine sites at risk from infrastructure development and identify the spectral signatures of these ancient communities. Documenting Chacoan community signature profiles and determining which areas are at risk of being affected by encroaching development will help project partners to better understand the Chacoan landscape and better protect and preserve these ancient sites.

Previous story | Main Page | Next story

Topic:

31 Comments

Julia Marrs 07-09-2016, 17:36

This sounds like a very exciting project, and I really enjoyed your video. When creating your final risk maps, were you able to take into account sites that were at higher risk because of existing damage or age, etc.? Was this something that the remotely sensed data could give you insight on, or did you happen to get any field-collected data on that project? The risk map in the video looks like it will be a great resource for your project partners. Congratulations!

Reply
Kelsey Herndon 07-09-2016, 20:40

Hi Julia,

Thanks for your comment! This iteration of the risk map did not take into account existing conditions of the sites. It would be very interesting to incorporate the stability (existing preservation), materials (are features made from stone, adobe, wood, compact earth?), and other in situ observations. We did use in situ locations for our suitability map, however we did not have in situ data for sites that had been destroyed or disturbed. We recognize this as a limitation of our project and have already been in talks with our project partners to try and find a solution. Many state historic preservation offices have records of some sites that have been disturbed or destroyed, so that is where we will start our search in the continuation of this project!

Thanks again for your question!

Kelsey

Reply
Julia Marrs 07-09-2016, 22:33

Hi Kelsey, thanks for your reply and for the update! That’s great that more in situ data exists in other state offices. Sounds like a very exciting continuation project is in the works.

Reply
Alison Thieme 07-09-2016, 09:46

Fascinating project! The risk map makes it look like most of the region is at extremely high risk from expanding infrastructure. Can you incorporate a mask for areas that are currently protected into the risk map?

Reply
Kelsey Herndon 07-09-2016, 12:20

Hi Alison! Thanks for your comment! We actually did end up incorporating a mask for areas that are already protected by National Park boundaries or NGOs. Unfortunately we completed the final Chacoan Site Risk Map after completing our VPS video. Additionally, we hope that a continuation of this project might develop a more refined risk map by considering the known locations of disturbed or destroyed sites. When we presented our results to the project partners, they were very excited about the implications of an accurate risk map, not only for direct protection of at-risk sites, but also as a tool to engage the public, politicians, and other decision makers.

Thanks again!

Kelsey

Reply
Diego Pons 02-09-2016, 11:59

Great project! I wonder if something similar has been done or could be done in Mesoamerica with other archaeological sites? and perhaps with other relevant threats to the region different than oil exploration?

Reply
Good question Diego! I can recall reading an article about Archaeologist using remote sensing to explore ancient Mexico (Source: Rochester Institute of Technology) 06-09-2016, 17:15

Good question Diego! I can recall reading an article about an Archaeologist, named Bill Midddleton, using remote sensing to explore ancient Mexico (Source: Rochester Institute of Technology) The article claimed that “Satellite imagery obtained from NASA will help archaeologist Bill Middleton peer into the ancient Mexican past. In a novel archaeological application, multi- and hyperspectral data will help build the most accurate and most detailed landscape map that exists of the southern state of Oaxaca, where the Zapotec people formed the first state-level and urban society in Mexico.”

Reply
Kelsey Herndon 07-09-2016, 12:31

Hi Diego! Thanks for your question! Similar work has been conducted in the Peten region of Guatemala. Archaeologists and geospatial scientists have used Ikonos satellite imagery to try and identify archaeological sites by assessing vegetation health. The Maya used enormous amounts of lime plaster in the construction of their monumental architecture, and some argue that high concentrations of lime in the soil (from the ancient buildings) affect the health of the vegetation that grows on top. Different regions where these studies were carried out had varying results; for example, work at the site of San Bartolo and surrounding areas resulted in the successful identification of many structures, while work at Ceibal and El Zotz had less definitive results.

Thanks again!

Kelsey

Reply
Amber Jones 22-08-2016, 13:29

Great job on this video! It was fun to watch, and easy to understand. Unique topic too.
I’m curious how data sets with varying spatial resolution can be used together? It seems like combining 4m – 30m data can skew results. Is there someway your team can reconcile this?

Thanks!

Reply
Dashiell Cruz 06-09-2016, 17:09

Good Question Amber! So when we used spatial data with varying spatial resolution, tools in ArcMap 10.3 are able to re-sample all of the data to the same spatial resolution. What we did for our project is re-sampled all the data to 30m resolution. However, the re-sampled 90m resolution data layers do not show as much variance as the 30m resolution layer. Let me know if you have any further questions or would like any further information!

Reply
Kelsey Herndon 07-09-2016, 20:30

Hi Amber,

Thanks for your question! Just to follow up Dash’s response: The 4m HyTES data and the 5m TIMS data were analyzed independently and were not combined with any other datasets. We manipulated these two datasets by examining different band combinations, band ratios, stretching methods, and filters to see if we could identify archaeological features. In our risk map and suitability map there were inconsistencies with the resolution of our inputs, so we resampled the rasters in ArcMap.

Thanks again for your question!

Kelsey

Reply
Nancy Herndon 18-08-2016, 22:41

What a fun video! Were you able to validate your suitability map? How did you determine what inputs to use with the suitability map?

Reply
Dashiell Cruz 06-09-2016, 17:22

Thanks for watching Ms. Herndon!
So our suitability map will be vetted by employees of the National Park Service (NPS). Our Impact Analysis team will follow up with the NPS to determine the validity of the suitability map. That being said, the inputs used to create the suitability map were determined by great literature review about the Ancient Chacoan people! It took a great understanding of social, spiritual, and natural processes to determine each input for the suitability map. Thanks for asking this question!

Reply
Kelsey Herndon 07-09-2016, 12:34

Hi Nancy! Thanks for your question!
We were able to access the accuracy of our suitability map. When we rant the MaxEnt model, we trained the model on 80% of the in situ points provided by our project partners and then used the remaining 20% to test for accuracy. We had an AUC for the test data of 81%. We hope that future work on this project can help improve the accuracy of the model.

Thanks again!

Kelsey

Reply
Brian Ramsdell 18-08-2016, 22:34

Great video! I noticed that you examined imagery from several different satellites and sensors. Did you find that one of these worked best for identifying sites and roads? Was there any data that did not work for identifying sites?

Reply
Kelsey Herndon (Chaco Canyon Cross-Cutting) 18-08-2016, 22:45

Thanks for your comment Brian! We did find that the HyTES land surface temperature data with a decorrelation stretch worked best for delineating known sites. In two cases we were able to potentially identify archaeological features that were previously unknown. Unfortunately, the HyTES data was piecemeal and only overlapped with two known sites. Hopefully in the future we can obtain additional HyTES data.

Reply
Dash Cruz (Chaco Canyon Cross-Cutting) 07-09-2016, 16:58

Thanks for watching Brian! On the other side of what Kelsey Ramsdell mentioned, we found that the 90m ASTER data was far too coarse to identify sites in the San Juan Basin! The HyTES data was more beneficial because it comes in better resolution than the ASTER data.

Reply
Daryl Ann Winstead (Mekong River Basin Agriculture) 18-08-2016, 17:06

Very interesting use for the data! What were some limitations and what could be done in the future? Thanks in advance for your response!

Reply
Kelsey Herndon (Chaco Canyon Cross-Cutting) 18-08-2016, 22:20

Thanks for your comment Daryl Ann! One of the most glaring limitations was that we were not able to complete an accuracy assessment for the Chacoan Sites Risk Map. An accuracy assessment would require the known locations of destroyed or disturbed Chacoan great houses, which we were unable to acquire during this term. However, our project partners have suggested some possible sources for this type of data that we will hopefully be able to use in a continuation project.

I hope this answers your question!

Thanks!

Reply
Daryl Ann Winstead (Mekong River Basin Agriculture) 19-08-2016, 12:28

Hi Kelsey! Thanks! It did indeed answer my question!

Reply
Sara Lubkin 15-08-2016, 18:48

Cool project! Is LIDAR available for the area? Recently, I’ve read a lot about LIDAR being used to identify archaeological sites.

Reply
Sydney Neeley 16-08-2016, 12:01

Thanks, Sara! You are absolutely correct that LIDAR has become an important tool in identifying archaeological sites. We were not able to acquire LIDAR data for this study, however, our project partners are currently working on obtaining LIDAR for future studies.

Reply
Sara Lubkin 02-09-2016, 15:29

Congratulations on making the finals!

Reply
Shane Neeley 12-08-2016, 15:46

Digs along the roads could find new artifacts! We should dig the roads before someone puts an asphalt road over it. One question; when you say the roads exist outside of the protected boundaries of the Chaco Canyon National Park, do you mean they are outside the park, or that some parts of the park are not protected?

Reply
Sydney Neeley 16-08-2016, 11:56

Great question, Shane! We mean that some Chacoan sites and roads exist outside of the boundaries of the Chaco Canyon National Park, and therefore, are unprotected. All of the Chacoan sites and roads within the national park boundaries are protected from encroaching development.

Reply
Kelsey Herndon (Chaco Canyon Cross-Cutting) 18-08-2016, 22:27

Hi Shane! Just to follow up Sydney’s response: there are also other National Park boundaries within the study area that we did not specifically mention in the video, including Mesa Verde National Park and Aztec Ruin National Monument. Additionally, several NGOs, such as the Archaeological Conservancy Group, are buying up the land that these sites are on to facilitate protection and research.

Thanks for your question!

Reply
Kevin Mckinney 11-08-2016, 20:15

Wow! Excellent video!

Reply
Sydney Neeley 16-08-2016, 11:52

Thanks for watching, Kevin!

Reply
Kelsey Herndon (Chaco Canyon Cross-Cutting) 18-08-2016, 22:28

Thanks Kevin!

Reply
Didi El-Behaedi 11-08-2016, 15:47

Hi guys! Great project, very interesting stuff! What other factors would you take into consideration if the project were to continue for another term? Will you supplement your findings using archaeological excavation in the future?

Reply
Ryan Schick 18-08-2016, 13:45

Thanks for the question Didi! For future studies we would like to incorporate data of known Chacoan sites that have been disturbed/destroyed to help assess the accuracy of the products we created. We would also like to acquire high resolution data for the entire study area to identify/delineate more sites. Additionally, we would like to look at areas beyond the San Juan Basin, as there are some known sites beyond these boundaries. Excavations are a little beyond the scope of DEVELOP, but results from this research could help our project partners identify areas of interest to excavate.

Reply

Leave a Reply