.

.

Monday, December 11, 2017

Tuesday, November 14, 2017

Close-range photogrammetry models of rock art (2/3)

Image processing and model stitching is progressing. I've been experimenting with subsetting models to maximize mesh face count and not exceeding available memory. Agisoft insists on calling subsets "chunks", which always makes me laugh as a millennial.

Here are some more recent uploads:




Head to my Sketchfab profile to see the rest and provide feedback! Stay tuned for outreach materials in the coming year.

Sunday, September 3, 2017

Close-range photogrammetry models of rock art (1/3)

In fall 2016, Christian Driver and I began experimenting with close-range photogrammetry to document rock art. We help Boulder to manage rock art, specifically a series of 80+ rock art panel panels in eastern Boulder County. Many of the rock art panels in the area are too weathered to document or share satisfactorily with people. Since all of the currently detectable rock art is in relief, popular image transformations such as a principal components analysis or decorrelation are ineffective. Both of these processes are used to reduce autocorrelation (correlation of a signal with a lagged copy of itself) within a set of signals while preserving other (read: important) aspects of the signal. 
  • For explanation on how Dstretch works, read Dr. Ronald E. Alley's white paper. Before reading Dr. Alley's paper, I didn't realize that the Jet Propulsion Lab developed the algorithm used by Dstretch originally for ASTER, a 14-band multispectral sensor. I'm interested in learning more about how dramatically fewer data points in typical RGB camera sensors used for rock art change the dimensionality of the algorithm. 

While the spectral information of the rock art itself is not sufficiently different than that of the surrounding rock to distinguish the rock art, the spatial information of the rock art is sufficiently different than that of the surrounding rock to distinguish it if you can capture it. Traditional rock art documentation was completed between 1984 and 1997 by the Indian Peaks chapter of the Colorado Archaeological Society (Kindig 1997). These recordings, while infinitely valuable for identification and relocation of panels, have a few limitations. They are not to scale, do not capture the type or method of rock art creation, do not capture condition or imminent threats (cracks, sloughing, vandalism, infiltration), and do not document the surrounding rock surface (texture, bedding, color, shape/volume). What data model maximizes production of these data for the least amount of time and cost? 3-dimensional or 3D models is a great start. 

There are several ways to capture these 3D data, many of them are demonstrably "whiz-bang" as many of my bosses say. Terrestrial laser scanners are the fastest and most expensive method we found, which maximizes spatial extent and spatial differences but is less effective in capturing spectral information. We focused on close-range photogrammetry for several reasons: it's fast, it's cheap (price of sufficient quality DSLR camera (thanks Christian!), lenses (usually 50mm), and software unless open source), spectral, spatial, textural data captures are sufficient resolution, and it's reproducible. 

Check out some examples of our 3D models so far. The models are color balanced and to scale. Sometimes the spatial extent of data capture is limited by our computer processing capacity. I'm hoping to experiment with parallel processing/RAM and processor upgrades in the next few months. 

Please leave me comments on either this page or Sketchfab if you enjoy seeing the models, what you did and did not like about the models, and any changes you suggest. 

-------------------------------------------------------------------------------------------------------------

Oil derrick



"Kirkendoll"


"WFM/ 1904"

Friday, March 31, 2017

DATA SHARE: Boulder local landmarks, historic districts, State and National Register listed resources

As part of a new planning effort at work, I assembled geospatial data for designated historic properties in City of Boulder and its public lands (45,000 acres surrounding Boulder, Colorado). These data were aggregated from:

  • City of Boulder historic landmarks and districts (excluding contributing elements and potential districts),
  • Boulder County historic landmarks, 
  • Boulder County historic districts and cultural landscapes, 
  • Colorado State Register of Historic Places-listed properties, 
  • National Register of Historic Places-listed properties, and
  • National Historic Landmarks. 
You can download this zip file, which contains an ESRI Layer Package file and accompanying shapefile of assembled info. The EPSG is 2876. Metadata is FGDC compliant and documents data sources, limits, and processing. Tabular data include designation type, parcel number, resource name. 

NOTE: These data do not include protected information and are publicly available from their respective sources (City of Boulder, Boulder County, Colorado Office of Archaeology and Historic Preservation, and National Park Service). 

Thursday, February 2, 2017

Saving Places 2017 Presentation

In February 2017, I presented at Colorado Preservation Inc.'s Saving Places Conference about using geospatial data to study linear cultural resources. The slideshow is accessible here for those looking for geospatial data (remotely sensed particularly) sources as promised.

This presentation was prompted by a paucity of evidence from existing and free geospatial data in cultural resource management reports and resource forms. Given the deposition context of the Front Range of Colorado where you can find both prehistoric and historic components on the ground surface, CRM archaeologists are missing out on time efficient and defensible data sources for resource documentation. Linear resources especially benefit from a geospatial approach because they're usually often difficult to record in entirety and assess segment contribution to overall resource eligibility to the National Register of Historic Places and a geospatial approach gives the CRM archaeologist basic lines of evidence to generate recommendations based on integrity (NPS Guidelines on Evaluating Integrity).

My goal for this presentation was to give the audience questions to ask of existing geospatial data to inform evaluations of cultural resource integrity and historic significance. The questions increase in specificity from broad (What is my geographic area of interest?) to pointed (Have design, setting, associations changed?) to concurring data (Do multiple sources agree?). I show examples from my own work in Boulder featuring:
  • spectral information using multispectral aerial imagery with band indices, 
  • texture using high spatial resolution panchromatic satellite imagery, 
  • historic-age maps (from Civilian Conservation Corps in Boulder, 1935), 
  • LiDAR point clouds, 
  • LiDAR digital elevation models (DEM), 
  • LiDAR digital surface models (DSM), and 
  • change detection using panchromatic historical orthorectified and oblique aerial imagery (partial example below). 
More importantly, I've assembled a list with links of data sources that people should use! Free sources are noted. I also had to give a shout out to GlobalXplorer and Dr. Sarah Parcak and encourage people to work on imagery interpretation, crowdsource site locations and much more. If you haven't signed up and tried it out, go try it now because it's much more fun than me droning on about free data (did I mention that almost all of the data I use is free?!).


Feel free to contact me with questions about the materials. I'm always happy to help people get the most out of their time and money/free data.

-Katy

Friday, September 2, 2016

Interview with new Colorado Assistant State Archaeologist, Chris Johnston

History Colorado Blogs recently featured an interview with Colorado's new Assistant State Archaeologist, Chris Johnston. It's a pleasure to be able to work with Chris and Dr. Holly Norton (Colorado State Archaeologist).

Stay up to date with History Colorado's Blogs for interesting pieces on national, state, and local archaeological and preservation topics.

Monday, June 27, 2016

Viewshed and visual impacts assessment near Colorado Chautuaqua, Boulder, Colorado

One of the most distinctive areas of Boulder is the Colorado Chautauqua, located on the edge of Enchanted Mesa overlooking the Boulder Valley. It has served as a rural enclave to urban Boulder since 1898. The Colorado Chautauqua has provided a variety of programs for over a century, including concerts, debates, lectures, and other recreational activities. Read more about the offerings of the Colorado Chautauqua in Mary Galey's The Grand Assembly, considered the authoritative text on the Colorado Chautauqua.


History of the Colorado Chautauqua

In the late 1890s, a group of Texans (including University of Texas president G. F. Winston) wanted to open up their own Chautauqua and looked to the Rocky Mountains for a location. Boulder was chosen for the site of Chautauqua and Boulder citizens, thrilled to have a Chautauqua nearby, raised $20,000 towards construction costs. The city of Boulder purchased 171 acres from the Bachelder Ranch and the Austin-Russel tract for the new Chautauqua, the first purchase of private property by the city of Boulder. The Texans covered expenses of running the Chautauqua programs. 

"The First Chautauqua at Boulder, Colo. 1898. Texas-Colorado Chautauqua, looking west. H. F. Pierson photo, Denver, Colo." Boulder Carnegie Library for Local History, 511-3-9 PHOTO 1. 

The Colorado Chautauqua was part of the national Chautauqua movement, which was nationally known for its emphasis on intellectual pursuits, moral self-improvement, and civic involvement. The movement began in upstate New York in 1876 as a center for political, educational, and recreational programs. By 1924, nearly 40 million people were annually attending events at over 300 Chautauquas across the country. The Chautauqua movement died out by the mid-1930s. Most historians cite the rise of car culture, radio, and movies among the primary causes as well as social changes, including a sharp increase in fundamentalism and evangelical Christianity in the 1920s, alternative educational opportunities for women, and the economic impossibility of organizing programs. Many Chautauqua communities became camp meetings or church camps. 

By 1955, Boulder's Chautauqua was one of only six remaining in the country. The programs and activities continued at the Colorado Chautauqua as they had since the beginning, but many of the building began to deteriorate. By 1975, the city of Boulder considered tearing the buildings down, but concern about the future of the park inspired the citizens of Boulder and the Colorado Chautauqua Association to implement a program to preserve the park and its historic buildings, structures, and landscape. In 1978, Colorado Chautauqua was designated as a local historic district and was listed on the National Register of Historic Places. In 2006, Colorado Chautauqua was designated as a National Historic Landmark.


Managing the Viewshed of the Colorado Chautauqua

Recently, Open Space and Mountain Parks Department (land managers for Chautauqua Meadow and lands surrounding Colorado Chautauqua) proposed repairs to the Chautauqua Trail. Chautauqua Trail is the flagship trail that most visitors (over 3 million annually) hike during their visit to Chautauqua Meadow or Colorado Chautauqua. The current trail design is problematic and leads visitors to hike off trail, causing braiding, erosion, and expansion of the trail corridor. The trail was maintained to be 8 feet wide; it is currently 24 feet wide in spots (Figure 1). Additionally, there are several informal gathering areas where folks like to take photos with the Flatirons in the background. Open Space and Mountain Parks would like to improve the trail by stabilizing the trail tread and to designate low development level gathering areas to minimize existing resource damage.

Figure 1. Photo CT.7785: Overview of Chautauqua Trail below/northeast of Gathering Area 2 (GA2) looking northeast towards Colorado Chautauqua. Braiding, gullying, weed encroachment visible on trail.
During the planning phase of this and similar projects, we consult with neighbors and stakeholders on the proposed work. One stakeholder was particularly concerned that our trail work would negatively impact the viewshed of Colorado Chautauqua. In response to the concern, my crew devised a two part plan to assess the viewshed impacts to Colorado Chautauqua.


        1. 3D Viewshed Model 

I constructed 3D viewshed models with ArcGIS 10.4 of the lines of sight from our treatment locations (Figure 2), key observation points identified by the stakeholder (Figure 3), and key observation points identified by Open Space and Mountain Parks (Figure 4). These locations are listed below. The viewshed models were generated using a 1 foot spatial resolution digital surface model (DSM) that included obstacles (buildings, vegetation). The DSM was derived from post-flood aerial LiDAR point cloud for the City of Boulder. The height of the observer was chosen at 6 feet above ground surface, simulating the viewshed for taller individuals to maximize the visibility of surrounding features.

Creating 3D viewsheds in a geographic information system (GIS) have benefits and drawbacks compared to other methods of viewshed assessment. A GIS is better able to consider the effects of terrain relief and curvature of the earth, creating a quantifiable output of what areas are visible or not from a given location. Because of Earth's curvature, lines of sight can only extend up to 3 miles or 5 kilometers unless drastic changes in elevation are involved. A GIS makes it easier to assess object and observer heights. Our models did not consider atmospheric refraction and how visibility of an object is affected. The GIS models also do not address observer's visual acuity. An object may have a clear line of sight to an observer, but an observer may not have the ability to see that object.

Visualizations from the model illustrate the results. The viewshed model for our proposed treatment locations shows that little of the landmark shares clear lines of sight with the treatment locations (Figure 2). The viewshed from Colorado Chautauqua is not substantially impacted (Figure 3). In assessing the potential for adverse indirect effects of our proposed treatments, we expect that no areas of concern at Colorado Chautauqua showed potential for adverse effects. The viewshed towards Colorado Chautauqua from the Chautauqua Meadow is likely to be impacted by the proposed treatments (Figure 4). There are clear lines of sight from one key observation point at each gathering area.

Figure 2: Lines of sight from proposed gathering areas. Gathering areas construction will require application of StaLok surfacing and irregular placement of boulders to delineate from off-trail terrain. Cells in green are visible from both gathering areas, blue cells are visible from one gathering area, uncolored cells are not visible from any proposed gathering area.
Figure 3Lines of sight from stakeholder-identified key observation points. Cells are color coded for how many key observation points have clear lines of sight to them.
Figure 4: Lines of sight from Open Space and Mountain Parks-identified key observation points. Cells are color coded for how many key observation points have clear lines of sight to them. 
_________________________________________________________________________________
DESCRIPTIONS OF KEY OBSERVATION POINTS:
CCA1Vehicle entry at Kinikinik RoadOSMP17th Street and BaselineGA1Gathering Area 1
CCA2Chautauqua GreenOSMP2Divergence of Baseline and Flagstaff RoadGA2:Gathering Area 2
CCA3Gwenthean CottageOSMP3Armstrong Bridge
CCA4Chautauqua ArborOSMP4Ski Jump Trail at woodland edge
CCA5Bluebell Road (Cottage 26)OSMP5Chautauqua Trail and Bluebell Mesa Trail
CCA6Columbine Lodge (west hallway on 2nd floor)OSMP6Bluebell Mesa Trail at woodland edge
CCA7Columbine Lodge (northern elevation porch on 2nd floor)OSMP7Bluebell Road (near Cottage 811)
CCA8:Cottage 16OSMP8:Meadow Trail near Civilian Conservation Corps site
_________________________________________________________________________________


        2. Visual Contrast Rating Survey 

Figure 6: Triangular 12 inches by 6 inches target.
Targets were placed in proposed gathering areas.
Figure 7: Subset of VCR survey group, including OSMP
trail project manager, archaeologist, and Colorado Chautauqua
Association Facilities Manager
The second part of our approach to demonstrate the potential impacts on the viewshed of Colorado Chautauqua was to conduct a Visual Contrast Rating (VCR) Survey. This survey was organized by my co-worker (fellow OSMP archaeologist), Christian Driver. We used a modified format of the Bureau of Land Management's inventory form. The purpose of a VCR survey is to analyze potential visual impact of proposed projects and activities. It requires participants to apply the basic principles of design in the resolution of visual impacts. The VCR survey process provides a means for determining visual impacts and for identifying measures to mitigate these impacts.

The VCR methods holds that degree to which a management activity/treatment affects the visual quality of a landscape depends on the visual contrast created between a project and the existing landscape. The contrast can be measured by comparing the project features with the major features in the existing landscape. The basic design elements of form, line, color, and texture are used to make this comparison and to describe the visual contrast created by the project.  

A VCR survey was completed in two days by four people, including a representative from the concerned stakeholder. Data and findings are still being compiled from the inventory forms. Discussions during the survey found consensus in how few of the treatment areas were actually visible by our own sight from key observation points. 

Consultation for this project is still ongoing with 30% Plan Sheets anticipated to be released in the next few weeks. I will update this post periodically as we advance in the planning process, particularly as we may need to seek additional clearances with a local landmark board to proceed. This viewshed analysis and visual resource assessment provided robust framework for conversations about indirect effects to significant historic properties. I consider this process successful in achieving our goals.