Archaeological Research at Cooper's Ferry

We integrate new technological approaches to the measurement of different aspects of the Cooper’s Ferry site in hopes of resolving a pattern or finding a significant correlation between different parts of the archaeological record.  During weeks 4 and 5, students Jasmine Kidwell and Jamie Klotz worked to measure the magnetic susceptibility of Cooper’s Ferry stratigraphic units and the sedimentary contents of a pit feature exposed in excavation unit A.  Their work contributed to Jasmine’s field school graduate student project.  OSU graduate student Alejandra (Aleks) Jimenez caught up with Jasmine to discuss her use of the Bartington magnetic susceptibility meter.

Aleks: What is your interest in archaeology?

Jasmine: Right now I am looking at Geoarchaeology, it seems to be a field of archaeology that is really interesting to me and I would like to work more on the subject to determine if its something I would like to further pursue. This field school offers a good opportunity for that.

Aleks: What is your grad student project while working with Dr. Loren Davis and the Cooper’s Ferry field school?

Jasmine: I got the chance to test out the magnetic susceptibility gun and looking at the lithostratigraphic levels of Unit A, specifically pit feature 1. I am interested in looking at the magnetic properties of the soil.

Aleks: How will you do this?

Jasmine: By using the magnetic susceptibility gun I was able to take magnetic readings from the layers in the soil. I pointed a small gun-like instrument at the soil and my pit partner Jamie Klotz pushed the button on a netbook and the instrument shot a magnetic pulse which then bounced back and read the magnetic composition of the soil. The information then appeared in the netbook.

Aleks: Is it safe to use the magnetic susceptibility gun?

Jasmine: Yes, it’s 100% safe.  There is no radiation and it is quiet and easy to use. You simply point and push a button, the instrument does the rest of the work.

Aleks: Can you describe the pit feature?

Jasmine: The pit is composed of eleven different layers, which is more than were established in 1997. This means by looking at the profile, or the side wall of the pit we can see different color sediments. Each layer has a different composition and so with the magnetic susceptibility we can distinguish these layers from each other due to their magnetic difference. At least, that is what we want to establish.

Aleks: How did you proceed to do this?

Jasmine: We took 25 readings from each of the layers in the pit feature and then took 50 readings from each of the lithostratigraphic level (LU), excluding LU 1.

Aleks: So what will all of the readings tell you?

Jasmine: Hopefully, after the analyses are complete, we will be able to discern the LUs from each other using their magnetic properties.

Aleks: What were the results so far?

Jasmine: We have completed the testing and have taken enough information but the results have yet to be determined. We are still working on the data and with the help of Dr. Shane MacFarlan, one of the staff members, we should have the information soon.

Aleks: How did you like using the magnetic susceptibility gun? Did you value the integrity of the procedure?

Jasmine: I really enjoyed using this procedure I was not familiar with the magnetic susceptibility gun but the procedures were very carefully explained and any questions I had Dr. Loren Davis was always at hand to help me. I think a test such as this is good because it will allow us to understand the sediment layers in a different manner, which will then either support our previous hypothesis of disprove them. Either way we will definitely discover something new.

Aleks: Did you enjoy your time in such a deep hole looking at sediments?

Jasmine: It was very dusty and I was very deep but with my pit partner, Jamie Klotz.  I like to think we had a good time. We certainly learned a lot, such as using the Munsel book to properly identify coloring of the sediments, we looked at texture and learned about texture analysis and about sediment identification.

Aleks: Any final thoughts?

Jasmine: My unit partner and I definitely learned a lot about the magnetic susceptibility procedure and it also allowed us to start thinking about archaeology in a different manner. There are many possibilities to help us understand the nature of archaeological sites and by using procedures such as the magnetic susceptibility we begin to understand that there are no limits to how we can proceed to understand archaeological sites.  I cannot wait to see the results.

With three weeks under their belts, the students are becoming confident in their abilities.  Nearly all of our field methods remain unchanged from last year; however, we’ve added some new techniques and equipment that augment our usual procedures.

First, we purchased another Nikon Total Station laser transit, which is used to double the rate at which we can record artifacts and other objects exposed in our excavations.  Last year, three-point spatial coordinates from our lone total station were in high demand and students often had to wait 30 minutes or more for their turn to obtain readings.

We have also acquired a Bartington Magnetic Susceptibility meter that we’ll use to measure the magnetic properties of sediments and artifacts alike.  This meter allows us to generate descriptive physical data about different materials that we can use to reveal patterns in the site’s geoarchaeological record.

We’ve written in the past about the portable x-ray fluorescence (PXRF) device that we use to measure the geochemistry of artifacts, sediments and other objects relevant to our research.  The PXRF device is a sophisticated piece of field technology and has tremendous potential for solving different kinds of problems.  We’re awaiting the arrival of our new Olympus Innov-X Delta PXRF device, which is currently being built and calibrated at the factory.  Anticipated arrival of our new PXRF device is in two weeks and we look forward to putting the machine through its paces until we close the site for the season.

Prior to coming out to the field, staff members Davis, Macfarlan and Henrickson submitted a manuscript to the Journal of Archaeological Science that presents a method for establishing and using a chemostratigraphic framework at the Cooper’s Ferry site.  The method described in this paper harnesses the power of the PXRF device to discriminate between different deposits of undisturbed archaeological site sediments on the basis of their geochemical attributes.  Moreover, this method can also identify the presence of disturbed sediments.  Being able to tell whether a particular artifact is in direct contact with disturbed or undisturbed sediments gives us a powerful means establishing when materials are contemporaneous with one another and also identifies datable samples that are in undisturbed contexts.  Having such a tool at hand is not only critical to our work at the Cooper’s Ferry site, but elevates archaeological debates about contextual associations between sediments and artifacts from the qualitative to the quantitative (and scientifically replicable) arena.  Here’s the abstract from our submitted paper:

A PXRF-Based Chemostratigraphy and Provenience System for the Cooper’s Ferry Site, Idaho

Loren G. Davis, Shane J. Macfarlan, and Celeste N. Henrickson

Paper in review at the Journal of Archaeological Science

Correlating archaeological finds with their associated stratigraphic context is elementary to modern excavation methods. Ascribing primary depositional position and associative context is traditionally a qualitative exercise. To improve the empirical nature of this process as part of new excavations at the Cooper’s Ferry site in western Idaho, we established a chemostratigraphic framework that enables us to quantitatively relate the elemental geochemistry of sediments associated with artifacts and other samples back to a master lithostratigraphic sequence. We subjected a total of 151 portable x-ray fluorescence (PXRF) readings from six lithostratigraphic units to multiple discriminant function analysis, which led to the identification of characteristic elements. To test the site’s chemostratigraphic framework, we collected PXRF readings from an infilled rodent burrow, which visually appeared to contain mixed sediments from multiple lithostratigraphic units. Statistical analysis could not relate the rodent burrow’s sedimentary geochemistry well to any of the known lithostratigraphic units, thus marking it as disturbed. This methodological approach allows us to make quantitative correlations between stratigraphic units across the site and most importantly, enables us to independently evaluate whether objects from any part of the site are in direct association with disturbed or undisturbed sediments.

As we await the arrival of our new PXRF device, students work to collect sediment samples from the vicinity (ca. 2 cm west, directly underneath, and ca. 2 cm east) of artifacts found in the excavation floor.  We’ll measure the elemental geochemistry of these gathered sediments taken near with artifacts in order to assess whether they indicate the presence of disturbed or undisturbed contextual association.  Once the PXRF device is in our hands, we’ll take these readings from in situ sediments near and under artifacts.

Last week, we had an unusual visit from the air.  Mary Crommett—a graduate of our field school from two years ago—and her pal Toby flew over our camp and landed at a local airstrip.  They had brought Toby’s experimental two-seater airplane from Oregon on a tour of western Idaho’s canyonlands.  Mary took several photos of our camp, the top of the canyon, and our excavation shelter at Cooper’s Ferry.

During our second week, the staff and students of the Cooper’s Ferry field school had the opportunity to tour the Nez Perce National Historical Park, located in Spalding, about fifteen miles outside of Lewiston in Idaho. The students were given a tour of the National Park, which included many activities such as watching a video about the history of the Nez Perce people, atlatl throwing, archery, viewing the archives and collections, and receiving a walking tour of the park.

Upon arrival to the Nez Perce National Historical Park the students were given the opportunity to assemble a tepee. One by one the students all helped to place the canvas cover, the stakes and the door. The students then all gathered inside to experience what the inside of a tepee felt like.

Afterwards, the students were presented a short film that explained the Nez Perce culture, history, and how they have maintained their traditions to this day. Kevin Peters, a National Parks Service Interpretative Specialist and member of the Nez Perce Tribe, and discussed historical photographs of Nez Perce figures and was able to answer questions pertaining to the Nez Perce culture.

After this introduction to the Nez Perce culture, the students were treated to a demonstration of the dart thrower (atlatl (say “at-lat-l”)) and were given the opportunity to use the weapon on a cardboard target of Rocky and Bullwinkle. Of our group, only two were successful in hitting the target. Clearly, throwing the Atlatl takes years of training and precise technique, and had this been a real hunt we would have been pretty hungry.

Next, the students were shown how to string and use a bow and arrow system.  Although more familiar to some students than the atlatl, the results were pretty much the same: only a few were able to hit the target.  The bow was small and comparable to the same type of bow used traditionally by the Nez Perce peoples.

Once the students had their fill of trying to kill the cardboard animals, they were taken back inside to the museum portion of the park where they were given explanations of cultural materials from the Nez Perce culture.  For this part, Diane Mallicken, Nez Perce interpretative specialist explained various materials such as the basket caps woven out of hemp and sometimes cornhusks. These caps were valued by the Nez Perce women as they were worn only by women of influence. The students also heard about the importance of camas root and the Nez Perce harvesting technique, which embraced the sustainable ideal of taking only as much as one needed.

To end the tour, Jason Lyon, a Park Ranger with archaeological experience, took the students to the rest of the park, which extends along the Clearwater River, downhill from the museum. He showed and discussed various archaeological sites, including semi-subterranean house pit depressions. The park is rich with contact and pre-contact period archaeological sites and provides the public with many educational opportunities.

Students of the Cooper’s Ferry field school were given a unique and memorable adventure as well as a very informative and educational experience about the Nez Perce culture. The students were very thankful to receive such a wonderful tour and day of educational activities.  Everyone gained invaluable knowledge that they will apply to their work at the Cooper’s Ferry site.  Many thanks to the staff of the Nez Perce National Historical Park!

On June 20, OSU staff and students converged once again on the town of Cottonwood, Idaho.  In addition those of us from Oregon State University, this year’s students also come from Georgia, Texas, Michigan, Ohio, Illinois, Washington, Idaho, California, and New Mexico.  Unlike last year, we arrived to clear skies and were able to set up tents in a dry field.  The first week in the field was initially spent setting up our field camp.  We’ve made a few improvements to our field camp this year, which should make camp life a bit better.  We’re using last year’s dig shelter as this year’s kitchen shelter and have added  kitchen sinks, shelving, and new propane stoves.

We set up our new excavation shade shelter (40′ x 60′) and set to work uncovering our excavation block.  Like last year, we had to re-excavate the thousands of sand bags we placed into the site, which protected our block over the winter.  There’s not getting past it: removing several thousand sand bags is hard work; however, our gang was able to empty the excavation block in about four hours.

With the excavation block opened once again, we set to work cleaning the floors and walls of loose dirt and fine roots.  We reviewed excavation procedures and how we record archaeological data in different ways.  By the end of the week, students had recorded the elevations of their excavation units and were ready to start this year’s excavations.

Today, students began excavating in their units and immediately uncovered numerous lithic flakes and fragments of bone and shell.  Everyone approached their first day of excavation at the site with care and enthusiasm.  The day passed quickly and we were forced to close up the operation just before 3:00 pm as a black cloud rolled through the canyon, bringing a short but intense thunderstorm.

In all, this has been a good week.  Everyone is settling in and cheerfully approaching this summer’s field season.  Tonight, we’ll have a lecture on the archaeology of Cooper’s Ferry and how it fits into the early prehistory of the Pacific Northwest.  Tomorrow, the crew will go to the Nez Perce National Historical Park for a tour of the facility’s collections and to learn through some hands-on activities.

People sometimes question the wisdom behind our choice to excavate at the Cooper’s Ferry site. Mainly this questioning is based on information about the recent history of the site and on things that people know about the archaeology of the Salmon River canyon. In this entry, I attempt to dispel a common misconception about the work that we’re doing at the site by addressing an important question that I’ve heard many times over the years.

Don’t you realize that you’re only digging through road fill?
In the 20th century, road construction took place in the lower Salmon River canyon in the vicinity of the Cooper’s Ferry site. To make the modern road, construction workers removed gravelly sediments and basalt bedrock deposits, cutting a notch into the toe of the canyon slope. The geological materials that they removed to create the road needed to be moved elsewhere and some of this material was pushed over the edge of the bank toward the river in the area of the Cooper’s Ferry site. Because road fill is a recent kind of deposit that will not contain intact archaeological evidence of ancient human occupation at the site, working only in these deposits would be a bad thing. So, before we spend our time and effort to carefully excavate at a potential site, we must know more about what the site contains. We can gain this advance knowledge of what’s in a site in two ways: first, we need to understand what’s already known about the site from any previous archaeological or geological excavations; second, we can dig a small hole into the site to quickly see what’s under the ground.

During several summers in the 1960s, an archaeologist named B. Robert Butler excavated a trench toward the site’s eastern end. Butler noted that they encountered thick deposits of fill dirt, likely from road construction; however, intact archaeological deposits were found toward the bottom of their excavation units. In the summer of 1997, Loren Davis directed the excavation of a single test pit (measuring 2 x 2 m or about 6 x 6 feet in surface area) on the western edge of the site to a depth of 2.5 meters (a little over 8 feet deep). This test pit gave us a window into the ground, so to speak, showing a series of different layers made of varying colors and kinds of geological materials—nearly all of which contained prehistoric artifacts. The 1997 test pit did encounter a layer of fill, which measured less than a foot thick, at the site’s surface. In addition to finding artifacts, our 1997 test pits also found the intact forms of several ancient campfire hearth and two different pits. One of these pits contained a collection of stone tools, including beautifully crafted spear points, which were placed in the ground for safekeeping. We also sent small fragments of animal bone and charcoal to scientists at a commercial laboratory called Beta Analytic (you can learn more about this lab at www.radiocarbon.com) who conducted radiocarbon analyses on our samples. The results of these radiocarbon analyses showed that organic items held in the layers at Cooper’s Ferry range in age between about 8,500 and 13,000 years old.

So, there clearly is road fill at the site, but the thickness of this deposit varies across the site. On the basis of information from prior excavations, we know that both disturbed road fill material and undisturbed archaeological deposits exist at the site and that the road fill deposits appear to be thicker on the upstream (east) end of the site’s landform. Based on our study of the site, we also understand how to tell the difference between road fill and undisturbed archaeological deposits. Unlike undisturbed deposits in the site, road fill deposits show swirled coloration (like a calico cat), lack fine sedimentary layering produced by ancient floods of the Salmon River, do not contain intact archaeological features like pits and campfire hearths, and commonly contain items from the 20th Century.

Geological cross section through the Cooper’s Ferry site from south (left) to north (right) through the site’s western (downstream) end.

This cross section model shows our expectation of what we would see if we cut a large trench through the site, based on previous archaeological and geological studies. The 1997 excavation unit is shown as the light orange rectangle and extended through deposits 3 and 5. Our recent excavations were built as an extension of the 1997 unit and have progressed through about half of the depth of the 1997 unit. The excavation units are actually oriented along an east-west line but are shown in a north-south orientation here in order to best illustrate how they relate to the site’s deposits.

It has been five months since the students left our Archaeology Field School at Cooper’s Ferry. Although students have all returned to their homes, our work continues. The artifacts and data collected during last summer’s excavations were taken to Oregon State University for further study by staff and students.
The process of post-field artifact processing and analysis involves many steps. In a series of new blog postings, we’ll track the progress of how artifacts are handled, recorded, and ultimately studied in the laboratory in an effort to build a database of the site’s archaeological materials.

For this week’s post, we’ll focus on the first step in our laboratory analysis: artifact cataloging. Before we can do any other kinds of analyses on the archaeological materials, we must account for all of the artifacts and samples collected during the previous field season. The catalog has information about each artifact recovered from our site including its catalog number, location within the site, contents, weight, basic morphological measurements, and any special notes. Building an artifact catalog requires us to examine each artifact or sample, collect new information about the item, and enter these data into a paper-based catalog. Recording these new data and other information recorded during the item’s original discovery at the site, will take several people hundreds of hours.

Student working hard on the catalog.

Although we are not certain just how many items we will have to catalog before the end of our project, we can make an educated guess about how large that number might be. As a result of the 1997 excavation of a single 2 x 2 m unit at the site, 52,025 items were grouped into 5,347 catalog entries. Our current project has opened up nine new 2 x 2 m excavation units, thus, if the new units produce the same amounts as the 1997 2 x 2 then we may ultimately need put something like 468,225 items into 48,123 catalog entries. Metaphorically, it’s like eating a whale and we’ll just have to start one bite at a time…

One page of many from the artifact catalog.

Here’s our cataloging procedure: starting with an excavation unit’s level bag, artifacts are given a gentle dry brush cleaning (it’s where old toothbrushes go to die). The artifacts may then be counted, measured and weighed depending on their type—some items, like bone or mussel shell fragments are weighed but not counted at this stage, given their fragile state (too much handling creates new fragments, which must then be counted, which makes new fragments—you get the picture). A formed stone tool, such as a projectile point, is counted and weighed; its length, width, and thickness are measured; and its rock type is determined (e.g., igneous, metamorphic, cryptocrystalline silicate).

A biface's width is measured.

Once we gather all of this information, the item is returned to its bag and placed back into the correct storage container. We’ll have completed cataloging the 2010 artifact collection before we return to Idaho for this summer’s field school session.

As week 8 of the field school comes to an end, students are looking forward to going home and saying sweet goodbyes to their new friends. Time has sped by once again this field season as the archaeology site was filled in and tents were taken down. The excavated and screened sediments were put into sandbags which were used to fill in the site. It is easier to pull sandbags out rather than dig out material in order to reopen Cooper’s Ferry for next summer’s season. First, plywood is cut and placed against the unit’s walls to protect them from the elements of winter and helps prevent collapse.

Plywood is placed along the walls of the unit to help prevent erosion.

Once the plywood was in place, students began to carry thousands of sandbags and placed them into the units. These sandbags could weigh up to 40 lbs.

Students begin to fill in the site with sandbags.

After only a few hours, the entire unit was mostly filled with sandbags. The large number of students greatly reduced the time it usually takes to fill in a site.

The site is filled in with sandbags.

Students stomp down the sandbags.

Next, students place a tarp over the sandbags.

In order for the excavated units to blend in with the environment, students pour the sediments from the extra sandbags and from the screening area on the surface of the sandbags.

The sediments in the extra sandbags are dumped onto the site.

The site is filled in and the tents are taken down.

After the speedy backfill and deconstruction of the site, students are able to gather around Cooper’s Ferry one last time before heading back to camp for some food and a shower.

OSU’s 2010 Field School at Cooper’s Ferry

Here is a video showing the backfilling process (no audio). The video is only 2 minutes long (we wish backfilling would have only taken 2 minutes!).

Throughout the fieldschool, students have been digging in 10 centimeter interval levels. For each level, a form is completed that describes the level’s surface elevation, mapped artifacts, rocks, and krotovinas. This form is a summary of the characteristics of that level, describing the sediments and features as well who was working on the level and the dates they worked.

A complete level form.

Every artifact that is excavated and mapped using the total station needs to be cataloged. To track the data associated with the artifact, bags are made with the data written on the front and the artifact is stored in this bag. In the field everything is keyed to unique number generated from the total station. This number is recorded on the level form with a brief artifact description. The data from the total station is downloaded to a laptop computer with the unique identifier, artifact description, and the coordinates of the artifact. These data will be analyzed later.

Each artifact is placed in a labeled bag.

Even though students label each artifact bag with a unique number that associates the artifact with its location stored in the total station, more information about each artifact needs to be recorded. Students learn the importance of cataloging, a system that summarizes all data about an artifact. A catalog number is assigned to each artifact recorded with total station. Catalog numbers are also assigned to artifacts or groups of artifacts recovered in the screens. Depending on the kind of artifact, information such as weight, size, shape, material type, and possible function is documented. This, in association with the catalog number, the unique number assigned by the total station, level, unit, and site number, are all recorded. The photo below shows an example page from the Cooper’s Ferry catalog. Many spaces are blank, left to be further filled out back at Oregon State University’s Archaeology Lab.

A sample page of the catalog.

This week has been full of interesting finds and experiences.  All students are excavating in lithostratigraphic unit 6, a geological deposit that is proving to be densely populated with artifacts and features.  The more artifacts and features that are located, the more time students spend on mapping and recording.  Along with an increased amount of paperwork comes the excitement of revealing a number of interesting finds.

Students found the base to a stemmed projectile point made from obsidian.  Obsidian is a material that was prized for the manufacture of stone tools; however, it is not found everywhere.  The closest obsidian sources to Cooper’s Ferry are found in northeastern and southern Oregon and at Timber Butte, which is located just north of Boise.  The presence of obsidian at the site suggests people were either traveling over long distances or trading with more distant people to acquire this desirable material.

A base of a projectile point made from obsidian.

Although complete stone tools are exciting to find, fragmentary pieces can provide significant information as well. When a tool is made, it is typically is kept and used. When a tool breaks, it is sometimes repaired and reused and is sometimes discarded. Because stone tools were useful to people of the past, it is more common for us to find discarded broken stone tools rather than complete ones. This past week, a few broken tools were found.

Some examples of broken stone tools.

The most common material that is used to make these stone tools is a cryptocrystalline silicate commonly known as chert. Chert is a silica rich rock that comes in many different qualities and colors. It is attractive for tool production due to its hardness and lack of large mineral crystals, which could cause the material to break in undesired places when struck. A typical characteristic of chert is its translucent nature in thinner sections, depending on the rock.

The material used to make stone tools is often translucent.

The OSU Cooper’s Ferry Archaeology Field School was featured in two local newspapers recently. Here are the links to these articles:

Idaho County Free Press
http://www.idahocountyfreepress.com/IFPNews6.shtml
http://www.idahocountyfreepress.com/IFPNews7.shtml

Cottonwood Chronicle
http://www.cottonwoodchronicle.com/070810/five.html

We were excited to be visited this week by Idaho’s State Archaeologist Ken Reid from the Idaho Historical Society. We hope all visitors have enjoyed their tours thus far.

The students revealed an interesting and exciting archaeological feature earlier this week. A feature is an arrangement of items or materials that in their shared spatial association, represent a cultural activity. Features can include things like pits, house foundations, organic stains in a site’s sediments, or dense piles of animal bone. Because these features represent particular activities, they are very important to the archaeologist. The feature discovered this week (known as Feature 3) appears as reddened sediments, caused by a chemical alteration (oxidation) of an ancient surface. We suspect that the creation of a large fire hearth, or a collection of smaller hearths, worked to alter the sediments, imparting the deep red color.

After the feature was discovered, students carefully worked to uncover the reddened sediments’ original surface and outlined the feature’s boundaries. Feature 3 is circular in shape and seems to have a slight depression at its center. Only a portion of Feature 3 lies within the excavated units, so it has an appearance of a half circle. The total station was used to capture spatial information about the feature’s surface by collecting readings in a grid pattern at 20 cm intervals.

Each flag marks a spot that was recorded by the total station. Notice the shape of Feature 3 as shown by the flags.

Once the spatial readings were collected, Loren Davis began recording the geochemical composition of each flagged spot using a portable x-ray fluorescence (XRF) meter, which is used to measure the chemical elements present within the site’s sediments. The data collected from the XRF meter can be used to compare the geochemistry of the reddened sediment within Feature 3 with the unaltered brown sediment outside of the feature, in order to make interpretations about the feature’s origins.

The XRF meter safely reads the geochemical composition of Feature 3

After the geochemical composition has been recorded for a particular part of the feature (marked by one of 185 different flags), students take a bulk sediment sample of that spot. The sample will be taken back to Oregon State University were a number of tests will be performed to determine its material composition.

A student takes a sample of Feature 3 sediment