Mapping, GPS, and GIS

 

 Mapping is an essential skill for archaeologists. Archaeology is the study of artifacts within a geographic context, and maps are necessary to provide that context. In the academic world, archaeologists use geographic data to study ancient settlement patterns across the landscape, and theorize about what those patterns might tell us about environmental adaptations, economic relations, or political power.

Mapping also has more practical applications for archaeologists within the CRM (cultural resources management) industry. The purpose of CRM is mainly to locate archaeological sites before they can be damaged by construction. Thus, accurately recording the locations of these sites is one of the most important duties of any CRM archaeologist. You cannot protect an archaeological site from an approaching pipeline or transmission line if you don’t know exactly where the site is.

 

Collecting Geospatial Data in the Field

In the early days of CRM archaeology—before the proliferation of GPS technology—mapping was a challenge. Archaeologists are generally not trained as land surveyors, and they often conduct pedestrian surveys in remote back country wilderness areas. In the old days of CRM, field archaeologists would trek out onto the open rangelands equipped with a compass and topographic map, and use their orienteering skills to plot newly found archaeological sites on their maps by hand.

This was not very precise or accurate. I was not a field archaeologist in the old days, as I did not enter CRM until 2011. But I’ve georeferenced* quite a few archaeological survey maps and site maps from the period prior to the year 2000, back when sites were often plotted by hand on USGS (U.S. Geological Survey) topographic maps. And in some cases, I’ve been able to compare these early maps with newer, more accurate maps that show the precise locations of these sites. On the old maps, archaeological sites are usually plotted in the wrong place, often by quite a wide margin. A good field archaeologist could plot a site on a map within about 40 meters of its correct location, but the plotting of some sites was off by more than 100 meters.

I should add that most of the maps I’ve personally georeferenced were drawn in northern Nevada, which is not an easy place to keep your bearings without modern GPS equipment. Northern Nevada is a sparsely populated desert covered in wide open rangeland, with many areas lacking distinct landforms that might be visible on a standard topographic map. Under these conditions, accurately plotting an archaeological site with no more than a topo map and a compass is extremely difficult. It’s a testament to the skill of the early field archaeologists that they were able to plot sites within about 40 meters of their correct locations.

But today, that level of accuracy is not quite good enough. A field archaeologist who accidentally plots an archaeological site 40 to 100 meters away from its correct location will end up protecting the wrong piece of land from construction, while inadvertently routing a pipeline or other construction project directly through the actual site.

Fortunately, the advent of GPS technology has added both accuracy and precision to our work. When the CRM industry began with the passage of the NHPA in 1966, GPS technology had not even been invented yet. But today, it is an indispensable tool for every archaeologist working in NHPA compliance.

The United States military invented GPS (Global Positioning System) in 1973, though it was not really operational until 1978. The military launched several navigational satellites into orbit, and these satellites now send signals to receivers back on earth. Trilateration of these signals allows a receiver to calculate its exact position on the earth’s surface. And thus, GPS was born.**

GPS technology was not available for civilian use until 1983, and even then, civilian access was restricted by the military’s policy of “selective availablity.” The military scrambled the signals being sent to civilian GPS receivers, ensuring that these receivers could not be accurate within less than 100 meters of the correct location. Selective availability remained in effect until Bill Clinton signed it away in 2000.

After the year 2000, now that civilians had access to much more accurate receivers, civilian use of GPS technology exploded. It became a handy tool for archaeologists and nearly everyone else. I was about eleven at the time, and I remember that this was the period when people who were much wealthier than me began to buy GPS navigation devices for their vehicles.

Twenty years later, the current archaeologist has access to a wide arsenal of GPS technology that can be used to record the real world coordinates of artifacts in the field, and upload those coordinates to a cartographic software program on a computer. The most popular devices (among archaeologists) have traditionally been made by the rival GPS companies Trimble and Garmin, but many archaeologists now use the internal GPS receivers found within their smart phones or tablets.

Garmin produces handheld GPS receivers that are accurate within about 3.65 meters of the real location under ideal circumstances (not accounting for clouds or trees that might obstruct the satellite signals). These devices are widely used in the Western states.

Trimble produces even more accurate devices. Any handheld device within the Trimble GeoExplorer series can be accurate within about 60 cm., or a little over half a meter. These are the devices that I have typically used during my career. A Trimble Juno receiver is accurate within about five meters, making it less popular among archaeologists.

Figure 1. Trimble GeoExplorer GeoXT GPS receiver displaying UTM coordinates collected in the field. This GPS receiver is capable of submeter accuracy.

All of the devices I’ve described so far are handheld devices that can be used on their own, but some companies produce special GPS receivers that can be paired with your personal smart device, such as a smart phone or tablet. The Trimble R1 and Juniper Geode are small GPS receivers that can pair with your smart phone or tablet via Bluetooth, offering submeter accuracy (much like a Trimble GeoExplorer). The internal GPS receiver inside your phone or tablet can't achieve submeter accuracy on its own. When used without an external receiver, your phone's GPS probably won't be accurate within less than about four meters (I've used a special app to test the accuracy of my field phone, and it was consistently accurate within about four meters—or so the app claimed).

If you don't need submeter accuracy, you can use a smart device on its own. A variety of data collection apps can be downloaded on any phone or tablet. One of the most well known is Avenza, but it is not the only one. ESRI has produced a geospatial data collection app known as Field Maps to replace its old app, known as Collector. 

All these devices have strengths and drawbacks. Any handheld device within the Trimble GeoExplorer series should be rugged enough to withstand the rigors of the field, while consistently providing submeter accuracy, but the data you collect is worthless unless it can be exported to a usable file on your computer. At the moment, the only way to do so is with a software called Pathfinder Office, which is very expensive. Garmin handheld devices are also very rugged, and it is cheap and easy to export data from them, but they are not very accurate (by archaeological standards). Smart phones and tablets are neither accurate nor rugged; you'll be lucky to achieve accuracy within four meters, and they are delicate little objects prone to breaking in the field. But you can use them to download a variety of different data collection apps to suit your needs. You can also pair them with a Juniper Geode or Trimble R1 to achieve high accuracy. These external GPS receivers are rugged, but the smart phones and tablets are not, and you can't really use the external receivers without a functioning smart device. Also, it is easy to lose the Bluetooth connection between smart device and external receiver without noticing, and then you end up using your smart device's internal GPS receiver to collect less accurate data than you need.

Also keep in mind that GPS receivers only work when they receive a signal from multiple satellites, and they aren’t effective if something obstructs those signals. Clouds and heavy tree cover can obstruct satellite signals. I once had to climb a tree in the field to get adequate satellite reception.

You can use any of this GPS equipment to collect real world coordinates in the field and upload those coordinates to a cartographic software platform on your computer. The study and usage of cartographic software is known as GIS (Geographic Information Science or Geographic Information Systems). For archaeologists, the most popular GIS software is ArcGIS, which is produced by a company known as ESRI. ArcGIS was originally designed for environmental scientists, but archaeologists frequently use it to process geospatial data and make maps. I was once a GIS technician specializing solely in archaeological data, and I used ArcGIS almost every day.

This is only a basic overview, because the process of uploading data and converting files can be complicated, and this is not meant to be a software tutorial.

 

Entering Geospatial Information Into Official Records

Now that you have a rough idea of how archaeologists record the locations of sites, it might be useful to know how those locations are entered into official records.

Every state in the United States is responsible for keeping records of the archaeological sites that are discovered during CRM surveys. Many states maintain online geospatial databases, where qualified archaeologists can log in and see where all the previously recorded sites have been plotted, as well as which areas have already been surveyed. For example, the state of Nevada maintains a geospatial database known as NVCRIS (Nevada Cultural Resource Information System), and when I was a GIS technician my job was to make corrections to the site boundaries and survey area boundaries within the database. Databases such as NVCRIS are restricted to qualified archaeologists. They are not open to the general public, out of the fear that members of the public might use this information to locate archaeological sites themselves and loot them for valuable artifacts. Looters commonly vandalize rockshelters or Native American graves in pursuit of artifacts that they can sell to collectors.

Before the introduction of geospatial databases, the individual states kept collections of paper maps on which the site boundaries had been drawn by hand. The state of Oklahoma (where I did fieldwork for my Master's thesis) did not have a geospatial database for archaeological sites until the end of 2022. Before then, anyone who wanted to know which sites had already been recorded in Oklahoma had to go to Norman, Oklahoma in person, and go through a collection of topo maps that had archaeological site boundaries drawn over them in pencil.

If you find an archaeological site in the field, you will need to submit your findings to the appropriate state government, and sometimes to a federal agency. Typically, this entails that the site boundary be plotted on a USGS topographic map. Originally, the site boundary would be drawn on the map by hand. Today, this can be done digitally, with ArcGIS or some other computer program. You may also be able to submit your geospatial data directly to the government agency, in the form of an ESRI shapefile.

 

Legal Descriptions

You will also need to fill out a site form. Different states use different formats for their site forms, but any state that uses PLSS (Public Land Survey System) will require geographic information in the form of a legal description.***

A legal description is what land surveyors use to keep track of parcels in their official records. In order to explain how to write a legal description for an archaeological site, I’m going to show you how to write the legal for the abandoned brick grain silo outside the house where I grew up in rural Illinois. I can’t show you the location of a real archaeological site that has been recorded in a database, because that information is confidential. As I’ve already explained, the locations of archaeological sites are not made available to the general public, due to the prevalence of looting. But there’s probably no harm in showing you the location of the old silo outside my dad’s former home (even though it could technically be considered a historical archaeological site, as it was built around the year 1900). It contains nothing of monetary value, and if you feel an urge to trespass there, be aware that the current landowners probably have a lot of guns, as is customary in rural Illinois.

Here is a satellite image of the old silo:

 

Figure 2. View of historical silo from space

Land surveyors in the United States write legal descriptions using a system known as PLSS (Public Land Survey System). I’ve noticed that even experienced archaeologists have trouble understanding this system, so I wanted to devote some special attention to it. The best way to explain PLSS is to start with the biggest units of land, and work our way down to the smaller divisions.

The silo outside my old home is located in Peoria County, Illinois, as shown below:

 

Figure 3. Map of historical silo's location imposed over Illinois counties

The state of Illinois is further divided by township. Essentially, it is divided into a set of perfect township squares, with each square covering 36 square miles (6 miles by 6 miles on each side). Each one of these squares is assigned two numbers—one for “township” and one for “range” (but the individual squares are generally just called townships). The "township" number indicates how far north or south the square is, and the "range" number indicates how far east or west it is. The township number is counted from an imaginary line called a baseline, which runs from east to west (for example, if the township unit is labeled as 9 N, that means its northern boundary is 54 miles north of the baseline). The range number is counted from an imaginary line called a principal meridian (for example, if the unit is labeled as 5 E, that means its eastern boundary is 30 miles east of the principal meridian).

That is how the term “township” is used in PLSS. It has a slightly different meaning in local politics. A “township” can also refer to an administrative division of a county, with its own small local government (road commissioner, accountant, etc.). These political units usually correspond perfectly with the 36-square-mile township squares I’ve discussed above. In some cases, the political boundaries of a township may be re-drawn to make governance easier, but the PLSS units are not supposed to change.

For example, I grew up in a remote area of Elmwood Township in Peoria County, Illinois. “Elmwood Township” is a political/administrative unit. It also happens to correspond with a 36-square-mile PLSS township unit located 48 miles north of the baseline and 24 miles east of the fourth principal meridian. Thus, the legal description so far would read as follows:

Township: 9 N

Range: 5 E

Principal meridian: 4th

Figure 4. Map showing silo's location and PLSS township boundaries in relation to baseline and principal meridian. The silo is located in the 9th township north of the baseline (hence, Township: 9 N) and the 5th township east of the principal meridian (hence, Range: 5 E). Note that the townships east of the Illinois River are counted from a different principal meridian, not shown in this map.

 

Figure 5. Map of silo's location imposed over PLSS township boundaries within Peoria County, IL. PLSS townships fit within county boundaries, but the county is not really part of the legal description.

Now, let’s work our way down to even smaller PLSS units. Each township/range square is further subdivided into 36 sections, with each section equaling one square mile. Section 1 is in the northeast corner, and the individual sections are numbered in alternating rows, ending at Section 36 in the southeast corner.

 

Figure 6. Map of silo's location imposed over section boundaries within Township 9 N Range 5 E

For example, I grew up in Section 23, meaning that the legal description so far is:

Section: 23

Township: 9 N

Range: 5 E

Principal meridian: 4th

Figure 7. Map of silo's location imposed over Section 23

In order to write an archaeological site’s legal description with the level of precision required by a state government, you generally need to divide the section into quarter sections three times. You will be narrowing down the site’s location to a square patch of land covering only 1/64 of a square mile. That is how precise your legal description needs to be.

First, I need to divide Section 23 into quarter sections: NW, NE, SW, and SE. The silo is located in the NE quarter. Then I divide the NE quarter into four quarters. The silo is in the SE quarter of the NE quarter. Finally, I divide that quarter into four more quarters. The silo is located in the SE quarter of the SE quarter of the NE quarter of Section 23. Thus, the full legal description reads as follows:

Quarter sections: SE SE NE****

Section: 23

Township: 9 N

Range: 5 E

Principal meridian: 4th

In all likelihood, you will need to do this for every site you record during a survey, and write the full legal description on the site form (Site forms generally do not require that you list the principal meridian, but you will have to name the county, and that should narrow down the location sufficiently).

UTM Coordinates

You will also need to learn a little about the UTM (Universal Transverse Mercator) grid system. This is a navigational coordinate system. When you use your GPS equipment to record the real world coordinates of artifacts in the field, you will often need to use UTM coordinates (as opposed to Latitude/Longitude coordinates, which belong to a different coordinate system).*****

Under the UTM grid system, the earth is divided into longitudinal zones. For example, the silo outside my old house is located in UTM Zone 16 North, which would be written as 16T. The precise UTM coordinates of the silo would be written as follows:

16T 255983 4514944

The first number (16T) indicates the UTM zone.

The second number (255983) indicates Easting.

The third number (4514944) indicates Northing.

It will be pertinent to know that the UTM grid system changed in 1983. The original system is known as NAD 1927 (North American Datum 1927). For some complicated geodetic reasons that I don’t fully understand myself, the 1927 system proved inadequate when being used with GPS technology, so the datum (frame of reference) was shifted, and the whole grid shifted with it. The new system is known as NAD 1983 (North American Datum 1983), and it was implemented due to the increasing usage of GPS technology at the time.

Why does this matter for archaeologists? Because the UTM coordinates written on old site forms often still use the 1927 system. In fact, archaeologists were still using the 1927 system until well after 2000. Old copies of USGS topographic maps use the 1927 system as well.

If I use the NAD 1927 grid system to determine the UTM coordinates of the old silo, I get the following coordinates:

16T 255976 4514736

Keep in mind that the silo itself has not actually moved. But there are two UTM grid systems, and they will give two completely different sets of coordinates to the same geographic location.

This issue is more likely to come up than you might think. There may be previously recorded sites within your survey area, and you will need to re-visit them during the course of your survey, possibly with the aid of the original site forms. But the UTM coordinates written on the site form may belong to the NAD 1927 system, and if you try to navigate to those same coordinates using the current NAD 1983 system, the coordinates will lead you to a completely different spot (about 200 meters south of the actual location). If you are reading or writing UTM coordinates, always be sure you know whether they belong to NAD 1927 or NAD 1983.

Unfortunately, many archaeologists during the old days were really bad at recording UTM coordinates, so the UTMs written on old site forms are often completely wrong, no matter which grid was being used. If you want to use a site form to locate a previously recorded site, but the UTMs on the site form are wrong, the best solution is to look for the physical features of the landscape that are drawn on the sketch map, which should be attached to the site form.

Quadrangles

Site forms generally require that you specify which "quadrangle" the site falls within. Personally, I don't think this is particularly useful information; this is mainly a holdover from the days when archaeologists plotted sites by hand on USGS topographic maps.

Quadrangles have nothing to do with PLSS units, and this can confuse archaeologists, so I'm going to break down what a quadrangle is.

A long time ago, the U.S. Geological Survey began producing topographic maps for the entirety of the United States. They created sets of maps at different scales, but archaeologists (and most other people) took to using maps made at the 1:24,000 scale (meaning that a unit on the map is 1/24,000 the size of its corresponding unit in real life). The USGS maps made at the 1:24,000 scale are known as 7.5 minute maps, because they cover a rectangular area corresponding to 7.5 minutes (a minute is 1/60 of a degree, and the globe is divided into 360 degrees). The rectangular area covered by a single 7.5 minute map is known as a 7.5 minute quadrangle.****** In effect, the USGS divided the whole United States into about 57,000 quadrangles, and named each one of them (they also created quadrangles at different sizes for maps made at different scales, but I'm not going to get into that, because archaeologists don't really use maps at the other scales).

These USGS quadrangles are perfect rectangles that cross state and county lines; they have nothing to do with the boundaries used by land surveyors or government administrators. However, the topographic maps themselves typically show PLSS units and UTM grid lines (be aware that maps made before 1983 will use the old UTM grid from 1927).

The historic silo I've been discussing falls within a 7.5 minute quadrangle known as Farmington East Quadrangle (named after the nearby town of Farmington, Illinois). The map below shows its location, imposed over the dozens of quadrangles that cover parts of the state of Illinois (note: this map does not show all quadrangles in the United States).

Figure 8. Location of historic silo imposed over 7.5 minute USGS quadrangle boundaries. The USGS has created a 1:24,000 scale topographic map for each of the rectangular units shown in the above map. The silo is located at the northern end of Farmington East Quadrangle.

Overview

This is the bare minimum of mapping knowledge necessary for conducting surveys and recording your findings. In fact, it’s less than the bare minimum, because I didn’t teach you how to use the various types of hardware or software. Ideally, if you want to make maps, you should also know a little about the principles of cartography, such as normalizing data, and knowing which map projections are appropriate for presenting which kinds of data. But I don’t have time for that here.

GIS and cartography have purely academic applications as well, both within and outside archaeology. I’m not going to discuss that here, but for those who are interested, there are plenty of studies about landscape archaeology and settlement patterns.

Footnotes:

*"Georeferencing" is the process of assigning geospatial coordinates to specific points in a pixelated image, such as a scan of a paper map.

**The term "GPS" refers specifically to the satellite system put into orbit by the United States military. Other countries have launched navigational satellites as well. The systems that include these satellites are known collectively as GNSS (Global Navigation Satellite System). Many of the devices that we colloquially call "GPS receivers" (and which I refer to as GPS receivers in this post) can also serve as GNSS receivers, if they are allowed to access international satellite signals.

***Not all states use PLSS, but if you are working in states that do, reading a PLSS legal description is an essential skill for archaeologists. Some states that used to be part of Mexico (such as Texas) do not use PLSS (for example, Nevada and California both use PLSS, and thus their site forms require legal descriptions, but Texas does not use PLSS, and thus does not include legal descriptions in its site forms).

****When I determined the quarter sections for this silo, I used the section boundaries on a USGS topographic map, not the Tigerline shapefile shown in the map I produced for this blog. The Tigerline shapefiles do not match up perfectly with the section boundaries drawn on USGS topo maps, but the topo maps are probably more accurate.

*****The numbers in a set of UTM coordinates represent meters on the earth's surface. For example, "northing" indicates the number of meters north of the equator. "Easting" is more complicated, which is why you also need to include the UTM zone in a set of coordinates. The Latitude/Longitude coordinate system is completely different—it divides the round globe into degrees, minutes, and seconds (in the same way that you can divide a circle into 360 degrees in a geometry class).

******The actual area that a 7.5 minute quadrangle map covers will vary by latitude. USGS topo maps use a conformal projection, meaning they preserve angular integrity (you can use these maps to follow a compass bearing), but in doing so, they must sacrifice areal accuracy (they do not show units as being the correct area, and this discrepancy becomes more pronounced at those latitudes closer to the poles). This is one of the complexities of trying to represent a round planet as a flat surface.

Other Resources

The sources below may be useful for students wanting to learn more about the academic applications of mapping in archaeology.

Barrett, J.C. and I. Ko. 2009 A phenomenology of landscape: a crisis in British landscape archaeology? Journal of Social Archaeology 9:275-294.

Hally, David J. 1993 The Territorial Size of Mississippian Chiefdoms. Archaeological Report 25:143-168.

White, Devin A. and Sarah L. Surface-Evans. 2012 Least Cost Analysis of Social Landscapes: Archaeological Case Studies. University of Utah Press, Salt Lake City.

 

 

 

Geology & Shovel Testing

 

Geology has always been an important aspect of archaeology, if for no other reason than that we often find artifacts buried in the ground, so we should probably know a little about the ground itself. That seems fairly intuitive to me, but many field techs seem to actively avoid learning about geology and its applications in archaeology. Most field techs find it boring. I once worked with a very intelligent field tech who took a geomorphology class as an undergraduate, but apparently was so bored that she failed to retain any of the information. Let’s be honest—people become archaeologists because they want to find cool things, not because they’re passionate about soils. Kids who dream of becoming Indiana Jones and grow up to enjoy writing anthropology papers are not interested in learning about how dirt moves slowly over time.

But archaeological field technicians desperately need to know more about geology, especially geomorphology. Some might argue that this is a niche specialty that has little to do with the everyday tasks of the average field tech. In fact, there is a disturbing number of young field techs who act as though their job is simply to find artifacts, without a thought for the geological context.

But my decade of experience has taught me that an understanding of geology is directly salient to the routine, everyday duties of the average field tech. If you are a field tech—no matter what anyone else claims—you need at least a basic understanding of geomorphology to carry out your duties correctly. Many field techs are unknowingly carrying out their duties in a very poor fashion.

Let me explain why an understanding of geomorphology is so important. First, you need to understand that one of the most basic duties of a field tech is a task known as “shovel testing.” Archaeological surveys use two main methods: shovel testing and pedestrian survey. Pedestrian survey simply entails that a team of field techs walk across their survey area and look for artifacts on the surface of the ground. This method is only effective in areas where artifacts are visible on the surface, such as in tilled agricultural fields, or arid rangeland where the vegetation is sparse. Meanwhile, some form of subsurface testing is required in areas where artifacts are buried by vegetation, such as grassy pastures or wooded areas where the forest floor is covered by a layer of duff. The most widely used form of subsurface testing is a technique known as “shovel testing” or “shovel probing,” whereby a field tech digs a series of small holes at regular intervals across the survey area, and sifts the excavated soil through hardware cloth in order to locate small artifacts.

That may sound fairly simple, and most of the time it is, but one of the recurring complexities of shovel testing is knowing at which depth to stop digging. Field techs often have trouble estimating how deeply they should penetrate into the ground. The most common advice given to field techs is to stop digging when they reach a “color change” in the soil. This is phenomenally bad advice. Soil profiles can contain multiple zones of differently colored soil, and on some landforms, the artifacts may be hidden in the lower layers. In fact, this advice will ensure that you miss all of the deeply buried sites in your survey area. And these deeply buried sites are likely to be much more intact than anything you will find on the surface or in the plow zone.

So, when should you terminate a shovel test? There is no “one-size-fits-all” answer, because the answer depends on the landform you’re surveying, and every landform is different. This is but one of many reasons why field techs need to learn a little about geomorphology, and geology in general. In fact, I don’t have enough room in this blog post to explain all the ways in which geology is useful to archaeologists.

In the interests of helping field techs learn how to shovel test adequately, I’m going to outline some basic concepts of geomorphology, and try to explain them in a way that makes the information easy to retain.

 

Let’s start with some basic terms:

Geology: the study of the earth (you should know this if you went to college)

Geoarchaeology: the study of geology as it applies to archaeology. Geoarchaeologists study a wide range of topics. They help explain the ways in which various soils form and move around over time, and how those sediments can bury artifacts. They can also locate the geographic sources of the geological materials used to make certain types of artifacts, such as stone tools or clay pottery.

Geophysical archaeology: the study of geophysics and the use of geophysical technology as an aid in archaeological studies. Geophysical archaeologists use magnetometry, GPR (ground-penetrating radar), and electrical resistivity studies to map buried features.

Geomorphology: the study of landforms and how they change over time. Geomorphology does not necessarily have to have anything to do with archaeology—the formation of a mountain range a billion years ago also falls under the purview of geomorphology. Geomorphologists often help archaeologists understand the formation of relatively recent landforms, such as floodplains, loess bluffs, and glacial till fields.

Pedology: the study of soil. It does not necessarily have to have anything to do with archaeology, but it can be applied to archaeology, because artifacts are often found buried in the soil.

Soil horizon: a horizontal band or zone of soil that has a different color or texture than the bands above or below it (or some other distinguishing factor, such as an increase in calcium carbonate or redoximorphic features). These distinctly colored bands usually indicate biological and chemical processes within the soil, such as the downward leaching of minerals. In some very specific cases, these bands may represent the deposition of new layers of sediment. Soil horizons are sometimes also referred to as pedologic horizons or mineral horizons. Archaeologists often mistakenly refer to soil horizons as “strata,” but this is usually not the correct term.

Stratum: a distinct layer of sedimentary rock, such as sandstone or limestone. Archaeologists often use this term (erroneously) to refer to soil horizons, which are horizontal bands within the soil profile. The study of rock strata has little to do with archaeology, but it is useful for geology and paleontology. You could arguably refer to a layer of soil as a "stratum" if it is a discrete deposit that was left at a different period of time than the soil above or below it. But most of the time, soil horizons are imprinted on existing geological material and do not represent discrete deposits, so the term "stratum" would be inaccurate by any standard.

Soil profile: the vertical plane bisecting the soil in any given area, revealing the distinct soil horizons.

Stratified site: an archaeological site in which the artifacts are stratified by time period, with the oldest artifacts in the lowest layers, and the newest artifacts in the upper layers. Most archaeological sites in North America are not stratified.

 

Now that that’s out of the way, let’s talk about the different kinds of landforms you may have to survey.

The most fundamental distinction you have to make, when thinking about soils during an archaeological survey, is between residual soils and depositional soils. A residual soil is a soil that formed in situ from the local bedrock, whereas a depositional soil consists of sediment that has been removed from its original location and deposited elsewhere. This distinction may not seem pertinent to archaeology at first, but I will do my best to explain why it matters. For one thing, sites located in depositional soils have the potential to be stratified or deeply buried, whereas sites located on residual soils have no such potential.

 

Residual Soil

In order to understand the formation of a residual soil, you need to understand what soil is made of. Soil is made of a combination of rock particles and organic material, but it is mostly rock particles. Wherever you are, the soil beneath your feet is probably made of tiny fragments of eroded bedrock.

In a residual soil, the soil particles have not moved away from the place where they once comprised a part of the local bedrock. Imagine you are standing on a slab of bedrock, a million years ago. Now imagine that the rock is being weathered and broken up into tiny particles over time, but these particles remain in roughly the same place. Then, imagine that these particles are being mixed with decomposing organic matter, such as feces and dead plants and animals. This is now a residual soil.

Residual soils are typically found on the ground surface in upland areas where no sediment from other areas has been deposited. For example, the ridges and hilltops of the Appalachian and Ozark Mountains are largely covered in residual soil, mixed with fragments of the original bedrock. Residual soils in such places are often very rocky (soil scientists refer to small slabs of stone in the soil as channers or channery, and large slabs as flagstones). Many field techs who work in these mountains mistakenly assume that the soil they find on the upland ridges has been deposited there, but this is incorrect. The soil most likely formed in situ.

If you dig a shovel test into a residual soil, you may see multiple horizons—or bands of differently colored soil—within the vertical profile. Most archaeologists erroneously assume that these horizontal bands are distinct layers of deposition, with multiple layers of sediment being deposited on top of each other. However, this is not true. There are no layers of deposition in a residual soil profile, because there is no deposition of new sediment at all.

These horizontal bands represent different kinds of biological and chemical change to the original bedrock. The different colors were imprinted on the existing geological material as it was weathered into soil. Let me explain:

The uppermost layer of true soil is known as the A horizon, or “topsoil.” The A horizon has been darkened by the organic material within it, due to its exposure to the biological activity at the surface. In other words, it is dark-colored because the particles have mixed with the decaying organic matter found at the surface.

Below the A horizon is the B horizon, or “subsoil.” The B horizon is also called the zone of illuviation, because it may contain minerals and clay particles that have illuviated (filtered) downwards from the topsoil. In many places, the B horizon has a red or yellow hue, due to the concentration of iron and aluminum oxides that have leached into it from above. The B horizon may also have a higher clay content than the soil above it, as clay particles (the smallest type of soil particle) filter downwards and accumulate in the subsoil.

Below the B horizon is the C horizon, or “parent material.” This is basically broken-up bedrock that has not been affected by the processes I’ve described above. It contains no organic material, unlike the A horizon. And it has not accumulated any downward-moving clay or oxidized minerals, unlike the B horizon.

And below the C horizon is the R horizon, or bedrock.

These are the four basic horizons in a residual soil profile: the A horizon (topsoil), B horizon (subsoil), C horizon (parent material), and R horizon (bedrock).

Some profiles will also contain an E horizon, or “zone of eluviation” (not to be confused with “illuviation”). The E horizon is typically a band of light brown to ashy white soil between the A horizon and B horizon. It contains little or no organic matter, and all of the iron and aluminum oxides have leached away into the B horizon.

In a naturally occurring residual soil, the topsoil may be a very thin layer, if it exists at all—in some places, it may be eroded away before it can form. The act of plowing will modify the A horizon, creating an Ap horizon (plowed topsoil) that is typically about 20-40 cm. thick.

When people drop artifacts on the surface of a residual soil, these artifacts will gradually become buried by biological activity, known as “bioturbation.” Leaf litter will bury artifacts along with various seeds that have fallen to the forest floor. These seeds will sprout roots, and the roots will push artifacts downwards into the soil. Burrowing animals such as ants and earthworms will cover the object with soil displaced from their tunnels. After a matter of decades, an artifact left on the ground in a heavily vegetated environment will be completely buried, swallowed up by the earth.

 

Figure 1. Illustration of residual soil profile

What this means for archaeologists is that sites on residual soils are not chronologically stratified—the older artifacts are not buried deeper than the younger artifacts. Many field techs assume that an artifact found in the B or C horizon must be older than the artifacts found in the A horizon, but this is not necessarily the case, because these horizons are NOT layers of sediment that can be said to be “older” or “younger” than each other. When artifacts are incorporated into the soil through bioturbation, they are mainly concentrated in the A horizon (or Ap horizon), where most of the biological activity is occurring. They may filter down into the B or C horizon as well, but that doesn’t mean that these artifacts are any older than the artifacts above them. Thus, if you find an archaeological site in a residual soil, it is almost certainly non-stratified—it is basically a surface site, where most of the artifacts, regardless of time period, are located at or near the surface.

What does all this mean for the average field tech, who simply wants to know when to stop digging a shovel test? In many cases, if you are shovel testing in a residual soil, it is sufficient to stop digging when you have penetrated 10 cm. into the B horizon. In my experience, most of the artifacts will be located in the A or Ap horizon. These artifacts have been buried by bioturbation, if buried at all, and most of the biological activity occurs in the A horizon, so it stands to reason that the A horizon would contain most of the artifacts.

Some older field techs or crew chiefs may tell you to stop digging when the soil becomes clayey. This can be good advice, but only within certain contexts. It is typically good advice if you are testing in a residual soil, but only if the B horizon contains more clay than the A horizon. As I’ve already explained, the artifacts will probably be located in the A horizon, so you can probably stop when you hit the B horizon. Because clay illuviates downwards into the B horizon, an increase in clay content (in addition to an increasingly red or yellow hue) is a good indicator that you have reached the subsoil.

Many poorly informed field techs have told me that I won't find artifacts in clay at all, but this is simply not true. I have found many artifacts, some very recent in manufacture, buried in solid clay. There is nothing about clay soil that makes it incapable of containing artifacts. If you're testing in a residual soil in which most of the clay has illuviated down into the subsoil, you will probably find most of your artifacts in the topsoil, but this has nothing to do with the clay content of the subsoil. The artifacts will be concentrated in the topsoil simply because the topsoil is close to the surface. In some soils, the topsoil is pure clay, and you can find plenty of artifacts in this clay if you bother to look. I certainly have.

I would also like to point out that artifacts are more likely to move downwards in sandy soil than they are in silt or clay. I have shovel tested in residual soils in which the A horizon and E horizon both consisted of very loose sand (having formed from weathered sandstone), and because the soil was so loose and sandy, the artifacts had all moved downwards out of the A horizon and into the E horizon. So in that sort of situation, it really is useful to keep digging until you hit some kind of clay (which would probably signify the B horizon), even if you have penetrated all the way through the A horizon. You might be finding artifacts up to a meter deep, but from a geological perspective, this is still basically a non-stratified surface site.

 

Depositional Soils

Now that we’ve discussed residual soils, we have to move on to the hard part. Geomorphology becomes much more complicated when soil starts to move away from its original location.

Soil that has moved from its original location and been deposited elsewhere is known as depositional soil. This kind of soil can be transported by many different agents, including wind, water, and ice. Here are a few of the different kinds of depositional sediment:

Alluvial: sediment that has been deposited by running water when a stream overflows its banks and floods the adjacent area. This kind of deposit is also known as an “overbank deposit.” Over time, alluvial sediment accumulates to form landforms known as "floodplains." These landforms are very flat, mostly devoid of large stones, and are found alongside creeks and rivers.

Aeolian: sediment that has been deposited by the wind. Sand dunes are aeolian landforms, which are constantly being shifted by the wind. Loess—or wind-blown silt—is a form of aeolian deposit that can be found on upland areas in many parts of North America, especially on blufftops overlooking major rivers such as the Mississippi, the Illinois, and the Ohio.

Glacial: sediment that has been deposited by a glacier—a moving sheet of ice. Much of North America was once covered in glaciers, and these glaciers left layers of “glacial till” strewn across the landscape. Glacial till can be a combination of many geological materials, including sand, clay, gravel, stones, large boulders, and anything else in the glacier’s path, all mixed haphazardly together. Most of the glaciers of the Pleistocene Epoch have melted, but surviving glaciers can still be found in alpine areas and the far north.

Colluvial: sediment that has been transported downhill by gravitational forces. Sometimes, soil slowly slumps downhill. Other times, it moves rapidly in the form of mudslides or rockslides. Colluvial sediment can be found at the bases of hillsides and mountain slopes.

Fluvial: sediment that accumulates underwater on a riverbed. Fluvial sediment can bury boats or other artifacts left in a river. In all likelihood, you will not need to shovel test in a river or other body of water. In fact, underwater archaeologists have their own methods. Underwater archaeology is not a common part of the CRM industry.

Lacustrine: sediment at the bottom of a lake.

Marine: sediment on the seafloor.

Soil can be transported by more than one agent. For example, during the Pleistocene Epoch, glaciers dragged sediment southwards across much of the Midwest. As the glaciers melted, the meltwaters washed away much of the finely ground glacial silt and clay, transporting this sediment onto the floodplains to the south. These silt and clay particles were then picked up by the wind and carried onto the upland areas south of the glaciers, where they still reside as periglacial loess. If you happen to be walking over a field in Illinois or western Iowa, the soil beneath your feet is probably loess that has been transported by a combination of glacial, alluvial, and aeolian activity.

Now that you have an idea of how soil moves around, I’m going to discuss some of these types of landform in a little more detail, as they pertain to archaeological surveys:

 

Alluvium

Rivers have soil particles suspended in their currents all the time, which is why their water tends to be murky and muddy. When rivers flood their banks, the floodwaters carry these soil particles onto the land adjacent the river. As the rushing water slows down, the particles fall out of suspension and settle on the surface of the ground. This is how floods cause new sediment to be deposited on the floodplains alongside streams and rivers.

 

Figure 2. Flooded agricultural field on alluvial plain alongside Kickapoo Creek, Illinois, after heavy rainfall. The floodwaters deposited fresh alluvial sediment before receding.

Alluvial sediment tends to be “finely sorted,” as particles of different sizes will fall out of suspension at different times, due to their different weights. Thus, sand particles are deposited with other sand particles, silt with silt, and clay with clay.

Most floods are not powerful enough to lift large stones over a river’s banks, which is why floodplains usually do not contain naturally occurring rocks bigger than a piece of gravel. There are some exceptions. For example, the floods within the Columbia Gorge are so intense that they can deposit massive boulders amidst the alluvium. But usually, floodplains are almost devoid of rocks. If you find a rock on a floodplain, there is a good chance it was introduced by humans. Even if it doesn’t look like an artifact, it may be a manuport (something transported by humans). I once found what I initially believed to be ordinary river cobbles in a test unit that I had excavated deep into alluvial sediment, and I made the mistake of thinking nothing of them. But a geomorphologist noticed immediately that they could not have occurred there naturally, and upon further inspection, he revealed to me that these were a type of artifact known as a “pitted stone” (these are often interpreted as nut-cracking stones, but their exact function is not known).

What does all of this mean for archaeologists? For one thing, it means that archaeological sites found on floodplains have the potential to be stratified by time period, with the older artifacts located in the lowest layers of sediment, and the newer artifacts in the highest layers. Rivers are continually flooding and depositing new sediment on their adjacent floodplains, burying old artifacts and creating new surfaces where younger artifacts can be dropped. The Koster site—one of the most significantly stratified sites in North America—is located on (or more accurately, within) a floodplain alongside the Illinois River. Stratified sites such as the Koster site can contain multiple layers of deposits, with artifacts located in each layer, possibly extending several meters underground.

Figure 3. Alluvial soil profile at the Nesquehoning site, a stratified archaeological site in Pennsylvania. Courtesy of Carr and Moeller (2015).

Of course, not all sites found on floodplains are chronologically stratified. Some floodplains do not flood much anymore, and seldom accumulate sediment. I have personally walked over floodplains where artifacts from completely different time periods were all located on the surface together, unburied.

I also need to clarify that the horizontal bands you see in the soil profile of a floodplain are not necessarily distinct layers of sediment. Sometimes they are, and you might even be able to see thin lenses of sand that represent singular flooding episodes. But often, you will see the same kinds of soil horizons that you would expect to find in a residual soil, because the same processes are at work. An A horizon at the top of a floodplain may not necessarily be a younger layer than the B horizon below it; it may just look different because it contains more organic matter, due to its proximity to the surface.

I once surveyed a very old floodplain that was created by a single massive flood at the end of the Pleistocene, and has not accumulated any sediment since, as far as I know. A distinct A horizon and B horizon had formed within the sediment after it was deposited, but the A horizon was not any “younger” than the B horizon—all of the sediment within that landform had been deposited during a single massive event about 11,700 years ago.

What does any of this mean for the average field tech, trying to shovel test on a floodplain? How deep should he or she dig? The truth is that there is no easy answer. You can dig through multiple “color changes” and still find artifacts. You can dig through solid clay and still find artifacts—after all, floodwaters can deposit clay particles in the same way that they deposit sand or silt particles, and these clay particles can bury artifacts too, meaning it's common for sites to be buried under alluvial clay. The advice that is usually given to field techs simply does not apply here (even if the advice would be good in other situations). In some floodplains, you can just about dig forever and still be finding cultural material. Hell, you can dig over a meter into the ground and find only modern plastic, without even approaching the older layers that might contain pre-Columbian artifacts.

I’ve used augers to conduct deep testing in alluvial clay deposits near the Skunk River in Iowa, and found Native American artifacts in thick gley (gray clay) nearly two meters deep. Most field crews would not even try to dig through such thick clay, but if you bother to do so, you might be surprised at what you can find.

One reason that you should not stop digging simply because you’ve encountered a color change is because that color change may indicate the beginning of an Ab horizon (buried topsoil) that is rich in artifacts. Many floodplains contain Ab horizons that are concealed beneath layers of more recent alluvium. These Ab horizons used to be stable ground surfaces, where people may have lived and dropped artifacts. I've seen too many field techs penetrate into an Ab horizon during a shovel test, only to stop digging, mistakenly assuming that the Ab horizon was subsoil.

I once surveyed a floodplain in Oklahoma, which contained a layer of dark, rich, heavily organic soil about 90 cm. below the surface. It was covered in a mantle of light brown alluvial sediment. The organic layer was clearly an Ab horizon, and it contained several artifacts and burnt animal bones, which revealed the presence of a significant pre-Columbian site. The only reason I mention this site is because the floodplain where it is located had been previously surveyed, but the original crew had not found anything. There could have been a lot of reasons for their failure to locate the site. They may have assumed that the Ab horizon was subsoil, and stopped digging when they found it. Most of the artifacts were about a meter below the surface, and the original crew was probably not required to test that deeply anyway. In addition, the ground may have been so hard at the time that they could not dig at all—some parts of the Southern Plains become as unyielding as concrete when the ground dries out in late summer. I once tried to survey another floodplain in Oklahoma where it was virtually impossible to shovel test because the earth was simply too hard and dry at that time of year, and I still wonder what might be buried there. 

My point is that it is very easy to miss sites that are contained within a deeply buried Ab horizon in an alluvial plain. When encountering a floodplain, you may need to adjust your methods and test more deeply than you otherwise would. Perhaps use an auger to penetrate farther into the ground than you can with a shovel. At the very least, you need to know not to stop digging if you reach an Ab horizon, because it is not "subsoil."

 

Figure 4. Illustration of alluvial soil profile with Ab horizon

It is also worth noting that the floodplain I discussed above was covered by a tilled agricultural field where the ground surface was completely visible, but there were no artifacts on the surface. They were all too deeply buried, about a meter below the ground. Many crews would have felt it sufficient to conduct a pedestrian survey here, without shovel testing at all, but if my crew had not implemented shovel testing, we never would have found the site. This is one of the complexities of trying to survey a floodplain. The ground surface may be bare enough to allow for pedestrian survey, but if all the artifacts are buried under alluvial sediment, pedestrian survey will not be effective, and you may need to resort to deep testing with a shovel or auger.

This is a case study in the importance of knowing how to read the earth. You should know when you’re standing on a floodplain, and if you are, you should be willing to dig deep, in case there are deeply buried artifacts. You should be willing to shovel test even if you have perfect visibility for pedestrian survey. While you’re shovel testing, you should know an Ab horizon when you see it, rather than mistaking it for subsoil, and you should understand that even if you encounter a B horizon (subsoil), there may be an Ab horizon below it. And you should know that just because the soil is clayey and difficult to dig through, or dry and compact, does not mean it is culturally sterile (devoid of artifacts). Floodplains make life more complicated for field crews, and unfortunately, many field crews are behind the learning curve, and probably end up missing archaeological sites.

 

Glacial Till

Layers of glacial till cover much of the ground surface across the Midwest and New England. Glacial till can contain everything from clay to sand to gravel to boulders.

These deposits were left during the Pleistocene, and since that time, many of these upland deposits have not accumulated any new sediment. Some of the upland glacial till deposits in Illinois have been buried under layers of loess, but in many other places (such as Ohio), the glacial till deposits are directly on the surface. Over the past 11,700 years, soil horizons have been forming within the glacial till after it was deposited, in the same way that similar horizons form within residual soils. The soil at the top accumulates organic matter, and the soil below it accumulates clay and iron oxide, causing visible horizontal bands to form. The topsoil is not necessarily more recent than the subsoil.

Artifacts left on the ground surface over the past 11,700 years will be concentrated at or near the surface. Most artifacts will be located in the A horizon. If you are shovel testing through glacial till, it is usually safe to terminate your test at the B horizon.

 

Periglacial Loess

I discussed earlier how silt and clay were washed out of melting glaciers, and then picked up by the wind and deposited on upland landforms as loess (wind-blown silt). Because these loess formations are affiliated with glacial activity, most of them stopped accumulating sediment after the Pleistocene. Artifacts made in the past 11,700 years will be on the surface, or in the topsoil.

There are some loess hills along the Missouri River that have probably continued to accumulate sediment into the early Holocene. In places such as these, very old artifacts (such as Paleoindian artifacts) could possibly be deeply buried, much as they would be in floodplains.

 

Other Landforms

Geomorphology is too complicated for all the salient information to be mentioned here, but there is a wide array of landforms you may have to survey over the course of a career in CRM. There are shifting sand dunes in southern Colorado, which can continually bury and then re-expose ancient artifacts. There are ancient Pleistocene lakebeds in northern Nevada, which have since turned to desert, and now have artifacts lying directly on the surface amidst the sagebrush. The coastal plain along the Gulf Coast contains sandy deposits that originally formed under the sea. In the badlands of North and South Dakota, ancient deposits of volcanic ash have transformed into slopes of bentonite clay.

 

Figure 5. Ancient Pleistocene lakebed in northern Nevada. What was once lacustrine sediment is now the surface of a desert, sparsely covered in sagebrush.

As a general rule, you will not need to shovel test much in the Western states, because the Western climate is dry, and artifacts can be visible through the sparse grass or sagebrush. Shovel testing is mainly implemented east of the Rocky Mountains. But even in the West, it can be useful to have some basic knowledge of geomorphology.

I haven’t even discussed the myriad of other ways in which geology can be applied to archaeology. Archaeologists can identify the geological material from which a stone tool was made, and trace it to a geographic area where that material naturally occurs. They can use pXRF analysis or neutron activation analysis to determine whether a sherd of pottery was made from local clay. They can even use a magnetometer to detect the buried remains of cooking pits or burned houses. But all of that falls outside the ordinary duties of the average field tech. The purpose of this blog post is simply to help field techs become more adept at the otherwise very simple task of shovel testing. The fact that such a long post can be written about this topic, while only offering the most simplified version of the geological sciences, should be a testament to the fact that a field tech’s job is more complicated than most people in CRM (especially field techs) are willing to admit.

 

 

Sources

 

Alex, Lynn M. 2000 Iowa’s Archaeological Past. University of Iowa Press, Iowa City.

Balek, Cynthia 2002  Buried Artifacts in Stable Upland Sites and the Role of Bioturbation: A Review. Geoarchaeology: An International Journal 17(1):41-51.

Butzer, Karl W. 1978 Changing Holocene Environments at the Koster Site: A Geo-Archaeological Perspective. American Antiquity 43(3):408-413.

 

Carr, Kurt W. and Roger W. Moeller 2015 First Pennsylvanians: The Archaeology of Native Americans in Pennsylvania. Pennsylvania Historical and Museum Commission, Harrisburg.

Luhr, James F. 2003 Earth. DK Publishing, New York.

 

Schaetzl, Randall J. and Michael L. Thompson 2015 Soils: Genesis and Geomorphology. Cambridge University Press, New York.

 

Schwegman, John E. 2016 The Natural Heritage of Illinois: Essays on its Land, Waters, Flora, and Fauna. Southern Illinois University Press, Carbondale.

Stein, Julie K. and William R. Farrand 2001 Sediments in Archaeological Context. University of Utah Press, Salt Lake City.

Targulian, V.O. and R.W. Arnold 2008 Pedosphere. In Global Ecology, edited by Sven Erik Jorgensen. Academic Press, Amsterdam.

Waters, Michael R. 1992 Geoarchaeology. A North American Perspective. University of Arizona Press, Tuscon.

Updated April 11, 2024