With both human societies and ecosystems worldwide now facing ongoing, and even accelerating, environmental change, both scholars and policy makers are increasingly concerned with predicting the future implications of climate change. Where will our coastlines, tree lines, and urban boundaries lie in 50 or 100 years? How will changes in the seasonality and intensity of precipitation, frosts, and heat waves affect the plants and animals on which we rely for food? And, most important, what are the consequences for us? 

One avenue for understanding human responses to dramatic environmental and climatic change is to look to the past when societies faced similar periods of rapid change. Paleoclimatologists and paleoecologists have developed numerous methods to identify ancient environmental change, creating rich records from glacial ice at the poles and on mountaintops, as well as cores drilled deep into seabeds and lakes that preserve hundreds or thousands of years of annually deposited sediments. Archaeologists who study the deep history of human-environmental relationships draw on these datasets, as well as archaeological records of social and economic change, to explore human adaptation to environmental change in the past. 

A variety of archaeological finds are useful in identifying climatic change, from mammal and fish bones to microscopic starch grains found on tools used in plant food processing. One material commonly found in archaeological sites from many different periods of the human past, nearly worldwide, is wood charcoal. Incompletely burned wood from fireplaces, ovens, kilns, and accidentally (or deliberately) burned buildings becomes inorganic charcoal, which is resistant to degradation from soil microbes and fungi and thus can survive for thousands of years within the soil. It is frequently possible to identify the type of tree that produced these charcoal remains and thus reconstruct patterns of wood use and forest change, both as a result of climatic change and deliberate or inadvertent human reconfiguration of woodlands. Scholars have developed methods for systematically recovering, identifying, and interpreting these remains to identify patterns of climate and environmental change in the past. 

Recently, Boston University and the Arnold Arboretum have begun a partnership to draw on the vast living collections of the Arboretum to improve the resolution of archaeological charcoal studies in the Environmental Archaeology Laboratory in the Department of Archaeology at Boston University. In this article, I describe how archaeologists study charcoal from archaeological sites and use it to reconstruct the human role in environmental change, highlighting how resources of the Arnold Arboretum enhance our teaching and research mission at Boston University. 

Recovering and Identifying Archaeological Plant Remains

Wood charcoal fragments from archaeological sites have been studied since the 1940s to address multiple questions about human wood use in the past. The first step in archaeological charcoal analysis is systematic recovery of charcoal remains from archaeological sites. Although not a universal practice, the recovery of plant remains is increasingly ubiquitous among archaeologists worldwide, even in remote areas of developing countries. We recover soil samples, generally 10 to 20 liters (2.6 to 5.3 gallons) in volume (equivalent to one or two buckets full), from every archaeological level and distinct feature (e.g., a pit or a hearth) identified during excavation. 

Archaeologists most commonly use a water flotation method to recover charred plant remains, including wood charcoal as well as carbonized seeds and other plant structures, from soil samples. Although flotation can be accomplished using only a pair of buckets and a fine mesh strainer, more common are systems that pump large volumes of water to process even large samples quickly. Clean water is pumped into the tank of the machine where the soil sample is held in a plastic window screen mesh. The water dissolves the soil, freeing carbonized plant remains, which float, and rinsing away sediment in the dirty effluent that is released from the bottom of the tank. Heavy components of the soil, including bone and pottery fragments as well as occasional heavy pieces of charcoal, are caught in the window screen and later dried and analyzed. The floating, or light, fraction consists of wood charcoal and carbonized plant remains, but also soil components lighter than water, including tiny roots and fine clay particles. The light fraction is allowed to overflow into a very fine polyester mesh, with holes less than 0.1 millimeter (0.004 inch) to catch even the smallest seeds. This fraction is then carefully air dried and brought to the laboratory for identification and analysis. 

We then pour the light fraction through a series of nested sieves, creating several size classes of material that can be sorted differently. In general, only wood charcoal fragments larger than 2 millimeters (0.08 inch) are analyzed, as smaller fragments are unlikely to be identifiable. Systematically sorting each size class under low-power stereomicroscopes, we remove each type of plant remain for subsequent identification and measurement, with wood charcoal, carbonized seeds and seed fragments, and nutshell distinguished and separated. Wood charcoal fragments are then weighed in aggregate and a representative number of those fragments are identified. 

The identification of wood charcoal can be challenging because fragments are often small and may be distorted by burning and subsequent deterioration in the soil. Fortunately, different species of woody plants vary considerably in their cellular anatomy, which allows wood (even charcoal) to be identified to varying levels of specificity depending on the wood type. Wood can be viewed from three planes, each of which presents a distinct set of anatomical structures for identification. All three are necessary for detailed identification, but the transverse, or cross section, is the most useful for charcoal identification and can be examined with a stereomicroscope at 20 to 100× magnification. Distinguishing hardwoods (angio-sperms) and softwoods (gymnosperms) can be easily accomplished using just low-power magnification of the transverse section; many families within these large categories can also be distinguished based solely on the transverse section. Using a combination of basic reflected light microscopy, high-power incident light microscopy, and electron microscopy, we catalog features of archaeological wood fragments and assign them tentative identifications based on their anatomy. Confirmation of these identifications, however, typically requires a comprehensive comparative collection of modern wood taken from properly identified and fully vouchered trees. Assembling such a comparative collection has been an ongoing effort of the Environmental Archaeology Laboratory and is the origin of our collaboration with the Arnold Arboretum. 

Using the Arboretum as a Research Collection

The Arnold Arboretum offers a tremendous opportunity to collect wood from a wide variety of temperate tree species from the Americas, Europe, and Asia. Each tree is properly identified and labeled, and considerable information regarding its life history is recorded in the Arboretum’s living collections database. For our partnership, since most woody plants are identifiable at the genus level, we preferentially collect wood from species native to the areas in which members of the Environmental Archaeology Laboratory work (mainly southern Europe, the Middle East, East Asia, and north-eastern North America). When the most relevant species are not available, we choose other species of those genera in order to obtain the most similar comparative specimens possible. 

Wood anatomy can vary based on the diameter and age of the branch collected, between branch and trunk wood, and because of unique growth conditions such as bending or disease. As a result, we attempt to collect wood from multiple parts of a tree when possible. The Arboretum facilitates our collection by allowing us to gather dead branches that have fallen from trees as well as gathering samples from trees that are trimmed or cut down during the course of routine tree maintenance activities. Members of the Environmental Archaeology Laboratory compiled a “wish list” of trees in the living collections that Arboretum arborists can refer to when tree work is done. The arborists then collect specimens from trees of specific interest to us. We periodically stop by the Arboretum to collect these wood samples for further processing at Boston University. 

Back in the Environmental Archaeology Laboratory, we interface with the Arboretum’s database and use the Arboretum Explorer web- site (http://arboretum.harvard.edu/explorer/) to gather information about trees that have been sampled. We record much of that information into the Environmental Archaeology Laboratory Collections Database, which is also searchable online (http://sites.bu.edu/ealab/ collections/database/). The wood sample is then divided between a wood specimen and a specimen to be converted into charcoal. Experimental carbonization of comparative wood samples is critical for two reasons. First, carbonization can modify the structure of the wood in predict- able ways, leading perhaps to certain patterns of cracks that can be diagnostic when examining archaeological wood charcoal. Second, charcoal can be easily broken to expose any of the three planes, facilitating rapid examination, while wood needs to be cut with an ultrathin blade so as not to crush the exposed cell walls, requiring additional equipment and time to prepare comparative slides. 

We carbonize wood using a muffle furnace capable of reaching temperatures of 1000°C (1832°F), although we typically carbonize wood around 400°C (752°F) to maximize speed of carbonization without incinerating the wood. It is critical that wood heat in an oxygen-poor reducing atmosphere because that promotes charcoal formation, while an abundance of oxygen would lead to ashing and destruction of the sample. We carefully wrap samples twice in heavy-duty aluminum foil to minimize contact with oxygen and pack them tightly in the muffle furnace. At 400°C, wood carbonizes in 10 to 40 minutes, depending on the thickness of the pieces. 

Finally, both charcoal and wood specimens are stored in labeled boxes within a specialized shelving system in the lab. The boxes include basic information on the wood and its location of origin, together with an identifier code that corresponds to its record in our database. A future project for the laboratory is to take microscopic images of the wood anatomy of all woods in the collection and to make them available online, both through the laboratory website and as a contribution to Inside Wood (http://insidewood.lib.ncsu.edu), a free, public, wood anatomy database created at North Carolina State University. Although extensive comparative collections of wood samples are preserved at other arboreta and herbaria worldwide, very few of these have been digitized to make them publically accessible. Because our collection includes specimens from many countries of the Middle East and Central Asia, as well as specimens from several arboreta in the United States, we aim to publicize our records as widely as possible as a research tool for archaeologists worldwide. 

Reconstructing Past Woodland Ecology and Wood Use, with Implications for the Future

Once it is possible to identify wood fragments reliably, we work to identify a statistically robust subsample of all wood charcoal fragments present in our archaeological samples. Recording both count and weight of these fragments, we are able to create diagrams that represent change in the prevalence and context of use for woods over time. For example, in my ongoing research at the ancient city of Gordion, in central Turkey, which was inhabited from the Early Bronze Age (3000 to 2000 BC) through the medieval period (fourteenth century AD), I was able to document changes in wood use practices and forest ecology over a span of 3,000 years. Gordion became a large city around 800 BC as the capital of the Phrygian kingdom, which grew from Gordion to control most of central Turkey. At that time the Phrygians began to construct monumental temples, massive city walls, and huge earthen burial mounds (the largest over 170 feet [52 meters] in height) containing royal burials inside elaborate wooden structures, including the oldest standing wooden building in the world. 

This amazing structure was fashioned from juniper (Juniperus spp.) wood, which was widely used within the city in roofing large public buildings. Juniper is a slow-growing tree, however, and the inhabitants of Gordion appear to have quickly exhausted their supply of easily cut large juniper trees. In later periods of occupation, charcoal samples from burned buildings indicate that oak (Quercus spp.) and pine (Pinus spp.) were the primary woods used in construction, both of which have the advantage of being fast-growing trees that often take over in sites where older juniper trees have been cut. Oak and pine, however, have inferior strength and rot resistance compared to juniper. Archaeological wood charcoal assemblages show a dramatic human impact on the landscape that led to considerable forest reorganization during the early history of the city. Later inhabitants of the region had to contend with a different landscape, and different availability of natural resources, than their ancestors. 

A color photo of a group of people in a lab.
Using the Arboretum as a Teaching Collection

Every summer I teach a weeklong intensive workshop on wood anatomy and wood charcoal identification for archaeologists. Participants come to Boston from universities nationwide, from Santa Barbara to Chapel Hill, and a few even join us from across the river in Cambridge. Participants are mainly doctoral students, but we have a number of faculty participants and even an occasional undergraduate. During the week we cover wood anatomy from initial concepts (e.g., the three planes of wood) to advanced structural variation (e.g., ray cell margin shape in gymnosperms). We also read and discuss a number of articles that illustrate best practices for sampling and recovery of wood charcoal from archaeological sediments, methods for quantifying and presenting results, and the challenges of changes brought about by both the initial burning of wood and its preservation in soils for hundreds or thousands of years. Students spend the last few days analyzing their own wood charcoal assemblages and learning how to identify the woods common to their areas of expertise, which have ranged from southwest China to Jordan, the California Channel Islands, the Yucatan, and the Andes.

One highlight of the week is our field trip to the Arnold Arboretum. The group receives an orientation and tour led by Michael Dosmann, Curator of Living Collections. During the tour, workshop participants learn about the unique collections of the Arnold Arboretum and about the life history of particular trees on the property. Following an orientation to the Arboretum Explorer web application, participants are able to use their smartphones to find particular trees of interest to them and spend the next two hours visiting those trees. We collect a few samples of dead wood from the ground under selected trees to bring back to the laboratory, where we then experimentally carbonize wood samples and study their microscopic structure. Participants then split their newly collected specimens, with a portion joining the permanent collection of the Environmental Archaeology Laboratory and the rest returning home with each participant. As a result, every participant returns to their home lab with the beginning of a personal comparative charcoal collection and the experience needed to expand their collection through fieldwork and collaborative ventures with local botanical gardens and arboreta.

Photo: Participants in the 2014 wood charcoal workshop analyzing samples in the Boston University Environmental Archaeology Laboratory. John M. Marston

Examples such as the case of Gordion parallel more recent human history, both in central Turkey and worldwide, in which human activity transforms a landscape for future inhabitants. When viewed from the perspective of later populations, we term these impacts “legacy effects,” and the implications of such changes are many. It has been argued by several scholars, including Jared Diamond, that the deforestation of Easter Island pushed its ecosystem beyond a tipping point that led to severely reduced resources and impoverishment of the isolated inhabitants. In contrast, legacy effects may also have been deliberate outcomes, designed to boost productivity and resource availability. The use of fire to maintain prairie habitats in the American Great Plains prior to European contact is an example of such “niche construction,” in which people modify their environment to boost productivity of desired resources to suit their cultural needs. 

Archaeologists have explored these environmental histories using wood charcoal analysis, and continue to search for a deeper understanding of not only when and how, but also why human groups manipulate their landscape in specific ways. These detailed studies offer cases of environmental disaster and social collapse, but also resilience and survival in even the most uninviting landscapes. As contemporary society faces environmental change on an unprecedented scale, archaeologists offer both cautionary and inspiring stories of human-environmental relationships that provide novel, proven effective tools for continued survival in a changing world. 

Additional Reading 

These include sources that outline the practice of archaeological wood charcoal analysis (Asouti and Austin 2005, Marston 2009); wood anatomy and identification (Panshin and de Zeeuw 1970, Schweingruber 1990, Schweingruber et al. 2006); frameworks for studying human-environmental interactions (Cumming et al. 2006, Marston 2015, Redman 1999, Smith 2007); and more about our team’s recent work at Gordion (Marston in press, Miller 2010, Rose 2012). 

Asouti, E. and P. Austin. 2005. Reconstructing woodland vegetation and its exploitation by past societies, based on the analysis and interpretation of archaeological wood charcoal macro-fossils. Environmental Archaeology 10: 1–18. 

Cumming, G. S., D. H. M. Cumming, and C. L. Redman. 2006. Scale mismatches in social-ecological systems: causes, consequences, and solutions. Ecology and Society 11: 14. 

Marston, J. M. 2009. Modeling wood acquisition strategies from archaeological charcoal remains. Journal of Archaeological Science 36: 2192–2200. 

Marston, J. M. 2015. Modeling resilience and sustainability in ancient agricultural systems. Journal of Ethnobiology 35: 585–605. 

Marston, J. M. (In press). Agricultural Sustainability and Environmental Change at Ancient Gordion. Philadelphia: University of Pennsylvania Museum Press. 

Miller, N. F. 2010. Botanical Aspects of Environment and Economy at Gordion, Turkey. Philadelphia: University of Pennsylvania Museum Press. 

Panshin, A. J. and C. de Zeeuw. 1970. Textbook of Wood Technology. New York: McGraw Hill. 

Redman, C. L. 1999. Human Impact on Ancient Environ­ments. Tucson: University of Arizona Press. 

Rose, C. B. (editor). 2012. The Archaeology of Phrygian Gordion, Royal City of Midas. Philadelphia: University of Pennsylvania Museum Press. 

Schweingruber, F. H. 1990. Anatomy of European Woods. Stuttgart: Haupt. 

Schweingruber, F. H., A. Börner, and E. D. Schulze. 2006. Atlas of Woody Plant Stems: Evolution, Structure, and Environmental Modifications. Berlin: Springer. 

Smith, B. D. 2007. The ultimate ecosystem engineers. Science 315: 1797. 


John M. Marston is Assistant Professor in the Departments of Archaeology and Anthropology at Boston University.