E. B. Banning

One of the field methods of archaeology, archaeological survey, is one that most people think of simply as a prelude to excavation. In fact, in modern archaeology, survey is some form is a field method in its own right, able to answer some kinds of archaeological research questions that excavations, alone, cannot answer. In addition, it is a key component of assessments in Cultural Resouce Management (CRM) or heritage management.

Many non-archaeologists assume that all the archaeological sites have been found by now (actually "new" ones are found every day), while others wonder how we manage to find sites, particularly when they are buried.

These questions involve one aspect of archaeological survey - prospection - but survey has various purposes, including:

• Prospection (search and detection of archaeological materials)
• Sampling and parameter estimation (estimating characteristics of some archaeological population)
• Detecting spatial structure (trying to find patterns in the distribution of artifacts, architecture, or sites)

Each of these goals is best accomplished with distinct sets of methods, although there is much overlap in method, especially as archaeologists often have more than one goal. These goals also overlap with such goals as managing heritage resources and minimizing the impact of modern development on those resources

• Sometimes called "purposive survey"
• Purpose: To discover particular targets or kinds of targets
• Effective or "successful" prospection finds sites or artifacts
• Constraints may include
  • Minimizing costs or search time
  • Operating within resource constraints
  • Minimizing the risk of missing targets
    
• Approaches to Prospection Optimization
  • Bayesian - optimal allocation of search effort among spaces, adapt to new information
  • Game Theory - treats search as a game between a searcher and a hider
• Factors Affecting Optimization
  • Cost (Search Time or Effort)
  • Resource Constraint
  • Range of the Sensor (Eyes or Instrument)
  • Sweep Width (Spacing)
  • Area or Number of Spaces to be Searched
  • Presence of False Targets
    

General factors affecting detection

• Properties of the target
• Type of signal that communicates information about these properties
• Medium of signal propagation
• Kind of sensor or method of inspection
• Our ability to recognize the signal and identify it

Specific factors affecting detection

• Visibility: What might obscure the target?
• Obtrusiveness: What characteristics of the target might give it more or less contrast with its environment?
• Crew members' skill, experience, motivation, tiredness, morale, etc.
• Accessibililty
• Method of Detection
• Time available to survey a given area
• Search geometries (e.g., transect shape, spacing, and orientation)
  • E.g., equilateral triangular arrangement of auger holes more efficient than square arrangement
• Search unit size, shape, and distribution
  • E.g., long, narrow units are more likely to intersect sites than square or circular units

How do Archaeologists find sites?

• Making use of existing information (maps, historical accounts, characteristics of past archaeological finds, etc.)
• Regional prospection by visual inspection on the ground or under water
• Subsurface testing with augers, test pits, shovel testing
• Aerial reconnaissance
• Satellite imagery
• Geophysical remote sensing (e.g., magnetometry for shipwrecks)
• Predictive modelling

• Good prospection strategies take advantage of any information we may have that could help us maximize the likelihood that we will find the sites or artifacts we seek
  • Predictive modelling based on past experience (sometimes with GIS)
  • Search patterns centred on location with highest probability of containing what we seek
  • Search patterns and methods designed so as to minimize the chance of overlooking the "target"
  • Allocate search effort to different spaces in the way that is likely to find the target most quickly, or minimize the chance of missing targets on a fixed budget

How do Archaeologists detect buried structures or artifacts?

• Surface survey
• Inspection of gullies, road cuts, construction sites, eroded surfaces
• Search for chemical anomalies (typically phosphate)
• Geophysical remote sensing
  • Magnetic
  • Electromagnetic
  • Resistivity
  • Seismic
  • Ground-penetrating Radar
• Test pits, augering, shovel-testing
• Aerial or satellite observation
  • Crop marks, etc.

• Sampling's purpose is NOT as a prospection technique (even though prospection searches can involve randomness)
• Sampling's purpose is to select a subset (sample) of a large or unobservable set (population) that is representative of that larger set
  • By "representative" we mean that the characteristics (statistics) of the sample should be closely similar to the characteristics (parameters) of the population (e.g., characteristic of average site size or proportion of pottery that has painted decoration)
• Sampling surveys use probability theory as the basis for selecting areas to search for sites
  • Random samples of polygons (usually quadrats) or transects
  • Systematic samples of points, polygons or, most typically, transects
  • Stratified samples that try to account for the fact that sites or artifacts may be distributed differently or with varying characteristics in different parts of the region or site to be surveyed
  • Combinations, including the popular (but not always justified) systematic, stratified, unaligned sample design
• Cluster Sampling
  • Can create difficulties when archaeological distributions arfe spatially patterned
    • "spatial autocorrelation" - if sites close together are more likely to be similar or identical in important characteristics than ones far apart, then the cluster sample will result in biased estimates of the population characteristics
  • Happens whenever analytical unit (e.g., site) is not the same as the sampling element (e.g., quadrat or transect)
• Sampling elements don't have to be square or rectangular or arranged on a grid
  • Rectangular quadrats are popular, but can be arranged along streams or shores, and not only on a grid
  • Other geometrical units (polygons, circles) are better in some contexts
  • Non-geometrical spaces are perfectly legitimate elements of a population
    • landscape elements (e.g., hilltops, stream banks, rock outcrops)
    • agricultural fields
    • caves

 Nor is sampling adequate for detecting pattern in the way artifacts and architecture within sites, or sites within regions, are distributed. Some kinds of sample would even systematically overlook this kind of pattern

For example, archaeologists might want to know:

• Are buildings on a site organized on a street grid?
• Are towns and villages distributed along a road or canal network?
• Are artifacts scattered across a field clustered and, if so, at what scale?

A random sample of little squares, for example, could miss most of the buildings or villages, making it impossible to see if they are arranged on a grid or along road networks.

• How effective was the survey?
• Did it accomplish its goals well?
• How confident can we be of its results?
• Do gaps in the map mean a real absence of targets?
  • "Exhaustion maps" to show how thoroughly the region was surveyed
• For sampling surveys, are the estimates of population parameters unbiased?
• To use survey results with confidence, ideally we should know detectors' detection functions (how efficient they are at finding things)
  • Resurvey
  • Or experiments on fields "salted" with artifacts in known locations

• Survey needs to be tailored to its goals
• Prospection is purposive, not haphazard or unscientific, survey
• We can optimize the recovery of target sites in various environments
• Sampling is effective for making generalizations, but still needs careful design with regard to units, spacing, arrangement
• Other surveys need extensive, contiguous observation over an area to detect spatial structure
  • Landscape archaeology, non-site survey, etc.
    
• All surveys need to consider the factors that affect detection of targets
  • Visibility, obtrusiveness, accessibility, budget, crew experience, etc.
• Evaluation is a necessary last step if we to have any confidence in a survey's results

Banning, E. B.
2002a Archaeological Survey. Manuals in Archaeological Method, Theory, and Technique 1. Kluwer Academic/Plenum Publishers, New York.

2002b Archaeological survey as optimal search. In Archaeological Informatics: Pushing the Envelope. CAA 2001 Computer Applications and Quantitative Methods in Archaeology, Proceedings of the 29th Conference, Gotland, April 2001, G. Burenhult and J. Arvidsson, eds. BAR International Series S1016. Archaeopress, Oxford: 341-50.

2001 Archaeological survey in Jordan. In The Archaeology of Jordan, B. MacDonald, R. Adams, and P. Bienkowski, eds. Sheffield Academic Press, Sheffield: 685-92.

1996 Highlands and lowlands: Problems and survey frameworks for rural archaeology in the Near East. Bulletin of the American Schools of Oriental Research 301: 25-45.

Bintliff, J., P. Howard, and A. Snodgrass
1999 The Hidden Landscape of Prehistoric Greece. Journal of Mediterranean Archaeology 12:139-68.

Carreté, J.-M., S. Keay, and M. Millett
1995 A Roman Provincial Capital and its Hinterland: The Survey of the Territory of Tarragona, Spain, 1985-1990. JRS Supplementary Series 15. Journal of Roman Archaeology, Ann Arbor, MI.

Cherry, J. F., J. L. Davis, and E. Mantzourani
1991 Landscape Archaeology as Long-Term History: Northern Keos in the Cycladic Islands. Monumenta Archaeologica 16. Institute of Archaeology, UCLA, Los Angeles.

Collins, James M., and Brian Molyneaux
2003 Archaeological Survey. The Archaeologist's Toolkit 2. Altamira Press.

Davidson, Thomas E.
1986 Computer-correcting historical maps for archaeological use.
Historical Archaeology 20:27-37.

Hawkins, Alicia, S. T. Stewart, and E. B. Banning
2003 Interobserver bias in enumerated data from archaeological survey. Journal of Archaeological Science 30: 1503-1512.

Ebert, J. I.
1992 Distributional Archaeology. University of New Mexico Press, Albuquerque.

Gaffney, V. L., J. Bintliff, and B. Slapsak
1991 Site Formation Processes and the Hvar Survey Project, Yugoslavia. In Interpreting Artefact Scatters: Contributions to Ploughzone Archaeology, edited by A. J. Schofield, pp. 79-77. Oxbows, Oxford.

Given, M.
2004 Mapping and Manuring: Can We Compare Sherd Density Figures? In Side-by-Side Survey: Comparative Regional Studies in the Mediterranean World, edited by S. E. Alcock and J. F. Cherry, pp. 13-21. Oxbow Books, Oxford.

Hirth, K. G.
1978 Problems in Data Recovery and Measurement in Settlement Archaeology. Journal of Field Archaeology 5: 125-131.

Koopman, B. O.
1980 Search and Screening: General Principles with Historical Applications. Pergamum Press, New York.

Lock, Gary, T. Bell, and J. Lloyd
1999 Towards a Methodology for Modeling Surface Survey Data: The Sangro Valley Project. In Geographical Information Systems and Landscape Archaeology, edited by M. Gillings, D. Mattingly, and J. van Dalen, pp. 55-63. Oxbow, Oxford.

Mee, C., and H. Forbes
1997 A Rough and Rocky Place. The Landscape and Settlement History of the Methana Pewninsula, Greece: Results of the Methana Survey Project. Liverpool University Press, Liverpool.

Plog, S., F. Plog, and W. Wait
1978 Decision-making in Modern Surveys. Advances in Archaeological Method and Theory 1: 384-421.

Schiffer, M. B., A. P. Sullivan, and T. C. Klinger
1978 The Design of Archaeological Surveys. World Archaeology 10: 1-28.

Shennan, S. J.
1985 Experiments in the Collection and Analysis of Archaeological Survey Data: The East Hampshire Survey. Academic Press, New York.

Schon, R.
2000 On a Site and Out of Sight: Where Have Our Data Gone? Journal of Mediterranean Archaeology 13:107-111.

2002 Seeding the Landscape: Experimental Contributions to Regional Survey Methodology. Unpublished PhD diss., Bryn Mawr College.

Shott, M. J.
1995 Reliability of Archaeological Records on Cultivated Surfaces: A Michigan Case Study. Journal of Field Archaeology 22:475-490.

Stone, L. D.
1975 Theory of Optimal Search. Academic Press, New York.

Terrenato, N.
2000 The Visibility of Sites and the Interpretation of Field Survey Results: Towards an Analysis of Incomplete Distributions. In Extracting Meaning from Ploughsoil Assemblages, edited by R. Francovich, H. Patterson and G. Barker, pp. 60-71. Oxbow Books, Oxford.

2004 Sample Size Matters! The Paradox of Global Trends and Local Surveys. In Side-by-Side Survey: Comparative Regional Studies in the Mediterranean World. edited by S. E. Alcock and J. F. Cherry, pp. 36-48. Oxbow Books, Oxford.

Terrenato, N. and A. J. Ammerman
1996 Visibility and Site Recovery in the Cecina Valley Survey, Italy. Journal of Field Archaeology 23: 91-109.

Thompson, Stephen
2004 Side-by-Side and Back-to-Front: Exploring Intra-Regional Latitudinal and Longitudinal Comparability in Survey Data. Three Case Studies from Metaponto, Southern Italy. In Side-by-Side Survey: Comparative Regional Studies in the Mediterranean World, edited by S. E. Alcock and J. F. Cherry, pp. 65-85. Oxbow Books, Oxford.

Wandsnider, LuAnn, and E. L. Camilli
1992 The Character of Surface Archaeological Deposits and its Influence on Survey Accuracy. Journal of Field Archaeology 19:169-188.

Wandsnider, LuAnn, and J. I. Ebert
1986 Accuracy in Archaeological Surface Survey in the Seedskadee Project Area, Southwestern Wyoming. In The Seedskadee Project: Remote Sensing in Non-Site Archaeology, edited by D. L. Drager and A. K. Ireland, pp. 211-226. Southwest Region, National Park Service, Albuquerque, NM, and Upper Colorado Region, Bureau of Reclamation, Salt Lake City, UT.