Archaeological Survey Methods: Finding Sites Before You Dig
Excavation is the most visible part of archaeology, but it is also the most destructive and the most expensive. Every trowel stroke is irreversible. The profession has responded to this reality by investing heavily in survey methods — techniques that identify, map, and characterise sites and landscapes without removing the ground. Survey data shapes excavation priorities, provides landscape context that excavation alone cannot, and increasingly delivers interpretable results without digging at all. The toolkit available to modern surveyors spans everything from systematic surface collection to satellite imagery to geophysical instruments sensitive to buried features.
Pedestrian Field Survey
The oldest survey method and still among the most productive. A team of walkers spaced at regular intervals (typically 10–25 metres depending on visibility and site density) traverses a study area, recording all artefacts and features visible on the surface. The method depends on ground cover, season, and crop stage: ploughed fields in dry weather offer the best visibility; pasture and woodland suppress it.
Field survey across the Mediterranean has transformed our understanding of rural settlement density. The Boeotia Survey in Greece, the Biferno Valley project in Italy, and the Labraunda landscape survey in Turkey all found that ancient settlement patterns were far denser than previously recognised from excavation alone — ancient farmers lived dispersed across agricultural land rather than concentrated in urban centres. The lesson transfers globally: systematic pedestrian survey consistently reveals more sites than any prior estimate predicted.
Geophysical Survey
Geophysical instruments detect subsurface features by measuring contrasts in the physical properties of soil and buried material. The main methods are magnetometry, ground-penetrating radar, electrical resistance survey, and electromagnetic induction. Each responds to different soil conditions and different types of buried features; experienced surveyors typically combine methods for the most complete picture.
Magnetometry measures magnetic field variations caused by the higher magnetic susceptibility of heated soil (hearths, kilns, burnt structures) and disturbed fills. It is the fastest method over open ground and has produced spectacular results at sites including Portus (the harbour of imperial Rome) and Stonehenge's wider landscape. Ground-penetrating radar (GPR) sends radar pulses into the ground and records reflections from interfaces between materials of different dielectric properties; it is particularly effective for detecting stone walls, floors, and voids. At Stonehenge, the Stonehenge Hidden Landscapes Project used both magnetometry and GPR across 12 square kilometres and found dozens of previously unknown monuments.
Electrical resistance survey measures the resistance of soil to an electrical current; stone walls and dry rubble have higher resistance than damp soil fills, making the method useful for mapping building plans. It is slower than magnetometry but produces cleaner wall-plan data in many soil types.
Remote Sensing: Aerial Photography and Satellite Imagery
Crop marks — differential growth of crops over buried features caused by variations in soil moisture and nutrients above walls (negative marks) and ditches (positive marks) — have been photographed from aircraft since the 1920s. The English crop-mark record, compiled over decades by the Royal Commission on the Historical Monuments of England, is one of the most comprehensive site databases in the world. The method requires hot, dry summers in which moisture stress is most acute; climate change has both improved some conditions and disrupted the predictable seasonal windows.
Satellite imagery extends the method globally. Commercial satellites with sub-metre resolution now allow analysts to map crop marks, soil marks, and earthwork shadows across continents. Sarah Parcak's work using multispectral imagery across Egypt identified over 3,000 potential settlement sites. The conflict in Syria enabled a grim form of landscape survey: before-and-after satellite images document the scale of looting damage at sites including Apamea, where agricultural tractor marks from clandestine digging are visible across hundreds of hectares.
LiDAR (Light Detection and Ranging)
LiDAR uses laser pulses fired from aircraft or drones to generate extremely precise three-dimensional models of ground surface. When processed to filter out vegetation returns, LiDAR reveals the bare earth beneath forest canopy — a surface that aerial photography cannot see. The results have been transformative in forested regions. The 2010 Caracol survey in Belize used airborne LiDAR to map 177 square kilometres of Maya rainforest, identifying causeways, reservoirs, agricultural terracing, and over 70,000 individual structures in three days — survey that would have taken decades on foot. The 2018 PACUNAM LiDAR Initiative covered 2,100 square kilometres of the Maya lowlands in Guatemala, revealing a landscape far more densely populated and politically organised than the forest floor suggested. In Cambodia, LiDAR surveys of the Angkor region have identified an entire low-density urban network surrounding the monumental core, fundamentally revising understanding of the medieval Khmer city.
Underwater Survey
The majority of the world's coastline has been submerged since the end of the last glacial maximum as sea levels rose roughly 120 metres between 20,000 and 6,000 years ago. The submerged coastal and shelf zones contain vast areas of former habitation that are archaeologically unexplored. Survey methods include side-scan sonar (producing acoustic images of the seabed), sub-bottom profiling (penetrating below the seabed to detect buried features), and multibeam bathymetry (generating precise depth models). The discovery of submerged Mesolithic sites on the Dogger Bank — once a land bridge connecting Britain to continental Europe — and in the North Sea has opened an entirely new chapter in European prehistoric research.
Systematic Sampling and Landscape Archaeology
Modern survey projects do not attempt to cover every square metre. Instead, they use stratified random sampling — dividing the study area into zones defined by geology, land use, or topography, then sampling each zone systematically. Statistical analysis of the sample data allows site densities and distributions to be estimated for the entire survey area. The result is a probabilistic landscape model rather than a definitive site inventory, but it is an honest representation of what the data can and cannot support.
The integration of survey data with GIS (Geographic Information Systems) allows spatial analysis of site distributions relative to environmental variables: proximity to water, soil type, elevation, slope angle. Predictive modelling based on GIS analysis guides where to focus fieldwork and allows managers to assess the likely archaeological sensitivity of areas threatened by development.
Non-Invasive Excavation Precursors
Before breaking ground, professional archaeologies now typically complete a site evaluation sequence: desktop study of maps, aerial photographs, and existing records; walkover survey; geophysics. Only when these stages indicate that excavation is warranted — or legally required before development — is the ground opened. This workflow reduces the proportion of excavated sites where nothing is found, focuses resources on the most significant contexts, and provides interpretive context for whatever the excavation reveals.