Why it matters
Without AI, misreads of patterns can lead to missed finds, uneven coverage, and unreported surface signals. Repeated survey work risks delays and unsafe trench decisions when services are present.
From design to safety
Design
AI aids survey design decisions
AI reviews existing records, predicts density and pattern likelihood from fieldwalking grids and GPS logs, informs written scheme of investigation and trench layouts, flags potential bias in sampling frames, and cross-checks against the HER and WSI constraints.
- fieldwalking grids
- GPS logs
- plough soil
Build
Guided reconnaissance during survey
AI flags priority transects, adjusts grid spacing, and schedules surface finds sieving with real-time density indicators and feedback to archaeologists, integrating photo capture and context capture routines, and notifying supervisors of deviations.
- fieldwalking grids
- surface finds
- GPS logs
- plough soil
Operation
Real-time pattern monitoring
AI aggregates metal detector signals with finds density and GPS traces to refine trench plans and recording in the field, exporting towards context sheets and Harris matrix updates for daily client reporting.
- metal detector signals
- GPS logs
- surface finds
- finds density
Maintenance
Pattern data maintenance
AI updates density models, re-calibrates detector thresholds, and synchronizes finds records with post-excavation datasets, uploading to HER deposition logs and archive-ready context sheets, while logging equipment calibration events.
- finds density
- surface finds
- GPS logs
Safety
Safety-focused pattern alerts
AI monitors buried services indicators, sensor readings, and plough soil residuals to trigger warnings and safe work orders, routing notifications to welfare, access control, and supervisor sign‑off, with audit trails.
- plough soil
- surface finds
- GPS logs
Part by part
- 1
fieldwalking grids
- Watch for
- Inconsistent spacing or gaps in grid coverage that reduce representativeness of surface patterns.
- AI action
- Detect coverage gaps and trigger re-survey or alt transect assignments.
- 2
surface finds
- Watch for
- Unlogged surface items or clusters that imply underlying contexts are underrepresented.
- AI action
- Flag clusters for immediate context recording and photographed documentation.
- 3
metal detector signals
- Watch for
- Anomalous or repetitive signals suggesting sub-surface anomalies or modern debris.
- AI action
- Flag signal clusters; assign rapid impact assessment trenching if warranted.
- 4
GPS logs
- Watch for
- Gaps or anomalies in GPS trace data that compromise plan accuracy.
- AI action
- Annotate log gaps and prompt re-survey with corrected coordinates.
- 5
finds density
- Watch for
- Uneven density estimates that could bias interpretation of site formation.
- AI action
- Normalize density surfaces and adjust context sampling plan.
- 6
plough soil
- Watch for
- Soil disturbance patterns indicating recent ploughing affecting surface context.
- AI action
- Warn on potential dilution of surface patterns; reroute fieldwalking.
Worked example
Survey Pattern Checks for a 3.2 ha Development Site
Illustrative scenario with sample data.
During early survey, AI integrates fieldwalking grid data with GPS and metal detector signals to optimize pattern coverage before trenching.
Signal in
siteAreaHa: 3.2; periodsFound: Early Bronze Age; gridSpacingm: 5; detectorThreshold: 0.8; gpsBounds: [(lat1,lon1),(lat2,lon2)]; density: 12 finds/ha; weather: dry; surveyors: 4 teams
System out
AI suggests increasing grid density along planned trench lines, issues a surface-find cluster alert, and generates a revised written scheme of investigation, plus an adjusted transect map for the manifest.
Human sign-off: Project manager reviews the AI-generated plan and approves the revised trench layouts and sampling strategy, with HER liaison confirmation.
How to roll it out
Phase 1
Weeks 1 to 2
Baseline data collection, system integration with fieldwork records and GIS, calibration against known benchmarks.
Phase 2
Weeks 3 to 6
Iterative refinement of survey patterns, density modelling, and protocol updates for field teams.
Phase 3
Weeks 7 onward
Operational monitoring, routine maintenance, and archival deposition of survey outputs.
Where people stay in charge
- All safety-critical decisions remain with the site safety officer or competent person under the written risk assessment.
- AI outputs may inform but not replace archaeologist oversight for Written Scheme of Investigation changes.
- The project manager must retain final authority over allocation of fieldwalking resources and trenching plans.
Questions owners and crews ask
What data streams feed AI in fieldwalking surveys?
AI ingests fieldwalking grids, surface finds records, GPS logs, metal detector signals, finds density, and plough soil data to identify patterns and guide sampling.
Who signs off on AI-generated survey changes?
The project manager and planning archaeologist sign off after review of rationale, risk, and conformity with the written scheme of investigation.
How does the system handle buried services risk?
AI correlates detector signals with known service maps and GPS traces to flag high-risk zones before trenching begins.
What records are produced for post-excavation archiving?
Context sheets, density models, survey plans, and metallic signal logs are compiled for HER deposition and archive submission.