Why it matters
Without AI, unanticipated anomalies can disrupt layouts, required sampling may be uneven, and blank areas may be overlooked, increasing risk to safety, missing contexts, and delays in reporting and compliance.
From design to safety
Design
AI informs trench geometry and sampling
The AI reads geophysics raster data, interpretations, and WSI constraints to propose trial trench footprints, sampling percentages, closure criteria, and context-recording requirements for single context methodology.
- Trial trench plan sheet
- Geophysics anomaly sensor array
- Sample percentage gauge
- Blank area demarcation flags
Build
Automated layout guidance during trenching
The system imports real-time survey marks, adjusts trench lines, flags unexcavated zones, updating the trench layout boards, pull-points, and stake logs, while maintaining versioned change records for QA.
- Trench layout stakes and boards
- Trial trench plan sheet
- Geophysics anomaly sensor array
- Blank area demarcation flags
Operation
Live validation of trench progress and records
Interpreted geophysics outputs sustain continuous checks against context sheets, enabling field staff to refine trench progression, sampling bands, and post-excavation notes in the moment, before reporting to HER.
- Trial trench plan sheet
- Geophysics anomaly sensor array
- Sample percentage gauge
- Trench layout stakes and boards
Maintenance
Data hygiene and archive synchronization
Periodic calibration of sensors, re-validation of WSI guidance, and synchronization of context sheets with the HER and post-excavation records, ensuring audit trails, archive deposition notes, and data integrity.
- Geophysics anomaly sensor array
- Written scheme of investigation binder
- Sample percentage gauge
- Trial trench plan sheet
Safety
Safety-aligned trench decisions and documentation
AI flags potential conflicts between trench lines and utilities, consolidates safety checks, and generates planning-compliance notes for sign-off by the qualified person.
- Trench layout stakes and boards
- Geophysics anomaly sensor array
- Blank area demarcation flags
- Written scheme of investigation binder
Part by part
- 1
Trial trench plan sheet
- Watch for
- Misalignment between proposed plan and geophysical targets, risking overruns or missed contexts.
- AI action
- Reads geophysics targets and updates plan pages; prompts review by field supervisor.
- 2
Geophysics anomaly sensor array
- Watch for
- Drift in anomaly coordinates or sensor drift affecting target validity.
- AI action
- Monitors anomaly coordinates; triggers recalibration and plan adjustment.
- 3
Sample percentage gauge
- Watch for
- Inconsistent sample fractions across trenches leading to biased results.
- AI action
- Calculates allocation and flags deviations for sampling officers.
- 4
Blank area demarcation flags
- Watch for
- Unmarked areas where trenching should not occur or where returns are needed.
- AI action
- Generates demarcation prompts and updates site maps.
- 5
Trench layout stakes and boards
- Watch for
- Shifted or displaced layout markers compromising trench alignment.
- AI action
- Tracks marker positions; issues corrective placement notices.
- 6
Written scheme of investigation binder
- Watch for
- Missing or outdated WSI sections that govern trench methods and reporting.
- AI action
- Alerts team to required revisions and stores revision history.
Worked example
Evaluation trench layout design for a 4.2 ha development-led site
Illustrative scenario with sample data.
Geophysics survey indicated multiple anomalies; AI-augmented trench layout guided initial trenches and sampling plan before excavation across a 4.2 hectare site with late Holocene deposits.
Signal in
siteArea_ha=4.2; survey_date=2026-04-12; anomalies=6; sampling_target=5%; WSI_version=3.1; terrain=loamy_sand; moisture=moderate; depth_model=shallower
System out
proposed_trench_footprints=[A1,A2,A3,B1,B2]; sampling_alloc=[A1:10%,B1:8%,A2:6%,A3:9%,B2:7%]; blank_zones=[C1,C2]; revised_footprints=[A1,A2,A4,B1]
Human sign-off: Project manager approves the final trench layout and sampling plan, ensuring alignment with HER, WSI, and client heritage conditions; curator reviews high-contrast anomalies before trenching begins.
How to roll it out
Phase 1
Weeks 1 to 2
Ingest baseline geophysics, align WSI, seed initial trench footprints and sampling bands, and log initial context sheets.
Phase 2
Weeks 3 to 5
Refine layouts with field checks, adjust blank areas, update context sheets, trench boards, and change-control records.
Phase 3
Weeks 6 onward
Consolidate trench progression data, archive outputs, and prepare post-excavation assessment inputs for final HER submission.
Where people stay in charge
- Decisions altering trench footprint or sampling fractions must be reviewed by a qualified person and recorded in HER.
- Critical safety choices, utility conflicts, and UXO considerations require sign-off by the site safety supervisor or planning archaeologist.
- WSI revisions impacting legal compliance must be approved by the client’s heritage consultant and the curator.
Questions owners and crews ask
What is the role of geophysics in trench layout design for archaeologists?
Geophysics identifies subsurface patterns; AI translates these into trench footprints, sampling bands, and blank areas, ensuring efficient layouts that respect WSI and statutory requirements.
How does AI handle sampling percentages across multiple trenches?
AI analyzes anomaly density, trench length, and context density to allocate sampling percentages evenly, flagging deviations for supervisory review.
What data sources feed the AI during design and build?
Geophysical rasters, anomaly coordinates, terrain models, WSI constraints, previous context sheets, and ongoing progress measurements feed the system.
Who approves changes to the trench layout produced by AI?
The project manager, site supervisor, and planning archaeologist consult and sign off; the curator or HER officer may review changes as required.