Site Investigation

Learning Objectives

  • Explain how site investigation supports project decisions and reduces subsurface uncertainty.
  • Describe the staged use of desk study, reconnaissance, geophysics, and direct exploration.
  • Select sampling and in-situ methods for the properties and materials of interest.
  • Interpret SPT corrections and distinguish field values from derived correlations.
  • Distinguish total core recovery, solid core recovery, and RQD.

Purpose and Scope

Site investigation

The planned collection, testing, and interpretation of information about subsurface materials, groundwater, and geologic structure for defined design, construction, and risk decisions.

Investigation Is Decision-Driven

An exploration program should identify the ground conditions that affect the proposed work and obtain information appropriate to the decisions being made. The required coverage, locations, depths, methods, and monitoring depend on project geometry, ground variability, consequences, existing evidence, and applicable project criteria. Uniform spacing or one test type is not a universal program.

Stages and Spatial Coverage

Progressive Investigation Stages

  • Desk study: review geological and topographic mapping, historical imagery, prior logs, land use, and available hazard information.
  • Reconnaissance: inspect terrain, exposures, drainage, distress, access, and evidence that may support or contradict the preliminary ground model.
  • Exploration: select test pits, borings, sampling, in-situ tests, geophysics, and groundwater observations to address specific uncertainties.
  • Interpretation: combine field and laboratory results into a ground model, distinguishing observations from interpolation and inference.
  • Supplementation: add targeted investigation or monitoring when results do not support the stated project decisions.
Plan and Section for Exploration Coverage

Original synthetic plan and section, without a survey scale. Borehole observations constrain dashed inferred contacts; the geophysical line is a separate investigation path. Actual spacing and depth follow project decisions.

Two boreholes on section line A–A′ project to matching section positions; a separate geophysical transect provides indirect coverage
Staged Site Investigation Planning

Plan exploration from project decisions and an initial ground model, then iterate until the site-specific information supports design and construction decisions. No fixed spacing or universal data-sufficiency threshold is implied.

Use direct exploration, in-situ tests, geophysics, sampling, laboratory work, and groundwater monitoring as complementary tools selected for the uncertainties that matter. Record observed versus inferred conditions and retain unresolved assumptions as explicit holds.

Staged Site Investigation PlanningPlan exploration from project decisions and an initial ground model, then iterate until the site-specific information supports design and construction decisions. No fixed spacing or universal data-sufficiency threshold is implied.. Use direct exploration, in-situ tests, geophysics, sampling, laboratory work, and groundwater monitoring as complementary tools selected for the uncertainties that matter. Record observed versus inferred conditions and retain unresolved assumptions as explicit holds.. Start → Define project decisions and required subsurface data; Define project decisions and required subsurface data → Review maps, records, prior logs, and site history; Review maps, records, prior logs, and site history → Reconnoiter terrain, exposures, drainage, and distress; Reconnoiter terrain, exposures, drainage, and distress → Build preliminary ground model and uncertainty register; Build preliminary ground model and uncertainty register → Study scope and criteria sufficiently defined?; Study scope and criteria sufficiently defined? — Yes → Select locations, methods, samples, and monitoring; Study scope and criteria sufficiently defined? — No → Hold: resolve missing scope or criteria; Select locations, methods, samples, and monitoring → Explore, test, sample, and log with quality controls; Explore, test, sample, and log with quality controls → Test samples and update the interpreted ground model; Test samples and update the interpreted ground model → Information supports the stated decisions?; Information supports the stated decisions? — Yes → Report findings, limits, and construction responses; Information supports the stated decisions? — No → Add targeted work; update model and recheck criteria; Add targeted work; update model and recheck criteria → Study scope and criteria sufficiently defined?; Report findings, limits, and construction responses → End

Start → Define project decisions and required subsurface data; Define project decisions and required subsurface data → Review maps, records, prior logs, and site history; Review maps, records, prior logs, and site history → Reconnoiter terrain, exposures, drainage, and distress; Reconnoiter terrain, exposures, drainage, and distress → Build preliminary ground model and uncertainty register; Build preliminary ground model and uncertainty register → Study scope and criteria sufficiently defined?; Study scope and criteria sufficiently defined? — Yes → Select locations, methods, samples, and monitoring; Study scope and criteria sufficiently defined? — No → Hold: resolve missing scope or criteria; Select locations, methods, samples, and monitoring → Explore, test, sample, and log with quality controls; Explore, test, sample, and log with quality controls → Test samples and update the interpreted ground model; Test samples and update the interpreted ground model → Information supports the stated decisions?; Information supports the stated decisions? — Yes → Report findings, limits, and construction responses; Information supports the stated decisions? — No → Add targeted work; update model and recheck criteria; Add targeted work; update model and recheck criteria → Study scope and criteria sufficiently defined?; Report findings, limits, and construction responses → End

  • Start: terminator
  • Define project decisions and required subsurface data: process
  • Review maps, records, prior logs, and site history: process
  • Reconnoiter terrain, exposures, drainage, and distress: process
  • Build preliminary ground model and uncertainty register: process
  • Study scope and criteria sufficiently defined?: decision
  • Hold: resolve missing scope or criteria: terminator
  • Select locations, methods, samples, and monitoring: process
  • Explore, test, sample, and log with quality controls: process
  • Test samples and update the interpreted ground model: process
  • Information supports the stated decisions?: decision
  • Add targeted work; update model and recheck criteria: process
  • Report findings, limits, and construction responses: document
  • End: terminator

Exploration, Sampling, and In-Situ Testing

Direct Exploration and Samples

Test Pits, Borings, and Rock Core

Test pits expose shallow soil and structure over a wider face but are limited by access, excavation support, groundwater, and safe reach. Borings provide deeper samples and in-situ test intervals. Rotary coring recovers rock for lithology, discontinuity, weathering, recovery, and RQD logging. Method selection and excavation precautions are project-specific.

Disturbed and Lower-Disturbance Samples

Split-barrel or auger samples are commonly disturbed and can support identification, classification, and selected index tests. Thin-wall tubes are intended to preserve soft cohesive-soil structure better for selected strength or compressibility tests, but no sample is perfectly undisturbed. Sampler geometry, material suitability, recovery, sealing, transport, and laboratory preparation all affect test quality.

Sample Disturbance and Test Purpose

The sketches compare idealized fabric preservation; select the sampler and laboratory test for the soil and property of interest.

Split-barrel sampling disrupts soil fabric more than a properly collected thin-wall tube specimen

Standard Penetration Test

SPT Procedure and Field N-Value

In the commonly described procedure, a 63.5 kg hammer with a 0.76 m drop drives a split-barrel sampler in three successive 150 mm increments. The first increment is seating; the field NN-value sums blows for increments two and three. Applicable test standards govern equipment, execution, recording, and handling of partial penetration or refusal.

SPT Seating and Recorded Increments

The labeled hammer and increment dimensions follow the cited FHWA guidance; execution and corrections must follow the governing project procedure.

The first 150 millimeters seat the sampler and the next two 150 millimeter increments define the field N-value

Energy-Corrected SPT Resistance

Normalizes a recorded field N-value to a reference energy ratio and selected equipment corrections.

N60=Nfield(ER60%)CBCRCSN_{60} = N_{\text{field}} \left(\frac{ER}{60\%}\right) C_B C_R C_S

Variables

SymbolDescriptionUnit
NfieldN_{\text{field}}Field blow count for the specified penetration intervalblows/300 mm
ERERMeasured hammer energy ratio, expressed as a percent%
CBC_BBorehole-diameter correction factordimensionless
CRC_RRod-length correction factordimensionless
CSC_SSampler correction factor under the adopted proceduredimensionless

SPT Values Are Not Direct Strength Measurements

N60N_{60} corrects a field measurement for selected equipment and energy effects. An overburden-normalized value is often written (N1)60=CNN60(N_1)_{60}=C_N N_{60}, where CNC_N comes from an adopted correlation. Correlations, limits, and applicability depend on soil type, stress state, procedure, and project basis; these indices are not direct measurements of strength or a substitute for appropriate calibration.

Keep the Record Before Applying a Correlation

Record the blows and actual penetration for each increment, together with depths, sampler, rods, borehole and hammer-energy information. A partial last increment such as 50 blows for 50 mm is not an ordinary 300 mm field NN; do not extrapolate it to 300 blows or silently add it to a complete increment. Stopping and refusal criteria come from the governing procedure. High resistance may reflect a boulder, cemented soil or other obstruction as well as rock.

Use ER=75%ER=75\% as 75/6075/60 in the percentage convention above, or 0.75/0.600.75/0.60 in a fractional convention; do not mix them. Equipment factors must match the adopted correlation and energy-measurement arrangement. Corrected indices can be fractional even though measured blows are whole counts. Zero blows with penetration need their actual self-weight/hammer observations recorded; a small positive NN is not a weight-of-hammer record.

SPT record and correction

Concept and model scope

Separate the seating count, field N and adopted energy/equipment corrections. Preserve an incomplete record without inventing a full N.

Model and limits: Synthetic counts and supplied factors, not a test-execution guide or a strength correlation. Blows: 0–50 each; measured ER: 30–100%; CB: 0.8–1.2, CR: 0.5–1, CS: 0.8–1.3. Factor ranges are illustrative inputs, not recommended project values. Governing procedures control stopping and recording.

Controls

Penetration record

Concept and model scope

Ordinary field N requires the complete final 300 mm interval after seating. The partial example retains actual third-increment penetration and holds ordinary N and N60. Changing this selector does not identify the material or prescribe a stopping criterion.

8 blows

Seating blows

First 150 mm is seating. Retain its count but exclude it from the ordinary field N.

Range: 0–50 blows. Step: 1 blows.

12 blows

Increment 2 blows

Count for this recorded increment. A partial interval retains its actual penetration and does not define a complete N.

Range: 0–50 blows. Step: 1 blows.

18 blows

Increment 3 blows

Count for this recorded increment. A partial interval retains its actual penetration and does not define a complete N.

Range: 0–50 blows. Step: 1 blows.

75 %

Measured energy ratio

Measured transferred energy as a percentage of theoretical hammer energy. ER/60 is the correction, not 60/ER; a hammer name does not establish its energy ratio.

Range: 30–100 %. Step: 1 %.

1.00

Borehole factor CB

Supplied dimensionless borehole factor from the adopted interpretation basis; not selected automatically from a generic table.

Range: 0.8–1.2 . Step: 0.05 .

0.85

Rod factor CR

Supplied dimensionless rod factor. Confirm compatibility with the energy measurement and adopted correlation to avoid double correction.

Range: 0.5–1 . Step: 0.05 .

1.00

Sampler factor CS

Supplied dimensionless sampler factor under the adopted basis. Preserve sampler and liner metadata.

Range: 0.8–1.3 . Step: 0.05 .

Complete illustrative record: N excludes seating. Corrected resistance is an index, not direct strength or a foundation decision.

Recorded blows per increment, with actual penetration.
1 · seating8 blows / 150 mm
Increment 1: 8 blows for 150 mmAll bars share a zero to fifty blow axis. The reference outline remains at zero blows. Incomplete penetration is not extrapolated.
2 · recorded12 blows / 150 mm
Increment 2: 12 blows for 150 mmAll bars share a zero to fifty blow axis. The reference outline remains at zero blows. Incomplete penetration is not extrapolated.
3 · recorded18 blows / 150 mm
Increment 3: 18 blows for 150 mmAll bars share a zero to fifty blow axis. The reference outline remains at zero blows. Incomplete penetration is not extrapolated.
01020304050

Common axis 0–50 blows. Counts and penetration are separate observations.

Field N = N2 + N3, seating excluded
30 blows/300 mm
N60 = N × ER/60 × CB × CR × CS
31.875 blows/300 mm
Combined correction factor
1.0625
Recorded total penetration
450 mm

Other In-Situ Methods and Groundwater Monitoring

Method Selection by Measurement

  • CPT/CPTu: continuous penetration resistance, sleeve friction, and, for CPTu, pore pressure; interpretation uses soil-behavior correlations and local calibration.
  • Vane shear: torque response in suitable soft cohesive soils, interpreted with attention to anisotropy, sensitivity, rate, and disturbance.
  • Piezometers or observation wells: hydraulic head at a screened interval. Installation, seals, development, equilibration, and timing determine whether readings represent the intended unit.
  • Geophysics: rapid, indirect coverage that can help interpolate between direct points; it does not uniquely identify material and should be checked against suitable ground truth.

Rock-Core Recovery and RQD

Core Logging

Record lithology, weathering, fracture orientation and condition, core run length, recovery, and relevant drilling observations. Distinguish what was directly observed from what was inferred between core runs or exploration locations.

Total Core Recovery (TCR)

The total recovered length of core and other recovered material divided by the drilled core-run length, expressed as a percentage.

Solid Core Recovery (SCR)

The recovered length of full-diameter solid core divided by the drilled core-run length, expressed as a percentage.

Rock Quality Designation (RQD)

The sum of centreline lengths of qualifying sound, intact core pieces at least 100 mm long, divided by the drilled core-run length, expressed as a percentage under the adopted logging procedure.

Core Recovery Indices

Use the same drilled interval for all recovered-length measures.

TCR=LrecoveredLrun×100%,SCR=Lfull diameterLrun×100%TCR=\frac{L_{\text{recovered}}}{L_{\text{run}}}\times100\%, \qquad SCR=\frac{L_{\text{full diameter}}}{L_{\text{run}}}\times100\%

Variables

SymbolDescriptionUnit
LrecoveredL_{\text{recovered}}Equivalent total recovered length, including recovered fragmentsmm
Lfull diameterL_{\text{full diameter}}Length logged as full-diameter solid core under the adopted proceduremm
LrunL_{\text{run}}Drilled interval, including unrecovered materialmm
TCR, SCR, and RQD Lengths

Original synthetic length account on one millimetre axis. Sound full-diameter pieces total 550 mm, an unsound full-diameter piece adds 150 mm, and fragments add 100 mm. Only 120, 250 and 100 mm pieces qualify for RQD. Packed strips do not locate the 200 mm loss.

An exact 1000 mm synthetic run has TCR 80%, SCR 70% and RQD 47%, including a sound piece exactly 100 mm long

Rock Quality Designation

Calculates the proportion of the core run represented by qualifying intact pieces.

RQD=∑Lqualifying pieces ≥100 mmLcore run×100%RQD = \frac{\sum L_{\text{qualifying pieces }\ge 100\text{ mm}}}{L_{\text{core run}}} \times 100\%

Variables

SymbolDescriptionUnit
Lqualifying pieces ≥100 mmL_{\text{qualifying pieces }\ge 100\text{ mm}}Centreline length of each sound intact piece meeting the adopted 100 mm criterionmm
Lcore runL_{\text{core run}}Length drilled during that core runmm

Conventional RQD Description Bands

The commonly reproduced Deere-style bands share written endpoints. For unambiguous exercises and the calculator here, use the unrounded value: 0≤RQD<250\le RQD<25 very poor; 25≤RQD<5025\le RQD<50 poor; 50≤RQD<7550\le RQD<75 fair; 75≤RQD<9075\le RQD<90 good; 90≤RQD≤10090\le RQD\le100 excellent. Thus 74.667% is fair even if a whole-percent display rounds to 75%. State the convention when reporting a boundary value.

These are descriptive index classes, not acceptance limits or stand-alone design ratings. RQD does not provide discontinuity orientation, infill, persistence, intact strength, groundwater pressures or a complete rock-mass design model.

Validate Recovery and Break Evidence

Check the run depths, units and recovered total before publishing any index. Recovered length greater than drilled length is an inconsistent record to reconcile, not a result to cap at 100%. Missing recovery has an unknown position and cause unless supported by drilling evidence; it does not automatically prove natural fracturing.

Fit and log pieces broken by documented drilling or handling as one original piece where justified. A fresh-looking face alone does not prove a mechanical break. Retain natural discontinuities and uncertainty in the log. Soundness and the adopted full-diameter qualification must be assessed; a long friable piece is not automatically RQD core.

Core recovery and RQD

Concept and model scope

Change a recovered length or logged condition and compare the three different numerators.

Model and limits: Synthetic packed-length accounting in centimetres, not a depth log. Sound full-diameter pieces at least 10 cm qualify. Document and rejoin confirmed mechanical breaks before entry. RQD classes describe an index, not strength or design acceptance. Run: 100–300 cm; up to 12 entries, each 0–50 cm.

Controls
100 cm

Drilled run

The denominator is the drilled interval, including unrecovered material. Excess recovery holds all indices for record reconciliation.

Range: 100–300 cm. Step: 10 cm.

12 cm

Entry length

Equivalent recovered centreline length. Exactly 10 cm qualifies only when sound and full diameter; zero length paints no recovery.

Range: 0–50 cm. Step: 1 cm.

Logged condition

Concept and model scope

Sound full-diameter core enters SCR and may enter RQD if at least 10 cm long. Unsound full-diameter core enters SCR but not RQD. Fragments or partial-diameter material enter TCR only in this teaching account.

The condition is supplied from logging evidence; the calculator does not diagnose soundness or break origin.

Valid length account. Missing material has no inferred position or fracture condition.

Drilled interval: 100 cm

Drilled interval on the common centimetre axisThe full drilled strip remains a reference even at zero recovery. The axis maximum is stated below the packed recovery.

Packed recovery: 80 cm

Packed recovered entries on the common centimetre axisEntries share the drilled strip scale. They are packed for length accounting, not located at observed depths. Use the entry ledger for lengths and logged conditions.Entry 1: 12 cm, Sound, full diameterEntry 2: 8 cm, Sound, full diameterEntry 3: 25 cm, Sound, full diameterEntry 4: 10 cm, Sound, full diameterEntry 5: 15 cm, Unsound, full diameterEntry 6: 10 cm, Fragments / partial diameter
0255075100
Common axis 0–100 cm. Packed lengths are not depths.
TCR = 80/100 × 100
80.00%
SCR = 70/100 × 100
70.00%
RQD = 47/100 × 100
47.00%
Unrounded descriptive class
Poor · no design conclusion
Recovered entry ledger
  1. Entry 1: 12 cm · Sound, full diameter · RQD included
  2. Entry 2: 8 cm · Sound, full diameter · RQD excluded
  3. Entry 3: 25 cm · Sound, full diameter · RQD included
  4. Entry 4: 10 cm · Sound, full diameter · RQD included
  5. Entry 5: 15 cm · Unsound, full diameter · RQD excluded
  6. Entry 6: 10 cm · Fragments / partial diameter · RQD excluded

Combining Core Runs

A combined index is weighted by drilled lengths, not by the number of runs.

RQDcombined=∑iLqualifying,i∑iLrun,i×100%RQD_{\text{combined}}=\frac{\sum_i L_{\text{qualifying},i}}{\sum_i L_{\text{run},i}}\times100\%

Variables

SymbolDescriptionUnit
Lqualifying,iL_{\text{qualifying},i}Qualifying sound length in run imm
Lrun,iL_{\text{run},i}Drilled length in run imm

Reporting Differences and Remaining Uncertainty

Subtracting two percentages gives a difference in percentage points. A relative percentage error instead divides that difference by the reference percentage before multiplying by 100. Always name the denominator.

A final investigation report should preserve coordinates and datum, dates, logs, sample quality and methods, correction assumptions, observed water levels and screened intervals, uncertainty between exploration points, and the decisions the evidence can support. One water reading is not a seasonal maximum. If an anomaly or important layer is unresolved, specify targeted follow-up and a decision hold instead of inventing a certain contact or foundation solution.

Key Takeaways
  • Plan investigation around project decisions and revise it when evidence leaves material uncertainty unresolved.
  • Direct exploration, geophysics, in-situ testing, sampling, and monitoring provide complementary evidence.
  • N60N_{60} and (N1)60(N_1)_{60} are corrected indices whose correlations require an applicable basis.
  • TCR, SCR, and RQD count different recovered lengths; RQD includes qualifying pieces exactly 100 mm long.
  • Report observed data, interpretation, uncertainty, and project-specific limitations separately.

References