# Field Density of Soils

Measuring in-place density and moisture of soils and base with the nuclear gauge and sand cone, then percent compaction and the oversize correction.

- Canonical URL: https://pavementinteractive.org/reference-desk/testing/soil-tests/field-density-of-soils/
- Last updated: 2026-11-01
- Standards: AASHTO T 255, AASHTO T 191, AASHTO T 224, AASHTO T 265, AASHTO T 310, ASTM D1556, ASTM D4718, ASTM D6938

Compaction of embankment, subgrade and base is accepted by measuring the soil's density and water content in place and comparing the dry density with a laboratory maximum. This article covers the two field methods most agencies use, the calculation of percent compaction, and the correction needed when the field soil contains more coarse particles than the laboratory test allowed. The laboratory side is described in *Moisture–Density Relationship (Proctor Test)*.

## Percent Compaction

Field density results are reported as [percent compaction](https://pavementinteractive.org/glossary/percent-compaction/) (Caltrans calls it relative compaction):

$$P_c = 100\,\frac{\rho_{d,field}}{\rho_{d,max}}$$

where:

- $$P_c$$ = percent compaction
- $$\rho_{d,field}$$ = dry density measured in place
- $$\rho_{d,max}$$ = maximum dry density from the specified laboratory test, corrected for oversize where needed

The field dry density comes from the measured wet density and water content in the same way as in the laboratory:

$$\rho_{d,field} = \frac{\rho_{w,field}}{1 + w/100}$$

Specifications set the required percent compaction, the laboratory test it refers to (standard or modified Proctor, or an agency method such as California Test 216), and often a moisture range around optimum as well (Caltrans, 2023[^1]).

## Nuclear Gauge Method

Most field density tests on soils are run with a nuclear moisture–density gauge (AASHTO T 310, ASTM D6938). The gauge measures wet density with a gamma source and detector and estimates water content with a neutron source that responds to hydrogen. How the gauge works is described in [Nuclear Density Gauge](https://pavementinteractive.org/reference-desk/construction/compaction/nuclear-density-gauge/); the points below are the ones that matter for soils.

- **Direct transmission.** For soils and base, the source rod is normally lowered into a hole made with a drive pin, to a depth matching the lift being tested. The gauge then measures the material between the source and the detectors, which suits a lift better than a surface reading.
- **Site preparation.** The gauge needs a flat, smooth seat. Small voids under the gauge are filled with a thin layer of fine sand or native fines, and the hole must be straight so the rod contacts the soil.
- **Standard count.** A reference count on the gauge's standard block is taken each day and compared with recent counts to confirm the gauge is stable.
- **Moisture correction.** The neutron reading responds to all hydrogen, not only free water, so soils with organic matter, gypsum, or bound water in some clays read high. Agencies correlate gauge moisture with oven-dry water contents (AASHTO T 255 or T 265) and apply a correction for each soil type (Texas DOT, 2025[^2]).

Gauge operation requires a radioactive materials license, operator training, dosimetry and secure storage. Those requirements come from the Nuclear Regulatory Commission or the state that regulates it, not from the test method.

## Sand Cone Method

The sand cone (AASHTO T 191, ASTM D1556) measures the volume of a hole directly. The technician digs a small hole through a base plate, saving every bit of the excavated soil, then fills the hole from a jar of uniform, dry sand whose density has been calibrated. The mass of sand that ran into the hole, divided by the sand's density, gives the hole's volume. The excavated soil's wet mass divided by that volume is the wet density, and an oven-dry sample of the same soil gives the water content.

The sand cone needs no license and no correlation, so it serves as a check on nuclear gauges and as the method where gauges aren't allowed. It is slower, and it is sensitive to vibration from nearby equipment (which settles the sand and overstates the volume), to loose soil falling into the hole, and to damp calibration sand.

## Correcting for Oversize Particles

A Proctor test is run only on the material passing one screen, the 4.75 mm (No. 4) or 19.0 mm (¾ inch). The field soil includes everything. Rock particles are denser than the compacted soil around them, so a field sample with more rock than the laboratory sample will show a higher density and may appear to pass even when the soil between the rocks is undercompacted. AASHTO T 224 makes an [oversize correction](https://pavementinteractive.org/glossary/oversize-correction/) to the laboratory maximum density for the oversize fraction actually found at the test site (Christopher et al., 2006[^3]):

$$\rho_{d,corr} = \frac{100\,\rho_f\,k}{\rho_f\,P_c + k\,P_f}$$

where:

- $$\rho_{d,corr}$$ = corrected maximum dry density of the total material
- $$\rho_f$$ = maximum dry density of the fine fraction, from the Proctor test
- $$P_c$$ = percent of oversize particles, by dry mass
- $$P_f$$ = percent of fine particles, by dry mass (100 − P_c)
- $$k$$ = density of the oversize particles: the bulk specific gravity of the oversize times 1,000 kg/m³ (62.4 lb/ft³)

The optimum moisture content is corrected in the same proportions, weighting the fine fraction's optimum and the oversize particles' absorbed moisture by their percentages:

$$w_{corr} = \frac{w_f\,P_f + w_c\,P_c}{100}$$

The oversize fraction is found by screening the material from the test location (for a nuclear test, a sample dug at the site; for a sand cone, the excavated soil). The bulk specific gravity of the oversize can be measured or taken from a known source value (Texas DOT, 2025[^2]).

The correction is meant for soils with a modest oversize content. As the oversize approaches a large share of the sample, the rock particles start to touch each other, and the equation, which assumes they float in the soil, no longer applies. Above that limit, agencies switch to a test on the total material or to a method specification.

Calculator: Field density and percent compaction (soils) — https://pavementinteractive.org/tools/soil-field-density/

## Reading the Results

- **Low percent compaction, wet soil.** If the soil is well above optimum, rolling may not help; it may pump or rut instead. The usual remedy is to dry the lift by discing or aeration, or to remove and replace it.
- **Low percent compaction, dry soil.** Adding water and re-rolling usually brings it up.
- **Very high percent compaction (over 100).** This can be genuine, especially on granular material under heavy rollers, but it should trigger a check that the right Proctor reference was used and that oversize was accounted for.
- **Moisture disagreement.** If the gauge moisture and an oven-dry sample disagree by more than the agency allows, rely on the oven-dry value and revisit the gauge correction.

## Standard Test Methods

- AASHTO T 310, ASTM D6938: In-place density and moisture content of soil and soil–aggregate by nuclear methods (shallow depth)
- AASHTO T 191, ASTM D1556: Density of soil in place by the sand-cone method
- AASHTO T 224, ASTM D4718: Correction for coarse particles in the soil compaction test
- AASHTO T 255 and T 265: Moisture content of aggregate and soil by oven drying

## Sources

[^1]: California Department of Transportation (Caltrans) (2023). Method of Test for Relative Compaction of Untreated and Treated Soils and Aggregates (California Test 216). Transportation Laboratory, California Department of Transportation. Sacramento, CA. https://dot.ca.gov/-/media/dot-media/programs/engineering/documents/californiatestmethods-ctm/ct-216-dec2023-a11y.pdf
[^2]: Texas Department of Transportation (TxDOT) (2025). Measuring In-Place Density and Moisture Content of Soils and Base Materials (Tex-115-E). Materials and Tests Division, Texas Department of Transportation. Austin, TX. https://ftp.dot.state.tx.us/pub/txdot-info/cst/TMS/100-E_series/pdfs/soi115.pdf
[^3]: Christopher, B.R.; Schwartz, C. and Boudreau, R. (2006). Geotechnical Aspects of Pavements Reference Manual. FHWA-NHI-05-037. National Highway Institute, Federal Highway Administration. Washington, D.C. https://www.fhwa.dot.gov/engineering/geotech/pubs/05037/05037.pdf