RMR Parameters and RQD Formula Explained

In Geotechnical Engineering

RQD is a core-quality index; RMR rates rock mass quality. Learn the RQD formula, RMR parameters and how the two measures differ.
Feb. 27, 2024
Mary Janine L. Kamenić Mary Janine L. Kamenić
Julianna Widlund P.E Julianna Widlund P.E
Stevan Lukic CEng Stevan Lukic CEng

RMR Parameters and RQD Formula Explained, over a photo of rock drill core laid out in core boxes at a drilling site.

In geotechnical engineering and rock mechanics, assessing the quality and stability of rock masses is crucial for tunneling, mining, dam construction, foundations on rock and slope stability analysis. Two of the most widely used tools are the Rock Quality Designation (RQD) and the Rock Mass Rating (RMR).

They are related but not interchangeable. RQD is a simple index of how intact the drill core is. RMR is a broader rock mass classification, and RQD is one of its inputs. This article explains the RQD formula, the RMR parameters and how the two measures differ, for estimators, contractors and civil or geotechnical readers working from borehole logs and site investigation reports.

Quick Summary

TopicKey Point
RQDA core-quality index: the percentage of a core run made up of sound pieces at least 100 mm long.
RQD denominatorThe full core-run length, not just the length of core recovered.
RMRA rock mass classification that sums weighted ratings for intact rock strength, RQD, discontinuity spacing, discontinuity condition and groundwater, then adjusts for discontinuity orientation.
How they relateRQD is one input to RMR. It is not an alternative to it.

RQD Formula and Calculation

RQD was introduced by D.U. Deere in the 1960s to give a quick, quantitative measure of rock quality from diamond drill core. It is calculated for each core run:

RQD (%) = (sum of qualifying sound core lengths / total core-run length) × 100

  • Qualifying pieces — under the usual convention, only sound (intact, not highly weathered) core pieces at least 100 mm (4 in) long count. Lengths are measured along the core centerline.
  • Total core-run length — the full length drilled in that run. Core that was lost or not recovered stays in the denominator and counts against RQD.

Illustrative 1500 mm core run: sound pieces of 100 mm or more count toward RQD, short pieces, natural-fracture fragments and core loss do not, and a drilling break is refitted. RQD = 950 / 1500 x 100 = 63%, which feeds into RMR.

Which pieces count in RQD, using the illustrative example below. Conceptual, not to scale.

RQD is not total core recovery

Total core recovery (TCR) is the length of core recovered divided by the core-run length. It counts every recovered piece, however small. RQD only counts sound pieces of 100 mm or more, so it is always equal to or lower than TCR. Do not divide by the recovered length, and do not count every recovered piece: both would overstate rock quality.

Natural breaks vs drilling-induced breaks

RQD is meant to reflect natural fractures in the rock mass, such as joints, bedding planes and shears. Breaks caused by drilling or handling are usually fresh and rough, and the pieces on either side should be fitted back together and counted as one piece. Where it is unclear whether a break is natural, a common conservative practice is to treat it as natural and note the judgement on the log.

Worked example (illustrative only)

This example is made up to show the arithmetic. It is not a real project log.

  • Core run — 1500 mm drilled. 1200 mm of core recovered; 300 mm lost.
  • Recovered pieces — 250 mm, 80 mm, 220 mm (one piece broken into 90 mm and 130 mm by a fresh drilling break), 60 mm and 40 mm fragments between natural fractures, 300 mm, 180 mm and 70 mm.
  • Qualifying pieces — 250 + 220 + 300 + 180 = 950 mm. The 80 mm, 60 mm, 40 mm and 70 mm pieces are under 100 mm and do not count. The 220 mm piece counts in full because the drilling break is refitted.
  • RQD — (950 / 1500) × 100 = 63% (63.3%).
  • TCR for comparison — (1200 / 1500) × 100 = 80%.

Two common mistakes show why the rules matter. Dividing by the recovered length gives 950 / 1200 = 79%, which hides the lost core. Treating the drilling break as natural leaves only the 130 mm part counting, so the run is understated at 860 / 1500 = 57%.

Things that affect RQD

  • Borehole direction — RQD depends on the angle between the borehole and the joint sets. A hole drilled parallel to a dominant joint set may barely cross it and return a high RQD, while a hole drilled across it records the joints at closer to their true spacing. RQD does not record joint orientation as a parameter in its own right, so note the hole direction and drill at different angles where one joint set dominates.
  • Core size and drilling quality — centerline measurement is intended to make RQD largely independent of core diameter. A smaller core diameter does not in itself mean the rock has more natural fractures, but small core and poor drilling technique can cause more mechanical breakage and core loss, which can lower RQD if it is not recognized. Record the core size on the log.
  • Core recovery — poor recovery lowers RQD because lost core stays in the denominator. If recovery is poor, check whether the cause is weak or broken ground or the drilling method.
  • Timing of logging — some rocks, such as shales and mudstones, can break up as they dry or are handled, so RQD is best logged soon after the core is taken out of the barrel.

RQD is a useful screening tool for spotting poor-quality zones in borehole data, but used alone it does not describe a rock mass. It says nothing about joint condition, joint orientation, groundwater or the strength of the rock. That is where RMR comes in. For more on reading investigation data, see our article on the importance of geotechnical reports.

RMR Parameters and How the Rating Works

The Rock Mass Rating, also called the Geomechanics Classification, was developed by Z.T. Bieniawski, first published in 1973 and refined several times since. The ratings assigned to each parameter changed between versions, so a project should state which version it uses. The description below follows the widely used 1989 version (Bieniawski, 1989).

The rock mass is first divided into structural regions of broadly similar geology, and each region is rated separately. RMR then combines five basic parameters, each with its own weighted rating, plus an adjustment for the orientation of discontinuities:

  1. Strength of intact rock — the uniaxial compressive strength (UCS) of intact rock specimens, or the point-load strength index. This is the strength of the rock material between the joints, measured on core or lump samples. It is not the strength of the jointed rock mass, which is usually much lower. Stronger intact rock earns a higher rating.
  2. RQD — the Rock Quality Designation from drill core, calculated as above. Higher RQD earns a higher rating.
  3. Spacing of discontinuities — how far apart the joints and other discontinuities are. Widely spaced discontinuities earn a higher rating than closely spaced ones.
  4. Condition of discontinuities — the character of the joint surfaces: persistence, aperture (separation), roughness, infilling and weathering or alteration. Tight, rough, unweathered joints earn a higher rating than open, smooth, clay-filled or weathered ones.
  5. Groundwater conditions — the water conditions in the rock mass, from completely dry through damp and wet to dripping and flowing. Drier conditions earn a higher rating.

Orientation adjustment: the strike and dip of the main discontinuities relative to the proposed works can make them more or less favorable. RMR applies a deduction for unfavorable orientations, and the size of that deduction depends on whether the project is a tunnel, a foundation or a slope.

The five ratings are not each scored out of 100. They carry different weights. In the 1989 version, the maximum ratings are 15 for intact rock strength, 20 for RQD, 20 for discontinuity spacing, 30 for discontinuity condition and 15 for groundwater. Together these sum to a basic RMR of up to 100. The orientation adjustment is then applied, and the final RMR places the rock mass in one of five classes, from very good rock to very poor rock. The full rating tables are reproduced in Evert Hoek's Practical Rock Engineering chapter on rock mass classification and in this summary of the RMR system; use the published tables for the version your project specifies.

Two common mix-ups: the groundwater parameter in RMR has no separate index name. GSI stands for the Geological Strength Index, a separate rock mass characterization system developed by Hoek and colleagues; it is not part of RMR. Similarly, Jn is the joint set number in Barton's Q-system, another classification that also uses RQD. It is not an RMR parameter, and RMR does not use Q-system notation.

RMR vs RQD

RQD (Rock Quality Designation)RMR (Rock Mass Rating)
What it measuresHow intact the drill core is along a core runOverall rock mass quality for engineering
Typical inputsLengths of sound core pieces of 100 mm or more, and the core-run lengthIntact rock strength (UCS or point load), RQD, discontinuity spacing, discontinuity condition, groundwater, plus an orientation adjustment
OutputA percentage for each core runA total rating and a rock mass class
LimitationsDepends on borehole direction and drilling quality; ignores joint condition, orientation, groundwater and rock strengthNeeds more data and judgement; results depend on the version used and how the rock is divided into regions

Comparison of RQD (Rock Quality Designation) and RMR (Rock Mass Rating): what each measures, its inputs, output and limits, with an arrow showing that RQD is one input to RMR, not an alternative.

RQD describes the core. RMR describes the rock mass, using RQD as one of its inputs.

In practice, RQD is logged on almost every rock borehole, while RMR is assessed where a project needs a fuller classification, such as for tunnels, deep excavations, rock slopes or foundations on rock. When a report gives both, check that the RMR uses the same RQD values that appear on the logs.

The Bottom Line

RQD is the percentage of a core run made up of sound pieces at least 100 mm long, divided by the full run length, not the recovered length. RMR combines intact rock strength, RQD, discontinuity spacing and condition, and groundwater into a weighted rating, then adjusts for discontinuity orientation. Used together with good logging and engineering judgement, they give a consistent first picture of rock conditions. This is an overview, not a substitute for project-specific rock mechanics assessment.

Frequently Asked Questions

What is the RQD formula?

RQD (%) = (sum of the lengths of sound core pieces at least 100 mm long / total core-run length) × 100. Lost core stays in the denominator.

What is the difference between RQD and core recovery?

Core recovery counts every recovered piece. RQD only counts sound pieces of 100 mm or more, so it is equal to or lower than recovery for the same run.

What are the RMR parameters?

Intact rock strength (UCS or point-load index), RQD, discontinuity spacing, discontinuity condition and groundwater, plus an adjustment for discontinuity orientation.

Can RQD replace RMR?

No. RQD is one input to RMR. On its own it does not account for joint condition, orientation, groundwater or rock strength.

Skip the manual work on your next project

Civils.ai reads your PDF drawings and measures the quantities for you, reviewed by a person before it comes back.