Measuring Hydraulic Conductivity

SESL Australia

Understanding soil drainage rates, is vital for groundwater control during construction and sports field projects. With several methods of estimating hydraulic conductivity available this article explores the various methods of measurement.

permeability of a soil is the ability of water to move through it (permeate it). It depends on the physical and chemical properties of the soil, notably particle size distribution (the range of particle sizes present), pore space, pore size and pore size distribution (continuity of the pore spaces.

The formal name is hydraulic conductivity, and the most common way it is measured is when the soil is saturated with water or “Saturated hydraulic conductivity”. The amount of water that will flow through the soil in a given time is directly proportional on the cross-sectional area (the bigger the area, the more water flows through), the hydraulic “head” of pressure applied, and the amount of time (obviously the longer the time the more water will flow). It is also indirectly proportional to the amount of resistance the soil puts up to flowing water which is proportional to the length of the soil column through which the water flows. Thus the Darcy Equation for the flow of anything through any conductive medium can be applied –

Q (the amount of water flowing)  A.t.H/L

Q = Ksat.A.t.H/L

Where A = cross-sectional area
t= time
H = hydraulic head of pressure
L = length of the soil column
Ksat.= the proportionality constant.

Ksat is an intrinsic property of the soil and is used to calculate flow under different conditions of hydraulic head, surface area and soil depth (or length). Ksat is what we measure and report to the client.

There are two main types of flow situations that we want to know:
1. Surface infiltration rate (I). Also commonly called “permeability” this is the rate at which a soil surface will take in irrigation or rainfall without ponding occurring. It is the number an irrigation farmer or sports field manager wants to know so that he/she does not irrigate at a rate faster then soil can take it. Typically the surface infiltration rate starts off faster in dry soil and slows down as the soil swells, and the cracks close.
2. Drain spacing. Once we know Ksat, we can calculate, for a given rainfall event how much water will flow to drains, and thus, how many drains we need and at what spacing in order to cope with this amount of water.

Measuring Ksat.

Measuring Ksat is simple in principal, take a tube and fill it with soil, tape another tube on top of that and fill it with water and keep topping it up then using a measuring device and a stopwatch measure the volume of water that flows in a given time (see diagram). Since we know the cross-sectional area of the tube (by measuring the diameter), the length of the soil core and the height of the water table we have imposed, we can calculate Ksat.

However, measuring Ksat is fraught will all kinds of difficulties and causes of error. Let’s look at the different methods-

1. Field methods. These are various but usually involve driving a core or metal into the soil and imposing a water table inside of this. Alternatively, you can drive a corer in and then take the intact soil core back to the lab and do it there. The problem is that variability between samples even within a small area can be huge due to variation in pore spaces, cracking and structure. You usually need to do about 6 measurements and average them to be able to predict the overall permeability of a given area. Variability is less in constructed sand table playing fields than in natural soils, but some replication is always advisable as well as checking goalmouths and centre fields separately to outfields as they are more compacted.
2. Laboratory methods which can be divided into “intact cores” as mentioned above and “disturbed and repacked” samples.
3. Intact Cores. This is really the only way to estimate the true “in-situ” permeability of a soil. At least 6 cores around 75mm diameter and 150-200mm long are best. In uniform turf, sandy rootzones allow at least 3.
4. Repacked samples. Trying to re-pack disturbed bulk soils which are structured is pointless. For example, a structured clay which might have shown low permeability in-situ, will show the permeability of a gravel with the stable peds are packed into a core. The only time we do this is where we are measuring clays for liners in landfills and dams, and here we apply 99% compaction to the soil as would occur when rolled down as a landfill liner. The only things, in our experience, that can be measured reasonably reliably when repacked as completely structureless loamy sands and sands. Even so the results you get a very dependent on the degree of compaction you achieve.

Let’s look at the three methods we employ for measuring Ksat and compacting soils in disturbed lab samples-

1. AS 4419. The new 2017 revision has dropped the requirement for a permeability test. This was on our advice as the method, which did not employ and compaction at all, not even firming down, gave extremely poor reproducibility due to “slumping” which is where fine particles migrate and block the pores between the coarser particles. Very commonly we find better permeability when we give a degree of firming down to the soil.
2. McIntyre and Jacobsen methods. This method uses 6 plastic tubes which are loosely filled with soil in a damp condition and then the tubes are dropped from 15cm height 1, 2, 4, 8, 16 and 32 times to form a series of increasing compaction. Results are then prepared to show Ksat and density. Quite often the 2,4, and 8 drops will show higher permeability than the 1 drop due to slumping but as compaction increases to 16 and 32 drops density rises and Ksat drops. The 32 drop gives about 80% of full compaction and is representative of a moderately used playing field. Only loamy sands and sands will give adequate permeability at 32 drops.
3. USGA method. This compacts the soil using 14 blows from a 3 kg “hammer” and achieves about 95-99% compaction. It is indicative of a heavily used playing surface like a golf green and is also used for the high grade playing fields. No soil will pass the test, and only slightly loamy sands and sands will give adequate permeability under this compaction.

Prediction methods

a. Particle size analysis can be used to predict the hydraulic conductivity of unstructured sands and sandy loams. This relies on equations that take values measured in the lab. It is quite a good predictor for loamy sands on sports fields, for example. On the other hand, it is not suitable for well structured soils, especially clays, a structure can make a clay act more like a gravel.
b. Permeability class can be estimated from the texture and degree of structure. This relies on a look-up table and gives a range; for example, 2–20 mm/hour for a strongly structured clay loam but 1–5 mm/hour for a weakly structured clay loam.

2. Field measurements on intact soils
a. Single infiltration ring. A metal ring at least 30 cm across is driven about 5 cm into the soil surface, and water is poured into it. The time it takes for the water to soak in is timed. This method is adequate for measuring the surface infiltration rate of sandy and reasonably uniform soils.
b. Double ring infiltrometer. Two metal rings, a smaller one inside a larger one, are driven into the soil surface, and water is poured into both. The water in the outer ring blocks lateral (sideways) flow from the inner ring, so the water in the inner ring contributes solely to downward flow, which is what we measure. The method is suitable for soils that tend to swell when wet. It is no better on very sandy soils than the single ring method.
c. Well permeameter. A tubelike instrument is inserted into a hole drilled with an auger, and the rate at which water flows into the underlying soil is measured. This tool is good for measuring subsurface drainage in field soil at depths of 100 to 1000 mm. It suffers from “smearing” of clay soils if making auger holes when it is too wet.

All of these methods can be configured to calculate the intrinsic Ksat of a soil.

Constant head: This is where the method employs a “Marriot” device which allows water to replenish the water in the infiltrometer tube at an unchanging head.

Falling head: If we measure the time it takes for the water in the infiltrometer to fall from its starting level to the level of the soil surface this is called a “falling head”. This can be taken into account in the maths calculation to still arrive at the intrinsic property Ksat.

Practical implications

Comparing field and laboratory methods can be very challenging. Here is some basic advice-

Soil type Method notes
Intact structured soils, loam and heavier
Must use intact cores taken from the field.
Disturbed soils of poor structure that are to stripped and reused or landscape soils for commercial landscaping Use McIntyre & Jacobsen HC6 package
Sandy loams for passive recreation and low use turf areas Use McIntyre & Jacobsen 8, 16 and 32 drop compaction level (HC3).
Loamy sands and sand rootzones for higher end playing fields and for biofiltration uses Use McIntyre & Jacobsen 8, 16 and 32 drops compaction level (HC3) and USGA to see how the soil behaves under moderate and heavy compaction and use.