Calibrating and using a Neutron Moisture Meter to determine water use by urban trees

Author(s): Grant, Cameron

Publication Information

YearPublished: 2010

Journal: Treenet

Country of Origin: Australia

Is this peer reviewed?:Unknown

Resource

Dr Cameron Grant CPSS3, University of Adelaide

Some background

Neutron moisture meters, NMMs, measure the amount of water in a given volume of soil. The volume of water divided by the total volume of soil is called the volumetric water content, VWC, which we express in various ways, such as m3 of water per m3 of soil, or mm of water per m-depth of soil. Measuring the VWC of the soil at different depths over time allows you to calculate how much water a tree uses during the growing season. You can also combine information about the VWC with other information to calculate the total amount of water a soil can store.

Some theory

Neutron moisture meters operate by sending out ‘fast’ neutrons, 0h1, into the soil and then measuring the number of ‘slow’ neutrons that ‘bounce back’. ‘Fast’ neutrons are slowed down only by large atoms such as H, and water of course is largely composed of this (i.e. H2O).This means that the higher the count of slow neutrons, the more water there is in the soil.

The count of slow neutrons, called the count rate, needs to be normalised for each NMM because the radioactive source in each NMM produces a slightly different quantity of ‘fast’ neutrons depending upon the amount and age of the radioactive sources in each machine (eg. 95Am243 & 4Be9). Normalisation also needs to occur for different soils because soils contain different amounts of H and other large atoms that are not related to the amount of H2O in the soil (e.g. H in organic matter, Cl from salt, B naturally occurring in some subsoils – these do not vary with the soil water content). The normalised count rate is obtained by dividing the count rate measured at each point in the soil by a standard count rate measured in a large body of water (such as an oil drum full of water) – the ratio of these two readings is called the relative count rate, RCR. There is a linear relationship between RCR and the amount of water in the soil, the VWC, according to:

VWC = a x RCR + b

where ‘a’ and ‘b’ are different constants for each soil type and probe. If the values of ‘a’ and ‘b’ are determined carefully and precisely, this equation can produce very accurate estimates of the amount of water in a soil.

Some practicalities

To obtain good values of ‘a’ and ‘b’ for your soil and NMM, you need to collect soil samples when the soil is wet and when it is dry to measure the VWC at the same times and location as the measured count rates.

The count rates in the soil are easy to obtain (see demonstration) and the standard count rate in a large drum of water needs to be measured only once in a while to check the stability of the decaying radioactive source (95Am243 has a ½-life in the order of hundreds of years). I will provide you with an average standard count rate for the NMM used today.

Obtaining good soil samples to calibrate the probe is trickier than you might think, and requires that you collect undisturbed soil samples to measure how much water they contain. Collecting undisturbed soil samples takes some skill and patience because most methods for collecting soil can disturb it considerably and provide useless information.

Once you extract a ‘good’ soil sample you divide it into two parts:

Weigh one part carefully to obtain its moist and dry weights (dry it in an oven at 105 C overnight) – this information is used to calculate the gravimetric water content, GWC (see table) and to correct the weight of the other part of the soil samp
Tie a piece of thread around the other part so that it can be suspended from the thread by hand. Weigh it accurately without losing any bits of soil. You now need to determine the volume of this piece of soil, and this is done by dipping it briefly in molten wax to seal it, then weighing it in air and again in water (Archimedes’ Principle) – see table and demonstration.

Location:

Gravimetric water content (g g-1)

Bulk density (g cm-3)

Soil horizon

Depth (cm)

Pot

Weight of pot (g)

Weight of pot + moist soil (g)

Weight of pot + OD soil (g)

Weight of soil in thread (g)

Weight of soil in thread in wax (g)

Weight of

soil in thread in wax in water (g)

Calculate the bulk density, BD, of the sample as follows:

Wmoist = moist weight of soil sample, g

Wwax = weight of wax coating the soil sample, g

W”water = the apparent increase in the weight of the water in the beaker experienced when the soil core + wax is immersed in it, g

Dwax = 0.84 g/cm3 and Dwater = 1 g/cm3

Use the BD and the GWC of each sample to calculate its VWC as follows: VWC = GWC x BD/Dwater

Repeat this procedure with several samples of soil at different water contents at different times of the year. This will allow you to produce a calibration for your probe and your soil as shown in the example in the following page…..

Some advantages and disadvantages of Neutron Moisture Meters

Advantages	Potential disadvantages
Based on solid principals of physics and therefore highly accurate once the neutron probe is calibrated. If something happens to your neutron probe (e.g. electronics breakdown) and you have to use a different probe, you can easily correct the calibration equations by taking a new reading in a barrel of water.	Not useful for detecting sharp wetting fronts because the NMM takes a reading from a large volume of soil, so any sharp changes in water content can be lost through averaging. However, if you are monitoring water use over long periods of time, you are probably not interested in detecting such things.
Neutron access tubes can be left in the ground for many years so you can come back to the exact same location to measure the water content without disturbing the soil.	Because the NMM contains a small (sealed) radioactive source, the EPS requires you to obtain a licence to own and operate NMMs, and special precautions are required when using and transporting the instrument; this requires some time and money, but it’s worth it.
The measurements do not take much time – a few seconds.	Calibration is required in most soils, particularly if the soil contains a lot of organic matter, salt or boron.
The sample volume is relatively large (larger than for any other single instrument), and therefore more representative of the true soil water content than measurements taken by other instruments.	NMMs are relatively expensive but can be purchased second hand for < $10,000 and they last a long time (I’m still using a second hand instrument that is 25 years old, and I expect it will still produce neutron in 500 years time….).

Calculating how much water your soil can store

Once you know the bulk density, BD, of the soil down the profile of the soil your trees are growing in, you can estimate the maximum amount of water the soil can hold. Let’s say, for example, that you measure the bulk density of the top 100 mm of the soil and find it to be 1.5 g/cm3 (or 1500 kg/m3). You can reasonably assume for most soils that the true density of the soil particles, PD, is ~2.6 g/cm3 (or 2600 kg/m3), which allows you to calculate the pore space ratio, which tells you how much of the total volume of your soil is made up of empty pore-space that can be filled with water:

This means, for this example, that 42.3% of the total soil volume in the top 100 mm is ‘pore space’, which means you can add 100 mm x 0.423 = 42.3 mm of water in the top 100 mm of soil. You can repeat this procedure for the whole soil profile and sum the amounts to estimate the absolute maximum amount of water your soil profile will hold. An example for a soil having 4 horizons and which contains tree roots down to 1 m depth is shown in the table below.

Soil Horizon	Depth (mm)	BD (g/cm3) (measured)	PD (g/cm3) (assumed)	PSR (cm3/cm3) (calculated)	Maximum amount of water (mm)
A1 (sandy)	0 to 100	1.45	2.65	0.453	45.3
A2 (loamy)	100 to 500	1.58	2.65	0.404	161.5
B (heavy clay)	500 to 750	1.82	2.65	0.313	78.3
C (carbonates)	750 to 1000	1.63	2.65	0.385	96.2
Total					381.3

This depth of water can be converted to a total volume of water simply by multiplying the total amount of water (381.3mm or 0.3813 m) by the area in which you wish to store water (e.g. the area between street trees). For example, if the distance between the bitumen road and the concrete footpath in front of a house is 3 m and the distance between the trees (or the distance between driveways) is 20 m, the soil profile should be able to take in approximately 0.3813 m x 3 m x 20 m = 22.9 m3.

If, for example, you assume 25 mm rain falls during a typical summer thunderstorm in Adelaide and that all the water from a house of say 500 m2 roof-area is diverted onto the street, this amounts to 12.5 m3 of water in front of every house on the street. If all this water were somehow re-directed onto the verge around each street tree, it would take only two storms like this to completely saturate the soil profile down to 1m.

Considerations

Diverting all runoff/street water onto tree verges is a good idea if 1) all the water can infiltrate the soil surface and re-distribute itself down the soil profile quickly, and 2) the water is used by actively transpiring trees during summer. Good soil management on the verges could achieve this.

During winter, however, the soil would quickly become saturated under the scenario outlined above because the area of the verge on which we divert the runoff water is considerably smaller than the area of the roofs from which the water is collected. Once the soil is saturated, the rate of water infiltration will decline and all additional water would run off the verges back onto the street again. Furthermore, most street trees are not well suited to saturated soil conditions, so any technology designed to divert street water onto verges would need to be flexible so that saturated conditions do not last very long.