Water Content and Water Suction

===============================

To get some idea of what this means, take a short garden hose and push it all under the surface of a bucket of water so it fills with water. Then put your finger over one end and lift it out of the water. When the end of the hose is one meter above the water in the bucket the suction on your finger is 10kPa, 2 meters is 20 kPa and so on. If your lungs are in good working order, then you can suck at about 30 kPa. Most plants like to get their water at about 30 kPa; any less and it is too wet. Most plants grow nicely at suctions up to about 200kPa (yes, that is the equivalent of sucking water up 20 metres of hose) and can survive at suctions up to 1000kPa provided the weather conditions are not too hot.

Soils 'hold onto' their water with different force depending mostly on how much clay they have. A loamy soil with 20% water has ample water for plant growth. The water can be extracted easily by plants, so is said to be at 'low suction level' - typically, 10kPa. The same 20% water content in a clay soil would not support good plant growth. It would appear quite dry and is said to be at a 'high suction level' - typically 500kPa or more. A sensor measuring 'water content' would return the same reading in both materials, relying on the user to interpret that reading in terms of plant water status.

A sensor measuring 'soil water suction' would return readings which are very different in these two materials - more closely reflecting the plant water status in the reading obtained. An array of 'water content' sensors would provide a direct measure of water in the soil profile and a relatively simple base for calculations in relation to irrigation water additions. An array of 'soil suction' sensors would have to be 'translated' via a soil water versus suction graph to calculate water appropriate irrigation additions.

For a particular soil, each suction level can be related to a water content, but for different soils, the relationship is different. If you pick a suction, say 10kPa, often called 'field capacity' and compare the water contents, the clay soils are holding much more water than the sandy ones. On this graph, a typical 'refill point' (time to turn on the water) is about 100kPa. Plants in soil which is allowed to reach 1000kPa will not produce very well and might die.

Fortunately, most growers only have one soil clay content to contend with and can rapidly become familiar with the relationship between sensor readings and the soil water requirement in relation to irrigation quantity. More recently, research has focussed on the direct use of water suction readings for regulated deficit irrigation control in vineyards, in order to improve wine quality.

Soil Water Estimation by Electrical Conductivity Methods

========================================================

This sensor will tell the user if the soil is wet or dry, and with careful calibration will give some idea of how wet the soil is. This very simple water sensor suffers from two disadvantages. The first is that the size of the pores (and electrical resistance) between the nails is dependent on the relative amount of sand and clay in the soil. The second is that the conductivity of the solution between the nails depends on the quality of the water in use and the salinity of the soil, which may also depend on how much leaching and evaporation of water has occurred. Any of these factors can cause substantial changes in the measurement and apparent (but undefined and often wrong) changes in soil water. Another problem can occur if the DC current used to measure the resistance is applied for more than a second. Gasses form at the nails, and the apparent resistance changes if the reading is repeated, or if one operator is slower than another.

The Gypsum Block

================

While the soil is wet, the pores in the gypsum are filled with water which dissolves some of the gypsum - enough to make a saturated solution of calcium sulphate, and the water conducts electrical current in a manner independent of soil water salinity, except at extreme levels of salinity (Richards and Campbell 1950). In wet conditions, the resistance would be quite small. As the soil around the block dries, water is 'pulled' from the block, the largest pores emptying of solution first then smaller and smaller pores as the soil suction increases. Electric current travelling between the electrodes in the block must travel a longer path through smaller pores so the resistance becomes greater. The gypsum supplies a fixed environment for the electrodes which is independent of the soil in which the block is installed. A gypsum block, made to consistent specifications, has a fixed porosity distribution, ie the size range of the pores is the same for each sensor. So whether the instrument is buried in clay or sand, the porosity around the electrodes remains the same as does the relationship between soil water suction and electrical resistance. Hence, a single calibration applies irrespective of the soil material in which the sensor is placed.

The same 'gassing' problem exists for the gypsum block as for the simple 'nail' sensor described above. An AC current bridge should be used or the blocks should be read rapidly by a modern logger (in milli-seconds) provided readings are not taken too frequently.

Obviously, an unmodified gypsum block can be constructed easily with little equipment, however, the material and spacing of the electrodes, and the consistency and setting time of the slurry of gypsum need to be controlled to get a consistent calibration. The blocks used in these tests have been produced in Australia for over 20 years. Similar blocks were first calibrated in 1951 (Aitchison and Butler, 1951) and the calibration was published at that time (see Figure 3).

In summary, the gypsum block measures soil suction and can provide a measure of plant stress provided the block is located in the plant root zone, and the plant has an adequate root system (ie is mature). The calibration of a gypsum block is independent of the material in which it is installed.

The gypsum block measures soil moisture tension in the 60 to 600kPa range, and is ideal for use in regulated deficit irrigation in vineyards with heavier soils, such as the Barossa Valley and Coonawarra districts in South Australia.

Installation of Gypsum Blocks

=============================

MEA systems uses gypsum blocks in groups of four, to obtain a picture of the soil moisture profile (Figures 1 and 2). Typical depths might be 20cm, 50cm, 70cm and 100cm. This covers surface moisture levels, moisture in the root zone of the vine, and moisture in the drainage zone below the vine.

Methods of installation vary, but can be carried out by vineyard staff, without the need to call in trained personnel. Installation is much less critical than with many other sensors. As long as the block is touching the soil around it, it will work. A good installation means the block will follow the soil suction in the surrounding soil minute by minute. A very poor installation means that the block might be a day or so behind the surrounding soil, but it still reaches the right answer in the end.

Some golden rules for installation. 

-----------------------------------

2) Make sure that there is at least 5 cm of soil between the sensor and any bentonite you are using.

3) Make sure that you don't place the sensor directly under a dripper.

4) Make sure that surface water cannot flow down the hole you dig to install the sensor or else your sensor will be giving some rather strange readings (see use of bentonite below).

MEA recommends the following installation method:-

1) Each of the four blocks is located in its own hole; this avoids the tedious business of replacing carefully preserved back fill necessary when four blocks are placed in a single hole. This also avoids 'leakage' of rain or irrigation water down the extra wires, which would create artificial moisture levels at the deeper blocks. Holes are located on the circumference of a small circle of about 15cm in diameter, to limit the spatial separation of the blocks. (The hole being centred under the dripper).

2) The block is prepared by removing its protective foil wrapping, and soaking it for 10 minutes in distilled water or rain water.

Block size is cylindrical, 23mm in diameter by 50mm long. Therefore, augering a hole 25mm to 100mm in diameter (depending on availability of augers) is sufficient. Put the soil from the last 150mm (6 inches) of the hole into a container and add water to make a thick slurry.

3) Pour the slurry to cover the sensor to a depth of about 150mm; sufficient to completely surround the block after installation. Pouring water down the hole and leaving it to soak may be an adequate alternative.

4) Double check the depth of the hole. Label the (above ground) end of the sensor wire with the depth and lower the gypsum block to the bottom of the hole . The still saturated block is pushed down into the slurry until submersed Add a little extra soil to force the slurry into intimate contact with the block.

5) Then make a mix of bentonite - 20%-30% mix of bentonite with sand (or local soil if it isn't too lumpy or stony) and backfill the hole and tamp it gently. Bentonite is used because it swells to 17 times its dry volume when wet and will stop surface water from flowing down through the loose material in the hole, giving strange readings on the sensor. If the sensor shows increased water content at 1 meter within minutes of you turning on the sprinkler then that is what is happening.

6) Stop the bentonite 20-30mm (around 1 inch) from the surfacece. Fill rest of the hole with the material removed from the hole in.

7) Installation of the blocks in the autumn preceding a winter's rainfall will allow the blocks time to "settle in" properly.

8) Once all four blocks are in place, 1cm of insulation is stripped from the end of each wire, to accommodate connection to either an MEA hand held gypsum block reader, or to an MEA113 Radio Field Station, to be described in the following section.

MEA113 Radio Field Station

==================================

MEA's engineers began developmental work in 1995 to make possible low cost loggable gypsum block soil water suction measurements. The primary aim of this first stage was to get more "blocks per buck" to the growers, to bring the cost of automated soil moisture measurements within reach of smaller vineyards.

A major cost saving would also be incorporation of gypsum block data back into existing MEA weather stations.

This integration of soil moisture and climatic records into a single central data logger, represents a considerable cost saving. Solar panel, charger, battery, logger enclosures and digital/analog/landline telemetry systems are all shared by the weather and soil monitoring systems.

This approach leads to 'one stop shop' collection of data from a vineyard.

Getting data back to a central logger without blowing out installation costs required some sleight-of-hand electronically. The problem involved getting good signals back over several kilometres of the lowest cost standard irrigation cable (seven core, 0.56mm2), and to cram data from eight blocks on two Field Stations onto these seven wires. This was achieved.

MEA's cable linked systems allowed up to 16 Gypsum Block Field Stations to be connected to a central MEA weather station or data logger, on up to eight standard irrigation cables, for a total of 64 gypsum blocks covering an area of several kilometres in radius around the station. The problems with protecting such cabled systems against lightning, rodents, animals and machinery soon became evident and led to MEA embarking on the development of a Radio system.

An MEA Radio Field Station is now installed at each monitoring site. The field stations communicate via license free radio back to an MEA111 Radio Base Station. Up to 38 field stations, each with up to 4 sensors, may be installed. The base station may be fitted with its own dedicated data logger or can feed into the logger of an MEA weather station, to provide integrated soil moisture and weather data.

Since releasing the MEA Radio system, we have found application for it in a wide range of other areas. One example is to bring back water temperature readings from a raft floating out on a reservoir. The use of the industry standard SDI-12 interface in this application allows large numbers of sensors to be connected via a common 3 core cable.

Gypsum Block Calibration

========================

Blocks are not individually calibrated however, as tests have shown that careful manufacturing techniques allow the blocks to be field interchangeable within their specified accuracy range.

Since the last calibration of the blocks occurred in 1951, MEA contracted CSIRO Land and Water (Cliff Hignett) to run calibration on 16 blocks selected at random from a manufactured batch. One of the 16 blocks failed in calibration; the remaining 15 showed that the calibration has shifted little in the intervening 45 years.

Of critical importance is the wide variability between individual gypsum blocks for soil moisture tensions below 60cB, which makes them unsuitable for use in sandy soils which are just about dry at this suction.

Figure 3 shows the new resistance versus suction curve for the Australian-made gypsum blocks. Note that the suction scale on the bottom axis is logarithmic, not linear.

Figure 3 (Not included here): Resistance Versus Suction for Australian made Gypsum Blocks (1951, 1997)

Figure 4 shows the response of the MEA2175 Gypsum Block Field Station, plotting output voltage versus suction. Note the increased sensitivity at the "wet end" between 60cB and 200cB, and the scale compression between 500cB and 1000cB. This compression allows coarse resolution logging in very dry soils as well up into the "wilting points" of vines grown on heavy clays.

Figure 4 (Not included here): Voltage Output Versus Suction for an MEA2175 Gypsum Block Field Station

Because the output of the MEA 2175 is near linear in the range 60cB to 400cB, the current loop signals are ideal for operating into an irrigation controller.

Field Trials

============

Installed systems were part of a CRC-Viticulture trial to formulate improved irrigation management strategies using the practise of regulated deficit irrigation. Test sites were in the Barossa Valley, Coonawarra, Padthaway and Griffith. Each of the 6 sites monitored 32 gypsum blocks.

MEA Automatic Weather Stations at each site collected the data, provided power and control to the gypsum block stations, and provided remote data access via analog cellular phone or Telecom landline.

A sample of seasonal data from one of the vineyards is given in Figure 5. Data shows that rainfall/irrigation events are not penetrating to the 70cm and 100cm depths.

Figure 5 (Not included here): Gypsum Block Soil Moisture Data from a South Australian Vineyard 1996/97

New Developments

===================

Access to this data on MEA systems has been available for several years through the cellular phone network (no cables!), by direct connection back to an office PC up to 500m away, by laptop computer, by logger or memory card collection, or by Telecom telephone landline. This has now been further expanded to include CDMA modems.

The latest addition to MEA's gypsum block family is the GBug, a small low cost continuous data logger. The battery powered GBug reads the blocks every 2 hours and can store 20 days worth of data. To collect the readings from the GBug, a companion unit called the MEA Retriever is used: simply hold the Retriever near the GBug, tap the GBug to wake it up and the readings stored in its memory will be sent out via a low power radio link into the Retriever. The display on the Retriever lets you review the data in the field. Better yet, take the Retriever back to the office and download the data into your PC, where it is displayed on the FREE GBug software.

By careful use of appropriate technology, MEA's engineers have given the faithful old gypsum block a new lease on life.

Contributors to this Article are:-

	========================

1) Andrew Skinner is the Engineering Director at MEA (Measurement Engineering Australia) in Adelaide. MEA acknowledges the support of Aus Industry in development of this product.

2) Cliff Hignett is a Certified Practicing Soil Scientist at CSIRO, Land and Water, Adelaide, South Australia.

3) Jan Dearden is a Research Officer (Viticulture) with the SA Research and Development Institute (SARDI, Viticulture) and manages the CRC-V funded Irrigation Improvement Project.

This article was published in "Australian Viticulture" October/November 1997

See also
"A Vine's Eye View: Soil Moisture Monitoring with Gypsum Blocks"
"The Australian Grapegrower & Winemaker" December 1997
Andrew Skinner, Engineering Director, Measurement Engineering Australia (MEA) sales@mea.com.au