proudly sponsored by
|
archiving of SOWACS proudly sponsored by |
|
"Resurrecting Gypsum Blocks for Soil Moisture Measurement" Sender: owner-sowacs@aqua.ccwr.ac.za Precedence: bulk **************************************************************************** This is a reply from Andrew Skinner at Measurement Engineering Australia (MEA) re the "negative" response of this user to the temperature coefficient of gypsum blocks. **************************************************************************** [I have resent this to sowacs as it appears not to have been distributed yet. Bruce] This respondent has raised a good point; there is a temperature coefficient to gypsum blocks - as there is for most sensors. As we (at MEA) are generally using the blocks over the 0.6Bar to 6Bar range (60kPa to 600kPa), the range swing equivalent to "0 to 10Bars for a 20degC temperature variation" quoted by the respondent seems to be out by about two orders of magnitude. [1kPa = 1centiBar = 1/100Bar, so 1Bar = 100kPa] But putting that aside for the moment as a typo... Firstly, the field experience:- What temperature range are the blocks experiencing at the trial sites????? (Remember blocks were at 4 depths, 20cm, 50cm, 70cm and 100cm.) Regretably, we didn't log temperature at these sites, but we do have data from elsewhere. Diurnal data from Reeves Plan, north of Adelaide, South Australia (temperate climate zone) in mid-winter gave the following typical range of diurnal temperature changes:- 5 cm depth: 2 to 15 degC 10 cm depth: 7 to 11 degC 30 cm depth: 10 to 11 degC What we are seeing is that diurnal temperature variation decreases with depth, and may even change phase by 180 degrees due to thermal lags on a seasonal basis. The important thing to remember is that if there are substantial temperature changes (at ANY depth) then there will also be temperature gradients, and if there is any moisture in the soil then it will move RAPIDLY in response to these temperature gradients. Reliable measurement of either suction or soil moisture under such conditions is extremely difficult, partly because of the rapidity of changes, but also because the soil will be switching from the draining side of the moisture retention curve to the wetting side at least once a day. Under such conditions the normal assumption we all make of a non hysteretic moisture retention curve becomes invalid. As the blocks in vineyards are buried at or below 20cm, the expected temperature variation is probably below 5 degC per day, with gradual seasonal changes as the earth warms and cools. Soil suction responses by the gypsum blocks due to irrigation are on a much larger and more rapid scale than these slow background changes due to daily average temperature changes. Yet what of the daily temperature swings effecting the GB output?????? Perhaps this is the perfect argument for logging the gypsum blocks. There is some minor variations in tension during the day, as we can see from our logged records over the past year and a bit of field trials. Yet the daily averaging of these readings effectively filters out these variations. Some of this may be temperature coefficient but most of it will be a real change in soil suction as heat moves water up and down the soil profile. The majority of users would not have any interest in these short term changes. However, the natural corollary to this is that users of hand-held gypsum block readers should attempt to make their readings at the same time each day to avoid reading part of the diurnal fluctuation as a 'real' change in water content. The best time is early in the morning before the sun strikes the soil directly and when soil temperature (in near surface layers) is near the daily mean value. The second point coming out of this discussion is that caution must be exercised when using GBs close to the surface. Not only is the block affected by temperature, but temperature gradients in the soil (eg between exposed ground and shade) will also drive soil moisture movement due evaporations and re-condensation. These effects will be seen by the blocks. MEA doesn't have a recent temperature coefficient for the blocks, but the original paper (Aitchison, Butler and Gurr, 1950) covering the design of the Australian GBs we use, states the temperature dependence as R_c = R_o ^ (1 + 0.002 * (T_o - T_c)) where R_o is the observed block resistance R_c is the corrected block resistance T_o is the observed block temperature (degF) T_c is the corrected block temperature (degF) MEA will be running temperature coefficient tests on our gypsum blocks within the next 6 months; results can be published on SOWACS if anyone is interested. Copies of field data (showing little to no sign of temperature-induced variations) are available from MEA in the original paper. We can also make available some of the hourly data for those interested in the diurnal temperature effects referred to here Andrew Skinner, CPEng Engineering Director Measurement Engineering Australia (MEA) Magill South Australia ph (618) 8332 9044 fax (618) 8332 9577 email meaust@ozemail.com.au Cliff Hignett CPSS CPAg CSIRO Land and Water PMB 2 Glen Osmond South Australia 5064 ph (618) 8303 8459 fx (618) 8303 8551 ah(08) 8276 7706 email cliff.hignett@adl.clw.csiro.au =========== MEASUREMENT ENGINEERING AUSTRALIA =========== Environmental Monitoring and Data Logging Applications Engineers. 27 Rowland Rd., Magill, SA 5072 ph: (08) 8332 9044, fax: (08) 8332 9577 e-mail: meaust@ozemail.com.au