The first step of the installation was to test and assemble the lysimeters at a length specific for their installation. The enclosed lysimeter body is assembled and tested at the factory for leaks prior to shipping. However fittings may loosen in transit and were re-checked for a proper seal. In order to test for tightness, the units were submerged in a deionized water bath while a positive pressure of 15 psi was applied. Under pressure, the porous ceramic cup should give off small “champagne” type bubbles over its entire surface if no leaks exist. Large bubbles forming at any joints on the body indicate a leak. If a unit displayed signs of leaks (noted on a few of the units) the joints were sealed with Teflon tape, and rechecked. The most common leak was encountered along the joint between the ceramic cup and the PVC body. Leaks were not found with the acrylic body lysimeters, probably because the ceramic cup was attached with an epoxy at the factory.
Following the pressure test, the lysimeter units underwent a vacuum test to the units head assembly. The factory suggested that a plastic membrane (or condom) should be first placed over the ceramic cup, and then a vacuum applied while carefully monitoring pressure drops. For this installation we checked vacuum with the ceramic cups placed entirely in deionized water. This laboratory check tested the factory seal and also allowed some practice with the units outside of the field. While under a vacuum the units would pull water through the ceramic cup into the vessel. Once the vessel was full with water, the unit would hold the vacuum unless it leaked. This test helped to locate leaks that would have caused problems if encountered later in the field.
The complete assembly and sizing of the PVC lysimeter units was done prior to there installation at the sites. The depth of the lysimeter ceramic cup was determined based upon root zone depth and site soil type. The sites were designed to have an array of lysimeter depths to measure any change in nitrate nitrogen concentration through the soil profile. At depths beneath the root zone, nitrate found in the soil water has the capability to leach to the groundwater. The deepest lysimeter was therefore placed at 120 and 150 cm respectively at the SCWA and SUNY site. The location of the lysimeter at SUNY is well above the water table (over 50 feet), while the SCWA site is located just above the shallower upper aquifer of the South Shore (between 1.5 to 2.0 feet)
The assembly was performed in accordance to the manufacturer’s manual (copies of the manual are included in the reports Appendix). Although the assembly was generally straightforward, a few items were encountered during the installation, which were not addressed in the manual. One important detail was to allow the well head assembly base to slide down along the lysimeter extension casing and body to ensure that the Teflon and polyethylene tubing could be attached at the proper length (Figure 8). This is important because the tubing was rather stiff and proper length was required to allow the unit to fit together properly. It was also noted that the polyethylene tubing should not be grooved for a seal because of its tendency to cut unevenly, leading to leaks in the system. The Teflon tubing could be grooved smoothly in comparison.
When the units were fully assembled and sized, the ceramic cups were placed in deionized water and a vacuum of 15 inches of mercury was applied for one hour. This procedure pre-wets the porous ceramic cup, and force water out of the pore spaces. After being under a vacuum for the hour, the units were mostly filled with deionized water. This step was performed to reduce the tendency for the air filled lysimeters to rise in the water slurry. It also allowed for the cleansing of the ceramic cup with a non-contaminated solution. Because of this, the lysimeters required immediate sampling after the installation was completed to remove the deionized water and prevent the dilution of the soil water solutions.
Figure 8 – Head Assembly (Image courtesy Monoflex Brochure)
Due to the shallow nature of the lysimeter design, the units were installed by hand. A 15-cm diameter hole was dug using the hand auger and post hole digger. A larger 34-cm hole was then dug around the borehole to a depth of 36-cm. This hole was sized to fit the plastic sprinkler box, which covers and encloses the unit beneath grade.
The depth was measured to allow for 10 cm silica flour (#00 cleaned filter sand) slurry to be placed at the base, beneath the lysimeter unit. When the hole was dug to appropriate depth, the slurry was mixed at a ratio of approximately 20-kg silica flour to 4-liters of deionized water. Upon mixing, the slurry was poured into the hole to a depth of at least 15-cm and then the complete lysimeter unit was placed in the hole. The silica flour was then placed in the hole filling it to a height of at least 10-cm above the ceramic cup. At this point a bentonite grout was used to seal the hole at this height. This was done to prevent surface water from draining through the hole, altering the existing flow conditions. With the deeper units (greater than 100 cm), the pre-existing soil was used to fill the hole to grade. This sandy-soil was sorted, with any material greater than 2 mm removed, and the returned to the hole. When the hole was filled to the appropriate height, the sprinkler box was put in place and positioned around the unit. Then cement was added around the lysimeter sealing the unit in place and preventing water from flowing into the hole (Figure 9).
Figure 9 – Lysimeter Installation Profile
The lysimeters, filled with deionized water prior to installation, had to be sampled immediately following installation to remove this deionized water. This water was removed in the same matter as when the unit is sampled at later time. A two-port rubber stopper is used with an Erlenmeyer collection flask (Figure 10) for collecting the samples. A 36-cm piece of polyethylene tubing is placed in one of the holes in the rubber stopper, with at least 2 cm protruding through the stopper. The tube in then placed in the sample recovery valve (blue) of the head assembly. Next, the tube (approximately 36-cm) from the hand pump is attached to the other hole in the stopper and then the valve opened. The pump is used to place a gentle vacuum, removing water from the unit into the flask. It is important to always apply a vacuum in a gentle manner to avoid the slurry to lose continuity with the ceramic cup.
After removing the deionized water from the lysimeter, it must then be “charged” in order to collect soil water. First, the pressure vacuum valve is opened and the polyethylene tube, from the hand pump, is inserted into the fitting. It is important that the sample retrieval valve is closed at this point. The hand pump is used to gently draw a vacuum of 46-54 cm of mercury on the vacuum pressure gauge. Once the correct vacuum is noted, the valve is quickly closed and the line is detached.
The soil water samples collected in the field were stored in a cooler until they were taken to the lab. They were first filtered through a 40 mm paper filter using a vacuum filtration system. The solution was then either frozen or analyzed immediately.
Figure 10 – Sample Collection (Image courtesy Monoflex Brochure)
Nitrate nitrogen concentrations were calculated using a HACH DR-2000 Spectrophotometer, set to analyze for high range nitrate nitrogen. The procedure uses a cadmium powder to reduce the nitrate to nitrite (which can be analyzed using a spectrophotometer). The results obtained are thus for the total nitrite and nitrate (oxidized nitrogen) found in the solution. Because nitrite will be converted (under favorable conditions without loss by denitrification and volatilization) to nitrate in the vadose zone this analysis was sufficient to determine the potential for nitrogen leaching.