1、Field Procedures and Data Analysis for theCornell Sprinkle InfiltrometerFor questions, contact:Harold van Es, Associate Professor: (607) 255-5629, hmv1cornell.edu, orRobert Schindelbeck, Research Support Specialist: (607) 255-1706, rrs3cornell.eduDepartment of Crop and Soil Science, Cornell Universi
2、ty, Ithaca, NY 14853-1901The Cornell Sprinkle InfiltrometerSoil infiltrability is an important soil quality indicator, as it has important agricultural and environmental implications and is strongly affected by land management practices. Measurement of soil infiltrability is generally done through p
3、onded ring infiltration or simulated rainfall, each having specific advantages and disadvantages. The Cornell Sprinkle Infiltrometer (Ogden et al., 1997) was designed to combine the advantages of both. It also allows for easy and rapid measurement of soil infiltration, as this is essential to adequa
4、tely estimate spatially and temporally-variable infiltration behavior (van Es, 1993). The Cornell Sprinkle Infiltrometer system consists of a portable rainfall simulator that is placed onto a single 241-mm (9 1/2“) inner diameter infiltration ring (Fig. 1) and allows for application of simulated rai
5、nfall at a wide range of predetermined rates. The apparatus permits the determination of several important soil hydrological properties: Time-to-runoff, sorptivity, and field-saturated infiltrability. RitQROtSprinkleIfltomtrsoilurface rot VtrtitFig. 1. Infiltrometer setup rtitrot241mm diameter metal
6、 infiltration ringFig. 1 Infiltrometer setupVt2In contrast to most other ponded infiltration measurements, this approach: Wets the soil in a more natural manner and eliminates soil slaking as a result of instantaneous ponding Reduces unnaturally high contributions of macropore flow under ponded cond
7、itions Provides a realistic surface boundary condition, including the effects of soil surface roughness which can greatly influence infiltration behavior Is conservative with waterCompared to most other rainfall simulators, the Cornell Sprinkle Infiltrometer measures infiltrability for a relatively
8、small soil surface area. However, its main advantages are: Low cost High portability Allows for rapid measurements by a single person Easy calibration for a wide range of simulated rainfall rates Conservative water useThe Cornell Sprinkle Infiltrometer employs a single, rather than a double infiltra
9、tion ring, and makes adjustments for three-dimensional flow at the bottom of the ring based on research by Reynolds and Elrick (1990).Field ProceduresSprinkler PreparationAlthough the sprinklers are robustly built for use under field conditions, the user should be aware that the capillary tubes at t
10、he bottom of the unit are the most sensitive part of the equipment. Efforts should be made to minimize contact of the tubes with soil or debris. Use of water with high sediment content should be avoided as it may increase the potential for clogging of the capillaries. Since natural rainfall is low i
11、n soluble salts, it is recommended (but not always logistically feasible) to use water of low ionic strength. This may be especially critical for sodic and other soils that are subject to slaking. Fill the sprinkler when positioned on a stable flat surface. Remove the large rubber stopper and air-en
12、try tube, and pour water into the vessel. Then re-insert the stopper/tube, and place it firmly to insure that the stopper is air-tight. (This is important as air should only enter the vessel through the air-entry tube.) The interface between the large stopper and the air-entry tube should also be ai
13、r-tight. Some vacuum grease may be used to insure this, while still allowing for easy adjustment of the tube. Once the sprinkler vessel has been filled and the stopper/tube firmly reinstalled, blow gently into the air-entry tube for a few seconds to apply some additional air pressure to remove possi
14、ble bubbles from the capillaries. This only needs to be done at the beginning of a set of measurements, and does not need to be repeated with refills on the same day. Then, put the small stopper on the top of the air-entry tube. This will air seal the vessel and the capillaries will cease dripping a
15、fter a few seconds. The sprinkler is now on stand-by and ready for use without losing any water in the meantime. 3Sprinkler calibrationThe sprinklers are designed to apply water at a wide range of simulated rainfall rates. The rate can be changed by moving the air-entry tube up (for higher rates) or
16、 down (for lower rates). It is recommended to calibrate the sprinkler for a rainfall rate of 25 to 30 cm/hr. This generally insures that ponding will occur for every measurement, and still allows for a measurement period of one hour without requiring refills. Note: Alternatively, the sprinklers may
17、be calibrated for an event of known recurrence period for the region of the study (e.g., a 50-year, 1 hour event). This will generally not insure ponding for all measurements, in which case one might interpret the measurement location as having “sufficiently high“ infiltrability. This may create cha
18、llenges when trying to analyze the data statistically, as it will not provide quantitative data for those sites. The actual sprinkling rate in the field may vary slightly from the calibrated rate as a result of temperature variations in the water. This is not a problem, as the actual application rat
19、e is directly measured in the procedure. To calibrate the sprinkler, perform the following:1. Set the air-entry tube to the desired level. The 30 cm/hr or 0.5 cm/min sprinkle rate is generally achieved when the bottom of the air-entry tube is located at 10 cm above the bottom of the container. This
20、is therefore a good starting point for the calibration effort.2. Measure the height of water level in the sprinkler vessel (H1). It is easiest to measure and record it in cm with one decimal value (e.g. 41.2 cm).3. Remove the small stopper from the air-entry tube, while simultaneously starting a sto
21、pwatch4. Allow for 3 minutes of sprinkling and read the water level exactly at this time (H2)5. Calculate the rainfall rate (cm/min) asH1-H2/36. If the actual rainfall rate is below the desired rate, move the air-entry tube upwards. Move it down if it is above the desired rate.7. Repeat the procedur
22、e until the desired rate is achieved. Note that the calibrated rainfall rate does not need to be very exact, as the actual rate is determined for each field measurement, and variations are accounted for in the data analysis.Once the sprinkler has been calibrated for the desired rate, refill the vess
23、el and reinstall all stoppers. It is now ready for actual field measurements. Note that calibration generally does not have to be repeated. It can also easily be checked with each subsequent field measurement.4Ring InsertionThe infiltration ring should be inserted without causing significant disturb
24、ance to the soil. This is best performed with the use of a hydraulic device that pushes the ring into the soil with a steady and constant force. Pounding rings into the soil using a hammer tends to cause some soil disturbance, especially in dense soils, and is therefore less preferred. In all cases,
25、 it is recommended to lay a piece of 4“ by 4“ wood of about 30 cm length horizontally on top of the ring and apply the driving force to it. Before inserting the ring, carefully remove pieces of debris, crop residue and small rocks that are immediately below the edge of the ring, as they would cause
26、soil disturbance when the ring is pushed in. In rocky soils, multiple attempts may be required to insure that ring insertion occurred without excessive disturbance. The ring should be inserted to a depth where the lower edge of the round overflow hole is flush with the soil surface. Depending on whi
27、ch end is used, the rings can be inserted to a depth of 7 cm or 15 cm. The deeper insertion is preferable, but may not be feasible in many field situations, especially with dense or rocky soils, and when the rings are hammered into the soil. In soils with a rough surface, the rings should be inserte
28、d with the hole located at the level where overflow of microrelief would occur under natural rainfall conditions. This allows the infiltration measurement to account for the effect of surface storage capacity, which greatly affects infiltrability under those conditions.Once the ring has been install
29、ed, insert the overflow tube assembly (stopper and tubing) into the ring (Fig. 1). At the end of the tube, dig a small hole to place the beaker. The hole for the beaker should be sufficiently distant (30 cm or more) from the infiltration ring to not interfere with water flow patterns. The tubing sho
30、uld slope away from the ring to insure that overflowing water does not back up and readily empties into the beaker. The beaker itself should therefore also be positioned sufficiently low. The sprinkler may now be placed on top of the ring in preparation for the measurements (as in Fig. 1). Alternati
31、vely, the sprinklers may be suspended above the ring (e.g., off a tripod). This will allow the simulated raindrops to gain velocity and more closely reproduce the energy of natural rains. MeasurementsThe following steps outline the measurement procedure:1. Measure the height of the water level in th
32、e sprinkler vessel (H1)2. Remove the small stopper from the air-entry tube, while simultaneously starting a stopwatch. Monitor the outflow tube to determine whether water is being discharged into the beaker. During this period, it is advised to slightly rotate the sprinkler every minute or so (more
33、often when the sprinkler is suspended) to prevent raindrops impacting the soil surface in the same location. 4. When water starts flowing out of the tube, record the time (TRO, time to runoff in minutes). The runoff water should now be flowing into the beaker.5. After three (or so) minutes, pour the
34、 water from the beaker into the graduated cylinder. This should be done while not spilling water that continues to come from the outflow tube (e.g., by quickly replacing the full beaker with another empty one, or temporarily blocking the outflow tube).6. Measure the runoff volume (Vt) in the graduat
35、ed cylinder (in ml). Record both Vt and the time at which water was collected.57. Repeat steps 5 and 6 for as long as desired (generally up to one hour), or until the water level in the vessel has reached the bottom of the air-entry tube. Do not continue beyond this point as the sprinkle rate will g
36、radually decrease. In most cases, steady-state conditions will have occurred within an hour. It may take longer with extremely dry soils and those that have shrinkage cracks that close very gradually during extended wetting.8. At the end of the measurement period, determine the water level in the ve
37、ssel (H2) and the time at which it is taken (Tf).Data AnalysisThe simulated rainfall rate (r, constant throughout the experiment) is determined by r = H1 - H2 / TfThe runoff rates (rot , cm/min) are determined by rot = Vt / (457.30*t)where 457.30 is the area of the ring, and t is the time interval f
38、or which runoff water was collected (3 minutes in our case). Infiltration rates (it) are determined by the difference between the rainfall rate and runoff rate:it = r - rotFigure 2 shows rainfall, runoff and infiltration rates for a typical measurement.6Estimation of Sorptivity Time-to-runoff (TRO)
39、is an important soil hydrological parameter that is dependent on the rainfall rate (r) as well as the initial soil water conditions. Runoff will occur earlier if r is higher and the soil is wetter. Sorptivity (S) is a more universal soil hydraulic property that describes early infiltration independe
40、nt of rainfall rate. It is estimated by (Kutilek, 1980):S = (2TRO)0.5 * rSorptivity also accounts for variable sprinkle rates which are difficult to avoid under field conditions, and provides an integrated assessment of early infiltration, including the effect of surface water storage with rough soi
41、l surfaces.Estimation of Field-Saturated Infiltrability Field-saturated infiltrability (ifs) reflects the steady-state infiltration capacity of the soil, after wet-up. It should be based on the data collected at the end of the measurement period, or whenever steady-state conditions occur. Since the
42、apparatus has a single ring, the measured infiltration rate needs to be adjusted for three-dimensional flow at the bottom of the ring. The required adjustment is generally greater when the ring insertion depth is shallower and the soil type is finer-textured. The adjustment factors suggested below a
43、re based on Reynolds and Elrick (1990) who used numerical modeling to estimate the effects of three-dimensional flow at the bottom of the ring. For the 7 cm and 15 cm ring insertion depth, multiply the measured infiltration rate by the constants listed in Table 1 to obtain the field-saturated infilt
44、rability:Fig. 2 Example Infiltration Data00.10.20.30.40.50.60 20 40 60time (min)fluxdensity (cm/min) .time torunoffitrotr7For example, for a ring insertion depth of 7 cm on a loam soil, the field-saturated infiltration rate is estimated asifs = it * 0.80Table 1. Conversion factors for field-saturate
45、d infiltrability to account for three-dimensional flow at the bottom of the ring (based on Reynolds and Elrick, 1990)Soil Type Ring Insertion Depth7 cm 15 cmsands and gravels 0.95 0.99loams 0.80 0.94clays and heavy clay loams 0.60 0.88Other UsesThe Cornell Sprinkle Infiltrometer can be employed for
46、other measurements of soil physical behavior. In a manner similar to the infiltration measurements, the sprinkler system may be employed to measure soil hydraulic conductivity in the field with rings inserted in different soil horizons in-situ. This can also be done in the laboratory using soil core
47、s, in which case no correction for three-dimensional flow would be required.The uniform droplet size allows the rainfall simulator to be used for measurement of soil aggregate stability under predetermined rainfall energy levels. This can provide relevant information on slaking potential, which stro
48、ngly relates to runoff and erosion.The sprinkle system may also be employed when natural soil wetting is required in the laboratory or field.ReferencesKutilek, M. 1980. Constant rainfall infiltration. J. Hydrol. 45:289-303.8Ogden, C.B., H.M. van Es, and R.R. Schindelbeck. 1997. Miniature rain simula
49、tor for measurement of infiltration and runoff. Soil Sci. Soc. Am. J. 61:1041-1043.Reynolds, W.D. and D.E. Elrick. 1990. Ponded infiltration from a single ring: I. Analysis of steady flow. Soil Sci. Soc. Am. J., 54:1233-1241.van Es, H.M. 1993. Evaluation of temporal, spatial, and tillage-induced variability for parameterization of soil infiltration. Geoderma. 60:187-199.
