| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Crop and Soil Sciences, 116 A.S.I. Building, The Pennsylvania State University, University Park, PA 16802
b School of Veterinary Medicine, University of Pennsylvania, New Bolton Center, 382 West Street Road, Kennett Square, PA 19348
* Corresponding author (yxz117{at}psu.edu)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONREFERENCES |
|---|
Abbreviations: D, soil water depletion • ET, evapotranspiration • LCE, leachate collection efficiency • PPL, percentage of precipitation collected as leachate • TAW, total available water
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONREFERENCES |
|---|
Zero-tension pan and passive capillary fiberglass wick lysimeters are much less expensive to install, easier to maintain, and require less soil disturbance during installation than monolith weighing lysimeters. However, pan lysimeters collect leachate only from macropore flow or when the soil above the pan is saturated. It is possible for downward water flow to bypass pan lysimeters because the soil above the pan has to be saturated before soil solution will drip into the pan. In this case, the soil water tension surrounding the pans will be greater than that above the pans and soil solution will diverge from the pans, resulting in low collection efficiencies (Jemison and Fox, 1992; Boll et al., 1991). Wick lysimeters use self-priming wicks to produce a hanging water column that exerts suction on the soil above the lysimeters. The suction is a function of wick length, flow rate in the wick, and wick hydraulic conductivity and can be designed to match that of the soils where the lysimeters are installed (Boll et al., 1992; Knutson and Selker, 1994). Investigators have shown that wick lysimeters enhanced leachate collection efficiency compared with pan lysimeters (Boll et al., 1991; Brandi-Dohrn et al., 1996).
Both pan and wick lysimeters need collection efficiency data to correct measured percolate volumes and masses of percolate constituents into actual flux in field situations. Russell and Ewel (1985) found that pan lysimeters with a collection area of 150 cm2 collected only 10% of precipitation. Radulovich and Sollins (1987) studied the pan collection efficiencies of different pan sizes and found that the collection efficiency and probability of collecting enough leachate for analysis were improved with larger pan sizes. In the first 3 yr of this lysimeter study, Jemison and Fox (1992) using Br tracer and water-balance methods determined that the pan lysimeters had collection efficiencies ranging from 13 to 92% with an average of 52%. Boll et al. (1991) found in a month-long rainfall simulation experiment that on average two wick lysimeters collected 103% of water and 65% of Br applied, compared with 25 and 7% for pan lysimeters. In a 2-yr experiment, Brandi-Dohrn et al. (1996) reported 66 to 80% collection efficiencies with wick lysimeters using a water balance method with ET estimated from pan evaporation.
In our study, the pan and wick lysimeter's LCEs, which are the measured leachate volumes for a certain period of time divided by estimated percolation water (precipitation + irrigation - ET) for the same period, were evaluated and directly compared using a water balance method for a 5-yr period. The Penman–Monteith method recommended by Food and Agriculture Organization (FAO) was used to estimate ET. The Penman–Monteith method is a new modification to the FAO 1977 Penman method (Doorenbos and Pruitt, 1977) and is considered the most reliable method for estimation of ET (Allen et al., 1998, Chiew and McMahon, 1992). To understand the possible errors of LCE inherited from the ET estimation, the effects of variations in parameter estimates used on ET estimation were analyzed.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONREFERENCES |
|---|
From spring 1988 to spring 1991, the field was used to estimate NO3 leaching from manured and unmanured corn (Zea mays L.) as influenced by N fertilizer rates (Jemison and Fox, 1992, 1994; Jemison et al., 1994a,b). From April 1991 to April 1994, it was used to study NO3 leaching under continuous N fertilized corn and alfalfa (Medicago sativa L.) (Toth and Fox, 1998). The plots that were planted with alfalfa during the years 1991 to 1994 were the manured corn treatments from 1988 to 1990. In April 1994 the alfalfa was killed with herbicide and plowed under and the entire field was planted with corn. Beginning in 1995, the plots that had been in manured corn and alfalfa became no-till, and the plots that had been in tilled continuous corn remained the same tillage treatment. The crop was corn in 1995, 1996, 1997, and 1999 and soybean [Glycine max (L.) Merr.] in 1998. Corn and soybean were planted in early to mid May. Grain was harvested in October when black layer had formed in the corn kernel and the soybeans were fully matured. All plant residue was left on the field.
In July 1997, the plots were irrigated five times using a drip-irrigation system. The drip lines were 152 cm apart with drip holes 61 cm apart. The total irrigation water applied was 6.1 cm with individual applications varying from 0.4 to 1.8 cm. No leaching resulted from the irrigation because of the low irrigation rate and dry soil.
Leachate was collected after significant rain events and on the last day of the month to provide monthly leachate data. Leachate collection was more frequent during the nongrowing season (November to April) because of low ET. Monthly and seasonal leachate volumes were calculated by summing the collected leachate for the corresponding period. The differences in leachate volumes between lysimeter types were analyzed by paired t tests. The leaching year was defined as from May of the year to April of the next year and was given the name of the starting year.
Lysimeter Design and Installation
Zero-tension 76 by 61 cm pan lysimeters were installed in spring 1988. Lysimeter pits were excavated with a backhoe and plywood support structures were placed in the pits. Lysimeter installation tunnels were excavated from one side of the pits at 1.2 m below the soil surface, and polypropylene bead-filled pans were inserted in the tunnels 30 cm from the pit edge and raised to the tunnel ceiling with turnbuckle supports. Detailed installation information can be found in Jemison and Fox (1992).
In April 1995, fiberglass wick lysimeters were installed at the same depth in the lysimeter pits but on the opposite side from the pan lysimeters. The wick lysimeter design was adapted from that published by Holder et al. (1991). A 30 by 30 by 1.3 cm thick plexiglass plate with a 3.2-cm diam. center hole was attached to a pressure treated plywood support structure (Fig. 1) . The support structures had four threaded rods and turnbuckles at the four corners that allowed the lysimeter surface to be raised to contact the tunnel ceiling. Wicks were 9.5 mm in diameter (PEP 3/8), 20-strand fiberglass ropes supplied by Pepperell Braiding Co. (East Pepperell, MA) (Knutson and Selker, 1994). Six wicks were used for each lysimeter. Hydrophobic manufacturing chemicals were removed from the wicks by combusting them at 400°C for 3 h (Knutson et al., 1993). Individual wick strands were frayed using dissecting probes and were spread evenly over the plexiglass creating a 900-cm2 soil water collection area. Spaces between strands were filled with extra wick material to achieve nearly 100% cover of the plexiglass plate surface. Collecting jugs were 15-L Nalgene polypropylene carboys (Nalge Co., Rochester, NY). The wicks from the support structure to the carboy were enclosed in a flexible plastic drainage hose that served as an evaporation barrier. The hose was connected to the support structure and collecting jug lids by polyvinyl Cl fittings and secured by stainless steel hose clamps. Tunnels for lysimeter installation were excavated 65 cm into the Bt horizon to accommodate the lysimeters plus a 30 cm buffer of undisturbed soil between the pit edge and the lysimeters. Immediately prior to installation, the ceiling of the excavation was carefully leveled and cleaned with a metal spatula. Lysimeters were inserted in the excavations and the wick surfaces were raised into firm contact with the ceiling using turnbuckles. The vertical distance from the top of the plexiglass to the bottom of the wick in the collecting jugs was 50 cm, which created up to 50 cm of water tension on the soil (Holder et al., 1991). The opening of the excavation was covered with plastic sheeting to limit evaporation.
|
 | [1] |
 | [2] |
 | [3] |
 | [4] |
For the crop residue cover adjustment, we assumed that crop residue cover was 90% after corn harvest and 85% after soybean harvest and winter decay was 15% for corn and 25% for soybean (Hill et al., 1997). We used a suggestion of Allen et al. (1998) that there is 5% reduction of evaporation for every 10% of ground cover. We assumed that crop residue decays in a linear fashion from harvest to the next planting.
The ETc estimation was adjusted for crops experiencing water stress (Allen et al., 1998). Whether water stress adjustment is needed for crops was determined by comparing monthly estimated crop ETo multiplied by Kc with monthly precipitation. When monthly estimated crop ETo multiplied by Kc values were greater than the precipitation, water shortage for crop growth was anticipated and the water stress coefficients (Ks) were calculated. Observed dry weather conditions and crop indications of water stress such as leaf curling helped us verify if the crop experienced a water stress. The Ks value is calculated as follows:
 | [5] |
 | [6] |
 | [7] |
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONREFERENCES |
|---|
|
Tillage treatments had no effect on 5-yr total leachate volume for either lysimeter type at the 5% level of significance (Table 1). For the wick lysimeters, the only significant difference in annual leachate volumes between tillages was the 1997 leaching year when more leachate was collected from the tilled plots. The 5-yr total leachate volumes of wick lysimeters were significantly different at the 10% level between the two tillages, but the leachate from the 1997 leaching year contributed most of the difference. If the 1997 data is excluded, the difference between the two tillages was not significant (P = 0.343). For the pan lysimeters, the annual leachate volumes were consistently higher in no-till than in tilled treatments for the five individual years and the differences were significant in 3 yr, but the 5-yr total leachate volumes were not significantly different for the two tillage systems.
|
|
Leachate Collection Efficiency
Annual, 4- and 5-yr LCEs were calculated by dividing the total leachate volume by the calculated total percolation water for the period of 1-, 4-, and 5-yr periods, respectively. Excluding 1997 data, the annual LCEs for the wick lysimeters were 99 to 129% for the tilled and 83 to 108% for the no-till treatments for the four individual years (Table 3). The 4-yr LCEs for the wick lysimeters were 106% for the tilled and 95% for the no-till with standard deviations of 25% for tilled and 28% for no-till, respectively. Since there was no significant difference in LCEs between tilled and no-till treatments for the leaching years 1995, 1996, 1998, and 1999, we calculated an overall average LCE of tilled and no-till for each of 4 yr. The average annual LCEs of 18 wick lysimeters varied from 93 to 119% with an average 4-yr LCE of 101% and a standard deviation of 27%. Boll et al. (1991) reported 98 and 108% collection efficiencies for two wick lysimeters after a month-long rainfall simulation and Brandi-Dohrn (1996) reported 66 to 80% collection efficiencies for their wick lysimeters. Recently, Louie, et al. (2000) reported an average LCE of 125% with a coefficient of variation of 36% for 30 wick lysimeters installed under a variety of soil managements (row crops, orchard, and other horticultural crops) and soil textures (fine sandy loam to silty clay loam).
|
Variation of Lysimeter Collection Efficiency
During the experiment period, no wick or pan lysimeter stopped yielding leachate or was flooded by a heavy precipitation event. However, LCEs differed greatly among the individual lysimeters. For both wick and pan lysimeters, some lysimeters consistently yielded higher or lower annual LCEs than others, but no lysimeter always had the highest or the lowest LCE throughout the 5-yr period. For the wick lysimeters, the LCE varied from a minimum of 31% for one lysimeter during the 1999 leaching year to 285% for another lysimeter during the 1997 leaching year. The 5-yr average standard deviation and coefficient of variation for LCEs of the wick lysimeters were 33 and 28%, respectively. For the pan lysimeters, the LCE varied from a minimum of 6% for one lysimeter during the 1996 leaching year to 127% for another lysimeter during the 1997 leaching year. The 5-yr average standard deviation and coefficient of variation for LCEs of the pan lysimeters were 13 and 26%, respectively. The number of lysimeters needed to obtain a reliable mean estimate of LCE was calculated based on the iterative procedure following Gilbert (1987)(p. 32). For an error margin to be within ±20% of the mean LCE with <5% probability exceeding this error margin, 13 wick lysimeters and 14 pan lysimeters would be required. For a 10% probability exceeding this error margin (20% of mean LCE), the number of lysimeters needed would be nine and ten for wick and pan lysimeters, respectively. Louie et al. (2000) and Holder et al. (1991) did similar estimations and found that eight and six lysimeters were required to obtain ±30% of error margin with 5% error probability. In our case, if we set the same requirements as they did, the number of lysimeters needed would be seven for wick and eight for pan lysimeters, respectively.
Evapotranspiration Estimation Error Analysis
The Penman–Monteith method had been shown to have the lowest error compared with measured ET values, ranging from +4% in humid regions to -1% in arid regions and was the best among 20 methods compared (Smith, et al., 1996). The standard procedure of Penman–Monteith method assumes that the crop has no limitations to its growth and ET and that soil is bare in the winter fallow period. In our case, crop residue was left on the field and the crop experienced water stress in 1995, 1998, and 1999. Therefore, adjustments of ET because of crop residue cover on the soil surface in the winter fallow periods and crop water stress during the vigorous growth stage in dry summers were necessary to have accurate ET estimates. Crop residue cover adjustments reduced annual ET by 7% for the 1998 soybean year and 11 to 14% for the corn years, compared with the ET estimates without the crop residue cover adjustment. The water stress adjustments reduced annual ET by 21% in 1995, 0.7% in 1998, and 16% in1999 compared with ET value without water stress adjustments. No water stress adjustment was necessary in 1996 and 1997 because of sufficient water supply for crop growth.
To perform crop residue cover and water stress adjustments, some estimates are needed. The accuracies of these estimates could contribute to error in our ET estimation. Varying estimates for crop residue cover adjustment within a reasonable range had very little effect on ET values. In our adjustment, we assumed that the crop residue covered 90 and 85% of the soil surface after corn and soybean harvest, respectively, and that 15% of corn residue and 25% of soybean residue decayed over the winter fallow period. If the plant residue cover percentage on soil surface varied ±10% from our assumed cover percentages, the ETc only differed by 1 to 2%. The ETc differed by <1% if the decay percentage deviated by ±10% from our assumed data.
Water stress adjustments are a function of TAW, the portion of the TAW that a crop can extract from the root zone (p), and soil water depletion in the root zone (D). In our case, the TAW for the test soil was estimated as 180 mm to a depth of 1.2 m using measured data for Hagerstown silt loam. If the real TAW in the soil deviates from this estimate by 25 mm (
14% of TAW), the error of ETc estimate would be 5%. Since TAW for a test soil can be measured, the error because of inaccuracy of TAW should be minimal. The effect of p on the ET estimation was insignificant. We used Eq. [7] to calculate p values for our ET estimates. Varying p estimates by ±20% of the calculated p values resulted in only 1 to 3% change in ET value.
Estimates of the value of D probably have the biggest effect on the magnitude of water stress adjustments among the estimates we used. If the estimated D value differed from the real field situation by 25 mm, estimated ET would differ by 7%. To estimate D, we assumed that D was zero after a series of rain events (Allen et al., 1998). For example, in 1995 we had 152 mm of precipitation within the previous 30 d and 65 mm of precipitation within 1 wk before June 12. We started the water stress adjustment calculation on June 13 by assuming that D was zero on June 12.
This analysis shows that the overall error in ET estimates resulting from variations of the estimates in crop residue cover and water stress adjustments over a reasonable range would probably be no more than 10%. Errors in our estimates of LCE because of errors of ET would be approximately the same as the errors in ET estimates, <10%. It can be shown mathematically that this similarity in error between ET and LCE occurs because the definition of LCE is leachate volume/(irrigation + precipitation - ET) and the ratio of (irrigation + precipitation)/ET is 2 in our case.
|
SUMMARY AND CONCLUSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONREFERENCES |
|---|
100%. Pan lysimeters collected about 20% of precipitation and had average LCEs of 40%. The 5-yr average standard deviations of LCEs for wick and pan lysimeters were 33 and 13%, respectively, but their coefficient of variation were similar (28% for wick and 26% for pan). Tillage treatments had no significant effect (5% probability level) on 5-yr total leachate volume, though consistently higher annual leachate volumes was observed in no-till for pan lysimeters. The high LCE for the 1997 leaching year was probably the result of the lysimeters acting as water sinks during an exceptionally wet and warm winter. Crop residue cover on the soil surface after harvest and crop water stress during the vigorous growth stage in dry summers reduce ET and, therefore, must be taken into account in performing ET estimation for water balance calculation. Deviations of estimates of crop residue cover and residue decay percentages by ±10% from our estimated values for crop residue cover adjustments did not cause significant differences in estimated ET values. Varying TAW and D estimates in crop water stress adjustments had bigger effects on the ET estimates than varying the crop residue cover and decay percentages. However, the overall errors of ET estimates caused by the inaccuracies of these estimates are probably no more than 10%. The errors of LCE estimates would be in the same range as that of ET based on the LCE definition and the approximate 2:1 ratio of (precipitation plus irrigation)/ET in our case. Received for publication October 27, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. D. Ritchey, D. G. Boyer, K. E. Turner, and J. D. Snuffer Goat Urine and Limestone Affect Nitrogen and Cation Distributions in an Acidic Grassland Soil Sci. Soc. Am. J., May 1, 2008; 72(3): 626 - 633. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Peckenham, J. A. Nadeau, A. Amirbahman, and R. S. Behr Release of nitrogen and trace metal species from field stacked biosolids. Waste Management Research, April 1, 2008; 26(2): 163 - 172. [Abstract] [PDF]
|
|
|
|
 |
|
 L. Weihermuller, J. Siemens, M. Deurer, S. Knoblauch, H. Rupp, A. Gottlein, and T. Putz In Situ Soil Water Extraction: A Review J. Environ. Qual., October 24, 2007; 36(6): 1735 - 1748. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Mertens, J. Diels, J. Feyen, and J. Vanderborght Numerical Analysis of Passive Capillary Wick Samplers prior to Field Installation Soil Sci. Soc. Am. J., January 1, 2007; 71(1): 35 - 42. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. D. Toth, Z. Dou, J. D. Ferguson, D. T. Galligan, and C. F. Ramberg Jr. Nitrogen- vs. Phosphorus-based Dairy Manure Applications to Field Crops: Nitrate and Phosphorus Leaching and Soil Phosphorus Accumulation J. Environ. Qual., October 27, 2006; 35(6): 2302 - 2312. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Morari Drainage Flux Measurement and Errors Associated with Automatic Tension-controlled Suction Plates Soil Sci. Soc. Am. J., September 20, 2006; 70(6): 1860 - 1871. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. van der Velde, S. R. Green, G. W. Gee, M. Vanclooster, and B. E. Clothier Evaluation of Drainage from Passive Suction and Nonsuction Flux Meters in a Volcanic Clay Soil under Tropical Conditions Vadose Zone J., November 11, 2005; 4(4): 1201 - 1209. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Mertens, G. F. Barkle, and R. Stenger Numerical Analysis to Investigate the Effects of the Design and Installation of Equilibrium Tension Plate Lysimeters on Leachate Volume Vadose Zone J., June 9, 2005; 4(3): 488 - 499. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. W. Gee Comments on "Improvements to Measuring Water Flux in the Vadose Zone" (K.C. Masarik, J.M. Norman, K.R. Brye, and J.M. Baker; J. Environ. Qual. 33:1152-1158). J. Environ. Qual., March 1, 2005; 34(2): 408 - 409. [Full Text] [PDF]
|
|
|
|
|
|
S. Czigany, M. Flury, J. B. Harsh, B. C. Williams, and J. M. Shira Suitability of Fiberglass Wicks to Sample Colloids from Vadose Zone Pore Water Vadose Zone J., February 1, 2005; 4(1): 175 - 183. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Siemens and M. Kaupenjohann Comparison of Three Methods for Field Measurement of Solute Leaching in a Sandy Soil Soil Sci. Soc. Am. J., July 1, 2004; 68(4): 1191 - 1196. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. C. Masarik, J. M. Norman, K. R. Brye, and J. M. Baker Improvements to Measuring Water Flux in the Vadose Zone J. Environ. Qual., May 1, 2004; 33(3): 1152 - 1158. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. R. Barzegar, S. J. Herbert, A. M. Hashemi, and C. S. Hu Passive Pan Sampler for Vadose Zone Leachate Collection Soil Sci. Soc. Am. J., May 1, 2004; 68(3): 744 - 749. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Zhu, R. H. Fox, and J. D. Toth Tillage Effects on Nitrate Leaching Measured by Pan and Wick Lysimeters Soil Sci. Soc. Am. J., September 1, 2003; 67(5): 1517 - 1523. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Zhu and R. H. Fox Corn-Soybean Rotation Effects on Nitrate Leaching Agron. J., July 1, 2003; 95(4): 1028 - 1033. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Vadose Zone Journal | Journal of Plant Registrations | ||||
| Journal of Natural Resources and Life Sciences Education |
Journal of Environmental Quality |
||||
