| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a College of Agriculture, California State Univ., Chico, CA 95929-0310
b Dep. of Land Resources and Environmental Sciences Montana State Univ., Bozeman, MT 59717-3120
* Corresponding author (mjohns{at}csuchico.edu).
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: BOD, biological oxygen demand • COD, chemical oxygen demand • CW, cannery water • DW, distilled water • EC, electrical conductivity • ESP, exchangeable sodium percentage • SAR, sodium adsorption ratio • SOM, soil organic matter • TOC, total organic carbon
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Effective soil remediation of organic wastes implies maximum decomposition of the organics by mineralization. Consideration must be given to the organic loading rate, the type of organic waste (i.e., labile vs. recalcitrant), soil residence time in the root zone, and soil environmental conditions and properties (Reddy et al., 1980; Magette et al., 1983; Crites et al., 2000). The key to maximizing soil mineralization is O2 availability inasmuch as aerobic microbial respiration is the most effective pathway for organic matter decomposition (Vela and Eubanks, 1973; Stevenson, 1994; Alverez and Alverez, 2000; Ajwa and Tabatabai, 2004). Thus, the residence time of organic waste in the soil, especially close to the soil surface where O2 availability is maximum, must be sufficiently long enough for reaeration processes and complete degradation to occur. Otherwise, untreated or partially treated waste can percolate downward below the root zone.
Coody et al. (1986) simulated a land treatment system that used grassed sand columns, COD loading treatments (glucose) ranging from 0 to 5760 mg L–1, and ponding for five sequential days, followed by 2 d of rest, for 25 wk. They reported >97% removal of added COD, although the highest COD treatments resulted in anoxic soil conditions. Smith and Hayden (1980) studied grass–legume cropped sites where sugarbeet (Beta vulgaris L. ssp. vulgaris) processing water having CODs ranging from 320 to 8000 mg L–1 was flood irrigated (28–301 cm yr–1) at intervals of 1, 2, or 4 wk. Weekly irrigated plots were discontinued during the study because of excessive loading. After an extended period of continuous irrigation, COD reductions in 150 cm of the soil were 48 to 84% in winter and 88 to 98% in summer. Tarquin (1976) studied land application of meat-packing plant wastewater having a COD of 9400 mg L–1 on grassed cropland. He reported that a 99% COD reduction was possible using sprinkler irrigation at an application rate of 10 cm wk–1 in Texas.
Some state (i.e., California and Washington) regulatory agencies overseeing industrial land application activities use a guideline that weekly average BOD loading rates should not exceed 112 kg ha–1d–1. This level is based on a recommendation by the USEPA (1977) for control of nuisance odors, although the actual basis comes from an earlier study on measured uptake of O2 by plants (Hagan et al., 1967). The underlying assumption to promotion of this metric is that an equivalent amount of soil O2 is necessary for microbial treatment of wastewater. If this guideline was to be broadly applied, it is likely that it would be highly restrictive to many current land application activities (Crites et al., 2000). For food processing wastewaters, BOD loading rates have often exceeded 112 kg ha–1 d–1 and sometimes exceeded 348 kg ha–1 d–1 (Reed and Crites, 1984; Crites and Tchobanoglous, 1998).
The BOD is not to be confused with COD. The BOD is a 5- or 7-d incubation assay in a water medium where selected bacteria are added to the solution to biodegrade all-inclusive labile C compounds by oxidation. The COD, in the context of the presently reported research, refers to oxidation, similar to the Walkley–Black process (Walkley and Black, 1934) used in soil, wherein labile and recalcitrant C are oxidized. The COD analysis is less time consuming and more reproducible, yet requires a different analytical approach than BOD analysis (Alley, 2000). Typically, BOD, COD, or total organic carbon (TOC) is applicable to wastewater analyses. Numerically, COD is always equal to or greater than BOD.
The study reported here is a simulation of limited-duration, high-rate COD loading with fruit processing cannery liquid wastes (CW) to soil columns using soil previously irrigated with CW. The primary objective was determination of the impact of high COD loadings on the chemistry of soil percolates leaching below the root zone. In addition, changes to root zone soil chemistry were evaluated. It was our intent to assess the extent to which COD constituents at high COD loading can be the limiting factor for land application.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Study Soil
Soil was collected before the 2005 CW irrigation season from a site where land application of CW from the cannery has been occurring annually for the past 8 yr. Irrigation with CW normally begins in mid-July and ends in early October, while most of the approximately 76 cm of annual precipitation occurs during nonirrigation winter months. The choice of this soil ensured that the microbial environment was that of a site with a long-term history of CW application. The soil classified is a member of the Kimball series (fine, mixed, active, thermic Mollic Palexeralfs) and has a capability classification of IIIe.
The upper 61 cm of soil was collected in bulk in two 30.5-cm increments. Subsamples were also collected from each soil depth increment and analyzed (Table 1 ). Textures for the 0- to 30.5- and 30.5- to 61-cm depth increments were determined by the hydrometer method (Gee and Bauder, 1986) to be sandy loam and sandy clay loam, respectively. The field soil had an average bulk density (Blake and Hartge, 1986) of 1.43 Mg m–3 (n = 34, SD = 0.14 Mg m–3) through the 0- to 61-cm depth.
|
Column Study Site Location
The study location (Paradise, CA) was approximately 32 km from the CW application site, at an elevation of 670 m elevation. The difference in the average September–October temperature in 2005 between the land application and study sites was 1.4°C. Maximum temperatures achieved at the land application and study sites were 35 and 34°C, respectively, while minimum temperatures were identical at 6°C.
The soil columns were placed in a three by five array wooden rack (61 by 97 by 109 cm) located outside in full sun. The sides of the rack were shaded with shade cloth (60% reduction), and only column tops were exposed to the sun. The soil columns were covered briefly during times of precipitation.
Throughout the entire study period, sprouting and growth of volunteer grasses, consisting of dallis grass (Paspalum dilatatum Poir.), Bermuda grass [Cynodon dactylon (L.) Pers.], and ryegrass (Lolium perenne L.), were allowed to better simulate conditions in situ, where processes such as evapotranspiration and uptake of nutrients occur. In addition, the presence of live roots should enhance the soil microbial environment and biochemical processes within the rhizosphere. Stand density was relatively uniform among the columns and deemed not of sufficient variability to affect treatment results. An estimate of evapotranspirational crop consumptive rates (ETc), using nearby CIMIS (California Irrigation Management Information System, 2005) weather station data, indicated ETc of 3.4 and 2.3 cm wk–1 for September and October 2005, respectively. These ETc values are comparable to the weekly treatment application rates, ranging from 3.4 to 6.8 cm wk–1. Treatment application rates reasonably equaled or exceeded ETc when applying a crop coefficient of 1 for grasses and provided a downward hydraulic gradient during dosing.
Cannery Wastewater Treatment
The CW was obtained from a holding pond of a cannery processing peach [Prunus persica (L.) Batsch var. persica], pear (Pyrus communis L.), and pineapple [Ananas comosus (L.) Merr.]. Sufficient CW for soil column dosing was collected twice (July and August 2005) and refrigerated at 2°C. Analyses of the CW indicated a COD range from 8235 to 14,424 mg L–1, with an average of 10,260 mg L–1 (Table 2
). These COD levels are consistent with food processing effluents and are more than one magnitude greater than what is normally found in untreated municipal wastewater (Crites et al., 2000). Most of the N in the CW is in organic forms. The BOD/COD ratios commonly range from 0.44 to 0.67 (July–October 2004 factory data, not shown) and are indicative of the potential soil biodegradability of the CW (Alley, 2000). The electrical conductivity (EC) of the water varied from 1.5 to 2.0 dS m–1, indicating a degree of restriction for use in irrigation (Ayers and Westcot, 1985). Soluble constituents were primarily Na+ and K+ salts, and the sodium adsorption ratio (SAR) ranged from 10.1 to 13.4. The Na hazard with the EC range was considered to be slight to moderate for irrigation (Ayers and Westcot, 1985; Hanson et al., 1999).
|
|
|
Percolates were collected in glass containers enclosed in a compartment below the rack supporting the columns, beginning the fourth week of dosing. A period of collection as long as 3 wk was required to collect sufficient volumes for analysis. Total volumes collected averaged 437, 445, 322, 744, and 702 mL for DW, SAR 11.4, and 467, 701, and 934 kg COD ha–1 d–1 treatments, respectively. Percolates were stored in sealed glass containers at 2°C before analysis. After the final (8th wk) treatment dosing, columns were allowed to drain for 1 wk. All columns were then dosed with 518 mL (6.82-cm depth) of distilled water and percolates were collected for analysis. Percolate treatment volumes averaged from 313 to 425 mL. Soil columns were then allowed to remain dormant and excluded from rainfall for 45 d. In late December 2005, the columns were exposed to approximately 20 cm of rain to simulate the "rest" period at the CW land application site. Percolates were collected during and immediately after this period. Percolate treatment volumes averaged from 834 to 1150 mL. The columns were then covered to prevent further rainfall infiltration.
In early February 2006, the columns were cut open. Rooting proliferation was observed and appropriately recorded. Soil samples were collected about (±2 cm) the midpoint column depths of 7.6, 22.9, 38.1, and 53.3 cm.
Water and Soil Analyses
All water and soil analyses were performed by Olsen's Agricultural Laboratory (McCook, NE). The CW and soil column leachates were analyzed by standard methods (Greenberg, 1995). Specific analysis included COD by Method 5220B, using K2Cr2O7 oxidation and a AgNO3 catalyst to minimize Cl– interference; 7-d BOD by Method 5210B; pH by Method 4500H B; EC and total dissolved solids by Method 2510B; alkalinity, CO32–, and HCO3– by Method 2320B; Ca2+, Mg2+, K+, Na+, SO42–, Fe, and Mn by Method 3120B; NO3– plus NO2– by Method 4500NO3F; and Cl– by Method 4500CL E. An adjusted sodium adsorption ratio (SARadj) was calculated (Ayers and Westcot, 1976). We made no attempt to measure odor consequential to the treatments imposed. The BOD and COD measurements were completed with respect to representation only of O2 supply levels.
All soil samples were analyzed by the chemical soil test procedures recommended for the North Central Region (Missouri Agricultural Experiment Station, 1998). Methods included pH (1:1); soil NO3––N; Olsen P; EC; loss-on-ignition organic matter; soil SO42– (calcium phosphate); exchangeable K+, Ca2+, Mg2+, and Na+ using NH4OAc (pH 8.2) extraction; and Zn, Fe, Cu, and Mn extracted by DTPA-TEA (Lindsay and Norvell, 1978). Cation exchange capacity (CEC) was determined by summation of exchangeable macrocations. Exchangeable sodium percentage (ESP) was calculated from exchangeable Na+ concentration and CEC (Hanson et al., 1999).
Statistical Analysis
One-factor ANOVA was applied to the soil column percolate data and two-factor ANOVA (treatments x depths) was applied to the soil analysis data. Equality of variance was not considered, since the number of cases per group was identical. For one-factor ANOVA, Tukey's post hoc test was used to determine statistical differences between paired means. All statistical analysis was performed using SPSS Version 13 statistical software (SPSS, 2005).
|
RESULTS AND DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Initial distilled water dosing raised the soil column moisture similar to irrigated field conditions and ensured percolate breakthrough, which occurred 10 d after initiation of dosing and within a 24-h time period for all the soil columns. Percolate volumes collected during the first 24 h of breakthrough averaged 293 mL (n = 15), with a CV of 13.2%.
Analyses of initial percolates are shown in Table 3. Elevated or varying concentrations of COD, NO3––N, and SO42––S are probably a reflection of the flushing effect caused by the wetting of dry soil (Birch, 1960; Lundquist et al., 1999; Fierer and Schimel, 2002). Presumably, flushing stabilized the soil chemical parameters and enhanced the subsequent evaluation of treatment effects. The pH and EC of percolates collected at this time averaged 7.7 and 1.2 dS m–1, with CVs of 0.9 and 24%, respectively. The SARadj values of the percolates quantitatively relate to the ESPs in Table 1 (McBride, 1994) and the SAR–ESP relationship is consistent with that reported by Hanson et al. (1999) and Miller and Gardiner (2001).
Chemical Oxygen Demand as the Land-Limiting Constituent
Percolate CODs are presented in Fig. 1
. Differences in percolate CODs were not significant (P < 0.05) among the CW treatments at any sampling time, and all percolate COD values were 0.4% or less of the applied CW COD (based on 9216 mg L–1 average). Percolate CODs for the CW treatments remained low throughout the study (Fig. 1), confirming that substantial mineralization or sequestration of C had occurred, even at the higher COD loading rates. Changes in percolate COD concentrations between the 8th wk after dosing and after 20 cm of post-treatment winter rains were minor. There were no significant differences in percolate COD concentrations after 20 cm of post-treatment winter rains.
|
A practical explanation for the higher CODs of the percolates of the two control treatments is greater soil dissolution of indigenous organic matter with low-ionic-strength water in the DW and SAR 11.4 treatments (Olsen et al., 1996). It is also possible that the COD dosing treatments contributed to a biochemical root zone environment somewhat different from that consequential to the DW and SAR 11.4 treatments.
The magnitude of percolate COD levels, when normalized to C using the relationship C/COD = 0.375 (Alley, 2000), suggests similarity to COD levels of pore water of most native soils (Neff and Asner, 2001), substantiating that COD constituents were either mineralized by microbial respiration or sequestered. Study conditions of relatively high temperatures and substantial soil moisture would favor high rates of mineralization (Vela and Eubanks, 1973; NRCS, 1992).
The CODs of the percolates of the two control treatments were the result of biochemical dissolution or decomposition of indigenous organic matter. It has been observed that lower ionic strength waters (i.e., control treatments) result in greater dissolved organic C leaching than saline waters during the decomposition of incorporated plant residues (Olsen et al., 1996). Furthermore, it was likely that the biochemistry influenced by the COD dosing treatments was much different because the CW is a complex substrate: soil microbial activity was much greater, and the active organic C pool was the CW organics rather than indigenous organic matter.
Other Soil Root Zone Leachate Chemistry
All treatment percolates throughout the study period had alkaline pHs (Table 5
), probably the result of hydrolytic exchange reactions between exchangeable Na+ and a proton source, either H2CO3 or H2O (McBride, 1994). Exchangeable Na+ certainly was introduced from CW additions. An additional source of exchangeable Na+ would have been past CW applications. Exchangeable Na+ levels in the soil at the end of the study also implicate hydrolytic reactions with exchangeable Na+. The lack of significant differences in SARadj between the DW treatment and the CW treatment percolates during the 8-wk dosing period further supports this interpretation (Table 5). The SARadj did decrease significantly for the DW treatment after the winter rains, however, compared with the SARadj resulting from the two higher CW treatments. No significant (P > 0.05) differences in dissolved Na+, Ca2+, Mg2+, or K+ concentrations in percolates from the DW and the CW treatments were measured (Table 6
). After winter rains, the dissolved cation concentrations were least in percolates from the DW treatment.
|
|
It appears that CW land applications similar to those investigated could result in some Na+ leaching beyond the root zone; however, salinity loading at soil depths below the 61-cm depth of this study were relatively low, even at higher CW loadings (Table 5). Relatively low salinity of percolates from CW treatments, EC 
1.8 dS m–1 (Table 2), after 8 wk of dosing implies that much of the CW ionic species was complexed or sequestered by organic ligands (i.e., fruit acids). In contrast were the percolates of the Na control (SAR 11.4 treatment), where an EC of 1.4 dS m–1 for the applied water resulted in much greater percolate ECs (Table 5).
Concentrations of NO3––N in percolates from CW treatments after the 4th-wk sampling indicated that potential is low for NO3––N 
10 mg L–1 in leachate under conditions similar to the present study (Table 7
). Mean treatment NO3––N concentrations at any sampling time were not significantly different (P > 0.05). The highest NO3––N concentrations were measured following initial dosing with the lower CW treatment (467 kg COD ha–1 d–1). Average NO3––N concentration of the CW was <0.1 mg L–1 (Table 2), while the soil used for the study had an average NO3––N concentration of 8.4 mg kg–1. Probably the greater volume of water associated with the two higher CW COD loading rates contributed to dilution of NO3– sourced from the soil. Additionally, it is reasonable to expect that denitrification may have contributed to some reduction of NO3– from the two higher CW COD treatments during the initial 4-wk dosing period.
|
Concentrations of Fe2+ and Mn2+ in percolate samples provide additional insight into the possibility of reducing conditions in the soil columns during and after CW dosing (Table 6). In general, breakthrough would occur first with Mn2+, followed by Fe2+, if reducing conditions predominated throughout the soil columns (Essington, 2004). Thus, relatively higher concentrations of these cations in percolates would be a reflection of reducing conditions, probably associated with saturated soil conditions.
For all CW treatments, Mn2+ percolate concentrations were <0.1 mg L–1 throughout the CW treatment period and were <0.3 mg L–1 after post-treatment winter rain on the columns. All soluble Fe2+ concentrations remained <0.3 mg L–1 for all treatments until after the winter rains. Following the rain events, the DW treatment resulted in an Fe2+ breakthrough concentration of 38.6 mg L–1. It is possible that the significantly higher percolate Fe2+ concentration associated with this treatment was reflective of colloidal Fe and mobilization of some Fe consequent to the lower pH rainfall addition. Soil analysis for Fe and Mn post-experiment revealed that the highest concentrations of these cations occurred near the soil surface (7.6 ± 2 cm soil depth).
Changes to Soil Chemistry
Post-treatment soil pH, saturated paste extract EC, and exchangeable Na+ were significantly different (P < 0.05) among treatments and soil depths. Soil organic matter (SOM) was not significantly different among treatments and soil depths (Fig. 2
).
|
All soil pHs (Fig. 2) were alkaline and similar to percolate pHs (Table 5). The post-study DW treatment soil had significantly lower pHs than the post-study soil of the other treatments. A trend of increasing pH with depth for the other treatments probably reflects downward movement of soluble species, accompanied by hydrolytic exchange reactions involving exchangeable Na+ and H+ (McBride, 1994). This can also be responsible for the decrease in exchangeable Na+ levels at all depths (Fig. 3 ) with DW. This displacement of Na+ with DW could also explain percolate Na+ concentrations (Table 6) and SARadj values (Table 5) early in the study at levels similar to the CW treatments. Estimated ESPs (data not shown) for the upper two sampled depths were 0.7 and 1.7%, respectively, compared with the initial ESP of 7% (Table 1).
|
All saturated paste extract ECs (Fig. 2) were low compared with the initial soil saturated paste extract ECs and compared with the CW EC. Considerable leaching of soluble salts occurred even in the SAR 11.4 treatment, as evidenced in the percolates (Table 5) and the relatively low saturated paste extract EC through the columns at the end of the study.
Post-experiment soil NO3––N was not significantly different among the treatments (Fig. 2). The highest percolate NO3––N was associated with the 467 kg COD ha–1 d–1 CW treatment (Table 7). The difference, however, between the NO3––N of this treatment and the other treatments was not significant, in part due to the variability of measured concentrations. All soil NO3––N concentrations were less than agronomic levels for forage production and were probably of little environmental significance.
|
CONCLUSIONS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
There was little evidence from this study that leaching of NO3––N and salt to deeper soil depths could cause environmental concerns, even at higher rates of loadings of CW. Hence, actual COD constituents did not appear to be limitations to land application of CW under the loading parameters and environmental conditions of this study. These observations, in conjunction with the chemical results, brings into question the 112 kg BOD ha–1 d–1 guideline for land application of organics presently used in California and elsewhere. Application of these soil column observations to the field would suggest that the guideline cannot be universally applied for prediction of effective biodegradation, the beginning of anoxic conditions, and impacts to plant health. The influence of Na+ in fruit CW on ESP and SAR was noteworthy and warrants attention and cautious land application management under conditions comparable to this study. Alkaline soil pHs could develop as an outcome of hydrolytic reactions in the soil as a consequence of Na added with the CW.
High fruit CW COD land application rates are possible, applying loading parameters used in this study. Outcomes of this study suggest greater loading range flexibility than is currently in practice, given the soil and environmental conditions of this study. A conservative CW land application best management practice would probably propose that COD loadings under these conditions should not, however, greatly exceed 467 kg ha–1 d–1, the rate currently assigned to this CW field land application in California. A best management practice should also ensure maintenance of sufficient soil residence time for COD mineralization to occur. Land application hydraulic rates of CW similar to that reported in this study should be matched to ETc requirements.
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Received for publication February 15, 2007.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. L. Corwin and S. A. Bradford Environmental Impacts and Sustainability of Degraded Water Reuse J. Environ. Qual., September 2, 2008; 37(5_Supplement): S-1 - S-7. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. R. Miller, Y. Rubin, K. U. Mayer, and P. H. Benito Modeling Vadose Zone Processes during Land Application of Food-Processing Waste Water in California's Central Valley J. Environ. Qual., September 2, 2008; 37(5_Supplement): S-43 - S-57. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
