| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Dep. of Civil, Geological, and Mining Engineering, Ecole Polytechnique de Montréal, P.O. Box 6079 Station Centre-Ville, Montreal, Quebec, Canada H3C 3A7
b NSERC Industrial Chair in Site Remediation and Management, Department of Chemical Engineering, Ecole Polytechnique de Montréal, P.O. Box 6079 Station Centre-Ville, Montreal, Quebec, Canada H3C 3A7
* Corresponding author (gerald.zagury{at}polymtl.ca)
Received for publication November 20, 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIAL AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: CCA, chromated copper arsenate • CEC, cation exchange capacity • PEG, polyethylene glycol
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIAL AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Despite its broad use, there is increasing concern about possible environmental contamination from leaching losses of components from CCA-treated wood (Hingston et al., 2001; Rouhi, 2001). In fact, the active ingredients of CCA can leach from wood utility poles in service depending on rainfall, pH of aqueous solutions, and wood species (Cooper, 1994). Other potential sources of damage are spillage, deposition of sludge, and dripping from newly impregnated wood at timber treatment facilities (Lund and Fobian, 1991).
Stilwell and Gorny (1997) evaluated soil contamination with Cr, Cu, and As under seven decks built from CCA pressure-treated wood. They reported moderately elevated Cu (75 mg kg-1), Cr (43 mg kg-1), and As (76 mg kg-1) content compared with the average in control soils. In all cases, Cu was released most readily, followed by As and then Cr. The Cu and Cr content rapidly decreased with depth, while the As was more mobile and tended to persist up to 15 cm in soil depth. Despite its extensive use in applications such as utility poles (especially electric and telephone), there is still little information available on the levels of contamination around CCA and CCA/PEG-treated wood poles. Mortimer (1991) found very low accumulations of Cu, Cr, and As in soil adjacent to CCA-C/PEG-treated utility poles in service from 3 to 36 months. On the other hand, Cooper et al. (1997), who evaluated the soil contamination with Cr, Cu, and As around more than 50 CCA-C/PEG-treated poles, found very high Cu and As content at the ground line, immediately adjacent to the poles. The latter study showed that metal levels in soil dropped rapidly with distance from the pole. Surprisingly, it was found that sand and loam soils had higher contaminant levels than clay soils at the ground line. Furthermore, Cooper et al. (1997) reported generally low CCA component levels in ground water samples taken below CCA-C/PEG-treated poles. In a few cases, levels were above the drinking water standards for all elements. In view of these differing results, the objective of this study was to thoroughly investigate the levels of soil and ground water contamination (Cr, Cu, and As) around various CCA-C/PEG-treated utility poles in service to assess their potential environmental impact. Moreover, additional soil physicochemical properties (pH, volatile solids, total sulfur, total inorganic carbon, buffer capacity, cation exchange capacity, and particle size distribution) left out by previous studies and that might have a strong influence on metal retention were determined. Ground water samples collected near the poles were also analyzed and bioassays with Daphnia magna were conducted to evaluate ground water ecotoxic potential.
|
MATERIAL AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIAL AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
All plastic instruments used for sample collection were cleaned sequentially with a phosphate-free detergent, rinsed with deionized water, then with 10% (v/v) nitric acid, and finally rinsed twice with deionized water. All metallic instruments (picks and shovels) used for excavation at depths of 0.5 and 1 m were rinsed with tap water, cleaned with a detergent, and then rinsed with deionized water (Ministère de l'Environnement et de la Faune du Québec, 1995).
Soil Properties and Metal Content
For each site, particle size distribution was performed with ASTM Methods D1140-92 and D422-63 (American Society for Testing and Materials, 1997a) and the soils were classified with the USDA classification system (gravel [>2 mm], sand [2 mm–50 µm], silt [50–2 µm], and clay [<2 µm]). The pH was measured in distilled water according to Method D-4972-95a (American Society for Testing and Materials, 1997b) with a soil to water ratio of 1:2 using a 8175 BN electrode (Orion Research, Beverly, MA) and Accumet Model 25 pH meter (Fisher Scientific, Hampton, NH). Volatile solids were determined at 550°C according to Karam (1993). Buffer capacity (expressed as the number of moles of H+ ions needed to lower the initial pH of 1 kg of soil by one pH unit) was determined with HNO3 according to Zagury et al. (1997). Cation exchange capacity (CEC) was determined on duplicate samples with the sodium acetate method (pH = 8.2) according to Chapman (1965). A phosphoric acid treatment followed by an infrared determination of CO2 evolved was performed to determine total inorganic carbon (Ministère de l'Environnement et de la Faune du Québec, 1996). Total sulfur was measured by combustion with an induction furnace (LECO Corporation, 1975).
Total metals (Cr, Cu, As) in soils (<2 mm) were determined with inductively coupled plasma atomic emission spectrometry (ICP–AES) (IRIS/Advantage model; Thermo Jarrel Ash, Franklin, MD) after digestion with HNO3 and HCl according to Clesceri et al. (1998). The method detection limit (MDL) was determined following the analysis of 10 replicates of a real soil sample. The arithmetic mean and the standard deviation of the 10 replicates were computed and the MDL (three times the standard deviation) was determined. The method quantification limit (MQL) was then established as five times the MDL. The MQLs for Cr, Cu, and As were 0.2, 0.5, and 5 mg kg-1, respectively. For quality-control purposes a procedure blank, a duplicate sample, a spiked sample, and a reference soil sample (Philip Services Corp., Anjou, QC, Canada) were included in each batch of 15 samples. The Cu, Cr, and As concentrations in all procedure blanks were all below quantification limits. The average relative percent deviation between duplicates was 11% for Cr, 5% for Cu, and 9% for As. These values show that there was a good homogeneity within a given soil sample. The average recoveries for the spiked samples (n = 12) were 100% for Cr (87–107%), 108% for Cu (103–112%), and 99% for As (85–112%). For the reference soil sample (n = 12), the average recoveries were 106% for Cr (104–113%), 106% for Cu (104–112%), and 98% for As (97–101%). The metal concentrations in soil samples were compared with the province of Québec's soil criteria system in which Level A represents background levels of metals naturally found in soils; Level B is the maximum acceptable concentration on residential, recreational, and institutional sites; and Level C is the maximum acceptable concentration on nonresidential commercial and industrial sites (Table 2) (Ministère de l'Environnement du Québec, 1999).
|
Ground Water Characteristics and Metal Concentrations
Total metals (Cr and Cu) in ground water were determined with ICP–AES, while As concentration was determined by ICP–AES–hydride generation (Model 3510ICP; Applied Research Laboratories, Ecublens, Switzerland) according to Clesceri et al. (1998). Alkalinity (H2SO4 titration) was measured according to Clesceri et al. (1998). Hexavalent chromium was determined colorimetrically by reaction with diphenylcarbazide (Ministère de l'Environnement du Québec, 1988). Quantification limits for Cr, Cr(VI), Cu, and As were 0.005, 0.02, 0.005, and 0.001 mg L-1, respectively. For quality control purposes a procedure blank, a reference water sample (Philip Services Corp.) and a spiked sample [for Cr(VI) analysis] were included in each batch of samples. All metal concentrations in all procedure blanks were below quantification limits. The average recovery for the Cr(VI) spiked samples (n = 3) was 100% (99–101%). The average recoveries for the reference sample (n = 4) were 94% for Cr (91–99%), 98% for Cr(VI) (97–100%), 94% for Cu (91–100%), and 101% for As (96–106%). Redox potential, pH, and temperature were measured in situ with an HI 3210S combination platinum redox potential electrode and a HI 1230B combination pH electrode connected to an HI 9025 C pH/mV meter (Hanna Instruments, Woonsocket, RI). According to the manufacturer's specification, the potential developed by the redox potential electrode relative to standard hydrogen electrode at 25°C is 207 mV.
Daphnia magna Toxicity Test
Daphnia magna survival tests were conducted according to the standard procedure of the province of Québec's Ministry of the Environment (Centre d'Expertise en Analyse Environnementale du Québec, 2000). Ground water samples were centrifuged on arrival in the laboratory and tested within 72 h. For each test dilution, 20 neonates (<24 hours old), separated from a stock population, were used in four groups of five. After checking the survival of D. magna in control ground water of each site (sampled 25 m from the pole), it was used as dilution medium (100, 75, 56, 42, and 32% [v/v]). For the negative control, the D. magna were kept in the water used for culture (180 mg CaCO3 L-1). Potassium dichromate was used as a positive control. The tests were performed in 15-mL test vessels, filled with 10 mL of ground water at 20 ± 2°C. After 48 h, the number of dead (microscopic observation of heart beat) and immobilized D. magna was recorded.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIAL AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The chemical properties for each of the six sites are given in Table 3. Despite the very different soil textures and chemical properties, the average pH of the six sites was slightly alkaline (7.66 ± 0.29) with a very low variation between the sites. As expected, sandy soils (Sites 1 and 2) had a lower CEC than clayey soils (Sites 3 and 4) while the organic soils (Sites 5 and 6) showed the highest average CEC value (35.9 cmol kg-1). Generally, the CEC is considered to arise principally from the organic matter and clay fractions. This is because the most negative charges responsible for the CEC originate from the dissociation of carboxyl groups in organic matter molecules and both permanent and variable charges on clay minerals. In this study, organic matter contribution to the CEC was much more important than the clay. Balasoiu et al. (2001) studied the influence of soil properties on retention of Cu, Cr, and As in soils artificially contaminated with CCA and found that organic matter content (peat moss) strongly influenced CEC (232 cmol kg-1) whereas clay contribution to the CEC was less important (38 cmol kg-1). This phenomena was observed on Site 5, which had a lower clay content (6.9%) than Site 4 (47.1%) but had a CEC 2.5 times higher (42.4 cmol kg-1) because of its elevated organic matter content (as measured by volatile solids) of around 30%. McLean and Bledsoe (1992) also reported an organic matter contribution to the CEC higher than 200 cmol kg-1 in surface mineral soils. Because of its high CEC, the soil surrounding the poles on Sites 3 and 5 is therefore expected to retain higher amounts of cationic metals like Cu.
|
In addition to visual and manual examination, the organic matter content of the soils was evaluated by the total volatile solids determination. Total volatile solids in sandy soils ranged from 53 ± 9 to 84 ± 10 g kg-1, whereas they ranged from 209 ± 42 to 306 ± 104 g kg-1 in highly organic soils. Total volatile solids of the six soils were primarily correlated to organic matter content and to a lesser extent to clay content (Sites 3 and 4). Total sulfur content was low (0.5–0.6 g kg-1) on all sites except on Site 4, which contained 1.2 g kg-1 of sulfur. This sulfur content is probably not high enough to generate acidic conditions that may favor metal mobilization, especially considering this site's buffer capacity.
Globally, the slightly alkaline pH combined with the high clay content of soil samples from Sites 3 and 4 and the elevated organic matter content of Sites 5 and 6 should increase the retention of metals in the soil around the poles. On the other hand, the low-organic-matter content, coarse-grained sandy soils from Sites 1 and 2 are not expected to highly retain metals that might leach from CCA-treated poles.
Soil Metals
The Cr, Cu, and As concentration gradients in soil samples collected around the six treated poles in service are listed respectively in Tables 4, 5, and 6. The variable standard deviations highlight the importance of performing the sampling in various axes from the poles to obtain a representative pattern of soil contamination. On all sites, Cu concentrations were the highest followed by As and Cr, even though the element concentration in the CCA-C/PEG treating solution is Cr > As > Cu. These results are consistent with the relative leaching tendencies previously reported (Cooper et al., 1997). For all metals, the highest concentrations were found at the ground line and immediately adjacent to the pole. The contaminant contents (especially Cr and Cu) then decreased considerably with distance and depth. Where comparisons could be made between jack pine and red pine poles (from a limited survey), no definite difference was found. Cooper et al. (1997), who observed higher contaminant levels around jack pine poles, suggested that this difference could be attributed to the larger average size of the jack pine poles sampled rather than a difference in susceptibility to leaching. Generally, contaminant levels varied with soil type rather than age of the pole in service. However, in organic soils all contaminant levels increased with age of the pole (4 vs. 7 yr). A contributing factor could be the presence of CCA extracting organic acids in the organic soils (Cooper and Ung, 1992b). As expected, sandy soils had lower contaminant levels than clayey and organic soils at the ground line. This coherent finding is contrary to the observations of Cooper et al. (1997).
|
|
|
Copper Concentration
As previously mentioned, contaminant levels varied with soil type rather than age of the pole in service, particularly with Cu (Table 5). Copper concentrations were in the same range in the sandy soils of Sites 1 and 2 despite the different ages of the poles (8 and 4 yr, respectively). Furthermore, average Cu contents were lower than Level C (500 mg kg-1) immediately adjacent to the pole at the ground line and decreased below Level A within 0.25 m from the poles. A Cu concentration gradient similar to the one found in sandy soils was found in the coarse-grained (73.5% gravel and sand) organic soil around the treated pole on Site 6. However, Cu levels were very different around the poles on Sites 3, 4, and 5. A Cu concentration of 1900 mg kg-1 (data not shown) was measured in a clayey soil sample at the ground line, immediately adjacent to the pole on Site 3. In fact, average Cu concentrations in soil samples collected at the ground line close to the poles on Sites 3, 4, and 5 largely exceeded 500 mg kg-1, with concentrations two to three times higher than Level C (Table 5). The elevated silt and clay content and the chemical properties (high CEC and high organic matter content) of these soils (Table 3) help explain the very high Cu levels found at the ground line. Copper is known to be mainly retained in soils through exchange and specific adsorption mechanisms and has a high affinity for organic ligands and humic compounds (Alloway, 1990; McLean and Bledsoe, 1992).
Despite the very elevated Cu levels found at the ground line immediately adjacent to the poles, Cu concentration gradients were similar to those for chromium and decreased radically (starting 0.1 m from the pole) with distance and depth. In general, Cu concentrations in the Level A and B range were attained at a depth of 0.5 m. Nevertheless, contrary to Cr, for which background levels were met 0.1 m from the poles, Cu concentrations exceeded the background concentration 0.5 m from the poles in clayey and highly organic soils, but met Level B criteria at this distance.
Arsenic Concentration
The data clearly show that As is the most problematic element of the CCA contaminants (Table 6). Actually, As content at the ground line and close to the pole on all study sites was much higher than Quebec's Level C criterion (50 mg kg-1). Depending on the sites, average As content was three to eight times higher than Level C (up to 410 ± 150 mg kg-1 on Site 5). An As concentration of 560 mg kg-1 was measured in one sample along the 240° sampling axis at the ground line and immediately adjacent to the pole on this site (data not shown). For the first time, the average contaminant concentration exceeded or was up to twice as high as Level C on several sites at a depth of 0.5 m. Furthermore, an average As content of 54 ± 26 mg kg-1 was measured immediately adjacent to the pole (Site 3) at a depth of 1 m compared with the control soil content of 5.7 ± 0.6 mg kg-1 (Table 6). Stilwell and Gorny (1997) investigated soil contamination under decks built from CCA pressure-treated wood and reported that Cu and Cr content rapidly decreased with depth, while As tended to persist up to 15 cm in soil depth. Our results, however, show that As can leach through the soil and persist up to 1 m in soil depth. The known strong interaction of arsenic with iron and aluminum oxides and clay minerals through ion exchange, specific adsorption, or coprecipitation (Moore et al., 2000) or the formation of arsenate-insoluble precipitates with iron, aluminum, and calcium (Cooper, 1994) failed to reduce the potential to leach through the soil along the pole. In fact, the iron content ranges in the sandy, clayey, and organic soils at the ground line were 8000 to 8400, 20600 to 43000, and 29000 to 30300 mg kg-1, respectively. Arsenate (HxAsOx-34) and arsenite (HxAsOx-33) adsorption are strongly pH-dependent and the six study sites had a similar slightly alkaline pH (7.66 ± 0.29). Under these conditions, the majority of pH-dependent surfaces (edges of clay minerals, oxides and hydroxides, acid functional groups of organic matter) should be negatively charged. This helps to explain the overall similar arsenic mobility on Sites 3 through 6. In a recent study on the influence of soil composition (organic matter, sand, and clay content) on retention of Cu, Cr, and As in CCA-C artificially contaminated soils at a constant pH (5.5 ± 0.1), the authors also found similar As retentions in all tested soils despite their very different soil compositions (Balasoiu et al., 2001). Depending on pH and redox potential of the soil environment, arsenic will occur as As(V) and/or as As(III). In general, As(V) predominates at high redox potentials. At a pH < 9.2 and as the redox decreases, the more mobile As(III) becomes the major dissolved As species (Sadiq et al., 1983; Masscheleyn et al., 1991). Knowledge of As speciation is therefore important to fully assess the mobility of this element in soil near CCA-treated poles. Nevertheless, it appears that sandy soils (Sites 1 and 2) had a lower As content at the ground line. Moreover, the soil adjacent to the 8-yr-old pole on Site 1 had the highest measured As content at a depth of 0.5 m, suggesting an increased mobility of As in this sandy soil. This is not surprising since light-textured sandy soils with low clay, organic matter, and iron content do not effectively retain arsenic oxyanions (McLean and Bledsoe 1992). Notwithstanding the very high As levels found immediately adjacent to the poles at the ground line and at a depth of 0.5 m, As concentration gradients (with distance) were similar to the other CCA contaminants. In general, As content in the Level A and B range was attained within 0.25 m from the poles.
Ground Water Characteristics
The chemical properties and metal concentrations of ground water samples collected immediately below the CCA-treated poles and 25 m away from the poles are listed in Table 7. The average in situ pH value observed for ground water under the poles was slightly acidic (5.71 ± 0.56). Higher pH values were obtained in laboratory measurements for these same samples with a mean pH of 6.95 ± 0.22. The discrepancy between in situ and laboratory measurement values can be explained by the higher CO2 partial pressure in the well (and the lower water temperature) compared with that found in the laboratory. Indeed, a higher CO2 concentration leads to a higher carbonic acid content in water (H2CO3) and hence, a lower pH. The redox potential values (ranging from 64 to 246 mV) attest to the fact that the studied ground water represented an oxidizing environment. Sites 4 and 5 exhibited a higher alkalinity (540 and 410 mg L-1 CaCO3) than for the other three sites and, hence, possess a higher buffer capacity.
|
Total As concentrations on Sites 1 to 3 were very low and comparable with the reference samples. On the other hand, the As content was higher on Sites 4 and 5 and the measured concentration under the pole on Site 5 (0.060 mg L-1) exceeded the Quebec criteria for drinking water. It should be kept in mind that the average As content was 410 ± 150 mg kg-1 in the soil collected at the ground line and immediately adjacent to the pole on this site. In July 1999, additional ground water samples were collected on Sites 4 and 5. The measured As concentrations were 0.079 and 0.012 mg L-1, respectively, on Sites 4 and 5. At this time, the As level in ground water under the pole on Site 4 was three times as high as the Quebec criteria for drinking water, whereas the As level on Site 5 was much lower. Hering and Chiu (2000) recently reported variable As concentrations over time in wells of a municipal ground water–based supply system. Seasonal changes in hydrologic conditions or in physicochemical soil conditions influence Fe stability and may directly influence the availability of As. Furthermore, changes in pH and redox conditions can modify the charge of pH-dependent surfaces and, as a result, influence the release of As in ground water. The variability of As concentrations in ground water below CCA/PEG-treated poles is currently under further investigation.
Cooper et al. (1997) reported generally low CCA component levels in ground water samples taken below CCA/PEG-treated poles. In a few cases, levels were above the drinking water standards for all elements, with Cu concentrations being highest. In the current study, As was the only CCA component that exceeded the drinking water criteria.
Ground Water Toxicity
Daphnia magna survival tests showed that none of the ground water samples taken below CCA/PEG-treated poles exhibited a toxic effect. No mortality was recorded in diluted and nondiluted samples, indicating the lack of acute toxicity under the test conditions. Furthermore, no mobility inhibition of D. magna was observed and, consequently, LC50 (concentration estimated to produce mortality in 50% of an organism test population) and EC50 (concentration estimated to inhibit mobility in 50% of an organism test population) values could not be determined. The lack of sensitivity of the test organism is not the reason for the absence of a toxic response, as D. magna proved to be an appropriate test organism for As-contaminated ground water (0.021 mg L-1) collected below CCA/PEG-treated poles (unpublished data, 2001). These results suggest instead that, despite the elevated total As content in ground water under the pole on Site 5 (0.060 mg L-1), the ground water below CCA-treated poles has a low ecotoxic potential.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIAL AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Generally, contaminant levels varied with soil type rather than age of the pole in service. Sandy soils had lower contaminant levels than clayey and organic soils at the ground line. However, all contaminant levels increased with age of the pole in organic soils.
Chromium concentrations in all soil samples were generally lower than 250 mg kg-1 and below the background level 0.1 m from the poles and at a depth of 0.5 m.
Copper concentrations in sandy soils were lower than 500 mg kg-1 immediately adjacent to the pole at the ground line and decreased below Level A (85 mg kg-1) within 0.25 m from the poles. However, Cu content close to the poles on clayey and organic Sites 3, 4, and 5 largely exceeded 500 mg kg-1, with concentrations two to three times higher than Level C (up to 1460 ± 677). The elevated silt and clay content and the chemical properties (high CEC and high organic matter content) of these soils explain the very high Cu levels found at the ground line.
The data show that As is the most potentially problematic element of the CCA contaminants. Average As content close to the pole on all study sites was three to eight times higher than Quebec's Level C criterion (50 mg kg-1) with an average content as high as 410 ± 150 mg kg-1 on Site 5. Furthermore, the average contaminant concentration exceeded or was up to twice as high as Level C on several sites at a depth of 0.5 m. The results also show that As can leach through the soil and persist up to 1 m in soil depth. Notwithstanding the very elevated concentrations, As concentration gradients (with distance) were similar to those of other CCA components. In general, As content in the 5 to 30 mg kg-1 range was attained within 0.25 m from the pole.
Copper and Cr concentrations in ground water samples were always below drinking water criteria and Cr(VI) was always below the quantification limit. Two out of the five ground water samples collected below the treated poles contained As concentrations higher than that obtained for the control sample and one sample exceeded the criterion for drinking water. The results also suggest that the As content in ground water is variable over time. However, D. magna survival tests showed that all ground water samples taken below CCA/PEG-treated poles did not exhibit a toxic effect.
Further studies are underway to assess Cr, Cu, and As partitioning in soil and As speciation in ground water. Bioassays are also being conducted on CCA-contaminated soils using earthworms (Eisenia foetida) and barley (Hordeum vulgare L.).
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIAL AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Related articles in JEQ:
This article has been cited by other articles:
|
|
 |
|
  J. J. Morrell, D. Keefe, and R. T. Baileys Copper, Zinc, and Arsenic in Soil Surrounding Douglas-Fir Poles Treated with Ammoniacal Copper Zinc Arsenate (ACZA) J. Environ. Qual., November 1, 2003; 32(6): 2095 - 2099. [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 | ||||
| Soil Science Society of America Journal | Journal of Plant Registrations | The Plant Genome | |||
in soil samples removed from selected distances and depths around chromated copper arsenate (CCA)–treated poles.