- Email: [email protected]
A mesocosm evaluation of processed drill cuttings for wetland restoration
Ecological Engineering 25 (2005) 41–50
A mesocosm evaluation of processed drill cuttings for wetland restoration Jonathan M. Willis a,c,∗ , Mark W. Hester a , Gary P. Shaffer b a
b
Coastal Plant Sciences Laboratory, Department of Biological Sciences and Pontchartrain Institute for Environmental Science, University of New Orleans, New Orleans, LA 70148, USA Wetland Restoration Laboratory, Department of Biological Sciences, Southeastern Louisiana University, Hammond, LA 70402, USA c Department of Oceanography and Coastal Sciences, Louisiana State University, Baton Rouge, LA 70803, USA Received 23 April 2004; received in revised form 2 January 2005; accepted 28 January 2005
Abstract Many areas of coastal Louisiana are experiencing rapid wetland loss, largely due to an elevated rate of relative sea-level rise. Efforts to restore these deteriorated wetlands often include some form of sediment addition for successful rehabilitation. The earthen materials generated during petrochemical exploration (drill cuttings) may prove to be an ancillary sediment source for restoration efforts, while also reducing the load of oil-field waste. This study assessed two differently processed drill cuttings (DC-A and DC-B), dredge spoil and topsoil (control) in conjunction with three hydrologic regimes and five vegetative conditions (Spartina alterniflora, Spartina patens, Spartina cynosuroides, Avicennia germinans, and an unvegetated control) at four salinity levels to determine sediment characteristics and capacity for supporting wetland plant growth. The DC-A and the dredge treatments demonstrated similar soil physico-chemical characteristics, which fell within acceptable ranges, and were comparable to the control in supporting plant growth. DC-B under fresh conditions demonstrated a high interstitial pH (∼11.0) that became ameliorated (∼8.3) under all saline conditions. Plant photosynthetic response was minimal in DC-B under fresh conditions, but comparable to the control under saline conditions. These results are promising and warrant further evaluations under a wider range of environmental conditions prior to field implementation. © 2005 Elsevier B.V. All rights reserved. Keywords: Wetland restoration; Drill cuttings; Louisiana; Spartina alterniflora; Spartina patens; Spartina cynosuroides; Avicennia germinans
1. Introduction Wetland loss in Louisiana has been occurring at the alarming rate of 40–56 km2 year−1 , the highest in the United States, due to a combination of natural geologic ∗ Corresponding author. Tel.: +1 504 280 1324; fax: +1 504 280 6121.
0925-8574/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.ecoleng.2005.01.018
processes and human influences (Salinas et al., 1986; Penland and Ramsey, 1990; Britsch and Dunbar, 1993; Boesch et al., 1994; Day et al., 2000). Subsidence due to dewatering and soil compaction and sediment deprivation stemming from anthropogenic activities (e.g. leveeing of the Mississippi River) in conjunction with increasing rates of global sea-level rise have resulted in elevated rates of relative sea-level rise (T¨ornqvist et al.,
42
J.M. Willis et al. / Ecological Engineering 25 (2005) 41–50
2002). This elevated rate of relative sea-level rise acts to increase stresses associated with flooding on emergent vegetation, eventually leading to open water habitat as the plants die and their stabilizing effect on the substrate is lost (Baumann et al., 1984; Mendelssohn and McKee, 1988; Nyman et al., 1993; DeLaune et al., 1990; Wilsey et al., 1992; Day et al., 1995; Mitsch and Gosselink, 2000). Thus, wetland restoration efforts within coastal Louisiana must often incorporate increasing marsh elevation through some form of sediment addition as a key component. The beneficial effects of such sediment additions in improving soil physico-chemical properties and enhancing plant productivity have been demonstrated in numerous studies (DeLaune et al., 1990; Wilsey et al., 1992; Webb et al., 1995; Mendelssohn and Kuhn, 2003). Controlled river diversions and targeted placement of dredged sediment are examples of current wetland restoration techniques in Louisiana that incorporate sediment addition (e.g. Lane and Day, 1999; Streever, 2000; Edwards and Proffitt, 2003). Drill cuttings generated in association with water-based drilling muds (fluids) may represent an ancillary sediment source that, if shown to be safe and efficacious, could aid in restoration efforts and reduce the oil-field waste load in landfills. Drill cuttings are a mixture of the earthen materials generated during the excavation of an oil or gas well and drilling fluids or “muds,” which are additives that lubricate the drilling bit, stabilize the borehole and lift cuttings out of the borehole (Gray and Darley, 1980; Scholten et al., 2000). Drill cuttings are traditionally treated onsite by physically separating the cuttings from the drilling fluids. The drilling fluids are then recycled for further drilling operations, whereas the cuttings and the residual muds generated in near-shore waters are disposed of as oil-field waste in landfills. This process generates the DC-A substrate employed in this study. An extended processing technique has been developed by SWACO Inc. (proprietary information of SWACO, Lafayette, LA) that can be used to further treat the remaining cuttings by the addition of stabilizing agents, which physically isolate the cuttings, and thereby any associated contaminants, in a silica matrix for greater environmental safety. This process generates the DC-B substrate used in this study. Kelley and Mendelssohn (1995) demonstrated that drill cuttings that had undergone this extensive
processing could, when amended with organic matter, support emergent wetland vegetation. Shaffer et al. (1996, 1998) examined drill cuttings that had been subjected to physical separation only as well as physical separation followed by isolation within a silica matrix. Shaffer et al. (1996, 1998) determined that the drill cuttings subjected to physical separation (DC-A) were capable of supporting vigorous wetland plant growth, but that plants grown in drill cuttings subjected to physical separation followed by silica isolation (DC-B) demonstrated minimal growth. The poor response of vegetation grown on this substrate was attributed to both the high pH of the interstitial waters of this substrate (mean pH ∼11.0) and the coarse texture of the soil (Shaffer et al., 1996, 1998). Nonetheless, the successful colonization and the high productivity of vegetation in the drill cuttings subjected to only physical separation indicate that drill cuttings have the capacity to support the growth of wetland vegetation, at least under freshwater conditions. Although the studies of Shaffer et al. (1996, 1998) are encouraging, no data are available on the response of these drill cuttings to saline waters or the growth of coastal wetland plants in these processed drill cuttings under saline conditions. The objectives of this study are two-fold: (1) to elucidate the physicochemical response, particularly interstitial water metal concentration, of two differently processed drill cuttings to a range of hydrologic regimes and salinity levels possible in a coastal wetland restoration scenario and (2) to evaluate the capacity of processed drill cuttings to support the survival and productivity of several coastal wetland plant species under various combinations of hydrologic regimes and salinity levels that may be encountered in coastal restoration scenarios. If processed drill cuttings can be shown to cause no environmental hazards, support the growth of wetland vegetation, and have a cost comparable to current oil-field waste disposal costs, and then the implementation of this method as an additional restoration strategy warrants consideration. 2. Materials and methods 2.1. Study facility The mesocosm facility employed for this study consists of 144 mesocosm vessels (200-l, polyethylene
J.M. Willis et al. / Ecological Engineering 25 (2005) 41–50
units) networked to four salinity reservoirs within a temperature-controlled greenhouse. An air lift water delivery system was devised to bring water into the mesocosms as described in Section 2.3. 2.2. Substrate type Each treatment mesocosm was filled to approximately 3/4 volumes with one of the four test substrates. The drill cutting materials used in the experiment were generated in Grand Bay, Louisiana (a tidally influenced, oligohaline marsh area). All drill cuttings were separated from the drilling fluids, which resulted in the DC-A substrate (processing accomplished by Cameron Corp.). Some of this material was subjected to further processing, resulting in the DC-B substrate (processing accomplished by SWACO Corp.). Drill cuttings substrates (DC-A and DC-B) and dredge spoil (generated in the same general area as the drill cuttings) were transported to the SLU mesocosm facility courtesy of Greenhill Petroleum Company.
43
local brackish marsh. Avicennia germinans (black mangrove) plants were purchased from a horticultural supplier. Three genotypes each of Spartina patens (wiregrass) and Spartina alterniflora (oystergrass) were selected to represent a range of salt tolerance based on previous research (Hester et al., 1996, 1998). Individual plants were rinsed to remove parent soil and then randomly allocated to the appropriate treatment and allowed to acclimate for approximately 2 weeks. 2.5. Salinity levels Salinity was sequentially increased from 0 to 9, to 18, and then to 36 ppt for all treatment vessels over approximately a 1-year period using Coralife scientific grade sea–salt. The durations of the salinity treatments were 4 months for the 0–9 ppt salinity levels, and 2 months for the 18–36 ppt salinity levels. All salinity increases were accomplished over a 2-week period to allow plants time for acclimation. Salinity levels of each treatment were monitored monthly with a salinity meter (YSI Model #30 salinity/conductivity meter) and adjusted accordingly.
2.3. Hydrology design 2.6. Plant photosynthetic response The mesocosm treatment vessels were networked to four, 3000-l reservoirs in a substrate-specific design that prevented cross contamination of water from different substrate sources. Water was brought into each mesocosm daily for 4 h by an air lift system, which aerated the water from gravity-fed reservoir waterlines as it was lifted by pumped air bubbles above the reservoir water level into each mesocosm vessel (see Shaffer et al. (1996) for further discussion of air lift system design). A system of PVC pipes and submersible pumps recycled the water in a substrate-specific fashion and created the following three hydrologic regimes: mesic (moist, but allowed to drain to 10 cm below the soil surface during daily water input), tidal (allowed to flood to 20 cm above the soil surface and drain to 10 cm below the soil surface during the daily tidal cycle) and flooded (flooded to 20 cm above the soil surface at all times). 2.4. Plant treatments Transplants of Spartina cynosuroides (big hogcane) were collected from the Bayou Lacombe area, a
Plant photosynthetic response was determined toward the end of each salinity level increase using a LICOR 6400 Photosynthesis System (LI-COR, Lincoln, NE, USA) with a built-in light source and CO2 controller with reference settings of 1500 mol m−2 s−1 and CO2 levels of 370 ppm. Measurements were performed on two representative leaves (first or second fully expanded leaf from top of stem) selected from each test plant and then averaged to provide one photosynthetic response value for that plant. 2.7. Substrate redox potential Substrate redox potential (Eh) was monitored toward the end of each salinity level increase using a pH/mV meter (Hanna model HI 9025) and brightened platinum redox electrodes that were constructed, calibrated and brightened as per Faulkner et al. (1989). Redox probes were placed at 2 and 15 cm depths below the substrate surface in each vessel, allowed to equilibrate overnight, and redox potentials determined the following day.
44
J.M. Willis et al. / Ecological Engineering 25 (2005) 41–50
2.8. Interstitial water analysis Interstitial water pH was monitored with a pH/mV meter (Hanna model HI 9025). Measurements were taken by placing the pH probe into a permanent interstitial well (slotted 1.875 cm PVC pipe extended 10 cm below soil surface) within each treatment vessel. Interstitial water samples for elemental analysis were collected by drawing fluid with a syringe from interstitial water wells. Samples were filtered (0.45 m), acidified with concentrated, reagent-grade nitric acid and subjected to elemental analysis by inductively coupled argon plasma mass-optical emission spectrophotometery (Jarrel-Ash ICP-OES Atom Comp Series 800), except for Ba, which was subjected to cold vapor atomic absorption (Perkin-Elmer Model 5000 spectrophotometer). Of the elemental concentrations obtained, statistical analysis and interpretation were limited to major toxic heavy metals (Cd, Cr, Pb and Ni, see Soil and Plant Analysis Council (1999)), elements thought to be elevated in processed drill cuttings due to additives (Ba and Al), and the macronutrients P and K.
5000). Target elements were the same as those chosen for interstitial water analysis (Al, Ba, Cd, Cr, K, Ni, P and Pb). 2.11. Statistical analysis Repeated measure ANOVAs, with salinity level treated as the repeat factor (Girden, 1992; Von Ende, 2001), were used to analyze all dependent variables that were measured at each salinity level (redox potential, interstitial pH, interstitial elemental concentration and photosynthetic response). Univariate ANOVAs were employed to test for differences within a salinity level when a significant interaction with salinity level was detected (Von Ende, 2001). Final biomass and plant tissue metal concentrations were analyzed using univariate ANOVAs. All data were tested for meeting the criteria of parametric analysis. Repeated measures ANOVAs were adjusted for departures from sphericity using the Hunyh–Feldt technique (Girden, 1992; Von Ende, 2001). All statistical analyses were performed using the appropriate procedure of SYSTAT version 8.0 (SPSS, 1998).
2.9. Biomass measurements Plant biomass was harvested at the end of the 36 ppt salinity level increment. Biomass was separated into aboveground alive, aboveground dead and belowground partitions, lightly rinsed to remove substrate and placed into paper bags at harvest. All biomass was dried in a plant-drying oven at 60 ◦ C until constant weight was achieved. 2.10. Plant tissue analysis A subsample of dried aboveground biomass (1 g of stem and leaf) was obtained from all vegetated test vessels in the mesic and flooded hydrology treatments. Plant material was ground in a Wiley mill and digested in 5 ml of concentrated, reagent-grade nitric acid in a block heater at 130 ◦ C for 3 h. Samples were allowed to cool overnight, brought up to 70 ml total volume with distilled–deionized water, inverted several times, and the supernatant collected for analysis The samples were then analyzed with an inductively coupled argon plasma mass-optical emission spectrophotometer (Jarrel-Ash ICP-OES Atom Comp Series 800) or cold vapor atomic absorption (CVAA Perkin-Elmer Model
3. Results 3.1. Elemental analysis (interstitial water) The results of substrate interstitial water analysis on all four substrates (topsoil, dredge spoil, DC-A and DC-B) at the end of the 0–36 ppt salinity level exposures are presented in Table 1 (for brevity the 9–18 ppt salinity treatments are omitted). There was a significant trend toward increases in the concentration of Ba, Pb and Ni, in the 36 ppt salinity level compared with the 0 ppt salinity level (Table 1; all P < 0.05). This trend was the most pronounced in DC-A and topsoil, and the least pronounced in DC-B. A significant interaction of salinity by substrate was present for interstitial Ni concentration, likely stemming from the greater concentration of interstitial Ni in the topsoil and DC-A treatments compared to the dredge spoil and DC-B substrates in the 36 ppt treatment, while no difference was evident in the 0 ppt treatment (Table 1; F = 3.217, P = 0.034). Interestingly, Al concentration actually decreased significantly in the 36 ppt salinity level compared with the 0 ppt salinity level (Table 1; F = 7.887, P = 0.002).
J.M. Willis et al. / Ecological Engineering 25 (2005) 41–50
45
Table 1 Results of elemental analysis of substrate interstitial water at the end of the 0 and 36 ppt salinity levels Element
Substrate Topsoil
Ba Cd Pb Cr Ni Al
Dredge
DC-A
DC-B
0 ppt
36 ppt
0 ppt
36 ppt
0 ppt
36 ppt
0 ppt
36 ppt
0.041 ± 0.013 BD ± 0.000 0.014 ± 0.014 BD ± 0.000 BD ± 0.000 0.555 ± 0.235
1.003 ± 0.081 0.010 ± 0.000 0.169 ± 0.075 BD ± 0.000 0.044 ± 0.033 0.710 ± 0.270
0.069 ± 0.009 0.007 ± 0.005 0.035 ± 0.012 0.003 ± 0.002 0.007 ± 0.005 0.889 ± 0.176
1.132 ± 0.276 0.018 ± 0.016 0.090 ± 0.022 0.001 ± 0.001 0.011 ± 0.004 0.222 ± 0.147
0.115 ± 0.016 0.003 ± 0.003 0.006 ± 0.004 0.001 ± 0.001 BD ± 0.000 1.020 ± 0.360
0.931 ± 0.126 0.014 ± 0.007 0.261 ± 0.189 0.002 ± 0.001 0.055 ± 0.029 1.177 ± 0.392
0.100 ± 0.012 0.005 ± 0.003 0.005 ± 0.005 BD ± 0.000 BD ± 0.000 7.569 ± 1.466
1.346 ± 3.290 BD ± 0.000 0.132 ± 0.041 0.001 ± 0.001 0.001 ± 0.020 0.096 ± 0.055
Values are mean concentrations (g g−1 ) ± standard errors (n = 36).
3.2. Elemental analysis (vegetation)
3.4. pH
Results of elemental analysis of acid-digested plant material are presented in Table 2. No significant effects of substrate, hydrology, species, or the interactions thereof were detected in plant tissue concentrations of Cd, Pb, Cr, Ni or Al. However, plants grown in the DC-A and DC-B substrates had higher concentrations of Ba than those grown in topsoil and dredge spoil (Table 2; contrast F = 6.32, P = 0.020).
There was a significant effect of salinity level and substrate type on pH (Fig. 2; F = 49.45, P < 0.001). As the prior studies indicated, the pH of DC-B was highly elevated relative to the other substrates under freshwater conditions. However, the pH of DC-B was significantly lower under all saline treatments than under freshwater conditions (Fig. 2; contrast F = 59.88, P < 0.001).
3.3. Soil redox potential
3.5. Photosynthetic response
Soil redox potential at 2 and 15 cm depths displayed a significant decrease with increasing salinity levels (Fig. 1; F = 108.74, P < 0.001 and F = 177.37, P < 0.001, respectively). A marginally significant salinity level by hydrologic regime interaction was detected, indicating that there was a tendency for the effect of hydrology to vary over time as salinity levels were increased (Fig. 1; F = 2.28, P = 0.038).
There was a significant effect of salinity level on net CO2 assimilation rate (Fig. 3; F = 14.43, P < 0.001). Analysis within salinity levels showed that at the 0 ppt salinity level, plants had significantly reduced net CO2 assimilation rates when grown in DC-B compared to the other substrates (Fig. 3; contrast F = 47.813, P < 0.001). This was probably due to the elevated pH that DC-B exhibits under freshwater conditions. No
Table 2 Results of elemental analysis of aboveground plant tissue at the end of the 36 ppt salinity level Element
Substrate Topsoil
Ba Cd Pb Cr Ni Al
6.030 0.280 0.630 0.440 1.470 49.90
± ± ± ± ± ±
Dredge 1.656 0.280 0.630 0.298 1.470 28.69
Values are means (g g−1 ) ± standard errors (n = 36).
6.728 0.219 0.814 0.429 1.689 73.237
DC-A ± ± ± ± ± ±
2.724 0.219 0.536 0.216 0.862 21.20
22.15 0.220 0.000 0.250 1.250 86.4
DC-B ± ± ± ± ± ±
6.917 0.220 0.000 0.162 0.813 20.76
20.25 0.411 0.971 0.560 1.146 131.6
± ± ± ± ± ±
7.803 0.270 0.660 0.281 0.751 60.91
46
J.M. Willis et al. / Ecological Engineering 25 (2005) 41–50
Fig. 2. The effect of salinity level and substrate type on interstitial pH (mean ± standard error) averaged across hydrologic regime and plant species.
level permanently flooded treatments were found to have lower net CO2 assimilation rates than tidal and mesic treatments (contrast F = 16.515, P < 0.001). These data indicate that drill cuttings are capable of supporting net CO2 assimilation rates comparable to topsoil, providing that the pH of DC-B is ameliorated. Overall, net CO2 assimilation rates were greatest in the 9 ppt salinity level. 3.6. Total biomass Substrate type, hydrologic regime, and plant species all had marginally significant effects on total plant biomass (Fig. 4; F = 1.77, P = 0.05), which appeared to be driven by the high performance of S. cynosuroides under all hydrologies in DC-B, and the high
Fig. 1. The effect of hydrologic regime and salinity level on soil redox (mean ± standard error) averaged across substrate type and plant species.
significant differences were detected among substrate types at the 9 or 18 ppt salinity levels. Also, at the 36 ppt salinity level plants grown in DC-A had significantly greater net CO2 assimilation rates compared with all other substrates (Fig. 3; contrast F = 9.125, P = 0.003). No significant differences were detected in net CO2 assimilation among substrate types at the 9 or 18 ppt salinity levels. Within the 0 ppt salinity
Fig. 3. The effect of salinity level and substrate type on net CO2 assimilation rate (mean ± standard error) averaged across hydrologic regime and plant species.
J.M. Willis et al. / Ecological Engineering 25 (2005) 41–50
47
forms (lower biomass of A. germinans). The main effects of substrate type and vegetative species, as well as the three-way interaction between substrate, hydrology and species, regarding aboveground biomass were significant (contrast F = 54.95, P < 0.001; F = 41.55, P < 0.001 and F = 2.90, P = 0.002, respectively). Also, a significant difference in belowground biomass was detected among species (F = 5.00, P = 0.005). All responses of total, aboveground and belowground biomass were consistent, therefore only the total biomass graphs are presented.
4. Discussion 4.1. Edaphic conditions
Fig. 4. The effect of substrate type, hydrologic regime and plant species on total biomass (mean ± standard error).
performance of S. patens in DC-B under mesic conditions. This limits the interpretation of the main effect of substrate type, but a contrast indicates that DC-B total biomass was significantly reduced compared to all other substrates (Fig. 4; contrast F = 8.40, P = 0.006). No significant differences were detected among hydrologies in terms of total biomass production. A significant effect of species was discerned (Fig. 4; F = 7.86, P = 0.001), but likely simply reflects the difference between herbaceous versus woody growth
As one would expect, substrate redox potentials became significantly more reduced as salinity levels were increased over time, likely reflecting the presence of alternative electron acceptors such as sulfate in the synthetic sea–salt mix and the increase in duration of flooding (Mitsch and Gosselink, 2000). Importantly, this study has allowed for the quantification of the behavior of the metal constituents in both the DC-A and DC-B substrates over a large range of substrate reduction (+400 to −100). Substrate interstitial pH in the DC-B substrate was dramatically lowered from extreme alkaline values (∼10.1) to moderate pH values (∼8.5) as salinity levels increased. The strong carbonate buffering capacity of seawater, which can buffer both acidic and basic systems was likely very important in this response (Dietrich, 1980). This buffering is of great significance because the high alkalinity demonstrated by the DC-B substrate was considered one of the greatest obstacles in regard to it serving as a substrate for the successful establishment of wetland vegetation. Interestingly, this pH-salinity phenomenon was not noted by Kelley and Mendelssohn (1995). However, this apparent discrepancy is likely a scale effect due to the large water reserves (3000-l) employed in this study, which provided a larger buffering capacity to ameliorate the high pH of the DC-B substrate. Previous drill cuttings research indicated that the use of processed drill cuttings to create wetlands may require additional amelioration of the high pH to be successful, which would reduce cost-effectiveness of
48
J.M. Willis et al. / Ecological Engineering 25 (2005) 41–50
the method (Kelley and Mendelssohn, 1995; Shaffer et al., 1996; DesRoches, 1998). However, an implication of this study is that if these processed drill cuttings are used for coastal wetland restoration where natural saline waters are present, and then there may be no need to amend the alkalinity of the substrate with organic materials as was previously thought necessary (Kelley and Mendelssohn, 1995). For the majority of heavy metals in aquatic systems, maximum potential release occurs at moderately low and extremely high pHs, and under slightly reduced or oxidized conditions (Fergusson, 1990; Bourg and Loch, 1995). The conditions reported by Gambrell et al. (1991) for minimal solubility of a suite of heavy metals (pH 8.0 and soil redox potential of −150 to −100 mV) are similar to the environmental conditions in the mesocosm at the 18 and 36 ppt salinity levels. Typical values for pH and Eh that would be expected in natural coastal systems were reached during this experiment under fresh and saline conditions, with no significant differences from the control (i.e. topsoil) detected in any of the metals of concern. Examination of aboveground plant tissue indicated significantly higher levels of Ba in the DC-A substrate (22.15 g g−1 ) and DC-B substrate (20.25 g g−1 ) compared to topsoil (6.03 g g−1 ) and dredge spoil (6.73 g g−1 ). However, normal levels of Ba in foliar tissue are considered to be between 10 and 50 g g−1 (Chaudhry et al., 1977; also see references in Cipollini and Pickering, 1986). Therefore, the levels of Ba in plant tissue grown on the two processed drill cutting substrates, although elevated compared to topsoil and dredge spoil, are not exceptionally high. No significant differences among substrates were detected for any other metals of concern (Cd, Cr, Pb and Ni).
Net CO2 assimilation rates under freshwater conditions indicated significantly lower plant performance when plants were grown on the DC-B substrate, corroborating the previous studies. However, a significant interactive effect of salinity level and substrate was detected, in which plants grown on the DC-B substrate had rates of net CO2 assimilation comparable to other substrates once salinity levels were increased and pH dropped to a more moderate level. Although analysis of plant biomass would indicate that overall plants were performing significantly worse in the DC-B substrate than in other substrates, one must take into account that biomass values were influenced by the time spent at the 0 ppt salinity level. No significant differences between substrate were detected in plant net CO2 assimilation rates under saline conditions, indicating that after 2 years in a simulated wetland environment drill cuttings processed by these techniques are suitable for supporting wetland plant growth in restoration projects. Two of the most wide-spread brackish and salt marsh plant species of Louisiana, S. alterniflora and S. patens, exhibited excellent growth, both in terms of net CO2 assimilation and total biomass, indicating that they would be good candidates for these types of wetland rehabilitation projects. Although no differences were detected in the performance of S. cynosuroides when grown on processed drill cuttings or on topsoil or dredge spoil, this species may not be as viable a candidate for creating wetlands in this fashion, because it exhibited high tissue mortality and relatively low photosynthetic rates throughout the experiment regardless of substrate type, possibly because of a low tolerance to transplant stress. Because S. alterniflora and S. patens are widespread throughout Atlantic and Gulf Coast marshes (Godfrey and Wooten, 1979) the results of this project may have relevance to many coastal areas.
4.2. Vegetation performance The two prime indicators of plant growth used in this experiment were net CO2 assimilation rate (photosynthesis) and biomass production. Net CO2 assimilation rates were measured at each salinity level, and can thus be used to evaluate plant responses to substrates and hydrologic regimes at different salinity levels. Biomass was harvested at the conclusion of the experimental period, and therefore represents the cumulative effects of substrate and hydrology across all salinity levels through time.
5. Conclusion In summary, this study has important implications for the use of processed drill cuttings in the rehabilitation of coastal wetlands. Previous studies on processed drill cuttings have indicated that the potential for their utilization in wetland rehabilitation existed, but that plant performance was often reduced from that observed on other substrates, or there were concerns about the safety of the processed material under various
J.M. Willis et al. / Ecological Engineering 25 (2005) 41–50
environmental scenarios. This study has illustrated that processed drill cuttings have the potential to serve as suitable and non-toxic substrates capable of supporting high rates of wetland plant growth under a range of environmental conditions. However, if processed drill cuttings are to become a viable option as a substrate source in wetland rehabilitation, we recommend that future research focus on further refinement in the formulation of environmentally safe, water-based drilling fluids, because many of the concerns in utilizing drill cuttings stem from the materials in the drilling fluids. Thus, rather than attempt to further develop encapsulation techniques to minimize potential contaminant release from the residual drilling fluid coating on the cuttings, it may prove more expedient to formulate water-based drilling fluids with wetland rehabilitation in mind, such that these drilling fluids are both environmentally safe and conducive to wetland plant growth.
Acknowledgements We would like to thank John Ford of the National Energy Technology Laboratory and John Veil of Argonne National Laboratory for all of their assistance throughout this project and the US Department of Energy who provided funding for this research.
References Baumann, R.H., Day Jr., J.W., Miller, C.A., 1984. Mississippi deltaic wetland survival: sedimentation versus coastal submergence. Science 224, 1093–1095. Boesch, D.F., Josselyn, M.N., Mehta, A.J., Morris, J.T., Nuttle, W.K., Simenstad, C.A., Swift, D.J.P., 1994. Scientific assessment of coastal wetland loss, restoration and management in Louisiana. J. Coastal Res. Spec. Issue 20, 103. Bourg, A.C.M., Loch, J.P.G., 1995. Mobilization of metals as affected by pH and redox conditions. In: Salomons, W., Stigliani, W.M. (Eds.), Biogeodynamics of Pollutants in Soils and Sediments: Risk Assessment of Delayed and Non-linear Responses. Springer-Verlag, Berlin, Germany, 352 pp. Britsch, L.D., Dunbar, J.B., 1993. Land loss rates: Louisiana coastal plain. J. Coastal Res. 9, 324–338. Chaudhry, F.M., Wallace, A., Mueller, R.T., 1977. Barium toxicity in plants. Commun. Soil Sci. Plant Anal. 8, 795–797. Cipollini, M.L., Pickering, J.L., 1986. Determination of the phytotoxicity of barium in leach-field disposal of gas well brines. Plant and Soil 92, 159–170.
49
Day Jr., J.W., Pont, D., Hensel, P.F., Ibanez, C., 1995. Impacts of sealevel rise on deltas in the Gulf of Mexico and the Mediterranean: the importance of pulsing events to sustainability. Estuaries 18, 636–647. Day Jr., J.W., Shaffer, G.P., Britsch, L.D., Reed, D.J., Hawes, S.R., Cahoon, D., 2000. Pattern and process of land loss in the Mississippi Delta: a spatial and temporal analysis of wetland habitat change. Estuaries 23, 425–438. DeLaune, R.D., Pezeshki, S.R., Pardue, J.H., Whitcomb, J.H., Patrick Jr., W.H., 1990. Some influences of sediment addition to a deteriorating salt marsh in the Mississippi River deltaic plain: a pilot study. J. Coastal Res. 6, 181–188. DesRoches, D.J., 1998. Restored Drill Cuttings: A Mesocosm Approach to Determine the Suitability as a Substrate for the Wetlands Restoration and Creation. Masters Thesis. Department of Biological Sciences, Southeastern Louisiana University, Hammond, LA, USA, 105 pp. Dietrich, G., 1980. General Oceanography: An Introduction. Wiley, New York, 626 pp. Edwards, K.R., Proffitt, C.E., 2003. Comparison of wetland structural characteristics between created and natural salt marshes in southwest Louisiana, USA. Wetlands 23, 344–356. Gambrell, R.P., Wiesepape, J.B., Patrick Jr., W.H., Duff, M.C., 1991. The effects of pH, redox, and salinity on metal release from a contaminated sediment. Water Air Soil Pollut. 57, 359–367. Girden, E.R., 1992. ANOVA Repeated Measures. Sage Publications, Newbury Park, CA, 77 pp. Godfrey, R.K., Wooten, J.W., 1979. Aquatic and Wetland Plants of the South eastern United States: Monocotyledons. University of Georgia Press, Athens, GA, USA, 712 pp. Gray, G.R., Darley, H.C.H., 1980. Composition and Properties of Oil Well Drilling Fluids, fourth ed. Gulf Publishing Co., Houston, TX, 630 pp. Faulkner, S.P., Patrick Jr., W.H., Gambrell, R.P., 1989. Field techniques for measuring wetland soil parameters. J. Soil Sci. Soc. Am. 53, 883–890. Fergusson, J.E., 1990. The Heavy Elements Chemistry, Environmental Impact and Health Effects. Pergamon Press, Elmsford, NY, USA, 614 pp. Hester, M.W., Mendelssohn, I.A., McKee, K.L., 1996. Intraspecific variation in salt tolerance and morphology in the coastal grass Spartina patens (Poaceae). Am. J. Bot. 83, 1521–1527. Hester, M.W., Mendelssohn, I.A., McKee, K.L., 1998. Intraspecific variation in salt tolerance and morphology in Panicum hemitomon and Spartina alterniflora. Int. J. Plant Sci. 159, 127–138. Kelley, S., Mendelssohn, I.A., 1995. An evaluation of stabilized, water-based drill cuttings and organic compost as potential substrate sources for marsh restoration and creation in coastal Louisiana. Ecol. Eng. 5, 497–517. Lane, R.R., Day Jr., J.W., 1999. Water quality analysis of a freshwater diversion at Caernarvon, Louisiana. Estuaries 22, 327–336. Mendelssohn, I.A., Kuhn, N.L., 2003. Sediment subsidy: effects on soil–plant responses in a rapidly submerging coastal salt marsh. Ecol. Eng. 21, 115–129. Mendelssohn, I.A., McKee, K.L., 1988. Spartina alterniflora dieback in Louisiana: time-course investigation of soil waterlogging effects. J. Ecol. 76, 509–521.
50
J.M. Willis et al. / Ecological Engineering 25 (2005) 41–50
Mitsch, W.J., Gosselink, J.G., 2000. Wetlands, third ed. Van Nostard Reinhold, NY, USA, 920 pp. Nyman, J.A., DeLaune, R.D., Roberts, H.H., Patrick Jr., W.H., 1993. Relationship between vegetation and soil formation in a rapidly submerging coastal marsh. Mar. Ecol. Prog. Ser. 96, 269– 279. Penland, S., Ramsey, K.E., 1990. Relative sea-level rise in Louisiana and the Gulf of Mexico. J. Coastal Res. 6, 323– 342. Salinas, L.M., DeLaune, R.D., Patrick Jr., W.H., 1986. Changes occurring along a rapidly submerging coastal area. J. Coastal Res. 2, 269–284. Scholten, M.C.Th., Karman, C.C., Huwer, S., 2000. Ecotoxicological risk assessment related to chemicals and pollutants in off-shore oil production. Toxicol. Lett. 112, 283–288. Shaffer, G.P., Hester, M.W., Childers, G.W., 1996. Restored Drill Cuttings for Wetlands: Year One Results of a Mesocosm Approach to Emulate Field Conditions Under Varying Hydrologic Regimes. National Petroleum Technology Program, U.S. Department of Energy, Technical Report, 46 pp. Shaffer, G.P., Hester, M.W., Miller, S., DesRoches, D.J., Souther, R.F., Childers, G.W., Campo, F.M., 1998. Restored Drill Cuttings for Wetlands: Results of a Two Year Mesocosm Approach to Emulate Field Conditions Under Varying Hydrologic Regimes.
National Petroleum Technology Program, U.S. Department of Energy, Technical Report, 86 pp. SPSS, 1998. SYSTAT for Windows. SPSS Science Marketing Department, Chicago, IL, USA, 1086 pp. Soil Plant Analysis Council, 1999. Soil Analysis Handbook of Reference Methods. CRC Press, Boca Raton, FL, USA, 264 pp. Streever, W.J., 2000. Spartina alterniflora marshes on dredged material: a critical review of the ongoing debate over success. Wetlands Ecol. Manag. 8, 295–316. T¨ornqvist, T.E., Gonz´alez, J.L., Newsom, L.A., Van der Borg, K., De Jong, A.F.M., 2002. Reconstructing “background” rates of sealevel rise as a tool for forecasting coastal wetland loss, Mississippi Delta. EOS 83, 530–531. Von Ende, C.N., 2001. Repeated-measure analysis: growth and other time-dependent measures. In: Scheiner, S.M., Gurevitch, J. (Eds.), Design and Analysis of Ecological Experiments. Oxford University Press, Oxford, UK, 415 pp. Webb, E.C., Mendelssohn, I.A., Wilsey, B.J., 1995. Causes for vegetation dieback in a Louisiana salt marsh: a bioassay approach. Aquatic Bot. 51, 281–289. Wilsey, B.J., McKee, K.L., Mendelssohn, I.A., 1992. Effects of increased elevation and macro- and micronutrient additions on Spartina alterniflora transplant success in salt marsh die-back areas in Louisiana. Environ. Manag. 16, 505–511.
A mesocosm evaluation of processed drill cuttings for wetland restoration Jonathan M. Willis a,c,∗ , Mark W. Hester a , Gary P. Shaffer b a
b
Coastal Plant Sciences Laboratory, Department of Biological Sciences and Pontchartrain Institute for Environmental Science, University of New Orleans, New Orleans, LA 70148, USA Wetland Restoration Laboratory, Department of Biological Sciences, Southeastern Louisiana University, Hammond, LA 70402, USA c Department of Oceanography and Coastal Sciences, Louisiana State University, Baton Rouge, LA 70803, USA Received 23 April 2004; received in revised form 2 January 2005; accepted 28 January 2005
Abstract Many areas of coastal Louisiana are experiencing rapid wetland loss, largely due to an elevated rate of relative sea-level rise. Efforts to restore these deteriorated wetlands often include some form of sediment addition for successful rehabilitation. The earthen materials generated during petrochemical exploration (drill cuttings) may prove to be an ancillary sediment source for restoration efforts, while also reducing the load of oil-field waste. This study assessed two differently processed drill cuttings (DC-A and DC-B), dredge spoil and topsoil (control) in conjunction with three hydrologic regimes and five vegetative conditions (Spartina alterniflora, Spartina patens, Spartina cynosuroides, Avicennia germinans, and an unvegetated control) at four salinity levels to determine sediment characteristics and capacity for supporting wetland plant growth. The DC-A and the dredge treatments demonstrated similar soil physico-chemical characteristics, which fell within acceptable ranges, and were comparable to the control in supporting plant growth. DC-B under fresh conditions demonstrated a high interstitial pH (∼11.0) that became ameliorated (∼8.3) under all saline conditions. Plant photosynthetic response was minimal in DC-B under fresh conditions, but comparable to the control under saline conditions. These results are promising and warrant further evaluations under a wider range of environmental conditions prior to field implementation. © 2005 Elsevier B.V. All rights reserved. Keywords: Wetland restoration; Drill cuttings; Louisiana; Spartina alterniflora; Spartina patens; Spartina cynosuroides; Avicennia germinans
1. Introduction Wetland loss in Louisiana has been occurring at the alarming rate of 40–56 km2 year−1 , the highest in the United States, due to a combination of natural geologic ∗ Corresponding author. Tel.: +1 504 280 1324; fax: +1 504 280 6121.
0925-8574/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.ecoleng.2005.01.018
processes and human influences (Salinas et al., 1986; Penland and Ramsey, 1990; Britsch and Dunbar, 1993; Boesch et al., 1994; Day et al., 2000). Subsidence due to dewatering and soil compaction and sediment deprivation stemming from anthropogenic activities (e.g. leveeing of the Mississippi River) in conjunction with increasing rates of global sea-level rise have resulted in elevated rates of relative sea-level rise (T¨ornqvist et al.,
42
J.M. Willis et al. / Ecological Engineering 25 (2005) 41–50
2002). This elevated rate of relative sea-level rise acts to increase stresses associated with flooding on emergent vegetation, eventually leading to open water habitat as the plants die and their stabilizing effect on the substrate is lost (Baumann et al., 1984; Mendelssohn and McKee, 1988; Nyman et al., 1993; DeLaune et al., 1990; Wilsey et al., 1992; Day et al., 1995; Mitsch and Gosselink, 2000). Thus, wetland restoration efforts within coastal Louisiana must often incorporate increasing marsh elevation through some form of sediment addition as a key component. The beneficial effects of such sediment additions in improving soil physico-chemical properties and enhancing plant productivity have been demonstrated in numerous studies (DeLaune et al., 1990; Wilsey et al., 1992; Webb et al., 1995; Mendelssohn and Kuhn, 2003). Controlled river diversions and targeted placement of dredged sediment are examples of current wetland restoration techniques in Louisiana that incorporate sediment addition (e.g. Lane and Day, 1999; Streever, 2000; Edwards and Proffitt, 2003). Drill cuttings generated in association with water-based drilling muds (fluids) may represent an ancillary sediment source that, if shown to be safe and efficacious, could aid in restoration efforts and reduce the oil-field waste load in landfills. Drill cuttings are a mixture of the earthen materials generated during the excavation of an oil or gas well and drilling fluids or “muds,” which are additives that lubricate the drilling bit, stabilize the borehole and lift cuttings out of the borehole (Gray and Darley, 1980; Scholten et al., 2000). Drill cuttings are traditionally treated onsite by physically separating the cuttings from the drilling fluids. The drilling fluids are then recycled for further drilling operations, whereas the cuttings and the residual muds generated in near-shore waters are disposed of as oil-field waste in landfills. This process generates the DC-A substrate employed in this study. An extended processing technique has been developed by SWACO Inc. (proprietary information of SWACO, Lafayette, LA) that can be used to further treat the remaining cuttings by the addition of stabilizing agents, which physically isolate the cuttings, and thereby any associated contaminants, in a silica matrix for greater environmental safety. This process generates the DC-B substrate used in this study. Kelley and Mendelssohn (1995) demonstrated that drill cuttings that had undergone this extensive
processing could, when amended with organic matter, support emergent wetland vegetation. Shaffer et al. (1996, 1998) examined drill cuttings that had been subjected to physical separation only as well as physical separation followed by isolation within a silica matrix. Shaffer et al. (1996, 1998) determined that the drill cuttings subjected to physical separation (DC-A) were capable of supporting vigorous wetland plant growth, but that plants grown in drill cuttings subjected to physical separation followed by silica isolation (DC-B) demonstrated minimal growth. The poor response of vegetation grown on this substrate was attributed to both the high pH of the interstitial waters of this substrate (mean pH ∼11.0) and the coarse texture of the soil (Shaffer et al., 1996, 1998). Nonetheless, the successful colonization and the high productivity of vegetation in the drill cuttings subjected to only physical separation indicate that drill cuttings have the capacity to support the growth of wetland vegetation, at least under freshwater conditions. Although the studies of Shaffer et al. (1996, 1998) are encouraging, no data are available on the response of these drill cuttings to saline waters or the growth of coastal wetland plants in these processed drill cuttings under saline conditions. The objectives of this study are two-fold: (1) to elucidate the physicochemical response, particularly interstitial water metal concentration, of two differently processed drill cuttings to a range of hydrologic regimes and salinity levels possible in a coastal wetland restoration scenario and (2) to evaluate the capacity of processed drill cuttings to support the survival and productivity of several coastal wetland plant species under various combinations of hydrologic regimes and salinity levels that may be encountered in coastal restoration scenarios. If processed drill cuttings can be shown to cause no environmental hazards, support the growth of wetland vegetation, and have a cost comparable to current oil-field waste disposal costs, and then the implementation of this method as an additional restoration strategy warrants consideration. 2. Materials and methods 2.1. Study facility The mesocosm facility employed for this study consists of 144 mesocosm vessels (200-l, polyethylene
J.M. Willis et al. / Ecological Engineering 25 (2005) 41–50
units) networked to four salinity reservoirs within a temperature-controlled greenhouse. An air lift water delivery system was devised to bring water into the mesocosms as described in Section 2.3. 2.2. Substrate type Each treatment mesocosm was filled to approximately 3/4 volumes with one of the four test substrates. The drill cutting materials used in the experiment were generated in Grand Bay, Louisiana (a tidally influenced, oligohaline marsh area). All drill cuttings were separated from the drilling fluids, which resulted in the DC-A substrate (processing accomplished by Cameron Corp.). Some of this material was subjected to further processing, resulting in the DC-B substrate (processing accomplished by SWACO Corp.). Drill cuttings substrates (DC-A and DC-B) and dredge spoil (generated in the same general area as the drill cuttings) were transported to the SLU mesocosm facility courtesy of Greenhill Petroleum Company.
43
local brackish marsh. Avicennia germinans (black mangrove) plants were purchased from a horticultural supplier. Three genotypes each of Spartina patens (wiregrass) and Spartina alterniflora (oystergrass) were selected to represent a range of salt tolerance based on previous research (Hester et al., 1996, 1998). Individual plants were rinsed to remove parent soil and then randomly allocated to the appropriate treatment and allowed to acclimate for approximately 2 weeks. 2.5. Salinity levels Salinity was sequentially increased from 0 to 9, to 18, and then to 36 ppt for all treatment vessels over approximately a 1-year period using Coralife scientific grade sea–salt. The durations of the salinity treatments were 4 months for the 0–9 ppt salinity levels, and 2 months for the 18–36 ppt salinity levels. All salinity increases were accomplished over a 2-week period to allow plants time for acclimation. Salinity levels of each treatment were monitored monthly with a salinity meter (YSI Model #30 salinity/conductivity meter) and adjusted accordingly.
2.3. Hydrology design 2.6. Plant photosynthetic response The mesocosm treatment vessels were networked to four, 3000-l reservoirs in a substrate-specific design that prevented cross contamination of water from different substrate sources. Water was brought into each mesocosm daily for 4 h by an air lift system, which aerated the water from gravity-fed reservoir waterlines as it was lifted by pumped air bubbles above the reservoir water level into each mesocosm vessel (see Shaffer et al. (1996) for further discussion of air lift system design). A system of PVC pipes and submersible pumps recycled the water in a substrate-specific fashion and created the following three hydrologic regimes: mesic (moist, but allowed to drain to 10 cm below the soil surface during daily water input), tidal (allowed to flood to 20 cm above the soil surface and drain to 10 cm below the soil surface during the daily tidal cycle) and flooded (flooded to 20 cm above the soil surface at all times). 2.4. Plant treatments Transplants of Spartina cynosuroides (big hogcane) were collected from the Bayou Lacombe area, a
Plant photosynthetic response was determined toward the end of each salinity level increase using a LICOR 6400 Photosynthesis System (LI-COR, Lincoln, NE, USA) with a built-in light source and CO2 controller with reference settings of 1500 mol m−2 s−1 and CO2 levels of 370 ppm. Measurements were performed on two representative leaves (first or second fully expanded leaf from top of stem) selected from each test plant and then averaged to provide one photosynthetic response value for that plant. 2.7. Substrate redox potential Substrate redox potential (Eh) was monitored toward the end of each salinity level increase using a pH/mV meter (Hanna model HI 9025) and brightened platinum redox electrodes that were constructed, calibrated and brightened as per Faulkner et al. (1989). Redox probes were placed at 2 and 15 cm depths below the substrate surface in each vessel, allowed to equilibrate overnight, and redox potentials determined the following day.
44
J.M. Willis et al. / Ecological Engineering 25 (2005) 41–50
2.8. Interstitial water analysis Interstitial water pH was monitored with a pH/mV meter (Hanna model HI 9025). Measurements were taken by placing the pH probe into a permanent interstitial well (slotted 1.875 cm PVC pipe extended 10 cm below soil surface) within each treatment vessel. Interstitial water samples for elemental analysis were collected by drawing fluid with a syringe from interstitial water wells. Samples were filtered (0.45 m), acidified with concentrated, reagent-grade nitric acid and subjected to elemental analysis by inductively coupled argon plasma mass-optical emission spectrophotometery (Jarrel-Ash ICP-OES Atom Comp Series 800), except for Ba, which was subjected to cold vapor atomic absorption (Perkin-Elmer Model 5000 spectrophotometer). Of the elemental concentrations obtained, statistical analysis and interpretation were limited to major toxic heavy metals (Cd, Cr, Pb and Ni, see Soil and Plant Analysis Council (1999)), elements thought to be elevated in processed drill cuttings due to additives (Ba and Al), and the macronutrients P and K.
5000). Target elements were the same as those chosen for interstitial water analysis (Al, Ba, Cd, Cr, K, Ni, P and Pb). 2.11. Statistical analysis Repeated measure ANOVAs, with salinity level treated as the repeat factor (Girden, 1992; Von Ende, 2001), were used to analyze all dependent variables that were measured at each salinity level (redox potential, interstitial pH, interstitial elemental concentration and photosynthetic response). Univariate ANOVAs were employed to test for differences within a salinity level when a significant interaction with salinity level was detected (Von Ende, 2001). Final biomass and plant tissue metal concentrations were analyzed using univariate ANOVAs. All data were tested for meeting the criteria of parametric analysis. Repeated measures ANOVAs were adjusted for departures from sphericity using the Hunyh–Feldt technique (Girden, 1992; Von Ende, 2001). All statistical analyses were performed using the appropriate procedure of SYSTAT version 8.0 (SPSS, 1998).
2.9. Biomass measurements Plant biomass was harvested at the end of the 36 ppt salinity level increment. Biomass was separated into aboveground alive, aboveground dead and belowground partitions, lightly rinsed to remove substrate and placed into paper bags at harvest. All biomass was dried in a plant-drying oven at 60 ◦ C until constant weight was achieved. 2.10. Plant tissue analysis A subsample of dried aboveground biomass (1 g of stem and leaf) was obtained from all vegetated test vessels in the mesic and flooded hydrology treatments. Plant material was ground in a Wiley mill and digested in 5 ml of concentrated, reagent-grade nitric acid in a block heater at 130 ◦ C for 3 h. Samples were allowed to cool overnight, brought up to 70 ml total volume with distilled–deionized water, inverted several times, and the supernatant collected for analysis The samples were then analyzed with an inductively coupled argon plasma mass-optical emission spectrophotometer (Jarrel-Ash ICP-OES Atom Comp Series 800) or cold vapor atomic absorption (CVAA Perkin-Elmer Model
3. Results 3.1. Elemental analysis (interstitial water) The results of substrate interstitial water analysis on all four substrates (topsoil, dredge spoil, DC-A and DC-B) at the end of the 0–36 ppt salinity level exposures are presented in Table 1 (for brevity the 9–18 ppt salinity treatments are omitted). There was a significant trend toward increases in the concentration of Ba, Pb and Ni, in the 36 ppt salinity level compared with the 0 ppt salinity level (Table 1; all P < 0.05). This trend was the most pronounced in DC-A and topsoil, and the least pronounced in DC-B. A significant interaction of salinity by substrate was present for interstitial Ni concentration, likely stemming from the greater concentration of interstitial Ni in the topsoil and DC-A treatments compared to the dredge spoil and DC-B substrates in the 36 ppt treatment, while no difference was evident in the 0 ppt treatment (Table 1; F = 3.217, P = 0.034). Interestingly, Al concentration actually decreased significantly in the 36 ppt salinity level compared with the 0 ppt salinity level (Table 1; F = 7.887, P = 0.002).
J.M. Willis et al. / Ecological Engineering 25 (2005) 41–50
45
Table 1 Results of elemental analysis of substrate interstitial water at the end of the 0 and 36 ppt salinity levels Element
Substrate Topsoil
Ba Cd Pb Cr Ni Al
Dredge
DC-A
DC-B
0 ppt
36 ppt
0 ppt
36 ppt
0 ppt
36 ppt
0 ppt
36 ppt
0.041 ± 0.013 BD ± 0.000 0.014 ± 0.014 BD ± 0.000 BD ± 0.000 0.555 ± 0.235
1.003 ± 0.081 0.010 ± 0.000 0.169 ± 0.075 BD ± 0.000 0.044 ± 0.033 0.710 ± 0.270
0.069 ± 0.009 0.007 ± 0.005 0.035 ± 0.012 0.003 ± 0.002 0.007 ± 0.005 0.889 ± 0.176
1.132 ± 0.276 0.018 ± 0.016 0.090 ± 0.022 0.001 ± 0.001 0.011 ± 0.004 0.222 ± 0.147
0.115 ± 0.016 0.003 ± 0.003 0.006 ± 0.004 0.001 ± 0.001 BD ± 0.000 1.020 ± 0.360
0.931 ± 0.126 0.014 ± 0.007 0.261 ± 0.189 0.002 ± 0.001 0.055 ± 0.029 1.177 ± 0.392
0.100 ± 0.012 0.005 ± 0.003 0.005 ± 0.005 BD ± 0.000 BD ± 0.000 7.569 ± 1.466
1.346 ± 3.290 BD ± 0.000 0.132 ± 0.041 0.001 ± 0.001 0.001 ± 0.020 0.096 ± 0.055
Values are mean concentrations (g g−1 ) ± standard errors (n = 36).
3.2. Elemental analysis (vegetation)
3.4. pH
Results of elemental analysis of acid-digested plant material are presented in Table 2. No significant effects of substrate, hydrology, species, or the interactions thereof were detected in plant tissue concentrations of Cd, Pb, Cr, Ni or Al. However, plants grown in the DC-A and DC-B substrates had higher concentrations of Ba than those grown in topsoil and dredge spoil (Table 2; contrast F = 6.32, P = 0.020).
There was a significant effect of salinity level and substrate type on pH (Fig. 2; F = 49.45, P < 0.001). As the prior studies indicated, the pH of DC-B was highly elevated relative to the other substrates under freshwater conditions. However, the pH of DC-B was significantly lower under all saline treatments than under freshwater conditions (Fig. 2; contrast F = 59.88, P < 0.001).
3.3. Soil redox potential
3.5. Photosynthetic response
Soil redox potential at 2 and 15 cm depths displayed a significant decrease with increasing salinity levels (Fig. 1; F = 108.74, P < 0.001 and F = 177.37, P < 0.001, respectively). A marginally significant salinity level by hydrologic regime interaction was detected, indicating that there was a tendency for the effect of hydrology to vary over time as salinity levels were increased (Fig. 1; F = 2.28, P = 0.038).
There was a significant effect of salinity level on net CO2 assimilation rate (Fig. 3; F = 14.43, P < 0.001). Analysis within salinity levels showed that at the 0 ppt salinity level, plants had significantly reduced net CO2 assimilation rates when grown in DC-B compared to the other substrates (Fig. 3; contrast F = 47.813, P < 0.001). This was probably due to the elevated pH that DC-B exhibits under freshwater conditions. No
Table 2 Results of elemental analysis of aboveground plant tissue at the end of the 36 ppt salinity level Element
Substrate Topsoil
Ba Cd Pb Cr Ni Al
6.030 0.280 0.630 0.440 1.470 49.90
± ± ± ± ± ±
Dredge 1.656 0.280 0.630 0.298 1.470 28.69
Values are means (g g−1 ) ± standard errors (n = 36).
6.728 0.219 0.814 0.429 1.689 73.237
DC-A ± ± ± ± ± ±
2.724 0.219 0.536 0.216 0.862 21.20
22.15 0.220 0.000 0.250 1.250 86.4
DC-B ± ± ± ± ± ±
6.917 0.220 0.000 0.162 0.813 20.76
20.25 0.411 0.971 0.560 1.146 131.6
± ± ± ± ± ±
7.803 0.270 0.660 0.281 0.751 60.91
46
J.M. Willis et al. / Ecological Engineering 25 (2005) 41–50
Fig. 2. The effect of salinity level and substrate type on interstitial pH (mean ± standard error) averaged across hydrologic regime and plant species.
level permanently flooded treatments were found to have lower net CO2 assimilation rates than tidal and mesic treatments (contrast F = 16.515, P < 0.001). These data indicate that drill cuttings are capable of supporting net CO2 assimilation rates comparable to topsoil, providing that the pH of DC-B is ameliorated. Overall, net CO2 assimilation rates were greatest in the 9 ppt salinity level. 3.6. Total biomass Substrate type, hydrologic regime, and plant species all had marginally significant effects on total plant biomass (Fig. 4; F = 1.77, P = 0.05), which appeared to be driven by the high performance of S. cynosuroides under all hydrologies in DC-B, and the high
Fig. 1. The effect of hydrologic regime and salinity level on soil redox (mean ± standard error) averaged across substrate type and plant species.
significant differences were detected among substrate types at the 9 or 18 ppt salinity levels. Also, at the 36 ppt salinity level plants grown in DC-A had significantly greater net CO2 assimilation rates compared with all other substrates (Fig. 3; contrast F = 9.125, P = 0.003). No significant differences were detected in net CO2 assimilation among substrate types at the 9 or 18 ppt salinity levels. Within the 0 ppt salinity
Fig. 3. The effect of salinity level and substrate type on net CO2 assimilation rate (mean ± standard error) averaged across hydrologic regime and plant species.
J.M. Willis et al. / Ecological Engineering 25 (2005) 41–50
47
forms (lower biomass of A. germinans). The main effects of substrate type and vegetative species, as well as the three-way interaction between substrate, hydrology and species, regarding aboveground biomass were significant (contrast F = 54.95, P < 0.001; F = 41.55, P < 0.001 and F = 2.90, P = 0.002, respectively). Also, a significant difference in belowground biomass was detected among species (F = 5.00, P = 0.005). All responses of total, aboveground and belowground biomass were consistent, therefore only the total biomass graphs are presented.
4. Discussion 4.1. Edaphic conditions
Fig. 4. The effect of substrate type, hydrologic regime and plant species on total biomass (mean ± standard error).
performance of S. patens in DC-B under mesic conditions. This limits the interpretation of the main effect of substrate type, but a contrast indicates that DC-B total biomass was significantly reduced compared to all other substrates (Fig. 4; contrast F = 8.40, P = 0.006). No significant differences were detected among hydrologies in terms of total biomass production. A significant effect of species was discerned (Fig. 4; F = 7.86, P = 0.001), but likely simply reflects the difference between herbaceous versus woody growth
As one would expect, substrate redox potentials became significantly more reduced as salinity levels were increased over time, likely reflecting the presence of alternative electron acceptors such as sulfate in the synthetic sea–salt mix and the increase in duration of flooding (Mitsch and Gosselink, 2000). Importantly, this study has allowed for the quantification of the behavior of the metal constituents in both the DC-A and DC-B substrates over a large range of substrate reduction (+400 to −100). Substrate interstitial pH in the DC-B substrate was dramatically lowered from extreme alkaline values (∼10.1) to moderate pH values (∼8.5) as salinity levels increased. The strong carbonate buffering capacity of seawater, which can buffer both acidic and basic systems was likely very important in this response (Dietrich, 1980). This buffering is of great significance because the high alkalinity demonstrated by the DC-B substrate was considered one of the greatest obstacles in regard to it serving as a substrate for the successful establishment of wetland vegetation. Interestingly, this pH-salinity phenomenon was not noted by Kelley and Mendelssohn (1995). However, this apparent discrepancy is likely a scale effect due to the large water reserves (3000-l) employed in this study, which provided a larger buffering capacity to ameliorate the high pH of the DC-B substrate. Previous drill cuttings research indicated that the use of processed drill cuttings to create wetlands may require additional amelioration of the high pH to be successful, which would reduce cost-effectiveness of
48
J.M. Willis et al. / Ecological Engineering 25 (2005) 41–50
the method (Kelley and Mendelssohn, 1995; Shaffer et al., 1996; DesRoches, 1998). However, an implication of this study is that if these processed drill cuttings are used for coastal wetland restoration where natural saline waters are present, and then there may be no need to amend the alkalinity of the substrate with organic materials as was previously thought necessary (Kelley and Mendelssohn, 1995). For the majority of heavy metals in aquatic systems, maximum potential release occurs at moderately low and extremely high pHs, and under slightly reduced or oxidized conditions (Fergusson, 1990; Bourg and Loch, 1995). The conditions reported by Gambrell et al. (1991) for minimal solubility of a suite of heavy metals (pH 8.0 and soil redox potential of −150 to −100 mV) are similar to the environmental conditions in the mesocosm at the 18 and 36 ppt salinity levels. Typical values for pH and Eh that would be expected in natural coastal systems were reached during this experiment under fresh and saline conditions, with no significant differences from the control (i.e. topsoil) detected in any of the metals of concern. Examination of aboveground plant tissue indicated significantly higher levels of Ba in the DC-A substrate (22.15 g g−1 ) and DC-B substrate (20.25 g g−1 ) compared to topsoil (6.03 g g−1 ) and dredge spoil (6.73 g g−1 ). However, normal levels of Ba in foliar tissue are considered to be between 10 and 50 g g−1 (Chaudhry et al., 1977; also see references in Cipollini and Pickering, 1986). Therefore, the levels of Ba in plant tissue grown on the two processed drill cutting substrates, although elevated compared to topsoil and dredge spoil, are not exceptionally high. No significant differences among substrates were detected for any other metals of concern (Cd, Cr, Pb and Ni).
Net CO2 assimilation rates under freshwater conditions indicated significantly lower plant performance when plants were grown on the DC-B substrate, corroborating the previous studies. However, a significant interactive effect of salinity level and substrate was detected, in which plants grown on the DC-B substrate had rates of net CO2 assimilation comparable to other substrates once salinity levels were increased and pH dropped to a more moderate level. Although analysis of plant biomass would indicate that overall plants were performing significantly worse in the DC-B substrate than in other substrates, one must take into account that biomass values were influenced by the time spent at the 0 ppt salinity level. No significant differences between substrate were detected in plant net CO2 assimilation rates under saline conditions, indicating that after 2 years in a simulated wetland environment drill cuttings processed by these techniques are suitable for supporting wetland plant growth in restoration projects. Two of the most wide-spread brackish and salt marsh plant species of Louisiana, S. alterniflora and S. patens, exhibited excellent growth, both in terms of net CO2 assimilation and total biomass, indicating that they would be good candidates for these types of wetland rehabilitation projects. Although no differences were detected in the performance of S. cynosuroides when grown on processed drill cuttings or on topsoil or dredge spoil, this species may not be as viable a candidate for creating wetlands in this fashion, because it exhibited high tissue mortality and relatively low photosynthetic rates throughout the experiment regardless of substrate type, possibly because of a low tolerance to transplant stress. Because S. alterniflora and S. patens are widespread throughout Atlantic and Gulf Coast marshes (Godfrey and Wooten, 1979) the results of this project may have relevance to many coastal areas.
4.2. Vegetation performance The two prime indicators of plant growth used in this experiment were net CO2 assimilation rate (photosynthesis) and biomass production. Net CO2 assimilation rates were measured at each salinity level, and can thus be used to evaluate plant responses to substrates and hydrologic regimes at different salinity levels. Biomass was harvested at the conclusion of the experimental period, and therefore represents the cumulative effects of substrate and hydrology across all salinity levels through time.
5. Conclusion In summary, this study has important implications for the use of processed drill cuttings in the rehabilitation of coastal wetlands. Previous studies on processed drill cuttings have indicated that the potential for their utilization in wetland rehabilitation existed, but that plant performance was often reduced from that observed on other substrates, or there were concerns about the safety of the processed material under various
J.M. Willis et al. / Ecological Engineering 25 (2005) 41–50
environmental scenarios. This study has illustrated that processed drill cuttings have the potential to serve as suitable and non-toxic substrates capable of supporting high rates of wetland plant growth under a range of environmental conditions. However, if processed drill cuttings are to become a viable option as a substrate source in wetland rehabilitation, we recommend that future research focus on further refinement in the formulation of environmentally safe, water-based drilling fluids, because many of the concerns in utilizing drill cuttings stem from the materials in the drilling fluids. Thus, rather than attempt to further develop encapsulation techniques to minimize potential contaminant release from the residual drilling fluid coating on the cuttings, it may prove more expedient to formulate water-based drilling fluids with wetland rehabilitation in mind, such that these drilling fluids are both environmentally safe and conducive to wetland plant growth.
Acknowledgements We would like to thank John Ford of the National Energy Technology Laboratory and John Veil of Argonne National Laboratory for all of their assistance throughout this project and the US Department of Energy who provided funding for this research.
References Baumann, R.H., Day Jr., J.W., Miller, C.A., 1984. Mississippi deltaic wetland survival: sedimentation versus coastal submergence. Science 224, 1093–1095. Boesch, D.F., Josselyn, M.N., Mehta, A.J., Morris, J.T., Nuttle, W.K., Simenstad, C.A., Swift, D.J.P., 1994. Scientific assessment of coastal wetland loss, restoration and management in Louisiana. J. Coastal Res. Spec. Issue 20, 103. Bourg, A.C.M., Loch, J.P.G., 1995. Mobilization of metals as affected by pH and redox conditions. In: Salomons, W., Stigliani, W.M. (Eds.), Biogeodynamics of Pollutants in Soils and Sediments: Risk Assessment of Delayed and Non-linear Responses. Springer-Verlag, Berlin, Germany, 352 pp. Britsch, L.D., Dunbar, J.B., 1993. Land loss rates: Louisiana coastal plain. J. Coastal Res. 9, 324–338. Chaudhry, F.M., Wallace, A., Mueller, R.T., 1977. Barium toxicity in plants. Commun. Soil Sci. Plant Anal. 8, 795–797. Cipollini, M.L., Pickering, J.L., 1986. Determination of the phytotoxicity of barium in leach-field disposal of gas well brines. Plant and Soil 92, 159–170.
49
Day Jr., J.W., Pont, D., Hensel, P.F., Ibanez, C., 1995. Impacts of sealevel rise on deltas in the Gulf of Mexico and the Mediterranean: the importance of pulsing events to sustainability. Estuaries 18, 636–647. Day Jr., J.W., Shaffer, G.P., Britsch, L.D., Reed, D.J., Hawes, S.R., Cahoon, D., 2000. Pattern and process of land loss in the Mississippi Delta: a spatial and temporal analysis of wetland habitat change. Estuaries 23, 425–438. DeLaune, R.D., Pezeshki, S.R., Pardue, J.H., Whitcomb, J.H., Patrick Jr., W.H., 1990. Some influences of sediment addition to a deteriorating salt marsh in the Mississippi River deltaic plain: a pilot study. J. Coastal Res. 6, 181–188. DesRoches, D.J., 1998. Restored Drill Cuttings: A Mesocosm Approach to Determine the Suitability as a Substrate for the Wetlands Restoration and Creation. Masters Thesis. Department of Biological Sciences, Southeastern Louisiana University, Hammond, LA, USA, 105 pp. Dietrich, G., 1980. General Oceanography: An Introduction. Wiley, New York, 626 pp. Edwards, K.R., Proffitt, C.E., 2003. Comparison of wetland structural characteristics between created and natural salt marshes in southwest Louisiana, USA. Wetlands 23, 344–356. Gambrell, R.P., Wiesepape, J.B., Patrick Jr., W.H., Duff, M.C., 1991. The effects of pH, redox, and salinity on metal release from a contaminated sediment. Water Air Soil Pollut. 57, 359–367. Girden, E.R., 1992. ANOVA Repeated Measures. Sage Publications, Newbury Park, CA, 77 pp. Godfrey, R.K., Wooten, J.W., 1979. Aquatic and Wetland Plants of the South eastern United States: Monocotyledons. University of Georgia Press, Athens, GA, USA, 712 pp. Gray, G.R., Darley, H.C.H., 1980. Composition and Properties of Oil Well Drilling Fluids, fourth ed. Gulf Publishing Co., Houston, TX, 630 pp. Faulkner, S.P., Patrick Jr., W.H., Gambrell, R.P., 1989. Field techniques for measuring wetland soil parameters. J. Soil Sci. Soc. Am. 53, 883–890. Fergusson, J.E., 1990. The Heavy Elements Chemistry, Environmental Impact and Health Effects. Pergamon Press, Elmsford, NY, USA, 614 pp. Hester, M.W., Mendelssohn, I.A., McKee, K.L., 1996. Intraspecific variation in salt tolerance and morphology in the coastal grass Spartina patens (Poaceae). Am. J. Bot. 83, 1521–1527. Hester, M.W., Mendelssohn, I.A., McKee, K.L., 1998. Intraspecific variation in salt tolerance and morphology in Panicum hemitomon and Spartina alterniflora. Int. J. Plant Sci. 159, 127–138. Kelley, S., Mendelssohn, I.A., 1995. An evaluation of stabilized, water-based drill cuttings and organic compost as potential substrate sources for marsh restoration and creation in coastal Louisiana. Ecol. Eng. 5, 497–517. Lane, R.R., Day Jr., J.W., 1999. Water quality analysis of a freshwater diversion at Caernarvon, Louisiana. Estuaries 22, 327–336. Mendelssohn, I.A., Kuhn, N.L., 2003. Sediment subsidy: effects on soil–plant responses in a rapidly submerging coastal salt marsh. Ecol. Eng. 21, 115–129. Mendelssohn, I.A., McKee, K.L., 1988. Spartina alterniflora dieback in Louisiana: time-course investigation of soil waterlogging effects. J. Ecol. 76, 509–521.
50
J.M. Willis et al. / Ecological Engineering 25 (2005) 41–50
Mitsch, W.J., Gosselink, J.G., 2000. Wetlands, third ed. Van Nostard Reinhold, NY, USA, 920 pp. Nyman, J.A., DeLaune, R.D., Roberts, H.H., Patrick Jr., W.H., 1993. Relationship between vegetation and soil formation in a rapidly submerging coastal marsh. Mar. Ecol. Prog. Ser. 96, 269– 279. Penland, S., Ramsey, K.E., 1990. Relative sea-level rise in Louisiana and the Gulf of Mexico. J. Coastal Res. 6, 323– 342. Salinas, L.M., DeLaune, R.D., Patrick Jr., W.H., 1986. Changes occurring along a rapidly submerging coastal area. J. Coastal Res. 2, 269–284. Scholten, M.C.Th., Karman, C.C., Huwer, S., 2000. Ecotoxicological risk assessment related to chemicals and pollutants in off-shore oil production. Toxicol. Lett. 112, 283–288. Shaffer, G.P., Hester, M.W., Childers, G.W., 1996. Restored Drill Cuttings for Wetlands: Year One Results of a Mesocosm Approach to Emulate Field Conditions Under Varying Hydrologic Regimes. National Petroleum Technology Program, U.S. Department of Energy, Technical Report, 46 pp. Shaffer, G.P., Hester, M.W., Miller, S., DesRoches, D.J., Souther, R.F., Childers, G.W., Campo, F.M., 1998. Restored Drill Cuttings for Wetlands: Results of a Two Year Mesocosm Approach to Emulate Field Conditions Under Varying Hydrologic Regimes.
National Petroleum Technology Program, U.S. Department of Energy, Technical Report, 86 pp. SPSS, 1998. SYSTAT for Windows. SPSS Science Marketing Department, Chicago, IL, USA, 1086 pp. Soil Plant Analysis Council, 1999. Soil Analysis Handbook of Reference Methods. CRC Press, Boca Raton, FL, USA, 264 pp. Streever, W.J., 2000. Spartina alterniflora marshes on dredged material: a critical review of the ongoing debate over success. Wetlands Ecol. Manag. 8, 295–316. T¨ornqvist, T.E., Gonz´alez, J.L., Newsom, L.A., Van der Borg, K., De Jong, A.F.M., 2002. Reconstructing “background” rates of sealevel rise as a tool for forecasting coastal wetland loss, Mississippi Delta. EOS 83, 530–531. Von Ende, C.N., 2001. Repeated-measure analysis: growth and other time-dependent measures. In: Scheiner, S.M., Gurevitch, J. (Eds.), Design and Analysis of Ecological Experiments. Oxford University Press, Oxford, UK, 415 pp. Webb, E.C., Mendelssohn, I.A., Wilsey, B.J., 1995. Causes for vegetation dieback in a Louisiana salt marsh: a bioassay approach. Aquatic Bot. 51, 281–289. Wilsey, B.J., McKee, K.L., Mendelssohn, I.A., 1992. Effects of increased elevation and macro- and micronutrient additions on Spartina alterniflora transplant success in salt marsh die-back areas in Louisiana. Environ. Manag. 16, 505–511.