Using green seaweed (Ulva spp.) as a soil amendment: Effects on soil quality and yield of sweet corn (Zea mays L.)

In this study, fall and spring seaweed collection was completed at local sites with frequent and reliable accumulation of beach-cast seaweed biomass. Watch Hill Beach and Mackerel Cove are relatively protected inlets, which often supports high beach deposition of seaweed by natural currents and wave action. However, seaweed proliferation as a result of anthropogenic nutrient inputs may be less at these sites than sites located further from the open ocean. At the time of collection for this study, sites with excessive seaweed accumulation presumably due to anthropogenic causes were scarce, due in part to the season during which collection took place, since seaweed biomass usually reaches the highest levels in July and August (Thornber et al., 2008), as well as beach-clearing weather events, such as Hurricane Rita in September 2011. Consequently, the seasonal variation of excess seaweed accumulation may require monitoring and communication among beach managers and farmers in order to make optimal use of this resource. In lieu of using often hard-to-predict “problematic” seaweed biomass, beach sites with high seaweed accumulation due to inherent geographic or environmental conditions may offer a more consistent and reliable source.

Seaweed biomass collected in fall 2011 from Westerly, RI, was largely composed of brown and red seaweed species, including Ascophyllum nodosum (12.5% dry weight (DW), Laminaria digitata (2% DW), Chondrus crispus (15.2% DW), Fucus vesiculus (8.2% DW)), assorted filamentous red algae (10.5% DW), and mixed, non-algal plant material (e.g. eelgrass, 51.5% DW). Seaweed collected in spring 2012 from Jamestown, RI included Saccharina saccharina (0.5% DW), A. nodosum (9.5% DW), Fucus sp. (11.6% DW), Grinellia americana (2.5% DW), C. crispus (3.3% DW), Ulva sp. (0.95% DW), assorted filamentous red algae (69.7% DW), and mixed non-algal plant material (1.7% DW). In comparison to terrestrial plant biomass, the seaweed collected for this study can be characterized as a high-moisture material with relatively high N content (15.5 g/kg DW) and low C:N ratio (10:1 to 13:1). For instance, values of N content and C:N ratios of common crop residue materials are in the range of 4-5 g N/kg plant DW and C:N ratio of ~100:1 for stem biomass, and 12-15 g N/kg plant DW and C:N ratio of ~30:1 for leaf biomass, based on values for soybean (Glycine max), corn (Zea mays), and switchgrass (Panicum virgatum) stems and leaves (Johnson et al., 2007). Additionally, the total calcium (Ca) and potassium (K) content of the seaweed material was higher than many terrestrial plants, on the order of 4 times greater Ca content and 2 times greater K content (Tian et al., 1992). For comparison, biomass of the cover crop velvet bean (Mucuna pruriens) contains approximately 5-7 g Ca/kg DW and 18 g K/kg DW (Tian et al., 1992), whereas levels in the seaweed biomass in this study were 21-38 g Ca/kg DW and 31-50 g K/kg DW. Elements of concern (e.g. heavy metals) may also be found at higher levels in seaweed relative to terrestrial plant biomass; however, in the context of field application guidelines, these were not sufficiently high to raise concerns in this study. For example, As was present in seaweed biomass (maximum 9.8 ± 1.8 mg/kg DM), but at much lower levels than those for F. vesiculosus and L. digitata biomass collected from coastal Scotland (25 ± 7 and 74 ± 2 mg/kg DM, respectively) (Castlehouse et al., 2003). For sewage sludge, an agricultural amendment with particular risk of heavy metal contamination, the US EPA regulatory limit for As for soil application is 75 mg/kg (USDA NRCS, 2000). Although the seaweed biomass collected in this study does not pose a concern in terms of exceeding regulatory limits for As application, consideration of As is warranted for seaweed application, particularly when brown algal species such as F. vesiculosus and L. digitata constitute a higher percentage of the biomass applied. US EPA yearly application limits and long-term maximum cumulative loading also provide guidelines and perspective for potential As application risks. In the present study, the As level in seaweed biomass would result in an annual loading of 0.03 kg/ha/yr for the HDS treatment, well below the EPA regulatory limit of 2 kg/ha/yr. Additionally, the total maximum cumulative loading for As is 41 kg/ha, which would require ~1,300 years of seaweed application at the HDS rate to reach the maximum load, assuming no losses of As from the soil. Cadmium (Cd), chromium (Cr), mercury (Hg), and copper (Cu) were not present above the limit of detection. Pb and Zn were detected, but at concentrations substantially lower than the US EPA regulations for sewage sludge application. For Pb and Zn, the maximum levels present in seaweed (14.2 ± 2.8 and 68.4 ± 11.0 mg/kg, respectively), are ~30 times lower than the US EPA limits of 420 and 2500 mg/kg for Pb and Zn, respectively, in sewage sludge. Consequently, the heavy metal content of the seaweed used in this study does not pose a concern in terms of long-term accumulation of heavy metals. Since many seaweed species or ecotypes are tolerant of high levels of heavy metals (e.g. Cu in anti-fouling paint) (Reed and Gadd, 1990), amendment with seaweed biomass collected from areas likely to be affected by heavy metal contamination may require pre-application analysis. Seasonal variation in heavy metal levels in seaweed may also be important for timing of collection, since higher concentrations are generally found in winter and early spring, and lower concentrations in summer and autumn (Caliceti et al., 2002).

For all physical properties, no statistically significant differences (p≥0.05) were detected among the fertilizer treatments (Table 3). Consequently, the hypothesized positive effects of seaweed amendment on soil physical quality (increased aggregate stability, infiltration and available water capacity, and reduced bulk density) were not supported in this study. Overall, aggregate stability of soil at the study site is rated as moderate to poor, ranging from ~30-40% (Gugino et al., 2009), a value that could be expected for an agricultural soil with repeated, frequent cultivation. Available water capacity, at 0.224 g/g, is rated as medium to good (Gugino et al., 2009). Many soil physical properties, including aggregate stability, require several years after management changes before appreciable improvements are observed (Islam and Weil, 2000). The duration of this study was likely insufficient for development of uniform observable, significant improvements in aggregate stability across the study site. In order to provide a better perspective for adoption of alternative management strategies, including amendment with seaweed, it is important to recognize that many changes in soil physical quality may not be observed immediately after implementation of a management change. In contrast to aggregate stability, AWC, bulk density, and infiltration rate are more ephemeral properties, and may be expected to respond within days of change in management (Islam and Weil, 2000). Rapid changes in these parameters are generally associated with tillage (i.e. mechanical disruption and aeration), although they may also be affected by substantial inputs of organic matter (Gugino et al., 2009). In this study, the amount of seaweed biomass applied resulted in a layer approximately 0.25 – 1 cm thick, a small amount of biomass relative to the volume of soil in the plow layer, which is on the order of 15 cm. Additionally, the treatments were uniformly subject to mechanical tillage following seaweed application, potentially masking any changes in AWC, bulk density or infiltration as a result of organic matter addition. High variability in infiltration rates across the treatment replicates may also have precluded the development or detection of significant differences.

Significant (p<0.05) increases in SO42- (Figure 1), EC (Figure 2), and exchangeable K+ (Figure 3) with seaweed addition support hypothesized effects of seaweed amendment. Changes in these properties were transient, with increases observed soon after seaweed addition, returning to PFF fertilizer treatment levels by the end of the growing season. Additionally, a reduction in pH was hypothesized in response to seaweed amendment, and this effect was observed for a short period (~ 1 month) after addition (Figure 4). By contrast, NO3- (Figure 5) and PO43- (Figure 6) levels did not increase in seaweed treatments (LDS and HDS) compared to the PFF treatment, supporting the hypothesis of equivalent provision of primary nutrients. In contrast, hypothesized increases in total trace elements (Ca, Mn, Fe) and toxic metals and metalloids (Pb, Cd, Cr, Zn, Hg, Cu, As) in soil were not observed (Table 4, 5). Equivalent or increased provision of primary nutrients (in this case, NO3-, PO43-, and K+) is an essential factor for successful adoption of an alternative fertilizer management practice, because limitation of these essential nutrients generally results in the most recognizable, quantifiable differences in crop growth, yield, and ultimately, economic viability. Prior to seaweed or PFF addition, soil levels of both PO43- and extractable K+ were relatively high (~60 and 150 μg/g dry soil for PO43- and K+, respectively). Based on the Cornell Soil Health Guide, extractable K+ at the site is rated as very good, exceeding the published rating chart (Gugino et al., 2009). Potassium does not pose a leaching or toxicity risk, and does not contribute to poor soil quality at high concentrations. In contrast, PO43- and NO3- can be a concern at excessive levels due to leaching risk, and improvements in soil quality decrease above maximum concentrations, reaching “poor” rating at concentrations above 30 μg/g dry soil for both PO43--P and NO3--N (Gugino et al., 2009; Marx et al., 1999; Heckman, 2003). For the soils at the study site, high natural abundance of Fe oxides and hydroxides generally allows for substantial retention of phosphate by metal-P complex formation, and the amount of PO43- in the soil solution (i.e. water-extractable PO43-) was negligible (N. Winkler, unpublished data), suggesting that the majority of PO43- extracted with NaHCO3 was previously loosely sorbed to Fe and Al oxide surfaces (Schoenau and Karamanos, 1993). In contrast to PO43- and K+, NO3- was consistently low (close to 0) across fertilizer treatments prior to seaweed application in October 2011, and early in the growing season (April and May 2012). Presumably, as availability of N from the seaweed biomass and PFF increased, NO3--N levels increased to moderate (10-15 μg N/g dry soil), but leaching risk was likely minimal due to rapid crop uptake. Addition and subsequent decomposition of seaweed biomass high in total S (~23 g/kg DM) resulted in significant differences in SO42-, but returned to control levels by the end of the growing season. Increases in SO42- could have conflicting effects on soil quality and crop production, influencing both pH and plant nutrition. For example, microbial S oxidation to sulfate results in the production of hydrogen ions (H+), reducing soil pH, with the sulfate contributing to the soluble salt content (Janzen, 1993; Germida, 2005). Alternatively, as a component of amino acids (cysteine, cystine, and methionine) and vitamins (e.g. vitamin A), S is also be a plant nutrient, and may have positive effects on crop production. Reduction in pH was observed in this study soon after addition of seaweed biomass. In addition to S oxidation as a potential influence on pH, other microbial processes contributing to reduced pH as a result of organic matter addition include (1) C mineralization and carbonic acid (H2CO3) production (Simunek and Suarez, 1993), and (2) ammonia oxidation via nitrification (Myrold, 2005). With respect to plant growth, the pH at the site was initially low, rated as “poor”, with an average pH ~5.3 in October 2011, and increased to “moderate” in the spring, with an average pH ~6.0 in April 2012 (Gugino et al., 2009). pH is a critical variable for soil quality, particularly in relation to nutrient availability; consequently, with a low initial pH, acidification as a result of seaweed addition may be of particular concern (Gugino et al., 2009). Distinction between pH reduction specifically related to seaweed addition (i.e. S oxidation) and acidification as a result of organic matter addition and decomposition is necessary for elucidation of potential pH effects of seaweed amendment. As hypothesized, EC was significantly higher in seaweed-amended plots, with higher values in HDS than in LDS. EC, which represents the ability of the soil solution to conduct an electrical charge, has a well-supported relationship with soil salinity and crop growth (Janzen, 1993). While Na+ and Cl- are the predominant ions accounting for soil salinity, Ca2+, Mg2+, and SO42- are also important components of total soil salinity (Janzen, 1993), and are considered plant nutrients. However, the negative effects of increased inorganic ions on plant and microorganism physiology are generally of greatest concern, with negative effects observed at levels above 2000 μS/cm (Janzen, 1993). In this study, EC reached a maximum of ~350 μS/cm as a result of seaweed addition, well below the risk for crop damage, even for especially salt-sensitive crops. For instance, beans (Phaseolus vulgaris), have an EC threshold of 1000 μS/cm (Maas, 1990), and sweet corn is moderately salt-sensitive, with a threshold of 1700 μS/cm (Maas, 1990). Thus, although significant EC increases were observed with seaweed addition, they did not reach levels known to have a negative effect on crop physiology. In temperate climates, dissolved salts are generally mobile in the soil, as evidenced by EC returning to PFF fertilizer treatment levels at conclusion of the growing season. Consequently, long-term accumulation may not be of concern, but increased EC remains a potential short-term negative effect.

Among the biological properties analyzed, only active carbon (C) was affected by seaweed application (Figure 7). Soil respiration (Figure 8), earthworm abundance (Figure 9) and total SOM content (Figure 10) varied over the growing season, but the same trends over time were observed for all fertilizer treatments. PMN was consistent among the fertilizer treatments throughout the growing season, with the exception of a significantly greater value in the PFF treatment in July 2012 (Figure 11). Soil respiration, a measure of overall microbial activity, was hypothesized to increase as a result of seaweed addition due to the provision of C substrates for microbial metabolism, some of which may be new to the microbial community and readily-degradable. In some cases (e.g. May, June, and July 2012), average soil respiration was greater in the HDS fertilizer treatment, but these changes were not consistent over time, and high variability within treatment replicates precluded identification of significant differences among fertilizer treatments. Consequently, the hypothesis of increased soil respiration in response to seaweed amendment was not supported. Similarly, earthworm abundance – hypothesized to decrease due to changes in soil EC – was highly variable across treatment replicates, and overall abundance of earthworms was very low across the study area. Although significant differences in earthworm abundance were not identified, some changes in chemical properties (e.g. increased EC, reduced pH, and increased heavy metal content) would be expected to affect earthworm abundance, particularly through effects on osmotic balance. Earthworms, with high surface area exposed to the soil-water environment, are particularly sensitive to changes in soil EC and other chemical properties (Lee, 1985), and their population density and biomass would be a useful continued indicator of the soil environment’s suitability for macrofauna. Potentially mineralizable N is a measure of the microbial community capacity to mineralize organic N to NH4+, a plant-available form. This process is dependent on the presence of microorganisms involved in N mineralization and availability of organic N for mineralization (Myrold, 2005). With addition of N-rich organic matter, PMN levels would be expected to increase with seaweed amendment, but hypothesized increases in PMN were not observed. The opposite effect (decreased PMN in comparison to the PFF fertilizer treatment) was observed on one sampling date (July 2012), wherein PMN in both seaweed treatments was significantly reduced. Overall, PMN was very low for all fertilizer treatments (5-10 μg N/g dry soil/week), corresponding to a rating of “poor” to “moderate” (Gugino et al., 2009), except for the sampling date following the addition of side-dress N, with values increasing to “moderate” to “good” in this treatment. In general, PMN may be correlated with factors such as organic matter, active C, and aggregate stability (Gugino et al., 2009); given moderate to low ratings in these properties in this study, low levels of PMN are a reasonable finding. In some cases, negative values of PMN were observed, indicating net immobilization of NH4+ resulting from an N-limited environment. The hypothesized increases in active C were observed on sampling dates closer to the end of the growing season. However, while increases in active C were observed with seaweed addition, these did not correspond with seaweed biomass quantity. The LDS fertilizer treatment had either equivalent or greater values than the HDS treatment on the dates when differences between seaweed and non-seaweed treatments were observed. The time required for disintegration processes (e.g. physical reduction of seaweed particle size) may be a factor in the lag in active C changes, with increases apparent approximately 3 months after seaweed biomass application (July, August, and September 2012). The average active C level at the study site across all fertilizer treatments (~600 mg/kg dry soil) is rated as “moderate” and follows the general trend of other related biological properties (e.g. SOM and PMN) (Gugino et al., 2009). In comparison to other measures of soil C (e.g. total SOM), active C responds relatively quickly to changes in management (Weil et al., 2003), representing the fraction of soil C oxidizable by dilute KMnO4. This fraction includes C in both living and dead microbial biomass, as well as C in compounds readily available to the microbial community (e.g. C in functional groups at the edge of complex organic molecules) (Weil et al., 2003). Active C is generally correlated with increased microbial activity, and is well-supported as an indicator variable of potential crop yield improvement (Weil et al., 2003). Consequently, changes in active C may represent a substantial benefit for overall soil quality and potential yield improvement as a result of seaweed amendment.

As a soil fertilizer management practice, replacing a part of the total N supply with seaweed was equally effective in terms of yield (Figures 12, 13) and above-ground biomass production (Figure 14) as using only pre-formulated fertilizer. In this study, seaweed biomass was used to replace fertilizer added at the pre-seeding, or broadcast, fertilization step. At this step, PFF fertilizer was applied at a rate of 45 kg N/ha. Due to high N requirements of sweet corn, an additional side-dress N application is usually required based on soil test results (UMass Cooperative Extension, 2013). Since NO3- and NH4+ levels were uniformly low - indicating N deficiency - prior to the sidedress stage (plant height ~30 cm), an additional 68 kg N/ha was applied to all fertilizer treatments, according to standard recommended management practices (UMass Cooperative Extension, 2013). Thus, sweet corn and soil quality production results must be viewed with the perspective of seaweed as providing a portion of overall N supply. Sweet corn yield was either equivalent or significantly greater for seaweed treatments compared to the PFF fertilizer treatment for all yield measures (average ear biomass, hundredweight, bushels/ha, and above-ground biomass). For average ear biomass, the LDS fertilizer treatment was significantly higher than either the HDS or PFF treatments (Figure 15). The remaining yield measures, as well as ear quality (dissolved soluble solids) (Figure 16), were statistically equivalent across all fertilizer treatments. While differences in yield were not statistically significant, a trend towards higher yield in seaweed-amended plots was observed for all yield measures, which may have economic relevance. For instance, the average increase in hundredweight on a per- hectare basis is equivalent to an additional $575 in income. Regardless of potential economic impact, the equivalent yield supported by seaweed amendments may be sufficient for consideration of adopting this as an alternative management practice, assuming additional costs are not restrictive. Yield across all fertilizer treatments in this study was lower than average yield for sweet corn in Rhode Island in 2007 (~45 vs. 60 hundred weight) (USDA NASS, 2013). Consequently, it is likely that supply of nutrients was equally limited for all treatments. Since extractable PO43- and K+ levels were generally high across treatments, NO3- is more likely the limiting nutrient. K+ was significantly higher in seaweed-amended treatments after spring application, but returned to PFF treatment levels at the conclusion of the growing season. At an initial (October 2011) average K level of ~150 μg/g dry soil across all fertilizer treatments, K+ was unlikely to be initially limiting, with an Cornell Soil Health overall soil quality rating of over 100% (Gugino et al., 2009). In contrast, the timing of N availability is critical for sweet corn growth, and initial values of NO3- were uniformly low across the study site. While no obvious signs of N limitation were observed in the corn crop, addition of seaweed biomass and PFF may have not been sufficient to maximize crop growth. The potential for N limitation underscores the importance of reliable predictions of N availability, based on material decomposition rates, nitrification and denitrification processes, leaching losses, volatilization, and other components of N cycling relevant to plant-available N supply. A large body of research regarding prediction of N supply as a function of material composition and climatic factors has been developed over several decades, including laboratory and field evaluations (De Neve and Hofman, 1996; Trinsoutrot et al., 2000). For seaweed biomass, the prediction of N mineralization in the field is largely unknown. For this study, a general value of mineralized N of 50% of total N was assumed based on mineralization for high-N legumes (Fox et al., 1990; Sattell et al., 1998), but this value can differ greatly depending on climatic variables (e.g. precipitation, temperature) and does not take into account losses due to leaching. As a means of strengthening N availability predictions for seaweed biomass, stable isotope techniques (e.g. enrichment of seaweed biomass with 15N) could be implemented in the field to trace the fate and mineralization rate of applied organic N.

Seaweed biomass collected for this study was composed of a mixture of red, brown, and green algal species, and in comparison to terrestrial plant biomass, was relatively high in elements important for crop growth (e.g. N, K, Ca, and Fe). Primary plant nutrients were either equivalent (N and P) or greater (K) with seaweed amendment, so the potential for realizing target crop nutrition requirements is supported. No effects, either negative or positive, were observed for the soil physical properties evaluated, which may be a function of time required for physical property change to be observed. For chemical properties, hypothesized effects on pH, EC, and SO42- were detected, with short-term decreased pH and increased EC and SO42- after seaweed addition. Decreased pH, as a critical variable for soil productivity, may be of concern for farm application, but the observed decrease is also associated with decomposition of any organic amendment. Increased EC, which at high levels may negatively affect crop growth, did not reach levels of concern for this season. Finally, SO42- production may play a part in decreasing pH, but S is also a plant nutrient, so effects may be contradictory. No effects on the biological properties of soil respiration, SOM, or earthworm abundance were observed. PMN was significantly higher in the PFF treatment in July 2012, while at the same date seaweed treatments had net N immobilization, so the prediction and consistency of N supply may be a limitation of seaweed application, as is the case for most organic amendment materials. However, a positive biological quality effect was increased active C, a soil quality indicator with good correlation to plant productivity and crop yield. Overall, effects on soil quality are both negative (e.g. decreased pH and increased EC) and positive (e.g. increased active C), but should be viewed in light of the persistence of effects, as well as the distinction between seaweed-specific and general organic matter addition effects. When applied as a fraction of overall N crop requirements, no significant differences in above-ground biomass or yield were observed, indicating that equivalent crop productivity could be obtained by implementing seaweed amendment. Additionally, while no differences in dissolved soluble solids were observed, the average weight of corn ears was greater in seaweed-amended treatments, so potential for improved ear quality and marketability may be a positive benefit of seaweed amendment. However, labor requirements and transportation costs may limit the economic viability, especially when balanced with the limited financial benefit of increases in yield. The balance of potential for increased crop success and material costs requires facilitation of improved methods and timing of collection.

• Figure 1. Mean (n=4) sulfate-S concentration as a function of fertilizer treatment and time. Months with significant differences among fertilizer treatments are indicated with (*). Error bars represent one standard deviation. Solid and dashed arrows represent seaweed and PFF application dates, respectively.
• Figure 3. Mean (n=4) extractable K+ concentration as a function of fertilizer treatment and time. Months with significant differences among fertilizer treatments are indicated with (*). Error bars represent one standard deviation. Solid and dashed arrows represent seaweed and PFF application dates, respectively.
• Table 5. Average (n=4) total nutrient element content of soil (± standard deviation) prior to seaweed addition (October 2011) and at the end of the growing season (September 2012). Within columns, values with the same letter are not significantly different.
• Figure 4. Mean (n=4) pH as a function of fertilizer treatment. Months with significant differences among fertilizer treatments are indicated with (*). Error bars represent one standard deviation. Solid and dashed arrows represent seaweed and PFF application dates, respectively.
• Figure 5. Mean (n=4) nitrate-N concentration as a function of fertilizer treatment and time. Months with significant differences among fertilizer treatments are indicated with (*). Error bars represent one standard deviation. Solid and dashed arrows represent seaweed and PFF application dates, respectively.
• Figure 9. Mean (n=4) earthworm abundance as a function of fertilizer treatment and time. Months with significant differences among fertilizer treatments are indicated with (*). Error bars represent one standard deviation. Solid and dashed arrows represent seaweed and PFF application dates, respectively.
• Figure 10. Mean (n=4) soil organic matter as a function of fertilizer treatment and time. Months with significant differences among fertilizer treatments are indicated with (*). Error bars represent one standard deviation. Solid and dashed arrows represent seaweed and PFF application dates, respectively.
• Figure 11. Mean (n=4) potentially mineralizable nitrogen as a function of fertilizer treatment and time. Months with significant differences among fertilizer treatments are indicated with (*). Error bars represent one standard deviation. Solid and dashed arrows represent seaweed and PFF application dates, respectively.
• Figure 12. Mean (n=4) sweet corn yield (in hundredweight/ha) as a function of fertilizer treatment. Treatments with the same letter were not significantly different. Error bars represent one standard deviation.
• Figure 13. Mean (n=4) sweet corn yield (in bushels/ha) as a function of fertilizer treatment. Treatments with the same letter were not significantly different. Error bars represent one standard deviation.
• Figure 15. Mean (n=4) ear fresh weight as a function of fertilizer treatment. Treatments with the same letter were not significantly different. Error bars represent one standard deviation.
• Table 3. Average (n=4) values for soil physical properties (±standard deviation). Within columns, values with the same letter are not significantly different.
• Figure 2. Mean (n=4) electrical conductivity as a function of fertilizer treatment and time. Months with significant differences among fertilizer treatments are indicated with (*). Error bars represent one standard deviation. Solid and dashed arrows represent seaweed and PFF application dates, respectively.
• Figure 6. Mean (n=4) phosphate-P levels as a function of fertilizer treatment and time. Months with significant differences among fertilizer treatments are indicated with (*). Error bars represent one standard deviation. Solid and dashed arrows represent seaweed and PFF application dates, respectively.
• Figure 14. Mean (n=4) above-ground plant biomass as a function of fertilizer treatment. Treatments with the same letter were not significantly different. Error bars represent one standard deviation.
• Table 4. Average (n=4) total toxic metal and metalloid content of soil (± standard deviation) prior to seaweed addition (October 2011) and at the end of the growing season (September 2012). Within columns, values with the same letter are not significantly different.
• Figure 7. Mean (n=4) active carbon as a function of fertilizer treatment and time. Months with significant differences among fertilizer treatments are indicated with (*). Error bars represent one standard deviation. Solid and dashed arrows represent seaweed and PFF application dates, respectively.
• Figure 8. Mean (n=4) soil respiration as a function of fertilizer treatment and time. Months with significant differences among fertilizer treatments are indicated with (*). Error bars represent one standard deviation. Solid and dashed arrows represent seaweed and PFF application dates, respectively.
• Figure 16. Mean (n=4) dissolved soluble solids as a function of fertilizer treatment. Treatments with the same letter were not significantly different. Error bars represent one standard deviation.