INTRODUCTION

Sand dunes are increasingly important to coastal resilience as sea levels rise and powerful storms become more frequent because of global warming (Del Genio, Mao-Sung, and Jonas, 2007). Low-lying coastal zones are some of the most heavily populated areas worldwide (McGranahan, Balk, and Anderson, 2007), and projected sea-level rise of up to 1.9 m by 2100 (Vermeer and Rahmstorf, 2009) may cause significant displacement and property damage. Although dunes are associated with many ecosystem services such as habitat provision, water regulation, and aesthetics (Everard, Jones, and Watts, 2010), their ability to attenuate flooding and reduce storm damage is arguably the most important to coastal property owners (Sigren et al., 2018). Native plants help maintain this service by stabilizing dunes through root and rhizome growth (Sigren, Figlus, and Armitage, 2014; Silva et al., 2016). Plant shoots also help to trap sand blown by wind and were found to play the key role in controlling maximum dune height (Durán and Moore, 2013). In New England, Ammophila breviligulata (American beachgrass) is the most common dune plant (Dunlop and Crow, 1985; Godfrey, 1977), but extensive die-off of A. breviligulata along the New Hampshire seacoast and North Shore of Massachusetts has raised concerns over the potential loss of productivity and dune stability.

Die-off of A. breviligulata and other dune plants has been well documented (Bourdreau and Houle, 2001; Seliskar and Huettel, 1993). In Delaware, Seliskar and Huettel (1993) showed large areas of dunes that were nearly devoid of vegetation and identified parasitic nematodes as the cause. However, a more recent genomic study suggests die-off in some New England dunes results from a combination of nematode and fungal infection (G. Moore, unpublished data). Because dunes are generally nutrient-deficient systems, fertilizer application may increase productivity of A. breviligulata (Boudreau and Houle, 2001; Day et al., 2004; Seliskar, 1995) and make it more resistant to infection. While A. breviligulata may benefit from fertilization in terms of total stem density and number (Day et al., 2004), overall plant diversity may decline as a result of chronic fertilization treatments (Day et al., 2018). A study in Delaware also found that dune soil was acidic in die-off areas and that the application of lime to increase pH resulted in greater A. breviligulata survival (Seliskar, 1995).

Seliskar's (1995) work in the mid-Atlantic warranted an examination of whether fertilizer and lime addition or their combination would have similar benefits to A. breviligulata and other dune plants in New England. A field experiment was conducted to determine whether these soil amendments could enhance recovery of A. breviligulata in replanted areas following a die-off event.

METHODS

The effects of fertilizer and lime were examined on dune plant communities at three sites on the northern end of Plum Island, Massachusetts: Reservation Terrace (42°48′51.9″ N, 70°48′47.8″ W), Penny Lane (42°48′36.0″ N, 70°48′34.2″ W), and Temple Blvd (42°47′28.5″ N, 70°48′28.0″ W) (Figure 1). The three sites were selected because of recent evidence of die-off (e.g., extensive areas of rapid defoliation spreading in a radial pattern outward across the dune over one growing season). Each of these recently barren areas were hand-planted with A. breviligulata at 30 cm centers in spring 2017 with peat moss and fertilizer. As documented by others in the region (Dunlop and Crow, 1985), surrounding vegetated areas of the dune contained typical native dune vegetation, primarily comprising A. breviligulata with scattered seaside goldenrod (Solidago sempervirens), beach pea (Lathyrus japonicus), seabeach knotweed (Lechea maritima), jointweed (Polygonella articulata), sea rocket (Cakile edentula), and beach clotbur (Xanthium echinatum), whereas beach heather (Hudsonia tomentosa) was present at only the Temple Blvd site (see supplemental material for percentage of cover of all species present).

View Figure

• Download as Powerpoint
• Download Figure

Experimental Design

A randomized block design was used for the experiment with three sites (blocks) comprising four treatments: (1) control (no amendments), (2) lime, (3) fertilizer, and (4) lime+fertilizer (Figure 2). The plot size was 16 m², with four treatment replicates per site. A commercially available slow release fertilizer (Osmocote_™ N14-P14-K14) was applied by hand in June 2017 at the rate of 720 mL or ≈765 g per plot (67 kg N/ha, 29 kg P/ha); pulverized Dolomitic Limestone (21.6% Ca, 10.0% Mg), hereafter referred to as lime, was spread using a sieve at a rate of 1.79 kg per plot (1120 kg/ha) (Seliskar, 1995). Fertilizer and lime were partially mixed into the sand after application to limit transport by wind or rain.

View Figure

• Download as Powerpoint
• Download Figure

RESULTS

Lime application resulted in higher soil pH, which ranged from 5.2 to 7.2 in controls. Plots treated with lime had a soil pH approximately 0.5 units higher than controls and fertilizer-only plots (Figure 3). The effect of lime on soil pH was statistically significant (p < 0.001) as well as the effect of site (p < 0.0001), with the highest pH measured at Penny Lane. The interaction between treatment and site was not significant; the buffering effect of the lime was similar across all three sites.

View Figure

• Download as Powerpoint
• Download Figure

Overall, plants appear to have benefited from fertilizer and lime application. Total plant cover was low (10–40%) but improved significantly as a result of soil amendments. Total plant cover was 72% higher in lime+fertilizer plots than in controls (p < 0.05; Figure 4), and cover was higher at Reservation Terrace than at Temple Blvd (p < 0.01). The percentage of cover of A. breviligulata showed a similar pattern, where the average was highest in the lime+fertilizer plots and lowest for controls and lime plots (Figure 5). The effect of soil amendments on A. breviligulata percentage of cover was significant (p < 0.05), and cover was higher at Reservation Terrace than the other sites (p < 0.0001). Diversity ranged from 0.55–0.62, and species richness ranged from 5.2–5.9 (Table 1). Neither metrics were affected by soil amendments. Although the percentage of live A. breviligulata plants differed between sites (p < 0.0001), there was no effect of soil amendments when data were averaged from 2017 and 2018 (p = 0.120; Figure 6a). However, when only 2018 data from the remaining two sites were analyzed, the effect was significant (p < 0.01), and the percentage of live plants was higher in lime+fertilizer plots by nearly 20% (Figure 6b). Means for each site and year are provided in the online supplemental material. The tallest A. breviligulata plants were found in lime+fertilizer plots (Figure 7), but the effect was not significant (p = 0.073). Fertilizer and lime-fertilizer plots contained the highest number of reproductive shoots, but differences were not significant because of high variability (Figure 8).

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

Table 1 Species richness (± standard error [SE]) and Shannon-Weiner diversity index calculated using percentage of cover. Data from all sites and years were averaged (n = 12 plots). No significant differences among treatments were found.

View Fullscreen

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

DISCUSSION

The results suggest that a combination of lime and fertilizer helped replanted dune communities recover from die-off events, as shown by the greater percentage of live plants and cover of A. breviligulata. In a similar study in Delaware, Seliskar (1995) found that nutrient addition resulted in greater plant height, culm bunch circumference, number of shoots per bunch, and percentage of cover of A. breviligulata. While Seliskar (1995) did not show the combined effect of fertilizer and lime, lime-only treatments increased A. breviligulata survival in dunes where initial pH had ranged from 4.0 to 5.5. Because the control pH in the present study sites ranged from 5.2 to 7.2, dune soils may not have been acidic enough for the lime-only treatment to significantly benefit plants; however, the combination of lime and fertilizer amendments resulted in a greater percentage of live plants in 2018, after plants had been exposed to a winter. In another field experiment in Quebec, Boudreau and Houle (2001) found that shoot biomass and density of A. breviligulata were higher in fertilized dune areas than in controls. Similarly, Day et al. (2004) showed total stem density and Ammophila-specific stem density increased during a 10-year fertilization period in a coastal barrier island in Virginia; however, Day et al. (2018) later noted that fertilizer in these plots had a negative effect on dune plant diversity. Soil amendments in the present study did not appear to decrease species diversity, nor was there a significant effect on species richness over the course of the study. The lack of effect on these parameters may be a result of low overall plant cover following die-off, the lower rate of fertilizer application, or the shorter study period. The beachgrass native to Europe, Ammophila arenaria, was also found to benefit from fertilizer, with one study showing greater productivity in fertilized areas (Van Der Putten, 1990) and another showing greater plant height and shoot density (Willis, 1965).

A series of Nor'easters resulted in the loss of Reservation Terrace over the winter of 2017–18. This site was located just north of a jetty extending from the mouth of the Merrimack River. The jetty was restored by the Army Corps of Engineers in 2014, resulting in increased water velocity that may influence beach scouring and rates of beach and dune erosion witnessed at the site. Quarterly dune profile surveys have documented extensive erosion resulting in loss of the beach front and dune crest at Reservation Terrace (G. Moore, unpublished data). Changes to water flow and sediment dynamics, combined with effects of more frequent severe storms likely contributed to the dune loss at Reservation Terrace. At Penny Lane and Temple Blvd, the effect of lime+fertilizer on all growth metrics increased between 2017 and 2018, suggesting additional supporting results if Reservation Terrace had not been destroyed in 2018.

By improving the health of dune plant communities through soil amendments, coastal resource managers may increase resilience to storm flooding. Studies have shown that belowground plant growth helps to stabilize dunes and prevent erosion (Sigren, Figlus, and Armitage, 2014; Silva et al., 2016). Although belowground growth was not measured in this study, the finding that fertilizer and lime increased the percentage of cover and percentage of live plants of A. breviligulata suggests there will be more plants in soil amendment areas and therefore more roots stabilizing the dunes. Tiller density and leaves per tiller of A. breviligulata have been shown to correlate positively with sand accumulation (Emery, Bell-Dereske, and Rudgers, 2015), suggesting that the higher plant cover found in amended dune soils will trap more sand and increase dune height. Because burial results in greater biomass of A. breviligulata (Disraeli, 1984; Harris, Zinnert, and Young, 2017) and other dune plants (Perumal and Maun, 2006), an enhanced sediment-trapping effect could improve plant growth while also building a taller dune barrier against wind-driven waves and storm surges as sea levels rise. Healthier plant communities will also improve habitat for small mammals, insects, and birds such as the threatened piping plover (Charadrius melodus), which has been shown to preferentially nest in areas with higher A. breviligulata shoot density at some sites (Flemming, Chiasson, and Austin-Smith, 1992). Although excess nutrients from fertilizer can contribute to eutrophication in nearby water bodies, this is more of a concern in enclosed areas such as estuaries and coastal lagoons (Bricker et al., 1999; Taylor et al., 1995) rather than open coasts. Whether planting because of die-off, blow outs, or following sand additions (i.e. nourishment) to beaches and foredunes, the addition of fertilizer and lime treatments at levels used herein is expected to stimulate growth of common native dune species and may be an effective approach to increase dune resilience.

CONCLUSIONS

The application of fertilizer and lime to coastal dunes appears to be a cost-effective way to promote regrowth following die-off and to build dune resilience to climate change. More long-term studies should be done to determine the optimal application frequency for dune plants in New England and potential impacts to water quality in sheltered areas. Dunes are dynamic environments that are constantly changing in size, shape, and position because of the forces of wind and water. No amount of soil amendments or vegetation can prevent erosion entirely, but healthier plant communities will benefit both the natural ecosystem and the property owners that depend on dunes for protection from storms and sea-level rise.