Differential recolonization of Atlantic intertidal habitats after disturbance reveals potential bottom-up community regulation

In the spring of 2014, abundant sea ice that drifted out of the Gulf of St. Lawrence caused extensive disturbance in rocky intertidal habitats on the northern Atlantic coast of mainland Nova Scotia, Canada. To monitor recovery of intertidal communities, we surveyed two wave-exposed locations in the early summer of 2014. Barnacle recruitment and the abundance of predatory dogwhelks were low at one location (Tor Bay Provincial Park) but more than 20 times higher at the other location (Whitehead). Satellite data indicated that the abundance of coastal phytoplankton (the main food source for barnacle larvae) was consistently higher at Whitehead just before the barnacle recruitment season, when barnacle larvae were in the water column. These observations suggest bottom-up forcing of intertidal communities. The underlying mechanisms and their intensity along the NW Atlantic coast could be investigated through studies done at local and regional scales.

high. For example, in wave-exposed habitats where algal canopies and sessile invertebrates (barnacles and mussels) were abundant before the arrival of the ice, only bare rock was visible after ice scour 4 .
To evaluate recolonization patterns, in the summer of 2014 we surveyed two wave-exposed locations that had been heavily scoured by ice in early April 4 : Whitehead (45.212° N, 61.174° W) and Tor Bay Provincial Park (45.183° N, 61.355° W; Figure 1). The surveyed intertidal habitats face the open Atlantic Ocean directly. On 23 June 2014, at each location we measured the density of barnacle recruits (Semibalanus balanoides) in 8 quadrats (10 cm × 10 cm) that we had randomly established along 30-m transect lines at the mid-to-high intertidal zone in late April. Because of the intense ice scour in early April, macroscopic organisms were absent at this zone in late April, so the substrate was then fully available for barnacle recruitment (barnacles are often the first sessile invertebrates to recolonize disturbed intertidal habitats 5,6 ).
Semibalanus balanoides is the only species of intertidal barnacle on this coast 1 . Every year, recruits of S. balanoides accumulate in intertidal habitats on this coast during May and June 7,8 . Our measurements (Dataset 1) on 23 June (after which no new recruits appeared) indicated that barnacle recruit density was significantly higher (Student's t 14 = 3.10, P = 0.017) at Whitehead (199.8 ± 62.0 recruits dm -2 , mean ± SE, n = 8 quadrats; Figure 2) than at Tor Bay Provincial Park (7.3 ± 4.2 recruits dm -2 ; Figure 3). This statistical   Number of barnacle recruits (Semibalanus balanoides) occurring on 23 June 2014 in 8 quadrats (10 cm × 10 cm) that had been randomly established along 30-m transect lines at the mid-to-high intertidal zone in late April (just before the beginning of barnacle settlement) at a wave-exposed site in Whitehead and in Tor Bay Provincial Park.
The greater density of barnacle recruits at Whitehead than at Tor Bay Provincial Park was related to a higher nearshore chlorophyll-a concentration during late March and April at Whitehead, according to MODIS satellite data (Table 1; National Aeronautics and Space Administration). Nearshore chlorophyll-a concentration indicates coastal phytoplankton abundance, and phytoplankton is the main food source for barnacle nauplius larvae 9 . For S. balanoides from the Atlantic coast of Nova Scotia, nauplius larvae occur in coastal waters for 5-6 weeks before metamorphosis to cyprids and then intertidal settlement 10 , which starts in early May on our studied coast 8 . Thus, it is possible that the higher food supply for larvae at Whitehead than at Tor Bay Provincial Park may have ultimately contributed to determining the higher barnacle recruitment at Whitehead. A positive relationship between nearshore phytoplankton abundance and intertidal barnacle recruitment was previously documented for NW Atlantic intertidal systems at a regional scale 7 .
To see whether barnacle recruitment could influence higher trophic levels, we measured the abundance of dogwhelks (Nucella lapillus; Figure 4) shortly after the end of the barnacle recruitment season. Nucella lapillus is the main predator of barnacles on the studied coast 1 , so presumably a higher barnacle recruitment could locally increase dogwhelk abundance. On 15 July 2014, at each of the  Relationships between coastal chlorophyll-a concentration, intertidal barnacle recruitment, and intertidal predator impacts have been identified on Pacific rocky shores 6 . The positive influence of prey food supply on predators mediated by prey recruitment is referred to as bottom-up regulation of community structure 12 . Coastal configuration and water column movements influence nearshore phytoplankton abundance 6 . What caused the phytoplankton difference between our two studied locations remains to be determined. However, the observed link between phytoplankton abundance, barnacle recruitment, and dogwhelk density does suggest that bottom-up forcing may also structure NW Atlantic intertidal communities. Understanding the underlying mechanisms and their intensity along the coast could be achieved with a larger spatial monitoring and field experimentation. Author contributions WP and RAS did the field surveys. RAS wrote the manuscript and WP provided critical comments to produce the final version.

Competing interests
No competing interests were disclosed. In this article, Petzold and Scrosati provide observational evidence suggesting that food supply drives the re-colonisation of benthic organisms in wave-exposed intertidal communities after being disturbed by drifting ices. The title and abstract of the article are appropriate for the content of the work and represent a suitable summary of it. Although the authors report the results of only one site for each "condition" (i.e. one site with high and one site with low Chlorophyll-a concentration), the differences in food supply between these sites are strong enough to propose further studies analysing the role of bottom-up forcing in this system. Accordingly, the conclusions of the study are sensible and justified on the basis of the results. As a follow-up approach, the authors may consider first, to expand the study to a larger set of sites in order to confirm the pattern, and second, to construct competing hypotheses (in addition to the "bottom-up hypothesis") that would be tested by means of manipulative experiments replicated at those sites. This article is a necessary benchmark from which further hypothesis-driven research should be conducted I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

Grant information
No competing interests were disclosed. Competing Interests: 23  This study is an interesting first approach to understanding the factors that affect the ability to recruit sessile benthic organisms after disturbances in an intertidal community. Petzold and Scrosati address the issue concerning the role of larval supply and its relation to phytoplankton abundance, which leads directly to considering that a bottom up mechanism may be acting in these communities. Furthermore, the idea that began the work is clearly stated and the theoretical context is appropriate, as well as the statistical treatment of the data. Although a more deep and permanent approach to the problem requires statistical treatment of the data. Although a more deep and permanent approach to the problem requires the carrying out of further studies, perhaps with the incorporation of field experiments, this initial statement is correct and I recommend its indexing.

I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.
No competing interests were disclosed. Competing Interests: