sign in sign up
<<
Home , Coast, Marine biology, Tide pool

UNIVERSITY OF HAWAI‘I AT HILO HOHONU 2020 VOL. 18

Effects of , tidepool shorelines, their biota are generally less -stud- ied compared to that of broader intertidal areas or nearby shallow- environments in assemblages at Lalakea and which they are generally nested (Almada & Faria Lehia Parks, Hilo, 2005; Cox et al. 2013; Wiegner et al. 2016). Little - Lisa L. K. Mason, Nikola Rodriguez, Sebastian and fewer studies have examined factors affecting Abraham Waiola and dynamics of tidepools Conservation Environmental Science 601 at either the regional or local scale (Cox et al. 2011). The high variability of tidepool structure Abstract and functionality from one locale to another poses Marine tidepools are unique that unique challenges for comparing tidepool char- provide critical and refugia to many near- acteristics between locations (Arakaki & Tokeshi . The and diversity of spe- 2019). Nonetheless, tidepools are important eco- cies within tidepool communities are highly vari- logical zones that are deserving of more focused pool size, water quality, and their proximity to the ocean and outlying reefs. In this study, we exam- Tidepools are distinct rocky marine or estua- ined the effects of three physical characteristics rine that are daily connected to the ocean - assemblages were assessed in terms of abun- pletely isolated from surrounding water bodies dance, , and species composition. during low tides. Tidepools are known to a We tested the effects of ocean tides (high vs low), diverse collection of marine species. Some species tidepool location (Lehia vs Lalakea), and tidepool may live their entire in tidepool areas, while others are only temporary residents that eventually - disperse to outer ocean areas as they outgrow the - - cies richness at high tides and lower at low tides. semblages and may disperse far beyond local We found that changes in ocean tides did not affect coastlines (Stamoulis et al. 2018; Friedlander et al. - 2019). Tidepools vary widely in their abiotic char- acteristics (Nakamura 1976; Todgham et al. the larger tidepools at each location, and higher 2006; Cox et al. 2011), community compositions (Metaxas & Lewis 1992; Shelton 2010), and en- a difference in species composition between loca- ergy dynamics (Trussell et al. 2004), and serve as tions and between pool sizes. We found that ma- unique between terrestrial and marine nini (Acanthurus triostegus), aholehole (Kuhlia ecosystems (Ray & Hayden 1992). Many taxa uti- sandvicensis), and mamo (Abudefduf adbomina- lize tidepools as breeding areas (Moring 1986), lis refugia (Underwood & Chapman 1996) from predators (Dethier 1980), and as integral parts of assemblages across sites similar to the ones in this study may serve as a critical strategy for sustain- Tidepools comprise rocky substrata (Arakaki et al. 2014) that provide ample shelter and resources Introduction One of the most prominent and diverse features 1990; Bezerra et al. 2017). in tidepool sys- of Hawaiian intertidal areas are tidepool systems. - Although tidepools are common along Hawaiian dents or transients (Machado et al. 2015) whose

108 UNIVERSITY OF HAWAI‘I AT HILO HOHONU 2020 VOL. 18

movements across tidepool systems can strongly of the calendar and provided inspiration for this - research. nity structure (Metaxas & Scheibling 1993) and - trophic dynamics (Castellanos-Galindo & Giraldo - 2008) across the system. Little is known about the degree to which phys- - - semblages at high and low at Lalakea Beach tion of biological communities in tidepool sys- abundance and species richness would be greater at high tide than low tide across locations because (i.e., abundance, species richness, and diversity) deeper pools and wider ocean connections at high in tidepool systems across the world are related to - factors such as geographical location and latitude (Arakaki et al. 2014), (Cox et al. 2011; would differ based on location and pool size, with Shelton 2010), pool areas and depths (Mahon & no effect based on high or low tides. At each site, Mahon 1994; Cox et al. 2011), microhabitat fea- tidepools were situated at varying distances from tures (Willis & Roberts 1996), and the physical makeup of the (Metaxas et al. 1994; Cox could be found at each of the pools at the different et al. 2011; Bezerra et al. 2017). Additionally, locations. Additionally, we predicted that larger shoreline topography (Archambault & Bourget pools would contain a larger number and higher & Scheibling 1993) may affect the biota of marine of resources in larger areas and deeper water. We coastal areas including tidepools. These types of interactions are well-studied for many other tropi- Beach and Lehia Beach parks, where we counted cal and temperate locations but not necessarily in - - pool at different tides during November 2019. line topography, their degree of connectedness to Methods the open ocean, and proximity to the waterline are Study sites considerably variable across Hawaiian coastlines. It is not known to what effect diel changes such as two sets of rocky intertidal tidepools along the - Beach (LALA), is situated near residential prop- A major source of traditional ecological knowl- erties and is exposed to moderately high levels (Friedlander et al. 2002). This tool was constructed Lehia (LEHI), is located away from residential by Hawaiians of old to reinforce understandings properties, yet, still experiences some level of of the interconnectedness between terrestrial and marine systems. Additionally, it describes many of sites due to the presence of the set of comparable the natural rhythms of physical and ecological pro- tidepool pairs (i.e., relatively similar sizes, depths, cesses that guide harvesting and replenishment of and ) at two different locations along the coastline. Additionally, these sites were easily ac- spawning, aggregation and feeding behaviors over cessible for surveys. the course of a moon cycle can be gleaned from The presence of apparent natural features the calendar making it a practical guide for many around each tidepool helped distinguish them as ecologically unique features within the . al. 2007). The inherent connection between tides Each set of adjacent tidepools was made up of one larger (Pool 1) and one smaller (Pool 2) tidepool

109 UNIVERSITY OF HAWAI‘I AT HILO HOHONU 2020 VOL. 18

connected at some parts of their borders to a more extensive pool system. Approximate measure- surveys in two phases. During phase 1, our team ments for pool areas were taken using Google Pro (2019). Large pools areas ranged from species presence. During phase 2, we counted the 106 m2 to 211 m2, and small pool areas ranged 38 to 54 m2 minutes. To prevent double counting, we did not of the pools were measured using a meter stick and - ranged from 0.48 m (small) to 1.91 m (large) at their deepest points during high tide. At each lo- for each species in all tidepools for each location, cation, these sets of tidepools directly connect to day, and tide. We also recorded the total number of species present in all tidepools for each location, day. However, natural lava rock boundaries gener- day, and tide. ally protect them from surges. Lehia’s pools Data analyses were consistently connected during high and low tides, while Lalakea pools were isolated from each other during low tides. The water at both sites was tidepool survey per day. We determined species brackish in some places because of underground richness as the total number of different species freshwater springs. Several natural features were observed per pool per day. For each site, we ana- similar between the two locations, including benthic substrates, pool sizes, and pool depths. for abundance and species richness. Checks for Benthic substrates at both sites were relatively het- normality revealed our data to be nonnormal. We erogeneous with sections of sandy bottom, small analyzed the effects of tides, locations, and pool boulders, and rocky rubble, and minimal vegeta- tion, or human-made . tests. We analyzed the effects of tides, locations, Field days and pool sizes on species richness and composi- tion using the chi-squared test for independence. (NOAA) to ensure surveys occurred within a one Results to a two-hour margin of time, encompassing the Fish abundance peak tide of the day. We surveyed all pools at each site for either the high tide or low tide within a during high tides and low tides (W = 79874, p- single day. We collected data at each site location value = 0.4401). We also found no differences in tides and low tides at Lehia (W = 14246, p-value before sunrise or after sunset. The duration of the = 0.8616) and at Lalakea (W = 27136, p-value = study lasted during one lunar cycle. We also refer- enced the Hawaiian moon calendar for supplemen- in Lalakea tidepools than in Lehia tidepools (W = - - cultural practices during this time (i.e., Makahiki) dances between the large and small tidepools were that may have impacted our surveys. similar (W = 79489, p-value = 0.7695). Fish surveys Species richness counts for all forty surveys. Completion of a total both sites during this study. There were thirty spe- for all four tidepools occurred during November 2019. We completed our underwater surveys by 1). We found that tides did not affect overall spe- in the tidepools. In some cases, sur- 2 = 19.937, df = 35, p-value = face observations and counts were possible at low 2 tide and during calm conditions. We divided each 2 = tidepool into three equally sized stations labeled 11.25, df = 9, p-value = 0.259, Figure 4). We also

110 UNIVERSITY OF HAWAI‘I AT HILO HOHONU 2020 VOL. 18

found that overall species richness was similar be- assemblage dynamics in these particular Keaukaha 2 = 10.533, df = 8, p-value = tidepool areas. 0.2296). Isolating the effects of tides in intertidal areas Fish community composition - - raphy of coastlines, benthic composition, size of 2 = pools, distance from ocean sources, and and 39.954, df = 34, p-value = 0.2226). We found dif- ferences in the types of species represented at each 2= 101.91, df = 34, p-value = 1.057e- 2011). On several days, we encountered unexpect- 08) and in the larger tidepools compared to the ed wave action and moderately strong currents in 2 = 57.394, df = 34, p-value our pools from ocean swells. Surges of swell water = 0.007306). The most abundant species for both were most noticeable at high tide, heightening wa- sites were the manini (Acanthurus triostegus), ter levels in pools, and possibly diluting any subtle aholehole (Kuhlia sandvicensis), and the mamo differences in water level caused by tide alone. (Abudefduf abdominalis). The least abundant The Lalakea tidepools received more direct ocean Bothus mancus Myripristis ber- ndti), and the roi (peacock grouper, Cephalopholis argus). the tidepools. Discussion The Hawaiian experience mixed tidal cycles throughout a lunar day. Differentiation be- and species richness in tidepools at high tide than tween the highest-high and lowest-low tides is es- low tide. Likewise, we predicted to see changes sential for studies designed around tidal surveying. in species composition within tidepools between high tide and low tide. Our predictions were based and low tide captured some of the tidal variations on the idea that higher water levels, larger pool within November. We failed to survey any of the surface area, and greater pool depth would facili- lowest-low tides due to their occurrence during non-daylight hours and were unable to collect data of the tidepools, increase food resources and mi- during the tide for November. We were also crohabitats to support a larger community, and al- unable to complete both high tide and low tide low transient or visitor species that were not pres- surveys for both sites on the same day which may ent at low tide into the tidepools. Additionally, it have introduced additional confounding variables. was also possible that sweeping changes in abiotic During analysis, we discovered that we had sur- veyed during a mid-tide and incorrectly recorded it as high tide. These inconsistencies in surveying may have impacted the results of this study. richness, or species composition between high tide We assumed at the beginning of our study that and low tide in any pools at either site. This effect was surprising, however, may be due to the nar- - ity at our sites. Freshwater springs are prevalent during this study was 0.64 m (SD 0.16), and the along the Keaukaha coastline. On days of low tide, mean low tide was 0.21 m (SD 0.02), according - pools at Lehia and along the inshore edge of the highest historical tide recorded in Hilo was 1.18 smaller tidepool at Lalakea. Despite this apparent m (date unknown; NOAA), which is a relatively phenomenon, we did not collect consistent abiotic low record compared to tidal swings along temper- data for the tidepools during our surveys. Our lack ate with steep topography and higher lati- of water quality data, particularly for water tem- tudes (Nakamura 1976; Metaxas et al. 1994). We perature and , left a gap in our analyses and limited the interpretations of our results. Although

111 UNIVERSITY OF HAWAI‘I AT HILO HOHONU 2020 VOL. 18

single location. This result suggests that the mech- no strong correlation between water temperature, anisms dispersing and maintaining marine species throughout the region (i.e., Keaukaha, reefs) may (Metaxas & Lewis 1992) we cannot assume this to be acting somewhat evenly across the broader be true for all places. Moreover, these factors may coastal marinescape. Future spatial and temporal not necessarily limit the distribution of certain studies that extend beyond the Keaukaha et al. (2011) modeled the effects of various abiotic further imply that human-induced , such - between the locations as we initially suspected. basalt composition of most coastal zones tends to- Further study on human impacts in these areas is wards higher thereby affecting water needed. temperatures. Further research may reveal insight - into the roles that different abiotic factors play dance or species richness between larger and smaller tidepools. However, we know this effect is like ecosystems similar to that of the Lehia and not common for many other types of tidepool sys- Lalakea areas. Certain types of vegetation (i.e., shading generally contain greater abundance and species richness (Mahon & Mahon 1994). In temperate lo- biological community. Shelton (2010) conducted a study on the effects of removal in tide- by increasing body size. But, greater can pools and found that areas lacking seagrass ex- only be supported in tidepool systems if more re- sources are available (Arakaki & Tokeshi 2019). within a tidal cycle. It is possible that vegetation During this study, we observed what we assumed in and along the waterline can have an effect on temperature, , or input in tidepools. Furthermore, the presence of tall trees once they reached their terminal size. Assemblage and grassy vegetation at Lehia may also have an characteristics are also likely due to avail- effect on tidepool water quality compared to the - open areas surrounding Lalakea tidepools. tition, which can potentially lead to a sort of over- crowding effect in tidepools (Castellanos-Galindo which may be due to more direct connections to the & Giraldo 2008). This overcrowding effect (i.e., ocean and an increase in wave action (Friedlander increased number of individuals with decreased et al. 2003), or more stable temperatures and pool size) may be due to fewer predator species higher salinity than at Lehia (James et al. 2008). in tidepools (Dethier 1980) and more complex re- A confounding factor also lies with differences source partitioning (Bezerra et al. 2017). Although we did not see a difference in spe- - cies richness in any of our tidepools, it is pos- - sible for high species richness to be sustainable intentionally but at low levels. From our surveys, within smaller tidepool systems through complex resource partitioning. Resource partitioning co- individuals of aholehole, Kuhlia sandvicensis) and incides with habitat preferences and depends, in a greater abundance of gobies (family ), part, on intertidal height. We noticed certain spe- - cies were present more often within certain strata ily Synodontidae) at Lalakea compared to Lehia. of the . Mamo (Abudefduf adbomi- Species richness and species composition were nalis) are generally that swim near similar between our locations. Of the total 34 spe- the surface to mid-depths of the water column. cies observed, only three species appeared at a ) was mostly

112 UNIVERSITY OF HAWAI‘I AT HILO HOHONU 2020 VOL. 18

observed at the bottom of each of the tidepools area (i.e., algal growth substrate) thereby affecting along the sandy bottom. Manini (Acanthurus tri- ostegus) are and were most frequently observed on rocks for . The most such as and (Arndt Keaukaha tidepools during this study were herbi- & Fricke 2019). - Tropical tidepool assemblages are shown to ing health and controlling algae overgrowth have more seasonal stability than in temperate ar- (Williams et al. 2016). High levels of herbivory eas (Harasti et al. 2016; Arakaki & Tokeshi 2019). can be supported in tidepools containing greater Albeit, to what degree this stability exists is some- heterogeneity of substrate types (Metaxas et al. what speculatory based on the vast number of dif- 1994). fering environments and numerous factors affect- ing tropical tidepool systems. The dynamic that species composition did vary by pool size. This of tidepool environments is not suitable for all ma- result may be due to shallower and more isolated rine species. Some tidepools may contain a greater number of eurytopic species with high tolerance to environmental changes, such as inclement and swell activity, within short periods. However, example, the Peacock Flounder (Bothus mancus) overall mild seasonal variation may help is one of the cryptic species we encountered. Its coloration is almost identical to the sandy bottom - where they sit still even when approached. Eels were another group that we assumed were present, which may support greater stability than their tem- although we were unable to account for them if we perate counterparts (Castellanos-Galindo 2005; did not see them in our pool. Additionally, some Bezerra et al. 2017). Yet, many anthropogenic species chose to reside just outside the boundary of our tidepools during our counts and therefore exacerbating these natural pressures. were not included as present in some surveys. There are many threats to Hawaiian coastal Repeated surveys may help account for these types and marine ecosystems, namely the establishment of disparities. of (Simberloff 2010). Invasive Manini, aholehole, and menpachi were com- species control is a challenging, yet, imperative monly observed in our study, are well-known as 2002; Duffy & Martin 2019). With the introduction and have a well-established economic and cultural value in many communities. Further study of the a growing need for active and aggressive manage- economic and cultural importance of nearshore ment to curb the onslaught of degra- marine species are needed in order to better moni- dation. An example of a coastal invasive species tor their abundance levels. We were unable to cal- is the red (Rhizophora mangle). Red culate index values during this were, at one point, the target of many eradication efforts across the Hawaiian Islands, research and monitoring of these valuable species and locations. to control mangrove resorted to the - use of herbicides that were feared to irreversibly cators of broader community scale changes in the ecosystem. For example, changes in tidepool as- - ter treatment of mangroves revealed negligible that surround the intertidal area (Almada & Faria 2005) and degradation of rocky microhabitats may (Mackenzie & Kryss 2013). Any negative impacts reduce the amount of shelter or food production that may have occurred were short lived and not

113 UNIVERSITY OF HAWAI‘I AT HILO HOHONU 2020 VOL. 18

detectable towards the end of the study. Despite Literature Cited the questionable resilience of tidepool commu- Allen J. 1998. Mangroves as alien species: the case nities, certain tidepool ecosystems can persist of Hawaii. Global & through episodes of acute stress, especially during Letters 7:61–71. the removal of these long-term threats. The cost- Almada VC, Faria C. 2004. Temporal variation warrants the removal of invasive vegetation that may ultimately worsen long-term effects on coast- - patterns and possible mechanisms with a al and marine environments. note on sampling protocols. Reviews in Fish of invasive species is one area in which further Biology and 14:239–250. knowledge of the biology and ecology of intertidal Almany GR. 2004. Does increased habitat complexity reduce and Understanding local community dynamics of 106:275– coastal and marine areas are essential for ecosys- 284. tem level conservation and management of natural Arakaki S, Tsuchiya M, Tokeshi M. 2014. and cultural resources (Tsang et al. 2019). Future research on topics of water quality, seasonality of assemblages: local substrate characteristics affect regional-scale trends. Hydrobiologia can build on this study. Improved management 733:45–62. of nearshore areas may help to support sustain- Arakaki S, Tokeshi M. 2019. Biomass compensation: Behind the diversity gradients 61:396– and more complex ecosystem-level community 405. dynamics can support policies that aim to protect Archambault P, Bourget E. 1996. Scales of coastal and the coastal gems that contribute to heterogeneity and benthic intertidal species richness, diversity and abundance. Marine resource management projects is a necessary com- Ecology Progress Series 136:111–121. ponent of Hilo’s coastal areas. Collaborative ef- forts that incorporate local ecological knowledge with special reference to shallow (i.e., Hawaiian moon calendar) into conservation tidepools. Biodiversity Data Journal 7. participation and ensure that future study of these Bezerra LAV, Padial AA, Mariano FB, Garcez types of unique systems may continue. DS, Sánchez-Botero JI. 2017. Fish diversity in Acknowledgments tidepools: assembling effects of environmental Mahalo nui to my amazing research team, N. heterogeneity. Environmental Biology of Fishes Rodgriguez and S. Wells, for your tireless dedi- 100:551–563. cation, immense intellectual contributions, and Castellanos-Galindo GA, Giraldo A, Rubio EA. friendly support during the entirety of this project. 2005. Community structure of an assemblage Mahalo T. Grabowski, R. Ostertag, and J. Sutton for materials, expertise, and valuable feedback , Colombia. Journal of Fish Biology that helped to improve the project. Mahalo H. 67:392–408. Mossman for your insights into the cultural im- Castellanos-Galindo GA, Giraldo A. 2008. Food plications of our project and Hawaiian perspec- tives of our system. Mahalo E. Karth and UH Hilo 153:1023–1035. Mahalo J. Savage and J. Mason for volunteering Cox TE, Baumgartner E, Philippoff J, Boyle KS. 2011. Spatial and vertical patterns in the for photos and video documentation. Environmental Biology of Fishes 90:329–342. 114 UNIVERSITY OF HAWAI‘I AT HILO HOHONU 2020 VOL. 18

Cox TE, Philippoff J, Baumgartner E, Zabin Kay AE. 1994. A of the Hawaiian CJ, Smith CM. 2013. Spatial and Temporal Islands: selected readings II. University of Variation in Rocky Intertidal Communities Hawaii Press, Honolulu, HI. 186-195. MacKenzie RA, Kryss CL. 2013. Impacts of Science 67:23–45. exotic mangroves and chemical eradication Dethier MN. 1980. Tidepools as refuges: Predation and the limits of the harpacticoid Marine Ecology Progress Series 472:219–237. Tigriopus californicus (Baker). Journal of Machado FS, Macieira RM, Zuluaga Gómez MA, Experimental and Ecology Costa AF, Mesquita EMC, Giarrizzo T. 2015. 42:99–111. Duffy D, Martin C. 2019. Cooperative natural National Park, southwestern Atlantic, with resource and invasive species management in additional ecological information. Biota Hawai’i. invasives: scaling up to meet Neotrop 15:epub. the challenge 497:497-502. Metaxas A, Scheibling RE. 1993. Community Friedlander AM, Parrish JD. 1998. Habitat structure and organization of tidepools. Marine Ecology Progress Series 98:187-198. Hawaiian reef. Journal of Experimental Metaxas A, Hunt H, Scheibling R. 1994. Spatial Marine Biology and Ecology 224:1–30. and temporal variability of macrobenthic Friedlander A, Poepoe K, Poepoe K, Helm communities in tidepools on a rocky shore in K, Bartram P, Maragos J, Abbott I. 2002. Nova Scotia, Canada. Marine Ecology 105:89– Application of Hawaiian traditions to 103. Moring JR. 1986. Seasonal presence of tidepool 813–815. Friedlander AM, Brown EK, Jokiel PL, Smith WR, northern California, USA. Hydrobiologia Rodgers KS. 2003. Effects of habitat, wave 134 :21–27. exposure, and status on tidepools of a boreal environment (Maine, . Coral Reefs 22:291–305. USA). Hydrobiologia 194:163–168. Friedlander AM, Donovan MK, Koike H, Poepoe KK, Bartram PK, Friedlander AM. Murakawa P, Goodell W. 2019. Characteristics 2007. The use of traditional knowledge in the contemporary management of a Aquatic Conservation: Marine and Freshwater Hawaiian community’s marine resources. Ecosystems 29:103–117. Green JM. 1971. High Tide Movements and and management. Coastal Management Homing Behaviour of the Tidepool Sourcebooks 4:119–144. Oligocottus maculosus. Journal of the Fisheries Ray GC, Hayden BP. 1992. Coastal Zone Ecotones. Research Board of Canada 28:383–389. Pages 403–420 in A. J. Hansen and F. di Castri, editors. Landscape Boundaries: Consequences Structure and Habitat Associations of Western for Biotic Diversity and Ecological Flows. Gobies (Teleostei: Gobiidae). Copeia Springer, New York, NY. Available from 1999:251–266. https://doi.org/10.1007/978-1-4612-2804-2_21 (accessed December 10, 2019). Shelton AO. 2010. Temperature and community in a warm-temperate South African . consequences of the loss of foundation Estuarine, Coastal and Shelf Science 76:723– species: Surfgrass (Phyllospadix spp., Hooker) 738. in tidepools. Journal of Experimental Marine Biology and Ecology 391:35–42.

115 UNIVERSITY OF HAWAI‘I AT HILO HOHONU 2020 VOL. 18

Simberloff D. 2010. Invasive species. Conservation Tables and Figures biology for all:131–152. Smith JE, Hunter CL, Smith CM. 2002. Distribution and Reproductive Characteristics of Nonindigenous and Invasive Marine Algae 56:299–315. Stamoulis KA et al. 2018. Seascape models reveal Figure 1. Our study sites were located at Lalakea and Lehia Beach park, which are both located Ecological Applications 28:910–925. along the Keaukaha coastline. The lalakea tide- Trussell GC, Ewanchuk PJ, Bertness MD, Silliman pools have a direct feed to the ocean and the tide- BR. 2004. Trophic cascades in rocky shore pools at Lehia are about 20-m from the ocean. tide pools: distinguishing lethal and nonlethal Image Google Earth. effects. Oecologia 139:427–432. Tsang Y-P, Tingley RW, Hsiao J, Infante DM. 2019. Identifying high value areas for conservation: Accounting for connections among terrestrial, freshwater, and in a tropical island system. Journal for Nature Conservation 50:125711. Underwood AJ, Chapman MG. 1996. Scales of spatial patterns of distribution of intertidal . Oecologia 107:212–224. Wiegner TN, Mokiao-Lee AU, Johnson EE. 2016. Identifying sources to thermal tide Figure 2. Survey stations at Lalakea pool 1 (a) and pools in Kapoho, Hawai’i, U.S.A, using a Lehia pool 1 (b). multi-stable isotope approach. 1. Mean species abundance and species rich- Bulletin 103:63–71. ness values for each location in Keaukaha. Five Williams ID, White DJ, Sparks RT, Lino KC, surveys for high tide and low tide at each location. Zamzow JP, Kelly ELA, Ramey HL. 2016. Species abundance was recorded in total counts. Responses of Herbivorous Fishes and Benthos Species richness was analysed using presence/ ab- to 6 Years of Protection at the Kahekili sence data. Area, Maui. PLoS ONE 11. Available from https://www. Abundance Species Richness ncbi.nlm.nih.gov/pmc/articles/PMC4963024/ Low High (accessed February 23, 2020). High Tide Tide Tide Low Tide Willis TJ, Roberts CD. 1996. Recolonisation and 168.8 14.8 193.8 (SD (SD (SD 15.8 (SD Lehia 26.9) 12.4) 1.1) 2.5) at Wellington, New Zealand. Environmental 483.4 17.6 Biology of Fishes 47 Lal- 423.6 (SD (SD (SD 17.8 (SD Zamzow JP. 2003. Ultraviolet-absorbing akea 171.8) 189.4) 2.2) 2.3) tidepool : variation over local and geographic scales. Marine Ecology Progress Series 263:169–175.

116 UNIVERSITY OF HAWAI‘I AT HILO HOHONU 2020 VOL. 18

- Appendix 1. List of species observed at each site at fer between tides (W = 14246, p-value = 0.8616). high tide and low tide, K=Lehia , L= Lalakea, fol- - lowed by tide (High and Low), followed by pool tween tides (W = 27136, p-value = 0.4). size (1=Large, 2=Small).

Figure 4. Mean species richness was not different at high tide and low tide for each site (Lehia, X2 = 3.3333, df = 4, p-value = 0.5037; Lalakea, X2 = 11.25, df = 9, p-value = 0.259).

117