Variation of the Structure of the Intertidal Fish Community of the Pacific Coast of Baja California Sur, México

The intertidal zone is a transition zone between terrestrial and coastal systems. During low tide, rocky pools are exposed to air for several hours throughout the day, and environmental conditions such as temperature and salinity are highly variable. There are not many studies of fish communities inhabiting the intertidal zone in México. Therefore, the aim of the present study was to analyze the variation of the community structure of intertidal fish considering temperature, salinity and dissolved oxygen of the water that remains trapped in the tide pools during low tide. Visual censuses were carried out monthly during spring tides on full moon days, when fluctuations of environmental variables are significant. From January to December 2015 visual census were performed in the intertidal zone of El Faro, a locality belonging to the common land of Conquista Agraria in the City of La Paz, west coast of Baja California Sur, México. Extension of censuses was 156*5 m, 145 tide pools were divided in three groups (small, medium and large) according to size and depths that ranged from 20 to 45 cm. Temperature, salinity and dissolved oxygen fluctuate throughout the year of study. Thermal difference between open ocean and tide pools ranged from 3.5 to 4.5°C. A total of 3,754 organisms from 22 species, 12 families, four orders and one class were recorded. To analyze community structure, the following ecological indices were used: species richness (SRI); evenness (J’); Shannon-Wiener’s diversity (H’) and Fisher’s alpha (α-Fisher) with overall values of 2.80, 0.91, 3.20 and 5.15, respectively. According to BVI, 12 species were dominant. All species recorded were grouped as abundant (2), frequent (11), common (5) and rare (4), according to their relative abundance and frequency.


Background
The intertidal zone is an area of transition between low and high tide, and it is characterized by key ecosystems such as sandy and rocky beaches. During a typical tidal cycle, there are variations of physicochemical variables such as temperature and salinity, as a result from solar incidence and evaporation during low tide. Intertidal zones are also a source of important ecosystem services; therefore, anthropogenic impacts may increase vulnerability of these ecosystems to climate change (Yá ñez and Day, 2010). Geomorphology of coastal areas regulates their structure and ecological dynamic, and is a useful reference to interpret impacts of climate change (Yáñez-Arancibia, 2005); particularly considering critical habitats, defined as the range of environmental conditions in which species can live and develop (Yáñez-Arancibia et al., 2009). These habitats include complex rocky areas which intertidal fish use as shelter to survive and withstand environmental change. During low tide, physicochemical variables such as temperature, salinity and dissolved oxygen can be highly variable during the day, particularly in exposed habitats such as tide pools. There are not many studies of fish communities inhabiting the intertidal zone in Mé xico, that include critical environmental parameters. Therefore, the aim of the present study was to determine the structure of the fish community trapped in tide pools during low tide and its relationship with physicochemical variables of the water such as temperature, salinity and dissolved oxygen.

Study area
The site "El Faro" is located 24 km west of the common land known as Conquista Agraria, at N23 57.351 and  Figure 1). Figure 1 Geographic location of the sampling sites of fish species sampled in Conquista Agraria Baja California Sur, Pacific coast From January to December 2015, monthly samplings were performed in the intertidal zone during full moon days. Visual censuses of fish were performed using transects of 156*5 m. 145 tide pools were recorded and divided in three groups (small: 2 m 2 , medium: 4 m 2 and large: 7 m 2 ) according to size and depths that ranged from 20 to 45 cm. Tide pools are located 122 cm above sea level. To determine structure of the fish community the following ecological indices were used: species richness (SRI); evenness (J'); Shannon-Wiener's diversity (H'); Fisher's alpha (α-Fisher); dominance (Biological value Index-BVI) which is an indicator of the general dominance by species (Sander, 1960). It is recommended to use 95% of the relative abundance of individuals per sample to eliminate data lacking relevant information (Loya-Salinas and Escofet, 1990). Finally, considering their relative abundance and occurrence, species were classified as abundant, frequent, common and rare. Physicochemical variables (temperature, salinity and dissolved oxygen) were recorded using a YSI 2030 Pro multiparameter instrument. Ecological indices were analysed using the software Primer-E & Permanova 6 and statistical analysis was performed using Statistica v.7. Species identifications were carried out using specialized literature (Meek andHildebrand, 1923-1928;Zahuranec, 1967;Miller and Lea, 1976;Whitehead, 1985;Whitehead et al., 1986;Fischer et al., 1995;Allen and Robertson, 1998;Thomson et al., 2000). (Table 1). Comparing between tide pools of different size and depths, there were no significant differences of salinity (p=0.33). Overall dissolved oxygen of the water trapped in the tide pools was 10.21 mg/L (Table 1). There were no significant differences between months (p=0.48). Comparing between tide pools of different size and depths, there were no significant differences of dissolved oxygen (p=0.26).

Community structure
A total of 3,754 organisms from 22 species, 12 families, four orders and one class were recorded and distributed in the tide pools as follows: 85 small tide pools (1,914 organisms), 48 medium tide pools (1,314 organisms) and 12 large tide pools (526 organisms).
Margalef's species richness (SRI) showed significant differences (p<0.05) (Margalef, 1969). April and July recorded the highest values (3.56 and 3.53, respectively), and the lowest values were recorded during December and February (1.66 and 2.20, respectively) ( Table 1). Considering the size of the tide pools, there were no significant differences (p=0.216).
There were also significant differences (p<0.05) of evenness recorded each month. September showed the highest value (0.95), and January showed the lowest value (0.81) ( Table 1). Considering the size of the tide pools, there were no significant differences (p>0.05).
Shannon-Wiener's diversity (H') mean value was 3.20 bits/ind. There were significant differences (p<0.05) between months, with April showing the highest H' value (3.66 bits/ind.), and December showing the lowest value. There were no significant differences when comparing between different size tide pools. Fisher's alpha was also used as a diversity index because it includes rare species and has higher biological significance. Its mean value was S=5.15. April registered the highest value (S=7.16), and December had the lowest value (S=2.89). There were significant differences between months and size of tide pools (p<0.05) ( Table 1).
According to relative abundance and frequency, 22 species recorded in this study were classified in four groups ( Figure 3): (1) Abundant species: A. declivifrons and A. troschelii with relative abundance of 26%.

Discussion
The intertidal zone is highly variable through time and space (Bridges, 1974;Little and Kitching, 1996) due to seasonal and daily variations of tide and solar radiation (Metaxas and Scheibling, 1993;1994). Our study area is characterized by a mixed tidal cycle, which is why physicochemical variables were measured during the afternoon low tide (between 16:00 and 17:00 hr.). Exposed time of pools during low tide was 6 h 20 min. According to our results, temperature variations during the year ranged from 24.2°C to 32.65°C (Table 1), similar to previous studies on Chile and El Salvador (Herná ndez et al., 2002;Gonzá lez-Murcia et al., 2016). Thermal difference between open ocean and tide pools was 4.6°C, maximum temperature (32.65°C) was recorded in September. According to Gonzá lez-Murcia et al. (2016), as the water gets warmer, fish increase their metabolic activity, and considering an increase of frequency and intensity of warming events due to climate change, fish and other ectothermic organisms may be challenge, particularly because their survival, growth and reproduction rely on critical temperature ranges. Organisms inhabiting the rocky intertidal zone are assumed to live very close to their thermal tolerance limits, therefore are considered as potential harbingers of the effects of climate change on species distributions in nature (Helmuth et al., 2002).
Tide pools in our study area are located 122 cm above sea level (lowest tide). According to , isolation of tide pools, as well as water evaporation, are important factors that determine composition and distribution of intertidal fish. These authors selected tide pools according to their position above the lowest tide (low, medium, high), and used larger and deeper tide pools, obtaining different temperature, salinity and dissolved oxygen values compared to our own. These results might be due to geographical differences, since our tide pools are in a subtropical region, where fish have tropical and temperate affinities, while their tide pools are in a temperate region.
Salinity showed significant differences throughout the year because of water evaporation of tide pools, similar to previous studies (Herná ndez et al., 2002;Gonzá lez-Murcia et al., 2016). Oxygen levels increased during the day (due to photosynthesis) and decrease at night (Velazco-Gil, 2006). Dissolved oxygen showed significant differences between months, ranging from 5.5 to 25 mg/L. These variations might result from difference on oxygen demand because of the presence of different organisms (molluscs (gastropods, polyplacophora, cephalopods), crustaceans, annelids, sipunculids, anemones, sea urchins and limpets; as well as brown, red and green algae) throughout the year (Velazco- Gil, 2006). Temperature, salinity and dissolved oxygen did not show significant differences when comparing size of tide pools, which suggests that there is no preference on size because environmental conditions are similar in all tide pools, and therefore they are all used as shelters during low tide.
Structure of the intertidal fish community of tropical zones in Mexico and other regions has not been well studied. According to Gonzá lez-Murcia et al., (2016), studies on the intertidal zone of tropical regions are limited. Therefore, the present study is an important contribution to the understanding of temporal changes of the structure of the intertidal fish community of the West coast of the Baja California peninsula (Ruiz-Campos and Mammann, 1987;Ruiz-Campos et al., 2010).
Species richness describes changes in the number of species according to sample size. Environmental variables and time fluctuations have a significant impact on the number of species. Overall species richness recorded in the present study was 2.80. Gonzá lez-Murcia et al. (2016) found no significant difference on species richness between tide pools, but did found changes between different water volumes. According to , structure of fish community of tide pools is regulated by biological factors such as shelter availability, feeding, and others. Considering these factors within our study area, species richness can be modified through time (Table 1). According to Pielou (1976), evenness is an important component of diversity that estimates a maximum value of diversity when all species are equally abundant. This ecological index describes the proportional distribution of abundance between species; its value ranges from 0 to 1, with 1 meaning that all species are equally abundant (Magurran, 1988). According to our results, there is a good level of evenness between the recorded species and their relative abundance reaching an overall value of 0.91. Previous studies of the fish community of tide pools do not include this ecological index in their analysis, therefore the present study is the first one using this index. Shannon-Winner's diversity index has not been widely used to describe structure of the fish community of tide pools. According to our results, an overall diversity value of 3.20 bits/ind., is considered a good indicator of diversity, although it is important to mention that this index may present some inconvenient at the time of analysis, such as departure from normality, related to sample size (Krebs, 2006). On contrast, Fisher's alpha diversity index (Fisher et al., 1943) evaluates diversity more efficiently considering number of organisms and number of species (Condit et al., 1996), therefore, from an ecological perspective, all species have the same weight (Moreno, 2001;Magurran, 2004). This index, alongside evenness and Shannon-Winner's diversity have not been used in this type of studies with intertidal fish. Grouping fish species according to relative abundance and frequency, might shed some light into their ecological role in the ecosystem, from the 22-species recorded in the present study, only two