Algal wrack deposits and macroinfaunal arthropods on sandy beaches of the Chilean coast

Four Chilean sandy beaches were sampled during the summer of 2000, to study the role of stranded algal wrack deposits on the population abundances of three detritus feeder species of the macroinfauna that inhabit the upper shore levels of that beaches: the talitrid amphipod Orchestoidea tuberculata Nicolet, the tylid isopod Tylos spinulosus Dana and the tenebrionid insect Phalerisida maculata Kulzer. The beaches were Apolillado (ca. 29° S), Quidico (ca. 38° S), Guabún and Mar Brava (ca. 42° S). Replicated samples were collected with a plastic cylinder (25 cm in diameter) from algal wrack deposits including the sediments beneath the wrack and nearby bare sand areas. Samples were collected at two beach levels, one closer to the sea with fresh deposits and other located on the upper beach with dry alga. Algal wrack deposits were mostly composed of the brown algae Macrocystis pyrifera (L.), Durvillaea antarctica (Chamisso) Hariot and Lessonia nigrescens Bory. O. tuberculata was found in the algal wrack deposits and bare sands collected from Apolillado, Quidico, Guabún and Mar Brava. On the other hand, T. spinulosus was just found at Apolillado, while P. maculata occurred in the sands beneath algal wrack deposits and bare sands collected from Apolillado, Quidico and Guabún. In general, the mean abundances of O. tuberculata, P. maculta and T. spinulosus were significantly higher in those samples collected from sands beneath algal wrack deposits; i.e., 56, 61 and 14 times higher (overall means) than in bare sands, respectively. It is concluded that stranded algal wrack deposits indeed promote an increase in population abundances of sandy beach detritus feeders, either because that deposits provide their main food source or shelter against variable environmental conditions (e.g., air temperature and humidity) during daylight hours. That might well explain the patchiness shown by these organisms, either across or along shore. This conclusion has important implications for sampling design of sandy beaches characterized by high inputs of algal wrack deposits.

Palabras clave: depósitos de algas varadas, artrópodos de playas arenosas, costa chilena.the macroinfauna were concentrated around the drift line of algal wrack deposits, but also species richness on an exposed sandy beach of the west coast of South Africa.Marsden (1991a) found a significant association between algal debris and distribution and density of the talitrid amphipod Talorchestia quoyana (Milne-Edwards) on a sandy beach of New Zealand.Moreover, this author found that the strength of that association was greater for freshly deposited algae than aged algal wrack deposits.Recently, Dugan et al. (2003) showed that in sandy beaches of California, subsidies of algal wrack deposits play a significant role in macroinfaunal community structure by increasing species richness and population abundance and thus, prey availability for shorebirds.
Three detritus feeder species inhabit the upper shore levels of Chilean sandy beaches: the talitrid amphipod Orchestoidea tuberculata Nicolet, the tylid isopod Tylos spinulosus Dana and the tenebrionid insect Phalerisida maculata Kulzer (Jaramillo 1987).While O. tuberculata and P. maculata have wider geographic ranges along the Chilean coast (about 22-42° and 18-42° S, respectively) that of T. spinulosus is restricted to a narrower coastal range extending between 23 and 30° S approximately (Jaramillo 1987, 1994, Schmalfuss & Vergara 2000).These detritus feeders are primarily active during the night when they leave the substrate and move over the beach surface (Jaramillo et al. 1980, 2000, 2003, Kennedy et al. 2000) to feed on algal wrack deposits stranded on the beach face during previous high tides; they bury again before or during dawn and remain beneath the sand surface or in sands located beneath algal wrack deposits during the day as do other talitrid amphipods, tylid isopods and tenebrionid insects elsewhere (e.g., Bowers 1964, Craig 1973, Kensley 1974, Stenton-Dozey & Griffiths 1983, Colombini et. al. 2000).
The population abundance and zonation of the macroinfaunal detritus feeders inhabiting the upper shore levels of exposed sandy beaches have been shown to respond to spatial and temporal variability in the supply of drift macrophytes.For example, Koop & Griffiths (1982) and Stenton-Dozey & Griffiths (1983) found that not only abundance and biomass of Subtidal and nearshore beds of large brown macroalgae (Macrocystis pyrifera (L.), Lessonia nigrescens Bory and Durvillaea antarctica (Chamisso) Hariot) develop all along the latitudinal range occupied by O. tuberculata, T. spinulosus and P. maculata (e.g., Guiler 1959a, 1959b, Santelices 1981, Moreno et al. 1984, Vásquez et al. 1998); thus, accumulations of stranded algae on the beach face are a ubiquitous feature of Chilean sandy beaches.Although several studies investigating the community level and behavioural aspects of O. tuberculata, T. spinulosus and P. maculata have been conducted (Avellanal et al. 2000, Jaramillo 2001, Jaramillo et al. 2000, 2003), no attempt to evaluate relationships between the population abundances of O. tuberculata, P. maculata and T. spinulosus and algal wrack deposits has been made.So far, the only study dealing with this question is that of Duarte (1974) who estimated the macroalgal consumption of O. tuberculata on a sandy beach of south central Chile.He concluded that the high consumption of adult amphipods (24.8 mg of dry algae per day) could account for the rapid decay of stranded algae on sandy beaches of southern Chile.
The present study aims to describe the effect of algal wrack deposits on the spatial distribution, composition and population abundance of upper shore detritus feeders at four sandy beaches located across a broad latitudinal range on the Chilean coast.Since some studies have shown that taxonomic composition of the detritus feeder guild of algal wrack deposits can be related to the age of the stranded algae (Griffiths & Stenton-Dozey 1981, Inglis 1989, Marsden 1991a, Colombini et al. 2000), we analyzed the population abundances of O. tuberculata, P. maculata and T. spinulosus associated with dry old algal wrack deposits located further up the drift line and with fresh macroalgae stranded lower down the beach face during the daytime at low tide.Since population abundance data of buried animals form the basis of the across shore zonation patterns described to date for the sandy beach macroinfauna (see McLachlan & Jaramillo 1995 for a review), we use the data of this study to consider the implications of distribution and amount of algal wrack deposits on our current understanding of zonation patterns on exposed sandy beaches.

Field samplings
Samples were collected during the spring low tides of late January (Guabún and Mar Brava) and early February 2000 (Apolillado and Quidico) from two bands of algal wrack deposits: an upper band, composed of dry algae desiccated by exposure to air and sun and stranded above the current high tide line or drift line, and a lower band with wet fresh algae deposited on the beach during the last high tide and located at the current drift line.While the upper bands of algal wrack deposits of Apolillado and Quidico were located on the retention zone, those from Guabún and Mar Brava were located on the dry zone of the beach (Table 1).Thus, the upper bands of algal wrack deposits in Apolillado and Quidico were located on wetter sands, than those studied at Guabún and Mar Brava.Lower bands of wracks from all beaches were all located on the retention zone (Table 1).The height above the low water line (LWL hereafter) of each band was measured using the Emery's profiling technique, i.e. eye-leveling of wave crest and the horizon from the low tide level (Emery 1961); also, the distance between bands.The sampling points were randomly chosen along the upper and lower band of algal wrack deposits.The along shore distance of the sampling area was 60-80 m.Twelve core samples were collected from each band during low tide with a plastic cylinder 25 cm in diameter (0.05 m 2 ) buried to a depth of 30 cm.At each band, six samples were collected from areas with algal wracks (thus, samples included these wracks and the sediments underneath); another six samples were collected from nearby bare sand areas (2 m apart from sampling points of algal wrack deposits).The samples collected were washed through a 1000 micron sieve; the macroinfauna obtained were stored in 5 % seawater-formalin until sorting, while algal wrack deposits collected were kept in closed plastic bags until determinations of fresh weight soon after.Later on, those samples were dried for two days (60 ºC).Water contents of algal wrack deposits were determined by calculations of weight differences.
Sediment samples (n = 3) were collected from the LWL of each beach to be used in estimations of Dean's parameter (see below).Wave height was estimated by measuring the height of breaking waves (n = 15) with graduated poles against the horizon, and adding the result to the height difference between the location of the observer and the lowest point where the backwash met the next incoming swash bore.The wave period (measured with a stop watch) was the time interval between breaking waves.

Analytical procedures
The grain size distributions of sands were analyzed using a settling tube (Emery 1938).Mean grain size was calculated according to the moments computational method (Seward-Thompson & Hails 1973).Mean grain size of sands from the LWL was used to estimate sand fall velocity according to the method described by Gibbs et al. (1971).From estimated mean wave height, wave period and sand fall velocity of sands, Dean's parameter (Ω) (Short & Wright 1983) was calculated to estimate morphodynamic type of each beach (Short & Wright 1983).
Fresh weight and water content of algal wrack deposits collected at the upper and lower band of stranded algae were compared with one way ANOVA on log and arcsin transformed data, respectively (Sokal & Rohlf 1995).The effects of algal wrack deposits and beach levels Values of Dean's parameter and distance from the drift line (present high tide) to the low water line (LWL) at each beach, height above the low water line of the upper and lower band of algal wrack deposits and distance between bands (RT, retention zone; DRY, dry zone) Valores del parámetro de Dean y distancia desde el nivel de marea alta a la línea de marea baja (LWL) en cada playa, altura sobre el nivel de marea baja de las bandas superior e inferior de los depósitos de algas y distancias entre las bandas (RT, zona de retención; DRY, zona de arena seca)  1995).The transformation of data allowed to fulfill the assumptions of both analyses.

Beach types and algal wrack deposits
The beaches of Apolillado, Quidico and Guabún were intermediate types (sensu Short & Wright 1983) with Dean's of 4.1, 4.7 and 2.2, respectively, while Mar Brava was a dissipative beach (Dean = 6.2) (Table 1).The width of the intertidal zone (i.e.distance from the drift line to the LWL) was lower at the beaches of Guabún and Quidico (22 and 34 m, respectively) and higher at Mar Brava (99 m) and Apolillado (53 m) (Table 1).
The upper band of algal wrack deposits was higher on the beach face at Apolillado and Guabún (288 and 338 cm above the LWL, respectively).The lowest difference between the height above the LWL of the upper and lower band of algal wrack deposits occurred at Quidico (5 cm), while the highest differences were found at Guabún and Mar Brava (72 and 66 cm, respectively) (Table 1).The lowest distances between both bands were measured at Apolillado and Guabún (7 and 10 m, respectively), while that measured at Quidico and Mar Brava were similar (16 and 18 m) (Table 1).
Algal wrack deposits collected at the upper and lower band of Apolillado and Mar Brava were composed of Macrocystis pyrifera, while that of Quidico were represented by Lessonia nigrescens.Dry rests of Durvillaea antarctica made up the algal wrack deposits collected at the upper band of Guabún, while rests of Macrocystis pyrifera dominated at the lower band of the same beach (Table 2).The mean weight of algal wrack deposits per unit area (0.05 m 2 ) was similar at the upper and lower band of stranded algae at Guabún; on the other hand, the mean weight of algal wrack deposits at Apolillado, Quidico and Mar Brava, was nearly twice at the lower band (Table 2).However, significant differences between weights of algal wrack deposits per unit area at both bands were only found for Apolillado (Table 2).Mean water contents of stranded algae of the lower band (mean ranges = 53.9-83.0%) were significantly higher than those calculated for the upper band of each beach (mean ranges = 18.9-22.9%) (Table 2).

The macroinfaunal arthropods
The amphipod Orchestoidea tuberculata was found in algal wrack deposits and bare sands collected from Apolillado, Quidico, Guabún and Mar Brava (Fig. 2).Mean population abundances of this amphipod were significantly higher in the lower band (sands underneath algal wrack deposits and bare sands included) as compared to the upper band at all beaches (1,320 versus 780, 16,305 versus 18 and 14,930 versus 560 ind m -2 for Apolillado, Guabún and Mar Brava, respectively) but Quidico, where no significant differences were found (Table 3, Fig. 2).The mean population abundances of O. tuberculata, were significantly higher in algal wrack deposits (upper and lower bands included) as compared to that in bare sands (2,010 versus 90; 2,508 versus 17; 15,221 versus 1,112 and 15,073 versus 417 ind m -2 for Apolillado, Quidico, Guabún and Mar Brava, respectively, Fig. 2).In Guabún and Mar Brava, the abundances of O. tuberculata was influenced by the interaction between beach level (i.e., upper and lower bands) and the presence of algal wrack deposits (Fig. 2 and Table 3).

The tenebrionid insect
Phalerisida maculata occurred in algal wrack deposits and bare sands studied in Apolillado, Quidico and Guabún (Fig. 3).Mean population abundances of this beetle did not differ significantly when beach levels (sands underneath algal wrack deposits and bare sands included) were compared at Apolillado (Table 4, Fig. 3).On the other hand, mean population abundances of P. maculata were significantly higher at the upper beach levels of Quidico (sands underneath algal wrack deposits and bare sands included) (Table 4, Fig. 3).The mean population abundances of this species, were significantly higher in algal wrack deposits (upper and lower bands included) as compared to that in bare sands (417 versus 10 and 13 versus 2 ind m -2 for Apolillado and Quidico, respectively) (Fig. 3, Table 4).In bare sands of Guabún, beetles were collected just from one replicate; thus, no statistical analyses were carried out for this beach.Nevertheless, it can be seen from Fig. 3 that the highest abundance of P. maculata at this beach occurred underneath algal wrack deposits of the upper beach sampled.

Abundance (ind m
-2 ) The tylid isopod Tylos spinulosus was found only at the algal wrack deposits and bare sands of Apolillado (Fig. 4).Mean population abundances of this isopod did not differ significantly when beach levels (sands underneath algal wrack deposits and bare sands included) were compared (Table 5, Fig. 3).The mean population abundances of T. spinulosus, were significantly higher in algal wrack deposits (upper and lower bands included) (1,283 versus 95 ind m -2 ) (Table 5, Fig. 5).

DISCUSSION
Population abundances of a variety of consumers can be potentially determined by food availability (e.g., Hunter & Price 1992, Polis & Hurd 1995, 1996, Menge et al. 1999), especially in habitats where food sources are of allocthonous origin, such as sandy beaches.The results of this study provide new evidence on this: in most of the samples collected from algal wrack deposits (either at the upper or lower band of stranded algae), the mean population abundances of O. tuberculata, P. maculata and T. spinulosus were significantly higher than those found in bare sand areas located nearby that deposits.These results are similar to that found by Dugan et al. (2000Dugan et al. ( , 2003) ) for sandy beaches of California, by Koop & Field (1980), Griffiths & Stenton-Dozey (1981), Griffiths et al. (1983) and Stenton-Dozey & Griffiths (1983) for the sand coast of South Africa, and by McLachlan (1985) for sandy beaches of Australia.

Abundance (ind m -2 )
As suggested by Koop & Field (1980), the supralittoral macroinfauna is rather affected by availability of food (such as algal wrack deposits) than by physical characteristics of the beach.This assertion has been recently demonstrated by Dugan et al. (2003) in sandy beaches of California by showing that algal wrack associated macroinfauna was significantly enhanced by cover of that detritus, while physical features of beaches such as sediment size, beach slope and beach type, did not explain significantly the variability of that macroinfauna.The last authors also found that beaches groomed regularly of algal wrack deposits harbored a less diverse and abundant macroinfauna than ungroomed beaches.Thus, the presence of algal wrack deposits, might well explain lack of relationships between population abundances of upper beach organisms such as talitrid amphipods and beach morphodynamics (e.g., Jaramillo & McLachlan 1993, Gómez & Defeo 1999, Contreras et al. 2003) or that between abundances of the whole macroinfauna and beach morphodynamics on sandy beaches where some of these detritus feeders (e.g., talitrids) make up most of the abundance figures (e.g., Jaramillo & McLachlan 1993, Jaramillo et al. 2001).Since upper beach arthropods usually live buried beyond the intertidal zone affected by the swashes of incoming waves, location and abundance of algal wrack deposits are probably more important in influencing population abundances of talitrid amphipods, tylid isopods and tenebrionid insects than variability in swash features associated to each morphodynamic beach type (see McArdle & McLachlan 1991, 1992).The talitrid amphipod O. tuberculata was the most abundant of the three species found at the algal wrack deposits of the beaches studied at the Chilean coast; this is similar to the findings of other authors which had found that talitrid amphipods usually dominate in sands located beneath the algal wrack deposits and in bare sands of the upper shore levels of exposed sandy beaches (MacIntyre 1963, Koop & Griffiths 1982, Griffiths et al. 1983, Stenton-Dozey & Griffiths 1983).Several authors have also noticed that talitrid amphipods are considered primary macroinfaunal colonizers of newly (fresh) algal wrack deposits stranded on sandy beaches (Griffiths & Stenton-Dozey 1981, Inglis 1989, Marsden 1991a, Colombini et al. 2000).Our results generally agree with the above, since in general, the highest abundances of O. tuberculata occurred at or underneath fresh stranded macroalgae suggesting food preference.On the other hand, the tenebrionid insect P. maculata showed the opposite pattern (more abundant on aged dry algal wrack deposits).Assuming that the higher occurrence of O. tuberculata and P. maculata at or underneath fresh and dry stranded macroalgae respectively, is actually the outcome of food preference (and not preference for example for microclimatic conditions), it can be hypothesized that opposite patterns are the result of differences in metabolic requirement of each species (Renaud et al. 1990, Colombini et al. 2000, Penning et al. 2000).
It is also possible that the lower abundances of O. tuberculata underneath the algal wrack deposits of the upper bands of Guabún and Mar Brava were related to the water content of sediments.While the upper and lower bands of algal wrack deposits of Apolillado and Quidico were located on the retention zone, the upper bands of Guabún and Mar Brava were located on the dry zone of the beach.Unpublished data (EJ) show that during summer, the ranges in water content of sands of the dry zone of Guabún and Mar Brava vary nearly 1-3 % while in the retention zone that variability is close to 5-18 %.Since mortality of O. tuberculata is close to 100 % in experimental sands with water contents close to 3 % (Jaramillo 1987), it can be hypothesized that the lower abundance of this amphipod in the upper bands of algal wrack deposits was due to the lower content of interstitial waters in sands located at the dry zone.Moreover, the height differences between bands were higher at Guabún and Mar Brava (72 and 66 cm, respectively) as compared to Apolillado and Quidico (21 and 5 cm, respectively), differences which probably also convey differences in water content of sands and further support the above hypothesis.Thus, the distribution and population abundances of O. tuberculata may be the result of several combined factors, such as water content of sands and the presence of algal wrack deposits.This is would explain the results of the two way ANOVA: i.e., the population abundances of O. tuberculata were affected by the interaction between beach levels and presence of algal wrack deposits at the beaches of Guabún and Mar Brava (see Table 3).
The talitrid amphipod O. tuberculata and the tenebrionid insect P. maculata were collected all along the latitudinal range studied; however, P. maculata was not found at Mar Brava.On the other hand, T. spinulosus was just found at the beach located further north (Apolillado).Earlier explanations of the along coast variability in taxonomic composition of the upper beach assemblage of the Chilean coast, included hypothesis related to latitudinal variability in environmental factors (e.g., air temperature and water content of sediments) and possible biological interactions (Jaramillo 1987).However, there is no present evidence to support hypotheses related to environmental variability along the Chilean coast, and it was not until recently that results of laboratory experiments demonstrated biological interactions among these species: while P. maculata affected negatively the survival of O. tuberculata, this amphipod affected negatively the survival of T. spinulosus (Jaramillo et al. 2003).Thus, these sorts of interactions were used to explain the partition in space and time of the locomotor activity of these detritus feeders over the beach surface (Jaramillo et al. 2003).Similarly, the negative interactions found between O. tuberculata and P. maculata may be used to explain the higher population abundances of this amphipod in Mar Brava where P. maculata was absent.Notwithstanding, this explanation must be taken with caution since the results of this study, come from snapshot samplings; in other words, we are not sure if during some times of the year the insect is present at that beach.
In conclusion, the results of this study suggest that the population abundances of upper shore arthropods of sandy beaches located along the Chilean coast, can be affected by algal wrack deposits.The distribution of algal wrack may also explain along shore patchiness and temporal variability in the zonation of that macroinfauna.As shown by Bowers (1964) and Craig (1970), the zone where the talitrid amphipod Orchestoidea corniculata Stout and the staphylinid beetle Thinopinus pictus Leconte burrow in the upper shore of Californian beaches, moves up and down following the lunar cycle, as do the stranded algal wrack deposits.Similarly, Colombini et al. (2000), found that the sandy beach macroinfauna of a Somalian beach (Indian Ocean) changed its zonation according the location of algal wrack deposits changed throughout time.Thus, future studies aimed to understand community organization of sandy beaches should consider that input and cover of stranded algal wrack deposits may be more important for some taxa than physical factors operating on these coastal habitats (i.e., beach morphodynamics) (Dugan et al. 2003), which do not necessarily affect directly the upper shore macroinfauna represented by semiterrestrial animals which usually live buried at or above the drift line.
Our results have profound implications for sampling design for sandy beaches; especially when studies of across shore zonation of the macroinfauna are the primary aims.The intertidal zonation of the sand beach macroinfauna is a well known pattern around the world (see review by McLachlan & Jaramillo 1995).Talitrid amphipods, tylid isopods, and tenebrionid insects are characteristic of the upper shore levels centred around the drift line, cirolanid isopods are usually the most typical organisms in the mid shore levels, while anomuran crabs (Emerita spp.), oedicerotid and phoxocephalid amphipods, and bivalves (Mesodesma and Donax spp.among others) dominate on the lower shore levels (e.g., Dahl 1952, Jaramillo 1994, McLachlan & Jaramillo 1995).These zonation patterns originate from population abundance data of buried organisms (see McLachlan & Jaramillo 1995 for a review), with causation primarily related to gradients of physical factors across the intertidal, including temperature, water content, compactness and grain size of sands (Salvat 1964, Rafaelli et al. 1991, Jaramillo & Fuentealba 1993, Jaramillo et al. 1993, McLachlan & Jaramillo 1995) but see e.g.Croker & Hatfield (1980) which showed that biological interactions related to the use of the intertidal space cause differential zonation of related species of haustoriid amphipods on the sand coast of eastern United States).Based upon the results of this study, allocation of sample units across and along shore, should considerer that algal wrack deposits affect the distribution, population abundances of detritus feeder arthropods, and consequently the whole community structure and the zonation patterns of the upper shore macroinfauna.Up to now, many sandy beach ecologists seem to have avoided sampling in areas of beaches that are covered by algal wrack deposits (but see Dugan et al. 2003); thus, published studies reporting species richness, population abundance figures and zonation patterns of the upper beach macroinfauna, may well be seriously flawed.

Fig. 2 :
Fig. 2: Mean abundances of Orchestoidea tuberculata collected from algal wrack deposits plus sands underneath (see Material and Methods) and nearby bare sand areas at the study sites.The values are means plus one standard deviation.Abundancias promedio de Orchestoidea tuberculata colectada en depósitos de algas varadas y arena subyacente (ver Materiales y Métodos) y en áreas cercanas en los sitios de estudio.Los valores son promedios más una desviación estándar.

Fig. 3 :Fig. 4 :
Fig. 3: Mean abundances of Phalerisida maculata collected from algal wrack deposits plus sands underneath (see Material and Methods) and nearby bare sand areas at the study sites.The values are means plus one standard deviation.Abundancias promedio de Phalerisidia maculata colectada en depósitos de algas varadas y arena subyacente (ver Materiales y Métodos) y en áreas cercanas en los sitios de estudio.Los valores son promedios más una desviación estándar.

TABLE 2
Dominant species of macroalgae, fresh weight and water content of algal wrack deposits at the upper and lower band of each beach.The values of weight and water content are means (n = 6) with standard deviations in parentheses F 1,10 = 4.18, P = 0.07 19.6 (0.9) 80.2 (3.6) F 1,10 = 1582.97,P < 0.01

TABLE 3
Results of the two way ANOVA carried out to test for differences in population abundances of Orchestoidea tuberculata at the Apolillado, Quidico, Guabún and Mar Brava beaches (see Material and Methods)Resultados de los ANDEVA de dos vías realizados para evaluar diferencias en la abundancia poblacional de Orchestoidea tuberculata en las playas de Apolillado, Quidico, Guabún y Mar Brava (ver Materiales y Métodos)

TABLE 4
Results of the two way ANOVA carried out to test for differences in population abundances of Phalerisida maculata at the Apolillado and Quidico beaches (see Material and Methods)Resultados de los ANDEVA de dos vías realizados para evaluar diferencias en la abundancia poblacional de Phalerisida maculata en las playas Apolillado y Quidico (ver Materiales y Métodos)

TABLE 5
Results of the two way ANOVA carried out to test for differences in population abundances of Tylos spinulosus at the Apolillado beach (see Material and Methods)Resultados de los ANDEVA de dos vías realizados para evaluar diferencias en la abundancia poblacional de Tylos spinulosus en la playa de Apolillado (ver Materiales y Métodos)