Paiva Beach (8°16’S, 34°56’W) is an open ocean tropical sandy beach approximately 7.8 km long located in the Cabo de Santo Agostinho municipality (Pernambuco State, northeast Brazil) (Holanda 2020). The sandstone reefs of Paiva Beach are elongated and occur parallel to the shore (Laborel 1970). The reef line closest to the beach is 2.5 km long and 1 km wide and densely colonized by macroalgal assemblages that are typically found in the tropical phycogeographic region (Horta et al. 2001). The climate is tropical hot/humid, with average annual temperatures above 25°C throughout the year (Domingues et al. 2017). The tidal regime can be classified as a mesotidal semi-diurnal type, with tide height averaging 0.7 m (neap tide) to 2.5 m (spring tide) (Domingues et al. 2017). The reefs and their diverse intertidal communities were severely impacted by the 2019 Brazilian Oil Spill. In only one week (19-28 October 2019), more than 1000 t of crude oil were collected from the Cabo de Santo Agostinho municipality alone (Craveiro et al. 2021). Although less perceptible, oil strains started appearing in Pernambuco’s beaches in August 2019 (Soares et al. 2020), months before the main sheet of crude oil arrived at Paiva Beach.

Sampling was performed monthly from July 2019 to July 2020. In March 2020, samples were not taken because of COVID-19 lockdown restrictions. On each sampling occasion, ten fronds of the red seaweed Jania capillacea were collected in intertidal tide pools. The fronds were wrapped with plastic bags to prevent motile animals from escaping and then removed from the substrate. Samples were fixed with 4% saline formalin. In the laboratory, the samples were sieved (0.3 mm mesh size) and the retained epifaunal organisms were sorted under the stereomicroscope. Specimens of C. dubia were separated, counted and preserved in 70% ethanol. For each month, approximately 120 to 170 individuals were randomly chosen (in some months fewer than 120 individuals were found) to be measured from the tip of the carapace to the end of the telson with a stereo microscope with an eyepiece reticle (body length, mm). For males, the length and width of chelae were measured. The individuals were classified into five categories, as proposed by Masunari (1983): juveniles/mancas (larval MII and MIII stages); pre-ovigerous (preparatory) females (those with oostegites); ovigerous females (those with marsupia); non-reproductive females (those with adult appearance but no sign of reproductive activity); and males (those with characteristic male features) (Fig. 1).

Monthly densities and lengths of C. dubia were log-transformed and compared by one-way ANOVA followed by pairwise Tukey tests. Chi-square tests (χ²) were performed to compare sex ratio between months and size classes. The parameters of the von Bertalanffy growth function (VBGF) were estimated through the ELEFAN_GA_Boot method, which is based on a bootstrapped genetic algorithm that provides a range of likely best fits for the VBGF parameters (Schwamborn et al. 2019). This approach is highly applicable for rapid assessment in data-poor situations such as data collected from short time periods (e.g. one year) (Mildenberger et al. 2017). Intervals of L _∞, k and t ₀ (pooled data) were chosen based on published studies about similar species from the southwestern Atlantic Ocean (Almeida 1994, Fonseca and D’Incao 2003, Leite et al. 2003, Pennafirme and Soares-Gomes 2017, Rumbold et al. 2015). Analyses were made in the TropFishR (Mildenberger et al. 2017) and fishboot (Schwamborn et al. 2019) packages.

The growth performance index (Φ’) was calculated from von Bertalanffy’s equation parameters as Pauly and Munro (1984):

The length at which 50% of females are sexually mature (L ₅₀) was estimated through a logistic function using two sets of variables: length-frequency (LFQ) data and categorical “gonadal maturation stages” (Torrejon-Magallanes 2020). Since the mean length of the individuals at MIII stage (1.75 mm) is lower than the smallest reproductive female (2.1 mm), all specimens in this category were treated as “immature”. L ₅₀ was calculated as

where Pi is the proportion of reproductive females for each (0.2 mm) size class; A is the curve asymptote; r is a rate parameter related to the speed of size change from non-reproductive to reproductive status; L _t is the total length (mm); and L ₅₀ is the size at first maturity (mm) (Fontoura et al. 2009).

Morphometric maturity in males was estimated using a regression analysis where X (body length) is considered the explanatory variable and the classification of maturity (juveniles, 0; adults, 1) is considered the response variable (binomial) through the form (Torrejon-Magallanes 2020):

where Pcs is the probability of an individual being mature at a determinate X length. β₀ (intercept) and β₁ (slope) are estimated parameters. Both gonadal (females) and morphometric (males) maturities were calculated using the “sizeMat” package (Torrejon-Magallanes 2020).

The instantaneous mortality rate (Z) was calculated through a linear capture curve based on length composition data (Sparre and Venema 1998). Estimates for maximum individual age within the population were based on Taylor (1958). The maximum longevity (t_max) was calculated from 95% of the asymptotic length, where

All data analyses were performed using R software (R Core Team 2022). For all tests a significance level of 5% was considered.

The maximum abundance of C. dubia in Paiva Beach (tropical Brazilian coast) occurred in drier months, particularly in August, November and December 2019. Despite the stable conditions of tropical environments, even slight variations in temperature, photoperiod, salinity and nutrient concentration might cause perceptible oscillations in population size and structure (Pandian 2016). Estuarine tanaid species have generally shown abundance peaks in rainy periods and under mesohaline conditions (Modlin and Harris 1989, Pennafirme and Soares-Gomes 2009). Marine species, in contrast, are generally more abundant during the dry season, when salinity is higher (de Souza et al. 2013, Tuya et al. 2001). Warmer temperatures in the dry season may also enhance microbial growth (Longo et al. 2019), which might contribute to the detritus chain, which is useful for both feeding and tube-building in Tanaidacea (Krasnow and Taghon 1997). In the rainy season, wave action, river discharge and sediment resuspension are considerably higher (Medeiros et al. 2001), which may cause higher mortality owing to physical and chemical stressors and reduction of available algal biomass.

There was a significant reduction in the size of individuals in August and October/November 2019. Despite the arrival of oil in these months, abundances continued to grow, so the lower body lengths might be a consequence of the recruitment of new juveniles into the population instead of a mass mortality event. The months of August (end of the rainy season), October and November (peak of the dry season) coincide with the highest salinities. However, other factors rather than just salinity might help to explain this pattern, such as the decrease in predators and/or competitors, or better conditions for the seaweed that provides refuges. However, a significant reduction in abundance was observed in September, the beginning of the dry season. This unusual decrease might have been caused by the contact with the oil that killed mainly juveniles, who are more vulnerable to harmful effects of pollutants than adults (Keesing et al. 2018). However, the population continued to grow and quickly recovered in the following months, which can be seen as the second peak in November/December. In laboratory experiments, C. dubia has shown to be relatively tolerant to changes in water chemistry (Araújo-Silva et al. 2022). Much of its resistance is attributed to the construction of tubes, which keep these animals protected from external agents and may help them colonize a wide variety of habitats within medium and infralittoral zones (Araújo-Silva et al. 2022).

Male size of morphometric maturity was 3.0 mm, and males larger than that were very rare. When males moult into their mature phase, they lose their functional mouthparts and cannot feed anymore, so they must die soon after reproduction (Highsmith 1983, Gardiner 1975), which explains the almost complete absence of males larger than 3 mm. Also, a positive correlation was observed between body length and chela length in males. The differential growth rate of chelae has an important role in crustacean reproduction (Ewers-Saucedo 2019). In Tanaidomorpha, chelipeds are used to access the female tube, hold females during mating and fight with other males (Buckle Ramirez 1965, Highsmith 1983, Johnson and Attramadal 1982). The copulatory male is considered the only free walking instar, so developed chelipeds are essential to climb and hold onto algae in turbulent environments (Johnson and Attramadal 1982). These features might give a reproductive advantage to larger males (Highsmith 1983). Another possible function of male aggression in leptocheliids is the exclusion of other males of sympatric species that might cohabit the algal frond (Stoner 1986).

A large size overlap between the sexes, mainly among more advanced developmental stages, was recorded. This is mainly because of the progressive decrease in growth rate with age (Sparre and Venema 1998). Pre-ovigerous females have a very small growth from the preparatory to marsupial stages because they allocate more resources to reproduction than to somatic growth (Masunari 1983). The largest individuals in our samples were female. This is probably because of the low survival rate and premature deaths of males (Mendoza 1982). More specifically, the largest individuals found were in fact non-reproductive adult females. It is very likely that they were multiparous females with interspaced “pregnancies” (Masunari 1983). The wide overlap between non-reproductive individuals with adult appearance (neuters) and reproductive ones suggests the existence of at least one intermediary instar between successive marsupial stages in C. dubia (Fig. 3). However, we did not find external marks (as in Johnson and Attramadal 1982) or any direct sign of previous “pregnancies” in large non-reproductive females.

The scarcity of reproductive individuals was notable. A few observations might explain this annual pattern, such as high natural mortality (5.77 year^-1) and possible competition among individuals to mate; the aggressiveness of males and possible formation of “harems” or highly condensed mating aggregations (Stoner 1986); interspaced “pregnancies” with intermediary instars (multiparity); and pre-ovigerous females with microscopic oostegites in early development, only visible through histological analysis (Masunari 1983). The last two factors might have produced a bias in the counting of reproductive/non-reproductive individuals.

The maintenance of a skewed sex ratio towards females during the entire year has also been observed in many previous studies on tanaids (e.g. Leite et al. 2003, Masunari 1983, Modlin and Harris 1989). Biased sex ratios have long been observed in several crustacean species (Ewers-Saucedo 2019). Some evolutionary and ecological mechanisms might explain the highly female-biased sex ratio in C. dubia (Ewers-Saucedo 2019):

The last strategy has been widely proposed to occur in many species of Tanaidacea (Gardiner 1975). It has been theorized that protogyny is an effective strategy for maximizing fitness (reproductive success) in all life stages (Highsmith 1983). Basically, males take too long to be able to reproduce (they must grow enough to fight with other males), whereas females mate as soon as they reach maturity. There is therefore a selective pressure for being a female at small (younger) stages and then becoming a male at larger (older) stages, because only the large males will be successful (Highsmith 1983). Alternatively, protogyny may also have evolved as a mechanism for counterbalancing the high mortality of males (Leite et al. 2003).

Although it is observed in many tanaids, there is no clear evidence that C. dubia is a sequential hermaphroditic species. If that were true, it would be expected to observe a dominance of one sex in smaller size classes and a gradual shift towards the other sex in larger size classes (Ewers-Saucedo 2019). The opposite occurred in Paiva Beach, where the sex ratio was strongly female-biased in all size classes. However, this might still be true if only a small percentage of females is able to become male. It has been observed in similar species that dominant males can inhibit sex reversal of nearby females (Highsmith 1983), which might keep the number of males in the aggregation low. The presence of two types of males based on their size differences has been treated as more evidence of protogynous hermaphroditism in Tanaidacea. The smallest males were considered gonochoristic (male-born) whereas the largest ones were protogynous (female-born) (Gardiner 1975). Interestingly, individuals identified as “Leptochelia dubia” collected from different parts of the West Coast of the USA have been treated either as dioecious (Mendoza 1982) or hermaphroditic (Highsmith 1983) on different occasions. Considering the lack of detailed natural observations, the taxonomic uncertainties and the geographic distances (tanaids are known to have low dispersive capacity; Bamber 2010), the best way to solve questions regarding the reproductive behaviour of C. dubia in the tropics would be to cultivate these organisms under controlled and realistic conditions.

The year-long presence of reproductive adults and juveniles indicates continuous reproductive activity, a common pattern observed in many crustacean species, especially in the tropics (Pandian 2016, Pennafirme and Soares-Gomes 2009). Specimens in Paiva Beach are much smaller (2.04±0.95 mm in length) than other tanaid species from the southwestern Atlantic (Table 2), and their natural mortality is relatively high. C. dubia also has multiple short cohorts, a relatively high growth rate (k) and a short lifespan (<11 months). All of these features indicate that it has a short and opportunistic life strategy (Pennafirme and Soares-Gomes 2009).

Latitudinal trends in life history parameters are commonly observed in Tanaidacea (Pennafirme and Soares-Gomes 2017). Warmer temperatures generally accelerate growth and maturity in ectotherms (Sudo 2003), and tropical marine invertebrates generally have shorter lifespans and smaller individual sizes than their temperate counterparts (Cardoso and Defeo 2004). Furthermore, unlike Monokalliapseudes schubartii and Tanais dulongii (Table 2), C. dubia is a seaweed-dweller, meaning that its size is highly dependent upon the interstitial size and volume of the host algae (Hacker and Steneck 1990). However, in a subtropical area in southern Brazil, Fonseca and D’Incao (2003) found a growth rate (k) of 4.54 year^-1 for M. schubartii. It is possible that other factors rather than temperature, such as food quality/availability (Berrigan and Charnov 1994), were responsible for the higher growth rate of this species than of our tropical species. The population studied by Fonseca and D’Incao (2003) inhabits a nutrient-rich protected estuarine bay (Patos Lagoon), which is very different from the open, nutrient-variable waters of tropical Brazilian beaches.

The absence of other studies that address the population parameters of C. dubia makes it difficult to evaluate the influence of environmental variables and regional variability in the growth of this species. However, compared with other Tanaidacea species of the southwestern Atlantic, C. dubia also showed the lowest growth performance index (Φ’). Zamora-Sarabia et al. (2022) found a negative correlation between Φ’ and sea surface temperature for a commercially important fish species in the southern Gulf of California. This correlation might explain the low Φ’ value compared with other tanaid populations from colder areas at higher latitudes (Table 2). Combined with the high mortality rate and low percentage of reproductive individuals, C. dubia apparently faces harsh environmental conditions at Paiva Beach, such as resource limitation, high temperatures, predation and strong hydrodynamics (Domingues et al. 2017). Moreover, the size of the largest individual found (4.6 mm in length) was about 87% of the theoretical maximum length (L _inf =5.26 mm). It is possible that these natural stressors are a strong limitation for the growth and survival of these organisms. This may make the studied population of C. dubia vulnerable to extreme climatic events caused by anthropogenic warming and/or El Niño oscillations (Zamora-Sarabia et al. 2022), although Araújo-Silva et al. (2022) showed that this species is still more tolerant to warming effects than other tropical peracarids. Growth parameters such as Φ’ are often used as an important reference for stock management and choosing appropriate species for aquaculture (Mathews and Samuel 1990, Pauly and Munro 1984, Zamora-Sarabia et al. 2022). Because peracarids have been recently studied as a valuable food resource for aquaculture (Abundez et al. 2021, Guevara et al. 2005), the present study should be taken into consideration for future decisions regarding the commercial use of this species.

In conclusion, C. dubia at Paiva Beach (tropical Brazilian coast) has a short-living, opportunistic life history that may be favoured by high-salinity conditions. The sex ratios are heavily female-skewed throughout all months and size classes. The species was apparently little impacted by the 2019 Oil Spill at the populational level, a sign of high tolerance to environmental impacts and great plasticity of lifestyle, which are interesting features of a possible bioindicator species. This study reinforces the need for cultivation techniques in order to solve key questions regarding the reproductive behaviour of the species and the high applicability of robust, bootstrapped length-based methods in population studies of small marine invertebrates.