Sandy beaches: state of the art of nematode ecology

In this review, we summarize existing knowledge of the ecology of sandy-beach nematodes, in relation to spatial distribution, food webs, pollution and climate change. We attempt to discuss spatial scale patterns (macro-, mesoand microscale) according to their degree of importance in structuring sandy-beach nematode assemblages. This review will provide a substantial background on current knowledge of sandybeach nematodes, and can be used as a starting point to delineate further investigations in this field. Over decades, sandy beaches have been the scene of studies focusing on community and population ecology, both related to morphodynamic models. The combination of physical factors (e.g. grain size, tidal exposure) and biological interactions (e.g. trophic relationships) is responsible for the spatial distribution of nematodes. In other words, the physical factors are more important in structuring nematodes communities over large scale of distribution while biological interactions are largely important in finer-scale distributions. It has been accepted that biological interactions are assumed to be of minor importance because physical factors overshadow the biological interactions in sandy beach sediments; however, the most recent results from in-situ and ex-situ experimental investigations on behavior and biological factors on a microscale have shown promise for understanding the mechanisms underlying larger-scale patterns and processes. Besides nematodes are very promising organisms used to understand the effects of pollution and climate changes although these subjects are less studied in sandy beaches than distribution patterns.


INTRODUCTION
Sandy beaches are dynamic ecosystems driven by prominent physical processes that shape the habitat for different functional and taxonomic groups. The term "sandy beach" can be used to describe a wide range of environments, from high-energy open-ocean beaches to sheltered estuarine sand flats (McLachlan 1983). Sandy beaches are, in general, dynamic environments occurring worldwide along ice-free coastlines, and located at the transition between the land and a waterbody such as oceans, seas or lakes. The beach sediment may be supplied by rivers or by the erosion of highlands adjacent to the coast, and the sea may also contribute to the sediment supply through input of biogenic brought to you by CORE View metadata, citation and similar papers at core.ac.uk structures (animal skeletons, coral and shell fragments) (adapted from McLachlan and Brown 2006). Here, the term sandy beach was used in a narrower context, applying only for high-energy open-ocean or pocket beaches ranging from the reflective to the dissipative extremes.
The sediment is mainly composed of quartz and/or carbonate sands of terrestrial and marine origin, respectively. Small amounts of feldspar, basalt and heavy minerals can also contribute to the sediment composition. Important features to characterize sediments are the grain size, the sorting coefficient and the angularity; all of them influence the porosity of sediments. The combination of permeability and penetrability determines the volume of water percolation, its drainage, and the oxygen penetration in sandy-beach sediments (McLachlan and Brown 2006).
The sandy beach ecosystem can be divided horizontally into different zones consisting of the foredune, backshore, swash/shoreline, surf zone and nearshore (Defeo and McLachlan 2005); vertically, it is represented by the pelagic and benthic systems. Sandy-beach endobenthic communities were largely neglected by ecologists until Remane (1933) initiated the first survey in Germany, with a focus on the benthic community. Today, sandy beaches are still less studied than most other coastal systems (Defeo and McLachlan 2011), and most research on intertidal sandy beaches has been concentrated on macrofauna and on birds (see Cornelius et al. 2001, Defeo andMcLachlan 2005, for a review). The less-prominent sandybeach meiofauna has received considerably less attention. For several years, meiofauna research on sandy beaches was dedicated to general surveys at higher taxonomic levels, and the first investigations focusing on the composition of sandy-beach nematode communities began only in the 1970s (Heip et al. 1985). More recent investigations have dealt with patterns of macro-(10 3 m), meso-(m) and microscale (10 -2 m) distributions and are more often framed in a context of coastal zone management. The macroscale pattern consists of differences between beach types and latitudes, the mesoscale pattern refers to distribution and community structure alongshore and across shore transects and finally the microscale pattern is related to distance between millimeters and few meters (McLachlan and Brown 2006). These three patterns of distribution are used in the context of this review and we chose to focus on nematodes because they are the most dominant group of the meiofauna in soft sediments and are supposed to act as a link for the flux of energy since they can be the food source for macrofaunal organisms and fishes (Esteves and Genevois 2006).
Although nematodes act as important nutritional resources for macroscopic organisms, the functional role of these organisms in sandy beach ecosystem is not well established since this environment is supposed to support three partially unconnected food-webs: a discrete food web consisting of interstitial organisms, a microbial loop and a macroscopic food web (McLachlan and Brown 2006). The relative importance of these three food webs is conceptually different among beach types; for instance, on cold-temperate sandy beaches, a diverse interstitial food web dominates on high wave energy while nematodes and the macroscopic food web are more important in dissipative shores (Menn 2002).
Beaches are squeezed between rising sea level on the marine side and expanding human population and development on the landward side (Schlacher et al. 2008). Therefore, this ecosystem faces numerous threats coming from both directions. Pollution, mining, disruption of sand transport and tourism development are threats mainly originated in the terrestrial side (Brown and McLachlan 2002) while the vulnerability to the impacts of the climate change is more related to the processes occurring in the sea side, such as sea level increase, shore erosion, acidification and increase of sea water temperature (McGlone and Vuille 2012).

NEMATODE ECOLOGY ON SANDY BEACHES 3
The aim of this review is to summarize what is known for sandy-beach nematodes in terms of spatial distribution patterns, temporal variability, food webs, pollution, and climate change. We believe that this review will open a new niche for future researchers to fill gaps in the understanding of nematode ecology.

MATERIALS AND METHODS
A bibliographic survey was done using Web of Science®, SCOPUS and Google Scholar considering works published until early 2015. Solely papers published in scientific journals and those with an ecological purpose related to meiofauna and nematode community from the intertidal region of sandy beaches were selected. Taxonomical articles and those ecological related to subtidal sampling design were largely excluded from the analysis. The selected studies were classified according to 1) type of benthic association (meiofauna or nematode), 2) pattern of distribution (macro-, meso-and microscale), 3) any other ecological approach including temporal variation, pollution, coastal management, colonization, natural impact, recreational activity, climate change and food web, 4) number of sampled beaches (Table I).

RESULTS AND DISCUSSION
In order to provide an overview of world investigations for meiofauna and nematode sandy-beach ecology, we have summarized this information in a world map showing where the studies were conducted (Fig.1) followed by the adopted approach (Table I). Of the 87 meiofauna studies, approximately 50% deal with nematodes (Table I). Our discussion is based on the main topics included in the built table and it was subdivided in the following sections.
The pattern of nematode distribution in sandy beaches is explained in detail in the following subsections. The variables are discussed below in relation to sandy-beach nematode communities.

Macroscale distribution of nematode communities
This topic has received relatively little attention. Two studies concerning the macroscale latitudinal distribution of sandy-beach nematodes and their diversity suggest that the present pattern follows the trend of increasing diversity toward the tropics Trueman 2005, Lee andRiveros 2012). Macroscale nematode studies exploring the full range of beach types have not yet been undertaken, but the limited information available shows that nematode diversity is highest in coarsegrained, intermediate sandy beaches  or in sheltered conditions (Hourston et al. 2005, Urban-Malinga et al. 2005. It indicates that the a high diversity can be found in the full range of beach types, but the above mentioned references do not use the same methodological strategy what limited the interpretation of the results. A comprehensive understanding of the nematode diversity pattern among different beach types will be only possible when an equal sampling design would be adopted. For instance, the sampling design used to understand the distribution patterns of polychaetes in different beach types could be adopted (Di Domenico et al. 2008); in this study, it is indicated that reflective sandy beaches are more diverse than intermediate and dissipative ones. Turbellarian is the dominant taxa in the sheltered beach; high meiofauna density in the exposed beach; no difference between meiofauna diversity of sheltered and exposed beaches   Table I for authors and a detailed information of the studies.

Mesoscale distribution of nematode communities
Concerning the horizontal zonation in across-shore transect on a single beach, granulometry, organic input, temperature and salinity are the key factors governing the horizontal nematode distribution (Platt 1977, Urban-Maligna et al. 2005, Moreno et al. 2006, Maria et al. 2012, whereas biological interactions between macro-and meiofauna and/or among meiofauna organisms have only an indirect effect (Maria et al. 2011b(Maria et al. , 2012(Maria et al. , 2013a. High diversity is often found where there is an optimal balance of these variables, i.e., physical and chemical conditions are intermediate , Hourston et al. 2005, Gingold et al. 2010). However, some, more heterogeneous beaches display intertidal sandbars intercalated by depressions, which retain water when tides recedes, across their wide intertidal zone (Masselink et al. 2006); this habitat is therein called runnel and those beaches that bear it are so-called macrotidal ridge-and-runnel beach. On these beaches, for example De Panne Beach in Belgium, studied by Maria et al. (2013b), the two microhabitats (runnels and sandbars) contain different nematode communities, which are reflected in dissimilar across-shore zonation. Three nematode associations (upper, middle and lower beach) are much more evident in the sandbars than in the runnels. These three biological zones on the sandbars were in accordance with the tidal zonation previously observed by Gheskiere et al. (2004) in the same study are contradicting the previous knowledge that the horizontal nematode zonation may just persist for short intervals of time and under calm conditions (Nicholas and Hodda 1999). The main difference between the two most recent studies is that the latter excluded the runnel microhabitat. The inclusion of the runnels by Maria et al. (2013b) showed that the shift over three horizontal nematode communities found in the sandbar was interrupted by the runnel communities. These results contrast somewhat with the results of Gingold et al. (2010) for a macrotidal ridge-and-runnel beach in the Gulf of California, where both sandbars and runnels showed the same pattern of across-shore zonation. We can conclude that next to the horizontal mesoscale, the presence of runnels may influence the nematode zonation.
Although both habitats (sandbars and runnels) are horizontally distributed over the sandy beach interface and did not show consistent differences in grain size or chlorophyll a content, they were not under the same horizontal gradient of air exposure during low tide. Based on these results, it is not so much the combination of a physical characteristic that determines the nematode community structure, but rather the degree of variation of these physical factors that affect the nematode horizontal distribution in the intertidal zone.
Another uncommon way to analyze the mesoscale distribution of nematodes is to examine an along-shore transect in the intertidal zone. In this case, samples that are far from each other are more heterogeneous Hodda 1999, Gingold et al. 2011); however, if the beach has different microhabitats along the across-shore transect, as in the case of a macrotidal ridge-and-runnel beach, the degree of patchiness is more accentuated in the microhabitat with calmer conditions (Gingold et al. 2011).

Microscale distribution
In relation to microscale vertical distribution, desiccation and oxygen availability are considered to be ultimate factors controlling the vertical distribution of the meiofauna (McLachlan 1978, Coull 1988, Maria et al. 2012. Oxygen is a significant limiting factor only in sheltered beaches. In this habitat, marine nematodes are restricted to the first centimeters of the sediment (McLachlan 1978), while in more exposed conditions, nematodes can be distributed more deeply (Urban-

10
TATIANA F. MARIA et al. Malinga et al. 2004). This phenomenon has led to the use of vertically undivided cores in meiofauna sandy-beach sampling, or the use of cores divided into large intervals of cm (i.e., larger than 1cm) (Blome 1983, Sharma and Webster 1983, Long and Ross 1999, Urban-Malinga et al. 2005. Moreover, samples are usually taken during low tide. This is a sampling strategy initially adopted for the macrofauna to avoid contamination by tidal migrants. It is however, also known that nematodes can migrate downward (Boaden and Platt 1971) or upward (McLachlan et al. 1977, Maria et al. 2012 in the sediment column at high tide. Therefore, to better understand the vertical distribution of nematodes, a finer-scale sampling scheme should be adopted, using cores subdivided into thin layers of a few centimeters, when dealing with protected and dissipative sandy beaches. Joint et al. (1982) has already demonstrated that nematodes showed different vertical distribution patterns at mm scale in an intertidal sandflat.
The relationship between the vertical distribution of sandy-beach nematodes and the tidal cycle was demonstrated by Boaden and Platt (1971) and McLachlan et al. (1977). These two studies were rather contradictory. The former authors showed that nematodes move down in the sediment, escaping from the upper layers and thus avoiding being washed away by the turbulent conditions. McLachlan et al. (1977) demonstrated that meiofauna living above the depth of the permanent water table undergoes vertical movements coupled with the tidal cycle, i.e., move up during submersion and vice-versa. Since the depth fluctuation of the groundwater level is directly linked to the different tidal stages (Urish and McKenna 2004), the upward movements of the meiofauna are closely linked to the tidal cycle. Indeed, the vertical distribution of many nematode species in the upper sediment layers of De Panne Beach during submersion was mainly explained by upward passive transport from deeper layers. Downward movements during emersion were explained by a combination of passive and active transport (Maria et al. 2012). Passive transport could be attributed to the drop in the underground water level, while active migration would occur to avoid harsh conditions created by drying of surface layers (McLachlan et al. 1977).
Although the investigation of the importance of the tidal cycle for the vertical distribution of the nematodes by Maria et al. (2012) was restricted to the upper five centimeters of the middle beach level, these results could be extrapolated to the other intertidal beach levels by combining the observations of the groundwater discharge in the beach sediments during the different tidal stages. The water table is very close to the sediment surface toward the low-water line during low tide. Therefore, here the vertical rise of the water table during the incoming tide would be more discrete than toward the high-water line. Consequently, upward passive transport of nematodes in the low intertidal beach would be less likely, so that nematodes would not dramatically modify their vertical distribution over the tidal cycle in this part of the beach. On the other hand, changes in the nematode vertical distribution over the tidal cycle would be more evident in the upper beach.
Biological interactions may also affect the vertical distribution of nematodes in the sediment, and may occur among different sizes of organisms. The role of biological interactions, such as competition and predation, in regulating macrofauna zonation remains little known (McLachlan and Jaramillo 1995). From the point of view of the meiofauna, biological interactions were only suggested to be important for atidal beaches (Hulings and Gray 1976), but Maria et al. (2012) showed that the species-specific vertical distribution of nematodes over a tidal cycle is clearly driven by biological interactions. The importance of predation was evidenced by the predacious nematode Enoplolaimus litoralis. This species was recorded in the layers just above the deposit-feeder Daptonema normandicum, which inhabited the subsurface 2-5-cm layers during submersion (Maria et al. 2012). The same pattern of predator-prey segregation was also observed in the coarsest sediments (<170µm and less than 5% silt) of a mudflat under high hydrodynamic stress, and was explained by food preferences (Steyaert et al. 2003). During exposure in De Panne Beach, E. litoralis was no longer restricted to the 1-3-cm layers as during submersion, but was found deeper in the sediment. This change may be the consequence of active migration, since another thoracostomopsid nematode, Enoploides longispiculosus, has shown high motility to actively catch deeper-living prey when the surface sediment was not water-saturated and the groundwater level was lowered (Steyaert et al. 2001). This migration took place during ebb tide, and also coincided with a low abundance of suitable prey species within the upper 5cm.
The importance of competition in driving the vertical distribution of species is confirmed to some extent by the vertical segregation of nematode species in the sediment. Species with a more 13 C-enriched diet, such as Sigmophoranema rufum, were abundant in the upper two centimeters of the sediment, whereas species belonging to epistrate (2A) and non-selective feeder (1B) groups that exploit a more 13 C-depleted diet showed higher densities in the subsurface layers -2 to 3 cm - (Maria et al. 2012). The importance of food competition between S. rufum and 1B/2A nematodes was supported by the results of an enrichment experiment, since the diatom uptake of epistrate nematodes was highest when the same food sources (diatoms) were offered to the nematode community. This may indicate that 2A nematodes are better competitors (Maria et al. 2011a).

NEMATODE COMMUNITIES IN BEACH FOOD WEBS
Food webs in sandy beaches differ with the beach type, and particularly with the degree of coupling between the beach and the surf zone (McLachlan and Brown 2006). Two distinct ecosystems are recognized: (1) beaches with little or no surf zone (reflective beaches), which are dependent on food input from the sea; and (2) beaches with extensive surf zones (dissipative beaches), which have high primary production from the surf diatoms that shows a vertical migration from the sediment deep in the surf zone to the water (McLachlan and Brown 2006). This ecosystem may support three partially unconnected food webs: the interstitial food web, the microbial loop (which is restricted to the surfzone and has been little investigated), and the macroscopic food web (Heymans and McLachlan 1996).
There are two studies that investigated the importance of nematodes in sandy beach food webs. Menn (2002) showed that a high wave energy beach has an impoverished food web due to a more diverse meiofauna while a less dynamic sandy beach showing a high dominance of nematodes in the meiofauna community can support a richer food web.
The second study assessed the integrative food web of the middle part of De Panne Beach by means of stable isotope analyses (Maria et al. 2012). Vascular plants were not included as a possible food source, since they are important primary producers of the supralittoral zone of Belgian sandy beaches (Speybroeck et al. 2008) and were assumed to be of less importance for the middle beach of De Panne, where a concrete dike separates the dunes from the beach . Most of the meiobenthic species (nematodes and copepods) utilize more carbon-enriched food sources than the macrofaunal organisms, indicating that there was no competition for food between these two classes of organisms. Phytoplankton and suspended detritus seem to be more important food sources for macrobenthic organisms on dissipative and reflective sandy beaches (Bergamino et al. 2011, Maria et al. 2012, while microphytobenthos 12 TATIANA F. MARIA et al.
(MPB) is utilized as a carbon source by nematodes, copepods, turbellarians and a few macrofaunal species, such as two species of the amphipod Bathyporeia (Maria et al. 2012). Laboratory experiments also indicated a higher uptake of benthic diatoms by these organisms (Maria et al. 2011a). Nonetheless, MPB is considered a limiting food source for sandy-beach organisms due to its low primary productivity, probably because of the strong hydrodynamic forces acting on the beach interface (McLachlan and Brown 2006); the preference for benthic over planktonic diatoms, as demonstrated by Maria et al. (2011a), leads to us to speculate that the absence of a diatom biofilm in sandy beaches, a condition extremely common on tidal flats, can be attributed to the rapid consumption of diatoms by benthic sandy-beach organisms. Although some field investigations and experimental approaches demonstrated that nematodes may serve as food source for large organisms, such as shrimps and fishes (see Coull 1990, for a revision), there was no explicit trophic link between macrofaunal and nematode species evidenced by isotopes analysis (Maria et al. 2012). It can be easily explained by the few numbers of species investigated in the study. POLLUTION A large set of features, including ease of sampling, omnipresence, high diversity, short generation time, and absence of a planktonic stage, make nematodes an excellent tool to evaluate the ecological condition of different environments (Giere 2009); however, very few studies have dealt with nematodes at lower taxonomic resolution (e.g. Fricke et al. 1981, Gheskiere et al. 2005b, Nanajkar and Ingole 2010. This may be related to the difficulty and tedium of species identification; therefore, pollution studies tend to use broader taxonomic categories and the two numerically dominant groups, nematodes and copepods, are the main focus of such studies. As seen in Table I, the four pollution-related studies focusing on sandy-beach nematodes dealt with different subjects: oil, recreational activities, and sewage discharge. In terms of oil pollution, the nematode density was still similar to reference sites when the sediment of the polluted site was not mechanically removed (Fricke et al. 1981). This finding contradicts the expected pattern of reduction in density at sites where an oil spill occurred (Wormald 1976, Giere 1979, Danovaro et al. 1995, Kang et al. 2014), but other field and experimental studies have indicated that nematodes are insensitive to oil pollution (Boucher 1980, Warwick et al. 1988. On the other hand, a mechanical treatment of the oil-polluted sediment had more impact in the nematode density than the pollution by oil per se, and more individuals were found in the deeper layers (Fricke et al. 1981).
There is little controversy around the time required for meiofauna recovery in areas subjected to oil spills. Fricke et al. (1981) found that meiofauna had recovered six months after an oil spill on the coast of South Africa (SA), while this time period was not enough for meiofauna recovery in three beaches located on the coast of Galicia (GC) (Veiga et al. 2010a). The lack of a similar response here is probably related to the difference in the magnitude of the oil spill (31,000 tons in SA x 50,000 tons in GC), and the local hydrodynamics (exposed in SA and under intermediate conditions in GC). In situ experiments have shown that sandybeach meiofauna can recover their density and community composition one month after an oil spill when very small quantities of oil are spilled (Kang et al. 2014).
The discharge of domestic sewage onto sandy beaches is a common problem in coastal areas, especially in developing countries, where a large part of the population may develop coastal areas in an unregulated manner. Nanajkar and Ingole (2010), studying the impact of chronic sewage discharge on a tropical sandy beach, observed that nematodes responded to the organic enrichment with an increase in density and dominance of one genus (Daptonema) in the areas closest to the sewage outflow. These authors believed that the non-selective deposit-feeding activity of Daptonema aided the bioremediation of the site.
From the point of view of beach management, this ecosystem has great socio-economic value as a recreational area and is therefore considered a key tourist destination during holiday periods. The recreational use of sandy beaches has a significant impact in terms of reduction of meioand nematofauna diversity, which responds by shifting to a large number of species with small size, rapid growth and high rates of reproduction (r-strategists). Nevertheless, this reduction may only be associated with the upper beach, which can be mechanically cleaned daily (Gheskiere et al. 2005b). On the other hand, in intense trampling activity in the intertidal area leads the meiofauna to migrate downward to layers deeper than 5cm (Moellman and Corbisier 2003).

CLIMATE CHANGE
Concerns about global warming have dramatically increased and it is widely understood that sandy beaches will be drastically affected, especially by sea-level rise. To date, very few studies have dealt with this subject (e.g. Kont et al. 2003, Lock et al. 2011, Mead et al. 2013. In terms of nematodes, only the study of Gingold et al. (2013) has concentrated on this group in sandy-beach environments. Laboratory microcosm experiments have shown that high temperatures lead to loss of important predacious and omnivorous nematodes that are important for the top-down control of the community, consequently leading to a change in the food web.
Concerning sea-level rise, the nematode intertidal community will be immersed more often or will become permanently submersed; therefore, we can speculate that many intertidal species will be unable to withstand long periods submerged.
This may eliminate many of these species and consequently reduce biodiversity in areas where the nematode biodiversity is presently very high. Besides, the rapid sea level rise will move beaches towards a more reflective morphodynamic state which is characterized by low meiofauna abundances (Yamanaka et al. 2010). In both cases, the reduction in nematode abundance may lead to a lost in ecosystem functioning. However, this topic requires further investigation.

CONCLUSIONS
Sandy beach ecosystems are one of the less studies coastal ecosystems; it can be easily exemplified by a simple search in the Web of Science which the number of studies in this environment correspond to 10% and 25% of those realized in estuaries and mangroves, respectively. The current state of knowledge shows us that over the decades, sandy beaches have been the scene of studies focusing on community and population ecology, both in relation to morphodynamic models. The combination of physical factors (e.g. grain size, tidal exposure and degree of drainage) and biological interactions (e.g. trophic relationships and competition) is responsible for the spatial distribution of nematodes. The degree of importance of these factors is related to the kind of distribution pattern analyzed; physical factors are more important in structuring nematodes communities over large scale of distribution (e.g. macro-and mesoscale) while biological interactions are largely important in finer-scale distributions.
Although nematodes are the dominant meiofauna group and are largely used as indicator of environmental conditions it is of primordial importance to understand the natural relationship between the environment and these organisms. Therefore, to understand better the ecological processes driving the sandy-beach nematode community is necessary to increase the efforts in understanding latitudinal and beach type patterns also taking into account the environmental heterogeneity. 14 TATIANA F. MARIA et al.
In terms of energy, nematodes are considered an important link between micro-and macrofaunal organisms; however, isotopic studies showed that there is a clear separation between the interstitial and macrofaunal food webs at least for temperate sandy beaches.
Even though the nematode diversity is related to physical and biological interactions, it is also related to the degree of impact occurring in the environment; under disturbed conditions (i.e., organic pollution, oil pollution, beach cleaning or trampling) the diversity decreases and the density of some dominant species (especially deposit feeders) increases, the natural recovery depends on the amount and duration of the impact.
And last but not the least, the relationship between nematodes and climate change is a promising research area, which many aspects must still be evaluated for sandy-beach organisms, but a common believe is that the nematode density must decrease.
This review elucidates the current knowledge of sandy-beach nematodes and as this ecosystem is easily reached and occurs worldwide we should stimulate more and more studies that can provide a more comprehensive approach toward a better understanding of the physical and biological interactions among sandy-beach organisms. This is essential in order to understand the possible effects of key human pressures on the beach ecosystem.

ACKNOWLEDGMENTS
Renata Alves and Alex Silva are acknowledged for their help with GIS Mapping Software and the map quality, respectively. The authors thank Jannet W. Reid, JWR Associates, for the critical review of the English. The data was obtained from the PhD thesis of the first author that was sponsored by Vlaamse Interuniversitaire Raad-VLIR-UOS. Additional funding was provided by the Special Research Fund of Ghent University (BOF-GOA 01GA1911W). André Esteves (CNPq 312143/2013-3) wishes to express his gratitude for the research fellowship granted by the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq). Authors are also grateful to the two reviewers (Maickel Armenteiros and the anonymous one) by their comments.