Printer Friendly

Monitoring bioeroding sponges: using rubble, quadrat, or intercept surveys?

Abstract. Relating to recent environmental changes, bioerosion rates of calcium carbonate materials appear to be increasing worldwide, often driven by sponges that cause bioerosion and are recognized bioindicators for coral reef health. Various field methods were compared to encourage more vigorous research on bioeroding sponges and their inclusion in major monitoring projects. The rubble technique developed by Holmes et al. (2000) had drawbacks often due to small specimen sizes: it was time-costly, generated large variation, and created a biased impression about dominant species. Quadrat surveys were most rapid but overestimated cover of small specimens. Line intercepts are recommended as easiest, least spatially biased, and most accurate, especially when comparing results from different observers. Intercepts required fewer samples and provided the best statistical efficiency, evidenced by better significances and test power. Bioeroding sponge abundances and biodiversities are influenced by water depth, sediment quality, and most importantly by availability of suitable attached substrate. Any related data should thus be standardized to amount of suitable substrate to allow comparison between different environments, concentrating on dominant, easily recognized species to avoid bias due to experience of observers.

Introduction

Worldwide coral reefs are suffering from anthropogenic environmental changes with consequences to local economies (e.g., Hoegh-Guldberg, 2009). While the science community largely investigates changes acting on reef calcification rates, the antagonistic process--calcium carbonate bioerosion--is not studied to the same extent despite having been discovered almost 50 years ago (Fig. 1). Neumann (1966) defined bioerosion as the destruction of hard substrates by living organisms, and it is a central part of decalcification processes on coral reefs (e.g., Glynn, 1997). Sponges are often a dominant cause of internal bioerosion and thus play a significant role in reef bioerosion (e.g., Hudson, 1977; MacGeachy, 1977; Highsmith, 1981; Perry, 1998; Mallela and Perry, 2007), able to drive the carbonate budget of a reef into an erosional state (Acker and Risk, 1985; Nava and Carballo, 2008). However, sponges are commonly excluded from related studies, because study methods are perceived as too difficult (e.g., Cantera et al., 2003; Scaps and Denis, 2008). Few empiric data are available on changes in sponge bioerosion over time and directly linking observations to environmental change (e.g., Ward-Paige et al., 2005), but a number of scientists provided reports on rising bioerosion rates and increasing densities of bioeroder communities worldwide, assuming this development to be related to deteriorating environmental conditions at a larger scale (Caribbean: Cortes et al., 1984; Rutzler, 2002; Taiwan: A. Chen, Academia Sinica, pers. comm.; Great Barrier Reef: Schonberg and Ortiz, 2009; A. Thompson, Australian Institute of Marine Science, pers. comm.). Focused aquarium experiments confirmed that bioerosion levels will rise significantly in present trends of environmental change (Tribollet et al., 2006; 2009; Duckworth and Peterson, 2013; Wisshak et al., 2012, 2013, 2014; Reyes-Nivia et al., 2013; Fang et al., 2013, 2014) and that this will impact coral reef health (Kennedy et al., 2013). This conclusion was further supported by field studies under naturally changed conditions (e.g., Fabricius et al., 2011). While many environmental changes negatively affect calcifiers, we are presently unable to show that bioeroding sponges are disadvantaged to the same extent. On the contrary, a number of studies provided evidence that unlike reef builders, bioeroding sponges often benefit from such changes (e.g., Rose and Risk, 1985; Rutzler, 2002; Schonberg and Ortiz, 2009; Wisshak et al., 2012; 2013; 2014) and can be used as bioindicators for decreasing water quality (e.g., Carballo et al., 1994; Holmes, 2000; Holmes et al., 2000; Chavez-Fonnegra et al., 2007). It is therefore reasonable to assume that with increasing damage to existing coral reefs, bioerosion caused by sponges may rise beyond healthy levels by exceeding calcification rates (e.g., Acker and Risk, 1985; Nava and Carballo, 2008), creating yet additional stress in reef environments and likely leading to reduced biodiversity due to reduced structural complexity (e.g., Kendall et al., 2009; Enochs and Manzello, 2012).

Despite the alarming prognoses, we lack widely accepted routine procedures and tools to collect data on sponge bioerosion that can be compared over time and space. Making preliminary recommendations and asking for observations by co-workers, Schonberg and Holmes (e.g., 2008 and 2010 [see Appendix], 2011) and Schonberg et al. (2012 [see Appendix], 2013) urged researchers during several scientific conferences to include bioeroding sponges in their monitoring programs. The presentations created much positive feedback and intensive interest, and the urgent need to understand related processes was widely acknowledged. However, to our knowledge, except for programs in the Caribbean/Gulf Region and neighboring areas of the Atlantic (Gilliam et al., 2004, 2006; Makowski et al., 2009; Ruzicka et al., 2009; Gilliam, 2010; Lang et al., 2010; Makowski and Keyes, 2011) bioeroding sponges are still not included in many of the larger, long-term monitoring programs. The present impression is that many monitoring agencies are already overcommitted and some of them cannot easily develop and test new approaches or protocols, or they rely on volunteer work or low budgets, which often prevents changes to existing procedures. What is therefore required is a ready-to-use, simple, quick, replicable method that can be employed to varying levels of detail and by comparatively untrained people--a method that still generates meaningful data (see also Wulff, 2001). It needs to be standardized to avoid spatial bias and to allow comparison between repeated data collections and between different locations. It should be cheap and should not require expensive technology so as to be available to agencies and nations of different incomes, some of which are important stakeholders and policy-makers for most of the world's coral reefs (Wilkinson, 1993; Holmes et al., 2000; Risk et al., 2001).

Katherine Holmes (Holmes, 2000; Holmes et al., 2000) proposed a method that may be suitable to obtain access to comprehensive spatial and temporal data on bioeroding sponge distributions and bioerosion per unit substrate. She and co-workers observed that in coral rubble, bioeroding sponge diversity and bioerosion caused by sponges increased toward nutrient-enriched environments (Holmes, 1997, 2000; Edinger, 1998; Holmes et al., 2000). She further suggested that this method can be used as part of a tool kit to evaluate coral reef health. The present study aims to test the suitability of the rubble method for monitoring purposes and to compare it to results generated by quadrator transect-based studies, which are more commonly used by monitoring agencies (e.g., English et al., 1997; Hill and Wilkinson, 2004). Comparing efforts needed and outcomes of the three methods resulted in a recommendation for monitoring purposes.

Materials and Methods

Aiming to recommend a simple method to monitor bioeroding sponges, a rapid study was conducted in Little Pioneer Bay at Orpheus Island, central Great Barrier Reef, Australia (Fig. 2A-D). The site was chosen because of its proximity to Orpheus Island Research Station (OIRS), because the author is able to identify the local dominant bioeroding sponges in situ, and because more than 10 years of background data on this sponge community are available, especially on their distributions in different reef zones (Schonberg and Ortiz, 2009; Schonberg, 2000, 2001; Schonberg, unpubl. data). Transects like those for the 2001 and 2009 references above were re-established and used as orientation for the present study: in the southern corner of Little Pioneer Bay, five 100-m transects were placed at right angles to the shore, spaced 10 m apart (Fig. 2D).

The present study was intentionally kept as simple as possible and employed a comparatively fuzzy approach to accommodate various possible conditions and to establish a method that can be readily used by non-experts with a low level of prior knowledge. Three methods were compared: obtaining data from collected rubble (Holmes, 1997, 2000) and from visual surveys of the reef surface using either quadrats (English et al., 1997; Schonberg, 2001; Hill and Wilkinson, 2004) or line-intercept transects (Loya, 1978; English et al., 1997; Hill and Wilkinson, 2004).

Rubble method

The suitability of the rubble method was evaluated by collecting five pieces of rubble per site or time (Fig. 3A). This process did not select for any particular type of rubble and included either coral debris of any form or unattached mollusc shells, because Holmes et al. (2000) did not find notable differences between bioerosion in Acropora compared to non-Acropora rubble. Rubble was taken from each of the local six reef zones defined by Schonberg (2001) and for each of the five 100-m transects (Fig. 2D). The first five available rubble pieces per site were picked up in a haphazard way without selecting for evidence of sponge erosion (5 pieces from 6 zones and 5 transects = 150 pieces). For each site, rubble was placed into a separate, sealable plastic bag. Opened plastic bags were kept in a flow-through aquarium at OIRS with edges folded down for a few hours until rubble was further processed.

In the laboratory the surface of each piece of rubble was first investigated, but evidence for bioeroding sponges was very rarely evident in this way. Thus, following the original instructions for the method (Holmes 1997, 2000; Holmes et al., 2000), each piece of rubble was fragmented and cut into five smaller pieces with a ratchet-action pipe cutter, and the resulting inner surfaces were again scrutinized for live bioeroding sponges. Found sponges were generally very small and contracted, and often no additional evidence was available from the surface. Species identification was thus impossible without further work, creating significant difficulties for the estimation of the volume of sponge tissue per piece of rubble and requiring a comparatively large effort for related investigations. Small amounts of tissue were plucked out with very fine watch-maker tweezers (e.g., no. 4 or 5 Inox, Switzerland), placed on a microscope slide, fixed in an aqueous mounting medium (GelMount, Biomeda, Australia), slightly squeezed with a cover slip, and sealed around the edges with clear nail polish after 2 h of drying. Resulting preparations were observed under a Laborlux 12 compound microscope (Leitz, Germany). Where possible, sponges were identified to species level, but the main aim was simply to distinguish and count the different species. Apart from scoring species counts per piece of rubble, no specific attempt was made to quantify the sponge bioerosion in rubble because the small specimen sizes would have significantly increased the work effort to a level unlikely to be suitable during monitoring studies.

Quadrat method

To evaluate the suitability of using quadrats, a continuous belt transect survey was conducted using a 0.5 X 0.5-m quadrat along the pre-established 100-m transects running at right angles to the shore (Fig. 2D). From this data set five subsamples = five quadrats from each reef zone and along each of the five 100-m transects were chosen at random, resulting in data from 150 quadrats or a total area of 37.5 [m.sup.2]. Each quadrat was subdivided into 25 smaller quadrats of 100 [cm.sup.2] (Fig. 3B). Each time the quadrat was placed onto the ground, the proportions of the following materials were estimated as counts of smaller quadrats (Schonberg, 2001):

*suitable calcium carbonate substrate available to bioeroding sponges (including coralline algae, calcium carbonate substrate that was invaded by these sponges, dead coral substrate that can be covered by algal turf, and rubble),

*live surface cover that was assumed to prevent settlement success of most bioeroding sponges (corals, soft corals, other sponges, etc.),

*sand (unsuitable for settlement),

*mud (unsuitable for settlement).

Biodiversity and abundance of bioeroding sponges were counted as number of smaller quadrats in which each perceived species occurred, even if there was only a small amount of sponge tissue, and including partial sponge specimens (type II observation according to Zvuloni et al., 2008, but only recorded for any material within the quadrat and including partial specimens). Sponge species were identified and distinguished in situ as far as possible or grouped as species complexes where species could not be distinguished beyond doubt or could not be recognized without laboratory studies.

Line-intercept method

The third approach was performed as a line intercept study (Fig. 3C). At the same sample site, per zone and along one of the 100-m transects, five separate, stretched 10-m line-intercept transects were laid out in parallel to the shore using a measuring tape (Fig. 2D; Birkeland, 1984), resulting in 30 line intercepts. The ground under each 10-m line intercept was visually inspected for evidence of bioeroding sponges that were crossed by the measuring tape (see Loya, 1978; English et al., 1997). The extent of each sponge was recorded as distance along the tape for every meter (Fig. 4). As for the quadrat method, linear extents of suitable, mostly dead substrate (including that inhabited by the sponge), live surface cover, sand, and mud were noted for the intercepts (Fig. 4). As for the quadrat method, sponge species or species groups were identified in situ, and species identifications are not presently supported by spicule mounts.

Analysis

In addition to generating total species counts per method and reef zone, I analyzed sponge data means from all three methods stepwise (means per individual sample unit of rubble, quadrat and line-intersect transect, means per zone and method, then means per available substrate and method) to compare patterns between data sets. Data were ultimately standardized to amount of suitable substrate to allow comparisons between different studies and to eliminate spatial bias caused by the fact that the sponges occur only in calcareous substrate, which is inhomogeneously distributed. In the first step, total values were calculated for species counts (biodiversity, omitting repeat occurrences when moving up in the data hierarchy), and total extent (abundances: number of occurrences in rubble, number of subquadrats in quadrats and total cm along line intercepts). Then mean biodiversity and abundance were calculated hierarchically by first obtaining mean values for each set of the five replicates per zone, which were then averaged over the five large transects (not necessary for intercepts), generating means per zone. In an attempt to bring them all to the same, comparable level, biodiversity data were divided by number of sub-units studied (5 rubble fragments, 25 sub-quadrats with a side-length of 10 cm, 10 1-m sections per intercept transect), while abundance data were expressed as percentages of study units occupied by the sponges (this was not possible for rubble). The last step standardized the data in a way that they can be compared between different sites and times where and when available amount of settlement substrate will differ: species numbers were divided by sub-units that represented suitable substrate, and abundances calculated as percent of suitable substrate occupied by bioeroding sponges. Again, this last step was not possible for rubble.

Substrate type data were available from the quadrat and intercept surveys. As only the upper surface was considered, not contours, the results were comparable between the two approaches and similar studies, but suitable substrate area was likely underestimated in comparison to other studies (Weinberg, 1981; Dodge et al., 1982; Beenaerts and Vanden Berghe, 2005). Stepwise means were calculated as described above, and substrates were expressed as percentages per zone. The results were then compared between the quadrat and the intercept methods, evaluating the accuracies and interchangeabilities of the two, especially with regard to proportions of suitable substrate.

After preliminary evaluations, the rubble method was regarded as the least suitable method, because the tiny size of the specimens necessitated detailed laboratory work looking at spicules and significant taxonomic knowledge about the sponges. Consequently, only the quadrat and intercept methods were further investigated. For one

data set each, the quadrat and the line intercept methods were applied by the author, by a person who had some experience with bioeroding sponges in the area (ca. 3 months), and by a person who was entirely new to the topic. A very quick introduction to the methods was provided to the two latter persons, and both had assisted during a few earlier dives involving sampling of a number of bioeroding sponges. They were told to note the occurrence and extent of every organism they could recognize as bioeroding sponge and to estimate the proportions of the different substrates (suitable, live, sand, mud).

Finally, results from the three methods were compared. The quality of the obtained data was tested in two ways. Species accumulation curves of estimated and observed biodiversities from all three methods were prepared in Estimates ver. 8.2.0 for Windows (2012) (Colwell and Coddington, 1994; Colwell, 2005; with settings as in Schonberg and Fromont, 2012) in order to achieve an estimate of the minimum sample size that would represent about 80% of the local number of species of bioeroding sponges (Loya, 1978; Scheer, 1978, recommends representing 1/2 to 2/3 of the local species). Statistical robustness of data resulting from the three methods was checked by conducting oneway ANOVAs on the biodiversities for each method with the factor "reef zone," including "effect size" and "test power" in the data output (SPSS 19, IBM, Armonk, NY). This approach assumed that the number of bioeroding sponge species would vary between zones, and accepted 0.5 for "effect size" and 0.8 for "test power" as desirable values according to general conventions (e.g., Field, 2009). Data were further assessed to estimate the feasibility of the methods--that is, how much time and effort were needed for each approach, costs necessary to conduct the studies, and how the results related to what was previously known of this particular bioeroding sponge community (Schonberg, 2001; Schonberg and Ortiz, 2009).

Results

One investigator, comparing all three methods over a range of samples

Results were difficult to compare between the three methods examined--collecting rubble and surveying the reef surface with quadrats or line intercepts--because the methods themselves created differences in the scale and behavior of the data. However, any of the three tested methods can be used to compare bioeroding sponge biodiversity between different sites, and as a consequence over time, but at present only the quadrat and the intercept methods are regarded as suitable for estimating abundances for monitoring approaches.

For estimating the local biodiversity of bioeroding sponges, the intercept method generated the best results, requiring fewer sample units than the other two methods (Fig. 5A-C). While both quadrat and intercept surveys reached calculated asymptotic species assimilation curves with 8 samples (Fig. 5B, C), the rubble method needed 28 samples (Fig. 5A). To represent 80% of the calculated local biodiversity of bioeroding sponges, 28, 13, and 9 sample units had to be taken with the rubble, quadrat, and intercept methods, respectively. The rubble method also had the largest variation and the quadrat method the smallest (Fig. 5). The presently recommended sample volume of n = 10 appeared to be good for the intercept surveys, less so for the quadrat, and not at all for the rubble surveys. Using total counts, the rubble method detected no significant difference of bioeroding sponge biodiversities between the reef zones and had a poor effect size and test power (Table 1). Both the quadrat and the intercept method recognized a significant difference in sponge biodiversities between reef zones, but only the intercept method had excellent values for effect size and test power (Table 1).

Effort and required resources also varied between the methods. All 150 pieces of rubble used for this study were collected within 30 min, but sifting through samples in the laboratory demanded the most processing time (Table 2) and could not be avoided, because the small size of the specimens made preparations and microscopy unavoidable. From rubble, no field or surface identifications could be made; therefore this approach required at least a basic knowledge of sponge taxonomy to enable species to be distinguished from the small tissue scraps of the pluck preparations, which is a disadvantage for general monitoring programs. If relying on field identifications only, the quadrat method was overall the fastest, and the intercept method took almost as much time in the field as the rubble method took in the laboratory (Table 2). In terms of equipment, sampling and processing requirements were quite similar for all three methods; the rubble method with its small sponge specimens made the use of laboratory equipment unavoidable, while this was optional for the other two methods. Due to the quick way rubble can be collected in the field, related work can possibly be conducted when wading or snorkeling, but scuba is strongly recommended for safety in quadrat or intercept sampling, significantly increasing the necessary costs and required time while conducting the census or optionally retrieving hammer-and-chisel samples (Table 2). In comparative data sets the rubble

method usually displayed the largest error bars due to a significant number of pieces of rubble containing no sponges and the occurrence of a large number of rare species, and the intercept method often had the least variation (Table 2, Fig. 6). The three methods identified different species as dominant and different total numbers of species for the entire data set, with results from the quadrat and intercept methods being similar, but distinct from those from the rubble method (Table 2). Quantifying extents of bioeroding sponges from rubble is time-intensive and was not attempted for the present study. The respective estimations from the intercepts were the most accurate values, while those from the quadrat method represented overestimations, although behaving in the same patterns as those observed from the intercept method (Table. 2, Fig. 6E-H).

For all three methods, total, mean, and proportional numbers of bioeroding sponge species per zone rose in a similar pattern from the sandy zone crestwards, peaked in the pavement zone, and then leveled out (Fig. 6A-D), the almost identical numbers possibly indicating that the sample size was adequate (Fig. 6A). The overall trend was most pronounced when relating biodiversity to available substrate, the step when data became standardized and comparable between different environments. Abundances showed a different development across the zones, also mostly regardless of the calculation steps, with specimen counts increasing across the reef flat with availability of attached substrate, peaking on the crest, and then decreasing again (Fig. 6E-H), also with clearest results in reference to available substrate. The intercept method detected the largest numbers of species and specimens as mean per main study unit (intercept line, Fig. 6B, F), but when data were displayed in comparable format (Fig. 6A, C), more species were found in rubble that was scrutinized in the laboratory. Data expressed as percent abundance per available substrate were exactly comparable to each other and confirmed that the quadrat method overestimated sponge abundances (Fig. 6F-H).

While the rubble method with laboratory-based data acquisition occasionally delivered different results concerning the species composition compared to the two surface-related and field-based methods, quadrats and intercepts generated very similar patterns when compared to each other for both biodiversities and abundances (Fig. 6). Data for substrate types across the reef zones obtained from the quadrat and intercept methods were also very alike (Fig. 7). For the substrate types, the intercept method again resulted in smaller error values than the quadrat method (Fig. 7).

Different investigators examining the same quadrat and intercept samples

Regardless of the method and investigator's experience, the accuracy of recognizing and quantifying the most dominant and more conspicuous species was very good (e.g., brown patches of Cliona orientalis in encrusting [beta] form, the bright red papillae of Cliothosa aurivillii, and the prominent black fistules of Siphonodictyon mucosum (represented by sponges 1, 2, and 4 in Fig. 4), while drab forms such as the papillate a form of C. orientalis, small specimens, and small-papillate forms such as Pione spp. and Cliona minuscula were more likely to be spotted and correctly identified with more experience (represented by sponge 3 in Fig. 4, Table 3; species authorities are provided in Table 2). Overlooking drab and small specimens influenced the species counts and estimates for abundances (Table 3).

Quantifying the proportions of the different substrate materials was more similar among the investigators when the quadrat was used, but still good along the intercepts (Table 3). When observed sponge occurrences were standardized to available suitable substrate, the intercept method was more consistent among the observers than when the quadrat method was used (Table 3) and compared well to figures obtained during the experienced "one-observer surveys."

Discussion

Rubble, quadrats, or intercepts: recommendation for monitoring bioeroding sponges

All three methods tested had positive and negative aspects for the proposed purpose (see also Dodge et al., 1982). The rubble method was repeatedly and successfully used before to compare sponge bioerosion between sites and over environmental gradients, and it was previously favored over other methods because it does not involve sampling of live corals and pieces are quickly collected (Holmes, 1997, 2000; Edinger, 1998; Holmes et al., 2000; Risk etal., 2001), maximizing the work time available on scuba (see Parravicini et al., 2010). Sampling of rubble can easily be done on a low budget and was recommended to be part of a coral reef health tool box (Table 2; Holmes et al., 2000; Risk et al., 2001). Other, isolated studies relied on the same principle when investigating reefal bioerosion (Perry, 1998; Sheppard et al., 2002; Carballo et al., 2008a), but otherwise the rubble method has largely been ignored by the scientific community, which generally has relied on quadrat studies or belt transects instead. The present study shows that, especially for total number of species per reef zone, results generated by collecting rubble are quite similar to those produced by surface surveys and therefore have significant value when used (Fig. 6A). However, and importantly, it created an entirely different impression about dominant bioeroding sponge species than did data collected from visual evaluation of the reef surfaces. The latter studies rely mainly on massive or at least attached substrate and will miss all species that are too small to be discovered from the surface. The rubble method mostly detected species that were less important in terms of aggressive overgrowth and undermining of coral polyps than the dominant species found with quadrats and intercepts (Table 2; an effect not noted by Holmes et ah, 2000; but see Schonberg, 2001; Carballo et al., 2008a). Total species counts from the rubble method across the entire reef flat were about twice as high as from the other two methods (Table 2). This may be related to biological reasons or, more likely, to the fact that only for rubble samples were species fully distinguished by microscopy, while the surface surveys relied purely on differences that can be distinguished in the field, thus grouping several species together. In further comparisons between the present approaches, the rubble method was the most time-consuming (Table 2), had the largest variation between samples (Fig. 6; see also Risk et al., 2001), required the most a priori experience to be able to distinguish tiny sponge specimens from characters painstakingly extracted from pluck-preparations, and presented the largest risks of overlooking very small specimens and thus of producing different results between different observers. It may also be important to consider that rubble can be moved, for example by storms such as Cyclone Yasi which crossed Orpheus Island half a year before sampling (Australian Institute of Marine Science, 2011), and related findings may not always accurately reflect conditions of a given sample site or zone. Although generating good results during targeted investigations (Holmes, 1997, 2000; Edinger, 1998; Holmes et al., 2000; Risk et al., 2001), quantification of sponge bioerosion from rubble is here considered too difficult and time-costly for large-scale and repeated studies that also involve other assessments. For all the above reasons, the rubble method is presently not considered to be the best approach for monitoring.

In comparison to the other two methods, the quadrat approach has been used most widely and traditionally as a means to study distributions of bioeroding sponges in different oceans. Quadrats were mostly placed along pre-established transects or were represented by full belt transect surveys, creating areal data (Rutzler, 1975; McKeever Targett and Schmahl, 1984; Acker and Risk, 1985; Rose and Risk, 1985; Sullivan and Chiappone, 1992; Schonberg, 2001; Rutzler, 2002; Callahan, 2005; Gilliam et al., 2004, 2006; Lopez-Victoria and Zea, 2005; Ward-Paige et al., 2005; Cebrian and Uriz, 2006; Chavez-Fonnegra et al., 2007; Zundelevich et al., 2007; Chiappone et al., 2007; Carballo et al. 2008a, b; Caballero et al., 2009; Parravicini et al., 2009; Ruzicka et al., 2009; Schonberg and Ortiz, 2009; Cebrian, 2010; Gilliam, 2010; Lang et al., 2010; Chavez-Fonnegra and Zea, 2011). Given some prior experience with the most common bioeroding sponges in the area or a quick introduction to a few target species, no extractive sampling would be required, making this method the overall most time-efficient choice (see also Dodge et al., 1982), and one that recognizes the same dominant sponge species as does the more time-costly intercept method (Table 2). However, the quadrat method over-estimated abundances of small, papillate bioeroding sponges that are counted as subunits of a given quadrat even if they contribute only a single papilla or fistule (e.g., as in Schonberg, 2001), unless single papillae are entirely omitted from the study (e.g., as in Callahan, 2005), or all components within a quadrat are estimated as percent area (e.g., as in Makowski et al., 2009; Makowski and Keyes, 2011). Percent area estimates would provide a better resolution than counting sub-quadrats, rendering data quality equal to that from line intercepts, but they could possibly introduce more bias by different observers. Counting sub-quadrats also means that the relationship between papillate [alpha] form sponges (overestimated) and endolithic-encrusting [beta] form sponges (more accurate estimate) may shift out of proportion (Figs. 3, 4). Also, even though variation between observations of a single observer was relatively small (Table 2) in comparison to those between observers of different experience, the quadrat method generated results that varied more strongly than those of the intercept method (Table 3). Furthermore, random placement of a set number of quadrats may not always include the required suitable calcareous substrate--that is, it may miss part of the patchily distributed bioeroding sponge community (Dana, 1976; Sudara and Snidvongs, 1984). Therefore, despite the fact that the quadrat method produced the most rapid results and is a successful, widely established, often recommended tool (e.g., Weinberg, 1981; Sullivan and Chiappone, 1992), it may not always be optimal for studies on bioeroding sponges and is presently not favored for monitoring them.

To my knowledge, to date only three studies included elements of line intercepts to quantify bioeroding sponges or related substrate proportions (Chavez-Fonnegra et al., 2007; Caballero et al., 2009; Chavez-Fonnegra and Zea, 2011). As here confirmed, this method can be comparatively time-consuming in the field (e.g., Hill and Wilkinson, 2004), has occasionally been described as an only mediocre technique to survey reefs (Weinberg, 1981), and may bias respective data toward larger sponge specimens (Hill and Wilkinson, 2004; Zvuloni et al., 2008), but is less susceptible to bias by inhomogeneous occurrences as observed for the use of quadrats (Dana, 1976; Sudara and Snidvongs, 1984). Unlike results presented by Dodge et al. (1982), in the present approach intercepts required fewer replicates than the other two methods, while still displaying less variability (Table 2, Figs. 5, 6). The determined number of required replicates of n = 10 is low and comparable to estimates established by earlier reef studies for coral cover (e.g., Nadon and Stirling, 2006). Where experience on the local sponge community is available or only the most obvious sponges are targeted (as e.g., in Gilliam, 2010; Lang et al., 2010; Fig. 4), a detailed and reliable data set can be produced in less than one day of work (Table 2) and can certainly be repeated over spatial and temporal scales. Like the above methods, intercept surveys can easily be used on a small budget. The intercept method correctly identified the dominant species that matter most for coral bioerosion: those that can invade live coral, display fast growth, and produce large individuals (Table 2). When our standardized sponge abundances were found to increase over the reef flat and then decrease in front of the reef crest, it confirmed results for earlier studies at the same site (Schonberg, 2001; Schonberg and Ortiz, 2009) and similar studies from other regions in the Pacific (e.g., Carballo et al., 2008a), although the situation in the Caribbean may be different (Perry, 1998). In the present study, the intercept method exhibited the least risk for bias between observers with different levels of experience (Table 3). Perhaps most importantly, this method will suit established routines of existing programs that are already working with approaches based on transect lines (e.g., Reef Check, see Hodgson et al., 2006). The intercept method is one of the most commonly used survey methods on coral reefs (e.g., Weinberg, 1981; Nadon and Stirling, 2006; Zvuloni et al., 2008); adaptable to various purposes (e.g., Loya, 1978; Marsh et al., 1984; Zvuloni et al., 2008); and supported by present results, it appears to be the most suitable approach for monitoring sponge bioerosion. It has previously been successfully used to observe the distribution of other bioeroding organisms in studies aiming to obtain data on the status of reef health and can thus be expected to have wider application (e.g., Scaps and Denis, 2008).

For monitoring programs that already include transect studies, I strongly recommend adding observations on bioeroding sponges, but not to include all species. Including two or three dominant species will be sufficient to recognize changes at the community level. To increase accuracy between replicate studies and to reduce the effort (Table 2), only the most common and most easily recognized species should be quantified (e.g., Gilliam, 2010; Lang et al., 2010) unless experienced personnel are available and more time can be expended. In addition to reducing observer bias, this also reduces possible bias by differences in the endolithic structures, because the most dominant species are usually larger and often have more coherent endolithic tissue than species with smaller specimens (as for sponges 1 and 4 in Fig. 4). As a quality control, an initial quick pilot study similar to the one described here should be conducted to compare results among the different observers involved and determine a minimum sample volume at a new location (English et al., 1997; Hodgson et al., 2006; Bakus et al., 2007). To correct for patchy spatial distribution of calcareous material that is suitable for sponge invasion (e.g., Schonberg, 2001), observations should always be standardized to amount of suitable substrate (including coralline algae, the dead substrate the sponge occupies, and substrate covered by algal turf; Fig. 4). This is necessary because the amount of suitable substrate available varies between oceans, reefs, and reef zones, as well as over time, and it significantly influences the sponge community structure on a given reef (e.g., Chavez-Fonnegra et al., 2007; Carballo et al. 2008a; Schonberg and Ortiz, 2009; Makowski and Keyes, 2011). As long as the estimates are based on continuous data rather than on scores (i.e., if quadrat data are estimates of percent area rather than counts of sub-quadrats), the method itself will not be as important for comparisons as is the reference to available substrate. The angle of observation can matter, however; for repeated observations of the same site, the position of the observer should always be the same. From the present experiences we know that differences between observations can result from swimming along the left or the right side of the line, when the observer on one side may have a good view on a sponge, while on the other it may be hidden by a structural irregularity. To create the most similar conditions between studies, the intercept line should be stretched horizontally and not follow all reefal contours. A stretched line may lead to an overestimation of live cover in comparison to contour studies (Leujak and Ormond, 2007), but using a slack line or following contours can result in almost 1/3 more variability between different observations (Bakus et al., 2007). In zones with much structural variability (e.g., our relief zone), the stretched line may be too far above the evaluated path to keep the observation angle constant, which can introduce bias (see also English et al., 1997). In such cases a plumb line can be suspended from the intercept line to better define which objects on the bottom should be counted, or a rod can be held vertically against the intercept (English et al., 1997; Leujak and Ormond, 2007). To facilitate planning of related work, I recommend an example data acquisition sheet that can be altered depending on local conditions (Table 4).

Rubble, quadrats, and intercepts: biological implications for the sample site

Different survey methods can generate similar results describing benthic patterns (e.g., Dodge et al., 1982). All applied methods showed that bioeroding sponge biodiversity decreased with risk of exposure to air in the shallow sandy and mixed zones. All three methods also indicated that compared to the other reef zones, bioeroding sponge biodiversity at Orpheus Island was slightly higher in the pavement zone--the zone with the highest availability of suitable, mostly dead, attached substrate that exhibits a higher degree of stability and durability than, for example, the substrate in the branching zone, which can largely disintegrate during bleaching events and storms (e.g., Berkelmans and Oliver, 1999; Baird and Marshall, 2002; Berkelmans et al., 2004; C. Schonberg, pers. obs. after Cyclone Yasi 2011). Under normal circumstances the pavement zone may thus best support the development of an established equilibrium eroder community. In the branching, crest, and relief zones, species numbers were insignificantly lower (Fig. 6D). This may be related to the sediment quality, which became finer toward the reef edge and may increasingly interfere with the sponges' pumping capacities or their ability to resist smothering (Fig. 7). While bioeroding sponges often benefit from particular organic matter in the water column, some species may be adversely affected by excessive sedimentation (Carballo et al., 1994; Chavez-Fonnegra et al., 2007). I assume this to be more important for species in rubble that can easily become buried in mud, and the drop in biodiversity and abundance after the pavement zone where sediments become finer is indeed most pronounced in data from the rubble method (Fig. 6). Studying bioeroding sponge abundances with the three different methods created a slightly different picture compared to the biodiversity data. Abundances steadily increased across the reef flat without a peak in the pavement zone, but were apparently related to circulation, water depth, and lower risk of desiccation. After that, abundances decreased again just behind the reef crest, where predators may have easier access and the fine sediments may again play a role in smothering small sponge specimens. This trend was especially clear when data were standardized to available substrate, which confirmed findings from earlier surveys in which only the quadrat method was used (Fig. 6H; Schonberg, 2001; Schonberg and Ortiz, 2009).

The above results imply that availability and durability of suitable, attached substrate is a key factor in determining the community structure and biodiversity of bioeroding sponges (e.g., Chavez-Fonnegra et al., 2007; Carballo et al. 2008a). Water depth, mixing, and fine sediments appear to have a critical influence on diversity as well (see also Schonberg, 2001), but probably play a more important role in controlling the abundances of bioeroding sponges, and responses appear to differ between different species.

Apart from the common, abundant, and dominant "usual suspects" in attached materials, such as Siphonodictyon mucosum, Cliona orientalis, and various brown-papillate clionaids (Schonberg, 2001; Schonberg and Ortiz, 2009), the rubble method showed that Cliona mucronata, Cliona ensifera, Cliona caesia, and Pione cf. vastifica are also very common at Orpheus Island, but usually have much smaller specimens and are not as easily detected unless the substrate is broken open, thus causing damage. Especially C mucronata and C. ensifera may mostly be overlooked in studies that visually investigate the substrate surface. The present results from the rubble method strongly resemble results from the Mexican Pacific (Carballo et al., 2008a), where the small Siphonodictyon crypticum, Clione vermifera, and Pione carpenteri were among the most com mon species, with the latter two being more closely taxonomically related to the four "rubble-species" from Orpheus Island than to "the usual suspects" in attached materials, such as C. orientalis (pers. comm. Y. Ise, Misaki Marine Biological Station). It thus appears that even though none of the observed species monopolizes a single substrate type, space competition, settlement strategies, or survival abilities generate patterns that confine one group more strongly to massive or at least fixed substrates, while the other group is especially successful in unattached substrates and may be able to survive occasional, temporal, or partial burial (see also Schonberg, 2001). Just as the species in the present study generated patterns across substrate types and reef zones, they will also generate patterns with environmental conditions, becoming useful bioindicators (e.g., Carballo et al., 1994; Holmes, 1997, 2000; Holmes et al., 2000). For this process we need to keep in mind that we investigate different communities when we regard only the substrate surface compared to when we also screen fragments, and when we concentrate on attached coral compared to unattached rubble. Apart from this and depending on the purpose of a given study, the rubble, quadrat, and intercept methods will all be useful tools for finding patterns and distributional trends.

Acknowledgments

The present study was fully funded by the Australian Institute of Marine Science, project "Monitoring of bioeroding sponges." The author wishes to acknowledge outstanding assistance by the staff at the Orpheus Island Research Station, especially by H. Burgess, and much appreciated help by various volunteers and collaborators present during the related field and laboratory work: C. Albrecht, C. Ansell, G. Ford, N. Lee, D. and R. Wisdom, M. Wisshak, A. Ullrich. This little study grew from earlier studies at the same location in which J.-C. Ortiz helped with the data analysis, and it is a direct result of ongoing discussions with K. Holmes who developed the rubble method. I further thank a large number of coral reef scientists for continuous interest and encouragement, and reviewers and editors for suggesting some changes to an earlier version of the manuscript. A. Hurman deserves thanks for quick help to extend conditions of the GBRMPA sample permit, and A. Thompson, A. Chen, and Y. Ise for sharing their observations on bioeroding sponges with me.

Literature Cited

Acker, K. L., and M. J. Risk. 1985. Substrate destruction and sediment production by the boring sponge Cliona caribbaea on Grand Cayman Island. J. Sediment. Petrol. 55: 705-711.

Australian Institute of Marine Science. 2011. Severe tropical cyclone Anthony and Yasi 21-Jan-2011-4-Feb-2011. [Online]. Available: http://data.aims.gov.au/cyclones/yasi-anthony.html 12012, August 9].

Baird, A. H., and P. A. Marshall. 2002. Mortality, growth and reproduction in scleractinian corals following bleaching on the Great Barrier Reef. Mar. Ecol. Prog. Ser. 237: 133-141.

Bakus, G. J., G. Nishiyama, E. Hajdu, H. Mehta, M. Mohammed, U. dos S. Pinheiro, S. A. Sohn, T. K. Pham, Z. bin Yasin, T. Shau-Hwai, A. Karam, and E. Hanan. 2007. A comparison of some population density sampling techniques for biodiversity, conservation, and environmental impact studies. Biodivers. Conserv. 16: 2445-2455.

Beenaerts, N., and E. Vanden Berghe. 2005. Comparative study of three transect methods to assess coral cover, richness and diversity. Western Indian Ocean J. Mar. Sci. 4: 29-37.

Berkelmans, R., and J. K. Oliver. 1999. Large-scale bleaching of corals on the Great Barrier Reef. Coral Reefs 18: 55-60.

Berkelmans, R., G. De'ath, S. Kinninmonth, and W. J. Skirving. 2004. A comparison of the 1998 and 2002 coral bleaching events on the Great Barrier Reef: spatial correlation, patterns, and predictions. Coral Reefs 23: 74-83.

Birkeland, C. 1984. Review of some coral reef assessment methods commonly used in South East Asia, the Pacific and elsewhere. Pp. 11-26 in Comparing Coral Reef Survey Methods. UNESCO Reports Mar. Sci. 21: 11-26.

Caballero, H., L. Busutil, Y. Garcia, and P. M. Alcolado. 2009. Variacion especial en comunidades de esponjas de la costa oriental de Bahfa de Cochinos, Cuba. Rev. Cienc. Mar. Costeras 1: 95-109.

Callahan, M. K. 2005. Distribution of clionid sponges in the Florida Keys National Sanctuary (FKNMS), 2001-2003. M.Sc. thesis, University of South Florida, Tampa, 79 pp. [Online]. Available: http://etd.fcla.edu/SF/SFE0001017/Callahan-Michael-Callahanthesis2005. pdf [2011, September 4].

Cantera, J. R., K. C. Orozco, E. Londofio-Cruz, and G. Toro-Farmer. 2003. Abundance and distribution patterns of infaunal associates and macroborers of the branched coral (Pocillopora damicornis) in Gorgona Island (Eastern Tropical Pacific). Bull. Mar. Set. 72: 207-219.

Carballo, J. L., J. E. Sanchez-Moyano, and J. C. Garci'a-Gomez. 1994. Taxonomic and ecological remarks on boring sponges (Clionidae) from the Straits of Gibraltar (southern Spain): tentative bioindicators? Zool. J. Linn, Soc. 112: 407-424.

Carballo, J. L., E. Bautista-Guerrero, and G. E. Leyte-Morales. 2008a. Boring sponges and the modelling of coral reefs in the east Pacific Ocean. Mar. Ecol. Prog. Ser. 356: 113-122.

Carballo, J. L., C. Vega, J. A. Cruz-Barraza, B. Yanez, H. Nava, E. Avila, and M. Wilson. 2008b. Short-and long-term patterns of sponge diversity on a rocky tropical coast: evidence of large-scale structuring factors. Mar. Ecol. 29: 216-236.

Cebrian, E. 2010. Grazing on coral reefs facilitates growth of the excavating sponge Cliona orienlalis (Clionaidae. Hadromerida). Mar. Ecol. 31: 533-538.

Cebrian, E., and M. J. Uriz. 2006. Grazing on fleshy seaweeds by sea urchins facilitates sponge Cliona viridis growth. Mar. Ecol. Prog. Ser. 323: 83-89.

Chavez-Fonnegra, A., and S. Zea. 2011. Coral colonization by the encrusting excavating Caribbean sponge Cliona delitrix. Mar. Ecol. 32: 162-173.

Chavez-Fonnegra, A., S. Zea, and M. L. Gomez. 2007. Abundance of the excavating sponge Cliona delitrix in relation to sewage discharge at San Andres Island, SW Caribbean, Colombia. Bol. Investig. Mar. Costeras 36: 63-78.

Chiappone, M., L. M. Rutten, S. L. Miller, and D. W. Swanson. 2007. Large-scale distributional patterns on the encrusting and excavating sponge Cliona delitrix Pang on Florida Keys coral substrates. Pp. 255-263 in Porifera Research: Biodiversity, Innovation and Sustainability, M. R. Custodio, E. Hajdu, G. Lobo-Hajdu, and G. Muricy, eds. Museu Nacional, Rio de Janeiro.

Colwell, R. K. 2005. EstimateS: Statistical estimation of species richness and shared species from samples. Ver. 7.5, User's Guide and Application. [Online]. Available: http://purl.oclc.org/estimates [2010, November 29].

Colwell, R. K., and J. A. Coddington. 1994. Estimating terrestrial biodiversity through extrapolation. Philos. Trans. R. Soc. B 345: 101-118.

Cortes, J., M. M. Murillo, H. M. Guzman, and J. Acufla. 1984. Perdida de zooxantelas y muerte de corales y otros organismos arrecifales en el Caribe y Pacifico de Costa Rica. Rev. Biol. Trop. 32: 227-231.

Dana, T. F. 1976. Reef-coral dispersion patterns and environmental variables on a Caribbean coral reef. Bull. Mar. Sci. 26: 1-13.

Dodge, R. E., A. Logan, and A. Antonius. 1982. Quantitative reef assessment studies in Bermuda: a comparison of methods and preliminary results. Bull. Mar. Sci. 32: 745-760.

Duckworth, A. R., and B. J. Peterson. 2013. Effects of seawater temperature and pH on the boring rates of the sponge Cliona celata in scallop shells. Mar. Biol. 160: 27-35.

Edinger, E. N. 1998. Effects of land-based pollution on Indonesian coral reefs: biodiversity, growth rates, bioerosion, and applications to the fossil record. Open access Ph.D. dissertation 2217, McMaster University, Hamilton, Ontario, 297 pp. [Online]. Available: http://hdl.handle.net/11375/6914 [2014, October 21].

English, S., C. Wilkinson, and V. Baker, (Eds). 1997. Survey Manual for Tropical Marine Resources, 2nd ed. Australian Institute of Marine Science, Townsville, Australia. Pp. 34-51.

Enochs, I. A., and D. P. Manzello. 2012. Response of cryptofaunal species richness and trophic potential to coral reef habitat degradation. Diversity 4: 94-104.

Fabricius, K. E., C. Langdon, S. Uthicke, C. Humphrey, S. Noonan, G. De'ath, R. Okazaki, N. Muehllehner, M. S. Glas, and J. M. Lough 2011. Losers and winners in coral reefs acclimatized to elevated carbon dioxide concentrations. Nat. Clim. Change 1: 165-169.

Fang, J. H. K., M. A. M. Athayde, C. H. L. Schonberg, D. I. Kline, O. Hoegh-Guldberg, and S. Dove. 2013. Sponge biomass and bioerosion rates increase under ocean warming and acidification. Glob. Change Biol. 19: 3581-3591.

Fang, J. H. K., C. H. L. Schonberg, M. A. M. Athayde, D. I. Kline, O. Hoegh-Guldberg, and S. Dove. 2014. Effects of ocean warming and acidification on the energy budget of an excavating sponge. Glob. Change Biol. 20: 1043-1054.

Field, A. 2009. Discovering Statistics Using SPSS, 3rd ed. Sage Publications, London. 822 pp.

Gilliam, D. S. 2010. Southeast Florida Coral Reef Evaluation and Monitoring Project: 2009 Year 7 Final Report. Florida DEP report RM085, Miami Beach, FL. 42 pp.

Gilliam, D. S., B. E. Ettinger, D. Fahy, E. Fahy, S. Gill, J. A. Monty, L. Shuman, B. Walker, and R. E. Dodge. 2004. Southeast Florida Coral Reef Evaluation and Monitoring Project: 2003 Year 1 Final Report. Florida DEP. Office of Coastal and Aquatic Managed Areas, Miami, FL. 14 pp.

Gilliam, D. S., B. E. Ettinger, D. Fahy, S. Gill, L. Klink, J. Monty, M. Philips, L. Shuman, N. Stephens, B. Walker, J. Walczak, and R. E. Dodge. 2006. Southeast Florida Coral Reef Evaluation and Monitoring Project: 2005 Year 3 Final Report. Florida DEP, Office of Coastal and Aquatic Managed Areas, Miami, FL. 25 pp.

Glynn, P. W. 1997. Bioerosion and coral reef growth: a dynamic balance. Pp. 68-95 in Life and Death of Coral Reefs, C. Birkeland, ed. Chapman and Hall, New York.

Highsmith, R. C. 1981. Coral bioerosion at Enewetak: agents and dynamics. Int. Rev. Hydrobiol. 66: 335-375.

Hill, J., and C. Wilkinson. 2004. Methods for Ecological Monitoring of Coral Reefs. A Resource for Managers, Ver. 1. Australian Institute of Marine Science, Townsville, Australia. 117 pp.

Hodgson, G., J. Hill, W. Kiene, L. Maun, J. Mihaly, J. Liebeler, C. Shuman, and R. Torres. 2006. Instruction Manual: A Guide to Coral Reef Monitoring. Reef Check Foundation, Los Angeles, CA. 86 pp.

Hoegh-Guldberg, O. 2009. Coral reef ecosystems and anthropogenic climate change. Reg. Environ. Change 11 (Suppl. 1): S215-S227.

Holmes, K. E. 1997. Eutrophication and its effect on bioeroding sponge communities. Pp. 1411-1415 in Proceedings of the 8th International Coral Reef Symposium, Vol. 2, H. S. Lessios and I. G. Macintyre, eds. Smithsonian Tropical Research Institute. Panama.

Holmes, K. E. 2000. Effects of eutrophication on bioeroding sponge communities with the description of new West Indian sponges, Cliona spp. (Porifera: Hadromerida: Clionidae). Invertebr. Biol. 119: 125-138.

Holmes, K. E., E. N. Edinger, Hariyadi, G. V. Limmon, and M. J. Risk. 2000. Bioerosion of live massive corals and branching coral rubble on Indonesian coral reefs. Mar. Pollut. Bull. 40: 606-617.

Hudson, J. H. 1977. Long-term bioerosion rates on a Florida reef: a new method. Pp. 491-497 in Proceedings of the 3rd International Coral Reef Symposium, Vol. 2, D. L. Taylor, ed. Rosenstiel School of Marine and Atmospheric Science, Miami. FL.

Kendall, M. S., L. J. Bauer, and C. F. G. Jeffrey. 2009. Influence of hard botom morphology on fish assemblages of the continental shelf of Georgia, southeastern USA. Bull. Mar. Sci. 84: 265-286.

Kennedy E. V., C. T. Perry, P. R. Halloran, M. Fine, J. P. Carricart-Ganivet, R. Iglesias-Prieto, A. Form, M. Wisshak, C. H. L. Schonberg, and P. J. Mumby. 2013. Avoiding coral structural collapse requires local and global action. Current Biol 23: 912-918.

Lang, J. C, K. W. Marks, P. A. Kramer, P. Richards Kramer, and R. N. Ginsburg. 2010. Atlantic and Gulf Rapid Reef Assessment protocols, Ver. 5.4 [Online]. Available: http://www.agrra.org/ [2012, August 10].

Leujak, W., and R. F. G. Ormond. 2007. Comparative accuracy and efficiency of six coral community survey methods. J. Exp. Mar. Biol. Ecol. 351: 168-187.

Lopez-Victoria, M., and S. Zea. 2005. Current trends of space occupation by encrusting excavating sponges on Colombian coral reefs. Mar. Ecol. 26: 33-41.

Loya, Y. 1978. Plotless and transect methods. Monogr. Oceanogr. Methodol. 5: 197-217.

MacGeachy, J. K. 1977. Factors controlling sponge boring in Barbados reef corals. Pp. 477-483 in Proceedings of the 3rd International Coral Reef Symposium, Vol. 2, D. L. Taylor, ed. Rosenstiel School of Marine and Atmospheric Science, Miami, FL.

Makowski, C, and P. Keyes. 2011. Using the Benthic Ecological Assessment for Marginal Reefs (BEAMR) method to quantify nearshore reef conditions in the Southeast Gulf of Mexico. J. Coastal Res. 27: 428-440.

Makowski, C, S. E. Prekel, M. J. Lybolt, and R. M. Baron. 2009. The Benthic Ecological Assessment for Marginal Reefs (BEAMR) Method. J. Coast. Res. 25: 515-522.

Mallela, J., and C. T. Perry. 2007. Calcium carbonate budgets for two coral reefs affected by different terrestrial runoff regimes, Rio Bueno, Jamaica. Coral Reefs 26: 129-145.

Marsh, M., R. H. Bradbury, and R. E. Reichelt. 1984. Determination of the physical parameters of coral distributions using line transect data. Coral Reefs 2: 175-180.

McKeever Targett, N., and G. P. Schmahl. 1984. Chemical Ecology and Distribubtion of Sponges in the Salt River Canyon, St. Croix, U.S.V.I. NOAA Tech. Memo. OAR NURP-1, Rockville, Maryland, 30 pp.

Nadon, M.-O., and G. Stirling. 2006. Field and simulation analyses of visual methods for sampling coral cover. Coral Reefs 25: 177-185.

Nava, H., and J. L. Carballo. 2008. Chemical and mechanical bioerosion of boring sponges from Mexican Pacific coral reefs. J. Exp. Biol. 211: 2827-2831.

Neumann, A. C. 1966. Observations on coastal erosion in Bermuda and measurements of the boring rate of the sponge, Cliona lampa. Limnol. Oceanogr. 11: 92-108.

Parravicini, V., C. Morri, G. Ciribilli, M. Montefalcone, G. Albertelli, and C. N. Bianchi. 2009. Size matters more than method: visual quadrats vs. photography in measuring human impact on Mediterranean rocky reef communities. Estuar. Coast. Shelf Sci. 81: 359-367.

Parravicini, V., F. Micheli, M. Montefalcone, E. Villa, C. Morri, and C. N. Bianchi. 2010. Rapid assessment of epibenthic communities: a comparison between two visual sampling techniques. J. Exp. Mar. Biol. Ecol. 395: 21-29.

Perry, C. T. 1998. Macroborers within coral framework at Discovery Bay, north Jamaica: species distribution and abundance, and effects on coral preservation. Coral Reefs 17: 277-287.

Reyes-Nivia, C, G. Diaz-Pulido, D. Kline, O. Hoegh-Guldberg, and S. Dove. 2013. Ocean acidification and warming scenarios increase microbioerosion of coral skeletons. Glob. Change Biol. 19: 1919-1929.

Risk, M. J., J. M. Heikoop, E. N. Edinger, and M. V. Erdmann. 2001. The assessment 'toolbox': community-based evaluation methods coupled with geochemical techniques to identify sources of stress. Bull. Mar. Sci. 69: 443-458.

Rose, C. S., and M. J. Risk. 1985. Increase in Cliona delitrix infestation of Montastrea cavernosa heads on an organically polluted portion of the Grand Cayman fringing reef. Mar. Ecol. 6: 345-362.

Riitzler, K. 1975. The role of burrowing sponges in bioerosion. Oecologia 19: 203-216.

Riitzler, K, 2002. Impact of crustose clionid sponges on Caribbean reef corals. Acta Geol. Hisp. 37: 61-72.

Ruzicka, R., K. Semon, M. Colella, V. Brinkhuis, J. Kidney, J. Morrison, K. Macaulay, J. W. Porter, M. Meyers, M. Christman, and J. Colee. 2009. Coral Reef Evaluation and Monitoring Project: 2009 Annual Report. Fish and Wildlife Research Institute and Florida Fish and Wildlife Conservation Commission, St. Petersburg, FL. 110 pp.

Scaps, P., and V. Denis. 2008. Can organisms associated with live scleractinian corals be used as indicators of coral reef status? Atoll Res. Bull. 566: 1-18.

Scheer, G. 1978. Application of phytosociologic methods. Monogr. Oceanogr. Methodol. 5: 175-169.

Schonberg, C. H. L. 2000. Bioeroding sponges common to the central Australian Great Barrier Reef: descriptions of three new species, two new records, and additions to two previously described species. Senckenb. Marit. 30: 161-221.

Schonberg, C. H. L. 2001. Small-scale distribution of Great Barrier Reef bioeroding sponges in shallow water. Ophelia 55: 39-54.

Schonberg, C. H. L., and J. Fromont. 2012. Sponge gardens of Ningaloo Reef (Carnarvon Shelf, Western Australia) are biodiversity hotspots. Hydrobiologia 687: 143-161. doi: 10.1007/s 10750-011-0863-5.

Schonberg, C. H. L., and K. E. Holmes. 2011. How to include bioeroding sponges into monitoring programs. 7th International Bioerosion Workshop, M. Carreiro-Silva and Sergio P. Avila, eds. 18-23 September 2011, Faial, Azores. [Online] Available: http://www.horta.uac.pt/intradop/index.php/pt/bioerosion-workshop [2014, November 11]. (Abstract).

Schonberg, C. H. L., and J.-C. Ortiz. 2009. Is sponge bioerosion increasing? Pp. 520-523 in Proceedings of the 11"' International Coral Reef Symposium, Vol. 1.7-11 July 2008, Ft. Lauderdale, FL. Schonberg C. H. L., M. Wisshak, A. Form, A. Freiwald, and K. E. Holmes. 2013. Impact of global change on sponge bioerosion: from experimental evidence to monitoring guidelines. Abstracts: 9"' World Sponge Conference, 4-8 November 2013, Fremantle, Australia. (Abstract).

Sheppard, C. R. C, M. Spalding, C. Bradshaw, and S. Wilson. 2002. Erosion vs. recovery of coral reefs after 1998 El Nino: Chagos Reefs, Indian Ocean. Ambio 311: 40-48.

Sudara, S., and A. Snidvongs. 1984. Problems in the study of coral communities. Pp. 70-73 in Comparing Coral Reef Survey Methods: Report of a Regional Unesco/UNEP Workshop, UNESCO Reports in. Marine Science, Vol. 21, Paris.

Sullivan, K. M., and M. Chiappone. 1992. A comparison of belt quadrat and species presence/absence sampling of stony coral (Scleractinia and Milleporina) and sponges for evaluating species patterning on patch reefs of the central Bahamas. Bull. Mar. Sci. 50: 464-488.

Tribollet, A., M. J. Atkinson, and C. Langdon. 2006. Effects of elevated pCO, on epilithic and endolithic metabolism of reef carbonates. Glob. Change Biol. 12: 2200-2208.

Tribollet, A., C. Godinot, M. Atkinson, and C. Langdon. 2009. Effects of elevated pC02 on dissolution of coral carbonates by microbial euendoliths. Glob. Biogeochem. Cycles 23: 1-7.

Van Soest, R. W. M., N. Boury-Esnault, J. N. A. Hooper, K. Riitzler, N. J. de Voogd, B. Alvarez de Glasby, E. Hajdu, A. B. Pisera, R. Manconi, C. H. L. Schonberg, et al. 2013. World Porifera Database [Online]. Available: http://www.marinespecies.org/porifera [2014, June 10].

Ward-Paige, C. A., M. J. Risk, O. A. Sherwood, and W. C. Jaap. 2005. Clionid sponge surveys on the Florida Reef Tract suggest land-based nutrient inputs. Mar. Pollut. Bull. 51: 570-579.

Weinberg, S. 1981. A comparison of coral reef survey methods. Bijdr. Dierk. 51: 199-218.

Wilkinson, C. R. 1993. Coral reefs of the world are facing widespread devastation: Can we prevent this through sustainable management practices? Pp. 11-21 in Proceedings of the 7"' International Coral Reef Symposium, Vol. 1, R. H. Richmond, ed. University of Guam Press, UOG Station, Guam..

Wisshak, M., C. H. L. Schonberg, A. Form, and A. Freiwald. 2012. Ocean acidification accelerates reef bioerosion. PLoS ONE 7: e45124. [Suppl. online]. Available: http://doi.pangaea.de/10.1594/PANGAEA. 831694 [2014, June 10J.

Wisshak, M., C. H. L. Schonberg, A. Form, and A. Freiwald. 2013. Effects of ocean acidification and global warming on reef bioerosion-lessons from a clionaid sponge. Aquat. Biol. 19: 111-127. [Suppl. online]. Available: http://doi.pangaea.de/l0.1594/PANGAEA.831660 [2014, June 10].

Wisshak, M., C. H. L. Schonberg, A. Form, and A. Freiwald. 2014. Sponge bioerosion accelerated by ocean acidification across species and latitudes? Helgol. Mar. Res. 68: 253-262. [Suppl. online]. Available: http://doi.pangaea.de/10.1594/PANGAEA.831657 [2014, June 10].

Wulff, J. 2001. Assessing and monitoring coral reef sponges: why and how? Bull. Mar. Sci. 69: 831-846.

Zundelevich, A., B. Lazar, and M. Ilan. 2007. Chemical versus mechanical bioerosion of coral reefs by boring sponges--lessons from Pione cf. vastifica. J. Exp. Biol. 210: 91-96.

Zvuloni, A., Y. Artzy-Randrup, L. Stone, R. van Woesik, and Y. Loya. 2008. Ecology size-frequency distributions: how to prevent and correct biases in spatial sampling. Limnol. Oceanogr. Methods 6: 144-152.

Appendix

Unpublished abstracts presented orally by author

[C]Christine H. L. Schonberg

Abstract for an oral presentation during the [11.sup.th] International Coral Reef Symposium in Fort Lauderdale, Florida, 2008:

Bioeroding sponges need to be monitored

Christine H. L. Schonberg', Katherine E. Holmes2

1Carl von Ossietzky University Oldenburg, Faculty V, Biology and Environmental Sciences, Department of Animal Biodiversity and Evolution, 26111 Oldenburg, Germany; 2Center for Biodiversity and Conservation, American Museum of Natural History, Central Park West at [79.sup.th] Street, New York, NY 10024-5192, USA

Bioeroding sponges are important in the balance of coral reef construction and destruction. Due to changes in environmental conditions, these sponges can become epidemic and have occasionally been recognised for their value as bioindicators. Recent studies revealed increases in bioeroding sponge abundances in the Caribbean and on the Australian Great Barrier Reef, which was largely explained with deteriorating conditions or disturbance events that led to an increased availability of suitable substrate for the settlement of bioeroders. The sponges appear to be hardier than corals, even if they are zooxanthellate. We propose to incorporate the most common species of bioeroding sponges into longterm monitoring projects. Species of the 'Cliona viridis' complex are thought to be most suitable for this, because they occur at all study sites, but species such as the Caribbean Cliona delitrix should also be considered. If the procedure is kept simple, replicate studies can be conducted at very different sites and in consecutive approaches, eventually leading to a sound database on the leading endolithic agents of warm water bioerosion.

Abstract for an oral presentation during the European Coral Reef Symposium in Wageningen, The Netherlands, 2010:

A(nother) call to include bioeroding sponges into monitoring programs

Christine H. L. Schonberg1, Katherine E. Holmes2

1Australian Institute of Marine Science, Oceans Institute at The University of Western Australia, 39 Fairway, Crawley, WA 6009, Australia; 2PNG Marine Program, Wildlife Conservation Society, PO Box 95, Kavieng, New Ireland Province, Papua New Guinea

Bioerosion is expected to rise with changing environments, and a number of anecdotal observations have been made, but quantitative information over time or in a format that can be compared between sites is extremely rare. Sponges are often the leading internal bioeroders and represent useful model organisms when studying reef bioerosion. During an oral presentation for the Florida ICRS we tried to encourage colleagues to include bioeroding sponges into ongoing monitoring programs. Additional information was distributed on CD, and the interest was extremely high. Afterwards however, feedback was poor, and we conducted some trials to make suitable suggestions how best to proceed. We would now like to encourage people to concentrate on common zooxanthellate, brown clionaids. We further propose the use of simple line-intercept transects. Any data will have to be normalised to available dead substrate per transect, because this varies significantly between zones, reefs and oceans and has an effect on distributions of bioeroding sponges.

Abstract for an oral presentation during the [12.sup.th] International Coral Reef Symposium in Cairns, Australia, 2012:

Changes in sponge bioerosion: from experimental evidence to monitoring guidelines

Christine H. L. Schonberg1, Max Wisshak2, Armin Form3, Andre Freiwald2, Katherine E. Holmes4

1Australian Institute of Marine Science, Crawley, WA 6009, Australia; 2Senckenberg am Meer, Abteilung fur Meeresforschung, 26382 Wilhelmshaven, Germany; 3GEOMAR, Marine Biogeochemistry, 24105 Kiel, Germany; 4Wildlife Conservation Society, Bronx, New York 10460, USA

Reef corals are seriously suffering from human and environmental impacts. However, to date we have no evidence that the bioeroders in reef ecosystems experience negative effects in the same way, whereas new experimental data for bioeroding sponges rather suggest the opposite. In an experimental setup at Orpheus Island (central Great Barrier Reef), the zooxanthellate bioeroding sponge Cliona orientalis was exposed to different combinations of lowered as well as elevated levels of carbon dioxide partial pressure and temperature. Our results prove a significant enforcement of the sponges' bioerosion capacity with increasing [p.sub.C][O.sub.2] (decreasing pH) inherent to ocean acidification, while temperature had comparatively little effect. This implies that in a high-[CO.sub.2] world, tropical reef ecosystems are facing the combined effect of weakened coral calcification and accelerated bioerosion, resulting in critical pressure on the balance between biogenic carbonate build-up and degradation. Despite the importance of the issue and an urgent demand for information for reef authorities and modellers of reef health, we lack data and particularly time series supporting impact scenarios over larger scales. Monitoring data are needed, and we recommend targeting only the most dominant species per area, which usually belong to the Cliona viridis species complex. These mostly occur in encrusting form and are easily spotted and quantified. Line intercept transects should be used, being the most simple, cheap, accurate and reliable of three tested methods. International monitoring programs in the league of Reef Check are highly encouraged to include this approach of recording bioeroding sponges in their existing protocols.

C. H. L. SCHONBERG (1*[paragraph])

(1) Australian Institute of Marine Science, Oceans Institute at The University of Western Australia, 39 Fairway, Crawley, WA 6009, Australia

Received 6 October 2013; accepted 26 June 2014.

(*) To whom correspondence should be addressed. E-mail: christine.schonberg@uwa.edu.au

([paragraph]) Current address: The University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia.

Table 1

Statistical efficiency compared for the three assessed monitoring
methods for occurrence of bioeroding sponges

                                Rubble       Quadrat     Intercept
Statistic                       method       method      method

Sample size (number of zones)     6             6           6
Degrees of freedom                5             5           5
Sums of squares                  11.900        10.967      76.767
Mean square                       2.380         2.193      15.953
Test statistic F                  2.164         3.463      10.183
P                                 0.092         0.017       0.000
Effect size                       0.311         0.419       0.680
Test power                        0.604         0.833       1.000

The same number of sample units were used for each approach (five
replicates for each of the six reef zones), and bioeroding sponge
biodiversities were compared across the different reef zones using
simple one-way analysis of variance. Conventionally acceptable values
in bold--that is, those confirming the suitability of the method and
a sufficient sample size when assuming the alternative hypothesis
that sponge biodiversities differ between zones.

Table 2

Comparison across three monitoring methods for bioeroding sponge
occurrences: by using rubble, quadrat, or line intercepts

Parameters compared          Rubble method

MINIMUM REQUIREMENTS
Field gear                   Dive or snorkel gear, or wading shoes,
                             slate, plastic bags,
                             catch bag

Laboratory requirements      Ratchet-action branch cutter, dissection
                             kit, slides and
                             cover slips, microscope
EFFORT(1)
Field [min]                  1: ca. 0.1
                             2: ca. 1
                             3: ca. 30
Laboratory [min]             1: ca. 10
                             2: ca. 60
                             3: ca. 360
Practical part of study [h]    =ca. 6:30
                             (with preliminary species identification)
PRESENT SAMPLE SIZE          No, error bars often > 1/3 of mean
SUFFICIENT?


RESULTS
The 3-6 most dominant        C. mucronata
species (by frequency,       Siphonodictyon sp. (probably more than
most common)(2)              1 sp.)
                             C. ensifera
                             C. cf. celata
                             C. caesia
                             P. cf. vastifica
Total species count          27 spp. +4, -5 (estimated) in 150 pieces
([+ or -] estimated          of rubble
error value, see
legend)(3)
Average species count per    0.7 spp. ([+ or -] 0.3) per piece
reef zone and study unit     of rubble
([+ or -] SD)
Average species count per  Ratio: 0.14 ([+ or -] 0.06) no. of spp./no.
study unit ([+ or -] SD)   of rubble fragment (5)

Average species count per  Ratio: 0.14 ([+ or -] 0.06) no. of
unit available substrate   occurrences/no. of
([+ or -] SD)              fragments (all fragments are 'suitable')

Total extent               104 occurrences in 150 pieces of rubble =
                           69.3%

Average extent of sponges  0.7 ([+ or -] 0.3) specimen counts per
per reef zone              piece of rubble
([+ or -] SD)
Average extent of sponges    NA
per study unit
([+ or -] SD)
Average extent of sponges    NA
per unit available
substrate ([+ or -] SD)

                             Quadrat method
MINIMUM REQUIREMENTS
field gear              Dive gear, quadrat, slate
                        If samples taken: hammer and chisel,
                        plastic bags, catch bag
Laboratory              Nothing, unless samples taken, then
requirements            as for rubble

 EFFORT'
 Field [min]            1: 2-3
                        2: 10-15
                        3: 60-90
Laboratory [mini        1-3: 0, unless samples taken, then
                        ca. 10 min per sample

Pr:tctical part of      = 1-1:30
the study [h]           (without laboratory species
                        identification)

PRESENT SAMPLE SIZE     Mostly good, but insufficient for
SUFFICIENT?             low-occurrence zones (sandy, relief)


RESULTS
The 3-6 most            C. orientalis
dominant                brown papillate Cliona sp.
                        (probably more than 1 sp.)
                        S. mucosum
                        C. minuscula

Total species count     15 spp. +6,-3 (estimated) in
([+ or -] estimated     150 quadrats = an area of 37 m2
error value, see
legend)(3)
Average species         1.3 spp. ([+ or -] 0.5) per quadrat
count per reef
zone and study
unit([+ or -] SD)
Average species         Ratio: 0.05 ([+ or -] 0.02) no.
count per study         of spp./sub-quadrat (25)
unit ([+ or -] SD)
Average species         Ratio: 0.09 ([+ or -] 0.02) no.
count per unit          of occurrences/subquadrats
available substrate     suitable substrate
([+ or -] SD)
Total extent            2180 of 3750 sub-quadrats
                        with sponges = 58.1%
Average extent          2.9 ([+ or -] 1.2) sub-quadrats
of sponges per          per quadrat with sponges = 11.6%
reef zone
([+ or -] SD)
Average extent          11.4% ([+ or -] 4.8) sub-quadrats
of sponges per          per quadrat with sponges
study unit
([+ or -] SD)
Average extent          20.8% ([+ or -] 10.3) of
of sponges per          suitable substrate under quadrats
unit available          contained sponges
substrate
([+ or -] SD)

                             Intercept method
MINIMUM REQUIREMENTS
field gear              Dive gear, tape measure, slate
                        If samples taken: hammer and chisel.
                        plastic bags, catch bag
Laboratory              Nothing, unless samples taken, then
requirements            as for rubble

 EFFORT'                1: 11.2 (calculated mean)
 Field [min]            2: ca. 360
                        3: NA
                        0 min, unless samples taken, then ca.
Laboratory [mini        10 min per sample

                        = 5-6
Pr:tctical part of      (without laboratory species
the study [h]           identification)
                        Mostly good, better than for
                        quadrats, but can be improved for
PRESENT SAMPLE SIZE     low-occurrence zones (sandy,
SUFFICIENT?             relief)

                        C. orientalis
RESULTS                 S. mucosum
The 3-6 most            CI. aurivillii brown papillate Cliona
dominant                sp. (probably more than 1 sp.)
                        C. minuscula

Total species count     16 spp. +3, -3 (estimated) on 30
([+ or -] estimated     line-intercept transects = crossing
error value, see        300 m
legend)(3)
Average species         4.4 spp. ([+ or -] 1.9) on a line-intercept
count per reef          transect
zone and study
unit([+ or -] SD)
Average species         Ratio: 0.004 ([+ or -] 0.002) no. of spp./
count per study         line-intercept transects (10 m)
unit ([+ or -] SD)
Average species         Ratio: 0.08 ([+ or -] 0.02) no. of spp./
count per unit          line-intercept transects with
available substrate     suitable substrate
([+ or -] SD)
Total extent            1935 cm of 3000 cm crossed
                        sponges = 64.5%
Average extent          64.5 cm ([+ or -] 25.21) of 1000 cm
of sponges per          crossed sponges = 6.5%
reef zone
([+ or -] SD)
Average extent          6.5 % ([+ or -] 2.5) of each m-section
of sponges per          line-intercept transects crossed
study unit              sponges
([+ or -] SD)
Average extent          11.1 % ([+ or -] 3.2) of suitable
of sponges per          substrate under line-intercept
unit available          transects contained sponges
substrate
([+ or -] SD)

1 Categories for time effort spent in the field or
laboratory: 1: smallest unit = one piece of rubble, one
0.5 x 0.5 m quadrat, or one 10-m line intercept transect;
2: effort per zone; 3: total effort per method. Standardized
values in bold.2 Abbreviations: C, Cliona; cf., similar to;
CI., Cliothosa; P., Pione; S., Siphonodictyon; sp(p)., species.
Authorities for the sponge species are as follows, for Tables 2
and 3 (van Soest et al, 2013):Cliona caesia (Schonberg, 2000),
Cliona cf. celata Grant, 1826-here sensu Schonberg (2000),
Cliona ensifera Sollas, 1878, Cliona minuscula Schonberg
et al., 2006, Cliona mucronata Sollas, 1878,Cliona orientalis
Thiele, 1900, Cliothosa aurivillii (Lindgren, 1897), Cliothosa
hancocki (Topsent, 1888), Pione cf. vastifica (Hancock, 1849)-here
sensu Schonberg (2001), and Siphonodictyonmucosum Bergquist,
1965. Further species mentioned in the Discussion: Clione
vermifera Hancock, 1867, Pione carpenteri (Hancock, 1867),
Siphonodictyon crypticum (Carballo et al., 2007).3 Total
species counts are based on analyses without spicule
preparations: Quadrat and intercept studies relate
largely to preliminary species identifications in
the field, while for all material foundin rubble
quick pluck preparations were made. Error values are
provided in brackets; for total species counts estimates
were generated by the author by considering possible
errors when eitherdistinguishing or lumping too many species.

Table 3

Comparison of accuracy of the quadrat and the intercept methods
between different investigators


Methods             Compared observations
Quadrat method
(conducted
on reef crest)      Total number of sub-quadrats with
                    bioeroding sponges
                    Proportion of bioeroding
                    sponges per suitable substrate
                    Total species count
                    Species recognized



                    Number of sub-quadrats with:
                    live cover
                    "suitable" substrate
Intercept           sand
method              mud
(conducted in       Total number of meter sections
pavement            with bioeroding sponges
zone)               Total extent of bioeroding sponges
                    along transect
                    Proportion of bioeroding
                    sponges per suitable substrate
                    Total species count
                    Species recognized






                    Extent of:
                    live cover
                    "suitable" substrate
                    sand
                    mud

                    Background of investigators
15+ years experience                1 + year experience
17 = 68% of investigated            11 = 44% of investigated
area                                area
90%                                 61%

4                                   4
C. orientalis ([alpha] and [beta])  C. orientalis ([alpha] and [beta])
C. minuscula                        "tiny black papillae"
Cl.. hancocki                       "orange Cliothosa"
orange bioeroding sponge            orange bioeroding sponge

6 = 24%                             7 = 28%
19 = 76%                            18 = 72%
0 = 0%                              0 = 0%
0 = 0%                              0 = 0%
8 of 10 m = 80%                     10 of 10 m = 100%

135 cm of line-intercept            155 cm of line-intercept
transects = 14%                     transects = 16%
17%                                 22%

4                                   4 or 5
S. mucosum                          S. mucosum

C. orientalis ([alpha] and [beta])  C. orientalis ([alpha] and [beta])
C. aff. orientalis                  "tiny brown papillae"
Cl. aurivillii                      "red Cliothosa" = Cl.
                                    aurivillii
                                    orange bioeroding sponge
                                    = Cl. aurivillii?
150 cm = 15%                        190 cm = 19%
810 cm = 81%                        690 cm = 69%
40 cm = 4%                          120 cm = 12%
0 cm = 0%                           0 cm = 0%

3 weeks experience                 mean [+ or -] SD per value
4.5 = 18% of                       43.3 [+ or -] 25.0% of
investigated area                  investigated area
27%                                59.3% [+ or -] 31.5

1                                  3 [+ or -] 1.7
C. orientalis [beta]




8 = 32%                            28.0% [+ or -] 4
17 = 68%                           72.0% [+ or -] 4
0 = 0%                              0.0% [+ or -] 0
0 = 0%                              0.0% [+ or -] 0
8 of 10 m = 80%                    86.7% [+ or -]11.5

130 cm of line-intercept           14.3% [+ or -] 1.5
transects = 13%
16%                                18.3% [+ or -] 3.2

3                                      4 [+ or -] 0.6
"black snorkel" = 5.
mucosum
C. orientalis ([alpha] and [beta])

"orange dots" = C.
aurivillii


155 cm = 15%                       16.3% [+ or -] 2.3
795 cm = 80%                       76.7% [+ or -] 6.7
50 cm = 5%                          7.0% [+ or -] 4.4
0 cm = 0%

Rows with standardized values in bold.
The species names used by each observer were given as
they appeared in the field notes to illustrate that
proper species identifications are not necessary at this stage.
Same species are arranged inthe same rows. Growth forms of
sponges: [alpha], papillate, entirely endolithic form with
only small patches of tissue exposed at substrate surface;
[beta], endolithic form with a coherent layer of tissue on
the substrate surface.Abbreviations: C, Cliona; Cl.,
Cliothosa; P., Pione; S., Siphonodictyon. Taxon authorities
for the sponge species are given in the legend for Table 2.

Table 4

Example of a recommended field sheet for data acquisition on
distributions of bioeroding sponges along line-intercept transects
employed at Orpheus Island using invented species names

Date: June-5-2011    Location: Little Pioneer Bay, Orpheus Island
                     Zone: pavement   Depth:  1.5 m
Observer: Christine    Notes: Four months after Cyclone Yasi.

Meter                   0-1    1-2   2-3     3-4     4-5
Live substrate            0    20      0      35      10
Suitable substrate       90    55    100      65      90
Sand                     10    25      0       0       0
Mud                       0     0      0       0       0
Encrusting brown                      20
Brown papillae          10             5   30 (25 ori/5 dk brn)
Black snorkel           10
Red papillae                     5    10
Meter                   5-6  6-7  7-8  8-9   9-10
Live substrate           25    0   10    5     45
Suitable substrate       70  100   90   95     55
Sand                      0    0    0    0
Mud                       0    0    0    0
Encrusting brown
Brown papillae                           5
Black snorkel            10
Red papillae              5         10         10

NOTE: Italics are examples of hand-written entries that can be filled
in on the standardized form.
COPYRIGHT 2015 University of Chicago Press
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2015 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Schonberg, C.H.L.
Publication:The Biological Bulletin
Geographic Code:8AUST
Date:Apr 1, 2015
Words:12524
Previous Article:Analysis of life-history traits in a sex-changing marine shrimp (Decapoda: Caridea: Rhynchocinetidae).
Next Article:Coelomic transport and clearance of durable foreign bodies by starfish (Asterias rubens).
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters