Portal de Periódicos da CAPES

Space-time variability in environmental thermal properties and snail thermoregulatory behaviour

2011; Wiley; Volume: 25; Issue: 5 Linguagem: Inglês

10.1111/j.1365-2435.2011.01859.x

1365-2435

Coraline Chapperon, Laurent Seuront,

Ocean Acidification Effects and Responses

Functional EcologyVolume 25, Issue 5 p. 1040-1050 Free Access Space–time variability in environmental thermal properties and snail thermoregulatory behaviour Coraline Chapperon, Corresponding Author Coraline Chapperon School of Biological Sciences, Flinders University, GPO Box 2100, Adelaide, SA 5001, Australia Correspondence author. E-mail: [email protected]Search for more papers by this authorLaurent Seuront, Laurent Seuront School of Biological Sciences, Flinders University, GPO Box 2100, Adelaide, SA 5001, Australia South Australian Research and Development Institute, Aquatic Sciences, West Beach, SA 5022, Australia Centre National de la Recherche Scientifique, Laboratoire d'Océanologie et de Géosciences, UMR WG 8187, Université des sciences et Technologies de Lille, Station Marine, Wimereux, FranceSearch for more papers by this author Coraline Chapperon, Corresponding Author Coraline Chapperon School of Biological Sciences, Flinders University, GPO Box 2100, Adelaide, SA 5001, Australia Correspondence author. E-mail: [email protected]Search for more papers by this authorLaurent Seuront, Laurent Seuront School of Biological Sciences, Flinders University, GPO Box 2100, Adelaide, SA 5001, Australia South Australian Research and Development Institute, Aquatic Sciences, West Beach, SA 5022, Australia Centre National de la Recherche Scientifique, Laboratoire d'Océanologie et de Géosciences, UMR WG 8187, Université des sciences et Technologies de Lille, Station Marine, Wimereux, FranceSearch for more papers by this author First published: 28 April 2011 https://doi.org/10.1111/j.1365-2435.2011.01859.xCitations: 56AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Summary 1. Behavioural adaptations of ectotherms to thermally heterogeneous environments are still overlooked in the literature despite the fact that organismal behaviour could enhance survival in the warming world. This is particularly critical in the intertidal where most ectotherms live at, or near to the upper limit of thermal tolerance. 2. This study investigated (i) the environmental factors determining the body temperatures of the intertidal gastropod Nerita atramentosa, (ii) the space–time variability in environmental and individual body temperatures and (iii) the potential variability in N. atramentosa thermoregulatory behaviours, i.e. microhabitat selection and aggregation. 3. Thermal imaging was used to assess the body temperatures of N. atramentosa and surrounding substrata over two seasons (autumn and summer), at two shore levels (low- vs. high-shore levels) within two habitats of different topographic complexity (rock platform and boulders) on the same rocky shore. 4. Snail body and substratum temperatures were significantly and positively correlated within each habitat at both seasons. Substratum temperature may thus be considered as a primary driver of body temperatures of organisms that attach to a substratum. Substratum temperature and other variables such as solar irradiance critically need to be integrated in climate-change models that use single climatic variables (e.g. air temperature) that are not necessarily correlated with individual body temperatures in nature. 5. The high space–time variability in both substratum and body temperatures reinforces the growing evidence that small spatial scale variations may surpass those observed at larger spatial scales. 6. Nerita atramentosa thermoregulatory behaviour under high thermal stress appeared to be habitat specific. 7. The small spatial scale heterogeneity in environmental and individual temperatures and in thermoregulatory behaviours has stressed the need to focus on body temperature patterns at the niche level and to integrate the organismal behaviour in climate-change models. Introduction Temperature determines a wide range of biological processes that are essential for animal life (Angilletta 2009). In particular, temperature has an effect on all physiological processes from the molecular to the organismal levels (Pörtner et al. 2006; Kingsolver 2009). Therefore, changes in temperatures affect organism fitness, performance and metabolism (Huey & Berrigan 2001; Dillon, Wang & Huey 2010), hence profoundly impact the structure, dynamic and functioning of populations and ecosystems (e.g. Morelissen & Harley 2007). However, the mean temperature and its variability have been predicted to increase in the warming climate (Planton et al. 2008). Nevertheless, levels of thermal tolerance and potential physiological and behavioural abilities of ectotherms to thermoregulate in the future climate, particularly in thermally heterogeneous environments, are still far from being understood. The mechanistic links between the body temperatures of ectotherms, which control local (Miller, Harley & Denny 2009) and global distribution patterns (Helmuth et al. 2002), and environmental variables are not as simple as previously anticipated (Helmuth 2009). More specifically, body temperatures of both terrestrial and marine ectotherms are determined by heat fluxes towards and from an organism (Gates 1980; Harley et al. 2009) which are subject to variations generated by the interaction between climatic heat sources (heat derived from ambient conditions, i.e. coarse-scale macroclimatic data such as air and water temperatures, Helmuth 2002; Vidal et al. 2010), non-climatic heat sources at the niche level (heat originated from the sun, i.e. solar irradiance, Díaz & Cabezas- Díaz 2004; Marshall, McQuaid & Williams 2010), and biotic factors (e.g. shell morphology, Harley et al. 2009; Polo-Cavia, López & Martín 2009). Therefore, the variability in a single factor may cause unexpected heterogeneity in body temperatures and leads to counter-intuitive patterns (e.g. Helmuth et al. 2002). Body temperatures of snails have also recently been demonstrated to be primarily controlled by non-climatic heat sources at the niche level (i.e. solar irradiance) instead of climatic heat sources (i.e. air and water temperatures; Marshall, McQuaid & Williams 2010). The space–time heterogeneity in organism and environment temperatures and the related physiological and behavioural adaptations require a better understanding in order to predict future species distribution ranges. This lack of knowledge is particularly critical in intertidal ecosystems that are thermally very heterogeneous over a range of scales, i.e. diel, tidal and seasonal variations within latitudinal and vertical clines, and microhabitats (Helmuth et al. 2006; Sinclair, Thompson & Seebacher 2006). Most intertidal invertebrates are close to the upper limit of their thermal tolerance (Somero 2002); hence, they are critically vulnerable to further changes in temperatures. Intertidal ectotherms have developed a range of physiological (Somero 2002) and behavioural adaptations (Munoz et al. 2005; Williams et al. 2005) to the natural thermal stress heterogeneity in order to maintain body temperatures within the species thermal tolerance window. Over the last decade, however, most attention has been given to the physiological responses of ectotherms and the development of new thermal sensors (e.g. biomimetic loggers, Shine & Kearney 2001; Schneider & Helmuth 2007) which have led to the establishment of new physiologically based mechanistic models, i.e. heat budget models. These models have successfully predicted individual body temperature patterns of sessile individuals such as limpets (Denny & Harley 2006) by integrating the morphology of organisms (e.g. shell shape). However, the potential buffering effect of behavioural thermoregulation of mobile ectotherms is still missing within climate change impact models (Kearney, Shine & Porter 2009). A few recent studies, although mostly terrestrial, have highlighted the importance of integrating the behaviour of mobile ectotherms (Díaz & Cabezas- Díaz 2004; Kearney, Shine & Porter 2009) that may increase the survival of mobile organism in a warming world (Huey & Tewksbury 2009). For example, locomotory abilities allow the exploitation of the ambient heterogeneity, hence the selection of thermally favourable niches (Huey et al. 2002). Because of their intrinsic complex topography, intertidal rocky shores abound with a variety of potential thermal refuges such as crevices, pits, rocks and pools that supply ectotherms with moisture and shade from solar radiations (Jackson 2010). Therefore, snails may actively select these microhabitats while travelling during the high tide to stabilize their body temperatures following emersion. Besides microhabitat selection behaviour, snails have displayed a range of thermoregulatory behaviours, e.g. mucous holdfast, raised posture and shell orientation (Garrity 1984; Munoz et al. 2005). Particularly, the formation of aggregates, commonly occurring among snails, is typically considered as a behavioural adaptation to desiccation and thermal stresses (e.g. Garrity 1984), although this is not always the case (e.g. Coleman 2010). In this context, the present study focused on the patterns of body temperatures and the thermoregulatory behaviours of the black snail Nerita atramentosa (Fig. 1a,c), a species particularly abundant on south Australian intertidal rocky shores at different spatial scales during cool and hot seasons. More specifically, the main goals of this work were (i) to explore the relationship between body temperatures and substratum surface temperatures at the individual scale, to analyse the space–time variability in (ii) substratum and body temperature patterns over two seasons (autumn vs. summer) in two topographically different habitats along the same rocky shore at small spatial scales (i.e. habitat, shore and niche levels), and in (iii) the potential thermoregulatory behaviours (i.e. aggregation and selection of thermally favourable niches). Figure 1Open in figure viewerPowerPoint Thermal images (b, d) and associated digital pictures (a, c) of Nerita atramentosa collected in summer on boulders, (a, b) at the high shore level within a crevice, and (c, d) at the low shore level on a flat rock. Average body temperature of snails and surrounding substratum surface within the crevice (a, b) were respectively 25·27 ± 0·19 °C (N = 7; mean ± standard error) and 23·4 ± 0·12 °C. On the flat rock, snails exhibited an average body temperature of 28·2 ± 0·40 °C (N = 2) and the surrounding substratum surface was 24·81 ± 0·09 °C. Materials and methods Studied Area and Species This work was conducted on a moderately exposed rocky shore located in Marino Rocks, South Australia (35°02′40S–138°30′30E), characterized by the presence of an alongshore gradient of substratum topographic complexity (i.e. rock platform to boulder field). This area supports great abundances of herbivorous gastropod species such as Bembicium sp., and Austrocochlea sp. and particularly the neritacean N. atramentosa (Reeve 1855). Here, we focused on N. atramentosa (Fig. 1a,c), which has specifically been chosen as (i) it is the dominant grazer and competitor for microalgae on Australian rocky shores (Underwood & Murphy 2008) and (ii) it is particularly subject to exposure to high temperatures because of its black pigmented shell and the related high absorption of solar radiation and retention during emersion (McMahon 1990). This mobile species is able to move between microhabitats during emersion (Chapperon & Seuront, pers. obs.). The study was undertaken at high- and low-shore levels (typically between the lower limits reached by the tidal flow at low tide in spring and neap tides, Seuront & Spilmont 2002) during low tides that occurred in the morning and mid-day times (i.e. between 10 am to 2 pm; Kuo & Sanford 2009) on four different days in both autumn 2009 (A, May 2009) and summer 2009–2010 (S, December 2009 and January 2010) on two topographically different habitats located 250 m apart. The field work started at the low tide time indicated by the Bureau of Meteorology of Australia and lasted for a minimum period of 2 h during the incoming tides. The first habitat was a rock platform (RP; 35°2′31·67″–138°30′35·37″) characterized by a flat, smooth rocky substratum with a few shallow pits and crevices at high shore level, and by the presence of pebbles and cobbles at low shore level. The second habitat was a boulder field (B; 35°2′38·04″–138°30′30·13″) mainly characterized by boulders (i.e. rock bigger than 256 mm) that provide a range of microhabitats such as pools, pits and crevices. Snail Density and Distribution Patterns In each habitat and shore level, 20 quadrats (25 × 25 cm) were haphazardly placed within a 51-m2 area. Digital pictures (digital camera Olympus J1 Tough-60; Olympus lmaging Corporation, Centre Valley, PA, USA) of each quadrat were taken to assess snail density and individual distribution at microscale. Individuals were classified either as being solitary or aggregated. An individual was considered aggregated when there was a direct shell contact with the shell of at least another conspecific. In addition, the microhabitat resting site of each snail was recorded. On both habitats, two microhabitats (flat rock and crevice) were defined in regards to substratum topographic complexity and exposure to solar radiations. Flat rock corresponds to a flat surface bereft of refuge to thermal stress, hence directly exposed to solar radiations. Crevice was defined as a depression wide and deep enough to fit at least one individual that may provide some protection from solar radiation, hence from thermal stress. On the rock platform, an additional microhabitat, under rock, was considered as a sheltered environment that provides entire protection from solar radiations. Snail Body Temperature and Substratum Surface Temperature Tissue temperatures of living animals have mainly been gathered using thermocouples or thermistors (Garrity 1984; Williams et al. 2005). In this study, we used infrared thermography as a non-contact and non-invasive method of temperature measurement (Helmuth 2002; Chapperon & Seuront 2011a). Thermal imaging has recently been shown to be an accurate and reliable tool to measure the mantle tissue temperature of N. atramentosa (Caddy-Retalic 2008). A preliminary approach (Caddy-Retalic 2008) was undertaken between N. atramentosa mantle temperatures measured with a thermistor probe and N. atramentosa dorsal shell temperatures assessed with a thermal imager Fluke Ti20 (Fluke Corporation, Everett, WA, USA). A significant positive correlation (Pearson correlation coefficient, R2 = 0·988, P < 0·001) was obtained between the mantle tissue temperatures and the dorsal shell temperatures (Caddy-Retalic 2008). The significant linear regression between mantle temperatures MTs and dorsal shell temperatures BTs was identified to be BT = 0·8875 × MT + 2·7044 (Caddy-Retalic 2008). Here, a thermal image of each individual observed in each quadrat was obtained using a thermal imaging camera Fluke Ti20 (Fluke Corporation). The thermal sensitivity of the thermal camera is ≤ 0·2 °C at 30 °C, and the temperature measurement accuracy is 2% or 2 °C, whichever is greater. Emissivity value (ε) was calibrated by applying a piece of masking tape characterized by a high emissivity (ɛ = 0·95) on 10 rocks and 10 snails. Specifically, when the temperature equilibrium was reached between tape and rock, and tape and snail, the emissivity value of the targets (i.e. rock and snail) was adjusted in order to obtain a temperature reading similar to that of the electrical tape of known emissivity. Mean emissivity values obtained for rock and body snail were respectively 0·954 ± 0·005 (; N = 10) and 0·946 ± 0·009, ranged from 0·94 to 0·99 and 0·91 to 0·98, and cannot be statistically distinguished (Wilcoxon–Mann–Whitney U-test, P < 0·05). Note that these emissivity values fall into the range of emissivity values employed for substrata (i.e. 0·95–1; Campbell & Norman 1998; Helmuth 1998; Denny & Harley 2006; Finke, Bozinovic & Navarrete 2009) and intertidal invertebrates (i.e. 0·96–1; Campbell & Norman 1998; Helmuth 1998; Denny & Harley 2006; Finke, Bozinovic & Navarrete 2009; Miller, Harley & Denny 2009). Emissivity value (ε) was consequently assumed to be fairly identical between organism and substratum and was hence set up at 0·95. Pictures of 307 and 203 individuals were collected on the boulder field and rock platform, respectively. Each individual was photographed once. Different individuals were used in the different habitats and shore levels. Images were subsequently analysed using InsideIR software version 4.0.1.10 (Fluke Corporation, 2006). For each thermal picture, snail body temperature (BT) and temperature of the surrounding substratum (ST) were assessed (Fig. 1b,d). A closed curve marker was drawn around each photographed shell in order to calculate the mean value of body (i.e. shell) temperature (BT). In addition, ST was averaged from four linear markers drawn on the substratum directly surrounding the individual shell. In particular, the distance between the linear markers and the shell was defined as approximately a quarter of the shell size measured on the picture. In addition, BTs and STs were measured during the incoming tides between 10 am and 2 pm which is the period that selects for heat tolerance (Somero 2010). It is therefore supposed that the temperatures measured corresponded to the maximal temperatures reached by both individuals and substrata, although this assumption may require further investigations. The mantle temperature MT was further calculated from the empirical relationship found between MT and BT (Caddy-Retalic 2008). Furthermore, a mantle-to-substratum temperature ratio (MSTratio) was defined to examine whether or not snail mantle temperature (MT) was closely related to that of the surrounding substratum (ST). The difference between mantle temperature and surrounding substratum (MSTdiff) was also calculated to quantify the potential difference in temperature between the snail mantle and its substratum. The distributions of the data MT, ST, MSTratio and MSTdiff at both seasons, in both habitats, in the different microhabitats and the distributions of snail density in both habitats were not all normally distributed (Kolmogorov–Smirnov test, P < 0·05). Nonparametric tests were consequently used throughout the manuscript. Spearman's correlation coefficients were used to assess the relationship between MT and ST at both seasons and in both habitats. All pairwise comparisons of MT, ST, MSTratio and MSTdiff between habitats, seasons, microhabitats, and aggregated vs. solitary individuals were conducted with the Mann–Whitney U-test. Comparisons of MT, ST, MSTratio and MSTdiff between the three microhabitats on the rock platform were performed with the Kruskall–Wallis test and subsequent nonparametric post hoc analyses (based on the Tukey test; Zar 2010) were performed to compare the different groups of measurements. All statistical analyses were carried out using pasw Statistics 18 (SPSS Inc., 2009, Chicago, IL, USA). Results Space–Time Dynamic of Nerita atramentosa Density and Distribution Patterns Boulder field The density of individuals was significantly higher in summer (Z = −4·212, P < 0·001; 64 ind m−2) than in autumn (48 ind m−2). The proportion of individuals in crevices was higher at the high shore level in both seasons (Fig. 2a). Instead, individuals rested more often on flat rocks at low shore levels (Fig. 2a). Overall, 58% and 29% of the total number of individuals were aggregated in summer and in autumn, respectively. Aggregation behaviour was more frequent at the high shore level in summer (Fig. 2a). In addition, N. atramentosa was found to aggregate more frequently within crevices than on flat rocks (Fig. 2a). Figure 2Open in figure viewerPowerPoint Nerita atramentosa microhabitat occupation and aggregation frequency in both seasons and shore levels on boulder field (top) and rock platform (bottom). A-LS, autumn low shore; A-HS, autumn high shore; S-LS, summer low shore; S-HS, summer high shore. The white, grey and black bars respectively correspond to flat rock, crevice and under rock microhabitats. The numbers at the bottom of each bar indicate the proportion of aggregation observed within each microhabitat. The italic numbers at the top of each bar represent the total number of individuals observed within each microhabitat. Rock platform No significant difference in density (Z = −1·121, P = 0·262) was observed between the two seasons. Overall, N. atramentosa was mainly observed on flat rocks (46%) rather than in crevices (35%) or under rocks (19%) over the two seasons. In autumn at low shore level, 96% of individuals were on flat rocks (Fig. 2b). In summer, however, organisms were mainly under rocks (Fig. 2b), particularly at low shore level (Fig. 2b; 63%). Most individuals were solitary in both seasons and at both shore levels (Fig. 2b) with the exception of the high shore level in autumn where 55% of individuals were aggregated. Moreover, individuals were found to be more aggregated within crevices and under rocks in both seasons (Fig. 2b). Space–Time Dynamics of Environment and Snail Body Thermal Properties Significant positive linear relationships were found between MT and ST in both habitats and seasons (Fig. 3a,b). More specifically, in autumn, MT, ST, MSTratio and MSTdiff were significantly higher on the rock platform than on boulders (ZMT = −12·183, ZST = −11·432, , , P < 0·001; Table 1, Fig. 3a,b). In summer, no significant differences in MT, ST, MSTratio and MSTdiff were found between the two habitats (ZMT = −0·005, P = 0·996; ZST = −1·619, P = 0·105; , P = 0·062; , P = 0·108; Table 1, Fig. 3a,b). Figure 3Open in figure viewerPowerPoint Nerita atramentosa individual mantle temperatures (MT) and substratum temperatures (ST) and air temperature (AT; only on c and d) on the boulder field (a, c) and the rock platform (b, d) in both seasons. (a, b) autumn and summer temperature values are respectively represented by white and grey circles (N = 510). The black line represents the first bissectrix, i.e. MT = ST. To improve the clarity of the graph, a value (47·93, 54·34) recorded in summer on the boulder field was removed. Positive and significant correlations were found between MT and ST in both habitats and at both seasons (autumn: ρB = 0·967, P < 0·001, n = 91; ρRP = 0·954, P < 0·001, n = 142; summer: ρB = 0·827, P < 0·001, n = 216; ρRP = 0·883, P < 0·001, n = 61). (c, d) Mean values of MT (black bars) and ST (grey bar) observed in both season and shore levels. A-HS: autumn high shore level, A-LS: autumn low shore level, S-HS: summer high shore level, S-LS. Errors bars are standard errors. Mean ATs were calculated from the data collected in Port Stanvac (i.e. closest meteorological station from Marino Rocks, ca. 10 km apart) during the studied low tides (source: Bureau of Meteorology of Australia). ***<0·001 (Mann–Whitney U-test). Table 1. Mean (standard deviation), minimum and maximum values in MT (mantle temperature), ST (substratum temperature), MSTratio (body-to-substratum temperature ratio) and MSTdiff (difference of temperature between the mantle of individuals and the surrounding substratum) on RP (rock platform) and B (boulders) in A (autumn) and S (summer) MT (°C) ST (°C) MSTratio MSTdiff (°C) RP-A (N = 142) Mean (SE) 22·13 (0·44) 19·64 (0·29) 1·12 (0·01) 2·49 (0·18) Min 15·84 15·44 0·89 −1·81 Max 36·19 32·34 1·44 9·31 B-A (N = 91) Mean (SE) 15·48 (0·13) 15·46 (0·13) 1·00 (0·93) 0·01 (0·04) Min 11·44 11·42 0·93 −1·08 Max 18·74 18·24 1·06 1·14 RP-S (N = 61) Mean (SE) 29·30 (0·43) 26·15 (0·32) 1·12 (0·01) 3·15 (0·22) Min 24·33 22·43 1·01 0·40 Max 37·55 34·29 1·26 7·05 B-S (N = 216) Mean (SE) 29·55 (0·32) 25·71 (0·22) 1·15 (0·01) 3·84 (0·17) Min 21·30 19·40 0·79 −5·98 Max 54·34 47·93 1·36 8·88 Boulder field In autumn, mantle and surrounding substratum temperatures were significantly higher at low shore level than at high shore level (ZMT = −6·849, ZST = −7·663, P < 0·001; Fig. 3c). However, no significant differences in MSTratio and MSTdiff were found between shore levels (, P = 0·398; , P = 0·407). In summer, MT, MSTratio and MSTdiff were significantly greater at low shore level than at high shore level (ZMT = −4·409, , , P < 0·001; Table 1, Fig. 3c). ST, however, was not significantly different between shore levels (ZST = −0·453, P = 0·651). Rock platform In autumn, MT, ST, MSTratio and MSTdiff were significantly greater at low shore level than at high shore level (ZMT = −6·665, ZST = −6·808, , , P < 0·001; Table 1, Fig. 3d). In summer, MT, ST, MSTratio and MSTdiff were significantly higher at high shore level than at low shore level (ZMT = −5·031, ZST = −4·332, , , P < 0·001; Table 1, Fig. 3d). Microhabitat Occupation and Thermal Properties Boulder field All results are summarized in Table 2. In autumn, MT, ST, MSTratio and MSTdiff were significantly higher on flat rocks than in crevices (Fig. 4a). At high shore level, no significant difference in MT, ST, MSTratio and MSTdiff was observed between microhabitats. At low shore level, MSTratio and MSTdiff were significantly greater on flat rocks than in crevices. MT and ST did not significantly differ between microhabitats. In summer, ST was significantly cooler on flat rocks than within crevices (Fig. 4c). However, MT did not significantly differ between microhabitats (Fig. 4c). MSTratio and MSTdiff values were significantly greater on flat rocks than within crevices. At high shore level, MT and ST were significantly warmer in crevices than on flat rocks. MSTratio and MSTdiff were not significantly different between microhabitats. At low shore level, no significant difference was found in MT, ST, MSTratio and MSTdiff between microhabitats. Table 2. Mann–Whitney U-test to investigate the variation in MT, ST, MSTratio and MSTdiff between the two microhabitats, i.e. crevice (C) and flat rock (FR) on the boulder field in autumn (A) and summer (S) in total (all), and specifically at the low and high shore levels (LSL and HSL, respectively). Results of the tests are indicated in the last column Boulders Z P A All MT −2·72 0·007 FR > C ST −2·85 0·004 MSTratio −2·94 0·003 MSTdiff −2·92 0·003 LSL MT −1·43 0·154 NS ST −0·92 0·358 MSTratio −3·30 C MSTdiff −3·36 <0·001 HSL MT 0·46 0·470 NS ST 0·58 0·581 MSTratio 0·51 0·524 MSTdiff 0·55 0·562 S All MT −1·03 0·303 NS ST −3·09 0·002 FR < C MSTratio −5·95 C MSTdiff −5·14 <0·001 LSL MT 0·59 0·600 NS ST 0·59 0·600 MSTratio 0·16 0·162 MSTdiff 0·28 0·286 HSL MT −3·33 <0·001 FR < C ST −4·21 <0·001 MSTratio −0·55 0·585 NS MSTdiff −0·17 0·864 NS, non-significant. Figure 4Open in figure viewerPowerPoint Nerita atramentosa individual mantle temperatures (MT) vs. substratum temperatures (ST) on the boulder field (a, c; N = 307) and the rock platform (b, d; N = 203) in autumn (a, b) and summer (c, d) in different microhabitats (white: flat rock, grey: crevice, black: under rocks). The black lines represent the first bissectrix, i.e. MT = ST. To improve the clarity of the graph, a value (47·93, 54·34) recorded in summer on the boulder field was removed. Rock platform All results are summarized in Table 3. In autumn, MT, ST, MSTratio and MSTdiff were significantly greater on flat rocks than under rocks than in crevices (Fig. 4b). At the high shore level, MT, ST, MSTratio and MSTdiff were significantly higher on flat rocks and under rocks than within crevices. No significant differences in MT, ST, MSTratio and MSTdiff were found between flat rocks and under rocks. At low shore level, N. atramentosa was only found on flat rocks with the exception of one observation under rock. In summer, MT, ST and MSTdiff were significantly warmer in crevices and on flat rocks than under rocks. No significant difference in MT, ST and MSTdiff was found between crevices and flat rocks. MSTratio was warmer on flat rocks than under rocks. No significant difference in MSTratio was found between crevices and under rocks, and flat rocks. At the high shore level, MT and ST did not significantly differ between microhabitats. MSTratio and MSTdiff were significantly higher on flat rocks than under rocks. At the opposite, at low shore level, MT and ST were significantly warmer on flat rocks than under rocks. MT within crevices was not significantly different from MT measured in the two others microhabitats. Similarly, ST in crevices was not significantly different from ST on flat rocks but was significantly higher than ST under rocks. MSTratio and MSTdiff were not significantly different between microhabitats. Table 3. Kruskall–Wallis test and subsequent multiple comparisons (post hoc based on the Tukey test) to investigate the variation in MT, ST, MSTratio and MSTdiff between the three microhabitats, i.e. crevice (C), flat rock (FR) and under rock (UR) on the rock platform in autumn (Au) and summer (Su) in total (all), and specifically at the low and high shore levels (LSL and HSL, respectively). Results of the tests are indicated in the last column Rock platform d.f. χ2 P Post hoc A All MT 2 31·74 UR > C ST 2 26·57 MSTratio 2 23·65 MSTdiff 2 25·94 HSL MT 2 41·07