As a consequence of the global warming, extreme climatic events (ECEs) have increased in frequency (Coumou and Rahmstorf, 2012). Marine heat waves (MHWs), prolonged discrete anomalously warm water events, occur when sea surface temperature (SST) exceeds for at least 5 days a climatological threshold (Hobday et al., 2016). Analyses of different sources of SST data have revealed a significant increase in MHWs duration, intensity and frequency (Oliver et al., 2018; Darmaraki et al., 2019a). MHWs have had different implications for marine ecosystems, as they have been associated with a variety of impacts, including alteration of ecosystem structure and function (e.g., Wernberg et al., 2016), shifts in species ranges (e.g., Wernberg et al., 2011), mass mortalities (e.g., Garrabou et al., 2009; Fordyce et al., 2019), local extinctions and economic impacts on seafood industries through declines in important fishery species and impacts on aquaculture (Madin et al., 2012; Hughes et al., 2017; Hyndes et al., 2017).

Seagrass meadows are among the planet's coastal ecosystems most effective for providing key ecological services including nursery grounds, nutrient cycling, sediment stabilization, trophic transfer to adjacent habitats and toward higher trophic levels (Hemminga and Duarte, 2000; Larkum et al., 2006) and coastal protection from erosion (Fonseca and Cahalan, 1992; Fonseca and Koehl, 2006). Moreover, seagrasses are highly efficient in sequestering carbon (C): together with saltmarshes and mangroves, they are responsible for capturing up to 70% of the organic C in the marine realm (Nelleman et al., 2009), making them one of the most effective Blue C sinks on the planet (Serrano et al., 2020). Stressors, such as sediments and nutrients inputs from terrestrial runoff, physical disturbance (e.g., trawling, anchoring), invasive species, diseases, aquaculture, overgrazing, algal blooms and global warming, have been shown to cause seagrass declines at scales ranging from square meters to hundreds of square kilometers (e.g., Munkes, 2005; Orth et al., 2006; Williams, 2007; Holmer et al., 2008; Waycott et al., 2009; Bockelmann et al., 2013; Giakoumi et al., 2015).

At the global scale, seagrasses are also influenced by climate change (CC) and understanding their response to the occurrence of ECEs, such as MHWs, represents a timely objective to determine the fate of the related ecosystems functions and services they provide. Indeed, relevant shoot mortalities have been correlated to MHWs occurrence (Marbà and Duarte, 2010; Shields et al., 2019; Smale et al., 2019; Strydom et al., 2020) and loss of the biodiversity linked to the seagrass mortality has been also documented (Nowicki et al., 2019). Responses to heat events of several seagrass species have been investigated at several biological levels (see Nguyen et al., 2021 for a review) and manipulative experiments have provided evidence that MHWs affect growth rate (i.e., Saha et al., 2020), promote leaf necrosis (i.e., Ontoria et al., 2019), reduce the photosynthetic capacity (Marín-Guirao et al., 2016), and alter fatty acid production by decreasing the proportion of polyunsaturated fatty acids (PUFA) and increasing the percentage of saturated fatty acids (SFA) (i.e., Beca-Carretero et al., 2018, 2020).

Posidonia oceanica (L.) Delile is a slow-growing seagrass, endemic to the Mediterranean, that is experiencing a widespread decline throughout the basin (Telesca et al., 2015). The regression of P. oceanica beds and the consequent expansion of alternative habitats (e.g., algal turfs or dead seagrass rhizomes) is particularly common in highly urbanized coastal areas, mostly due to eutrophication and increased sedimentation rates (Montefalcone, 2009; Tamburello et al., 2012). However, how meadows are responding to CC pressures is currently under investigation. Populations living in locally deteriorated conditions, such as high nutrient input, can be more fragile to the effects of MHWs (Pazzaglia et al., 2020). In general, phenological response of seagrasses to environmental changes is intrinsically related to genotype/population plasticity (i.e., the amplitude of the individual reaction norm), an evolutionary component which is affected by many external and internal interacting factors, involving local adaptation/acclimation and genetic/epigenetic diversity (Pazzaglia et al., 2021a). The temperature change projected in the Mediterranean Sea ranges between 0.81 and 3.71°C in the upper layer (0–150 m) by the end of the 21st century, depending on the greenhouse gas emissions scenario (Soto-Navarro et al., 2020). The rising SST has accelerated MHWs occurrence and effects on Mediterranean benthic communities have already been described (Garrabou et al., 2009; Marbà and Duarte, 2010; Rubio-Portillo et al., 2016), even though their effects on subtidal water conditions are anything but obvious: SST is undoubtedly a useful proxy for very shallow water temperature, but using loggers remains recommended to have more accurate and precise temperature estimates as deep water temperature can be scarcely predicted from SST (Ceccherelli et al., 2020). The most severe impacts of MHWs on benthic communities are expected in summer when the intensity of these extreme events can easily exceed the maximum tolerance limit of the species (e.g., those in 2003, 2012, 2015, Darmaraki et al., 2019b). However, contrasting results have been obtained on P. oceanica populations response to MHWs in relation to their distribution. A few mesocosm experiments have fostered the importance of local adaptations, highlighting the different levels of thermotolerance, defined as the ability to survive a normally lethal heat stress (Norris and Hightower, 2000), of shallow vs. deep P. oceanica plants (Marín-Guirao et al., 2016, 2017). The same has been shown for plants living in different thermal environments along latitudinal gradients: shallow and low-latitude plants better tolerate high temperature exposure (4°C above the summer average), in respect to shallow and high-latitude plants (physiological and molecular traits, Marín-Guirao et al., 2016, 2019). Data on fatty acid composition have corroborated these results providing evidence that populations living at warmest temperatures were more thermo-tolerant and exhibited a greater capacity to cope with heat events by adjusting their lipid composition (PUFA/SFA) faster (Beca-Carretero et al., 2018). A recent reciprocal translocation experiment does not support the higher vulnerability of cold-adapted populations to sea-water temperature increase (Bennett et al., 2022), highlighting the importance of further studies in this direction. In fact, despite these results, the resistance of P. oceanica to summer MHWs is still far to be fully understood, and more insights on their biochemical responses to MHWs are needed to address the potential consequences of an increasing occurrence of ECEs on seagrass resilience to climate change.

To provide further insights, we investigated the effects of (i) natural present-day summer MHWs and (ii) simulated longer-lasting and more intense summer MHWs on the morphology and biochemistry of P. oceanica plants with a different thermal history. To this end, we have conducted two parallel field experiments: (i) a mensurative experiment, where the effects of current MHWs were studied on plants at two sites with different thermal regimes, and (ii) a manipulative common garden experiment in the field, where the effects of simulated future MHWs were tested on transplanted plants from the two sites. In the mensurative study, plant responses were evaluated after one and then two MHWs in a row, to evaluate if effects could be cumulative. In the manipulative experiment the effect of acute MHWs, as those expected to occur in the coming decades, was assessed on transplanted plants to explore how the species will respond to future MHWs and whether the responses differ depending on the thermal origin of plants.

The aim of our experiments is to define early warning morphological and biochemical responses of P. oceanica to short-term intense heat events associated with MHWs. The results contribute to assess the morphological and biochemical changes involved in the P. oceanica performance to resist to summer MHWs so to predict the vulnerability and the adaptability of this species to heat waves that will affect the seagrass meadows over the coming decades.

Heat effects due to current and future summer MHWs were evidenced by both morphological and biochemical P. oceanica variables. Although in both experiments P. oceanica morphology was affected by MHWs, plants were able to overcome both the natural heat event and the simulated MHWs by the end of the study, as all the shoots were still alive. Indeed, all the plants showed only sublethal responses which followed a similar pattern in both experiments for most of the variables, revealing that current and future MHWs had similar effects, but different in magnitude (Supplementary Figure 2): in fact, the plants of the manipulative experiment that experienced higher intensity and longer lasting MHWs, resembling future scenarios (Darmaraki et al., 2019b), showed in general more intense responses, strongly depending on the origin of the plants. This similar pattern, even though different in magnitude, supports the hypothesis that dealing with cuttings for the manipulative experiment, rather than natural shoots (used for the mensurative one), did not introduce other variability: in fact, clonal integration among shoots, known for buffering against environmental changes (Ruocco et al., 2021), should be in P. oceanica about 15 cm (Marbà et al., 2002) and cuttings used in our study were always longer. Therefore, the consistent patterns between the current and future MHWs effects inform on how the seagrass will react to heat events in the future by decreasing the number of leaves, the maximum leaf length and lipid content and by increasing the leaf necrosis and the carbohydrate content. However, they also indicate that the local context will notably influence the performance of the plant.

The effects of current and future summer MHWs were clearly detectable, though results confirm the key role played by the local thermal condition (in terms of climatology) on plant performance. Here the effects of the current MHWs (mensurative experiment) are reported to depend on the site, and the effects of the intensity of MHWs (manipulative experiment) on the plant origin site. More specifically, during the current MHWs the factor site affected maximum leaf length, percentage of the total necrotic leaf length and all the biochemical variables, strongly supporting the hypothesis that P. oceanica has site-specific thermal acclimation responses. The importance of the site was evident also during future MHWs, when changes in the morphological attributes of plants from the warm origin showed, in general, a higher thermo resistance than plants from the cold one. The total number of leaves per shoot decreased homogeneously in the short-term (T2), while at the end of the experiment (T3) only plants of warm origin that experienced the MHW in the CT site (unaffected by the simulated MHW) recovered by producing new leaves. These results suggest that overcoming the effects of MHWs depended on the temperature intensity, but also on the origin of the shoots. Furthermore, maximum leaf length decreased sharply in both the short (T2) and the long-term (T3), but the duration of the heat event had a protrusive effect since it canceled the initial (T2) differences due to the origin and wave intensity. Similar sublethal responses of P. oceanica due to a prolonged induced heat stress were already found in mesocosm experiments (Marín-Guirao et al., 2018), suggesting that the declines reported in natural populations after anomalous warming events could be the consequences of prolonged heat-induced physiological alterations detectable in several morphological traits (Heckathorn et al., 2013).

In both experiments, plant origin also influenced the leaf necrosis, that in shoots from the warm site was much higher than in those from the cold site. This response was proportional in magnitude to the heat stress, as future MHWs triggered a higher leaf necrosis (but only in plants from the warm origin) than the current MHWs (Supplementary Figure 2). Necrosis is known to be a common higher plant response to abiotic stress (Van Doorn et al., 2011; Beca-Carretero et al., 2020), used as an indicator of irreversible heat-induced damage to the whole leaf tissue (since Kappen, 1981), and thus the results are not surprising. However, the fact that this phenomenon only regarded the warm site plants could not be generalized and keeps open the question whether these were the most stressed plants or, rather, those that most likely would have survived to the heat period. Further research is necessary to question to what extent the necrosis should be seen as a sign of leaf senescence and consequent death or activation of a process that could improve the tolerance of the plant by reallocating the resources. The diversion of resources from leaves to rhizomes have been suggested as a likely strategy in the species to conserve resource stocks and withstand biotic (e.g., hervibory; Ruocco et al., 2018) and abiotic (e.g., heat; Marín-Guirao et al., 2018) stress conditions. Unfortunately, as no data after T3 are available, whether the plants would have been destined to a site-dependent mortality remains an unanswered question also because seagrass shoot death is not necessarily anticipated by morphological changes (Ceccherelli et al., 2018) and to date heat resistance thresholds have not been clearly correlated to leaf necrosis. Though it is not possible to robustly infer about which between the two P. oceanica populations is the most vulnerable to the MHWs, the results suggest that they have different optimum temperature thresholds. This hypothesis is corroborated by the different reaction of the two seagrass populations to the same stimuli (common garden experiment), with the potentiality that the same condition can represent either a stimulation or an inhibition until plant death, depending on the local adaptation of individuals (Alexieva et al., 2003). Indeed, different populations of the same species can also be locally adapted (Kawecki and Ebert, 2004) and, since there is still ambiguity about genetic vs. plastic changes and adaptive vs. non-adaptive changes, tailored experiments employing appropriate inferential methods are needed to draw general conclusions on these issues (Merilä and Hendry, 2014; Reusch, 2014). Moreover, further experiments aimed at assessing the development and the survival rate of P. oceanica should be carried out to relate the seagrass performance to the effective resilience to the heat event.

A cumulative effect of two short summer MHWs in a row was not really appreciated neither in terms of shoots morphology nor of leaf biochemistry. Shoot morphology, in fact, was not affected by time in the mensurative experiment, while the biochemical composition varied only inconsistently. This result could be due to either a very rapid recovery of the before-MHW condition or to a complex compensatory metabolism of the plant (Reusch et al., 2005; Traboni et al., 2018). On the other hand, it is not possible to reject the existence of a thermal priming status that hardened the plants during the first MHW making them more resistant to the thermal stress of the second MHW, as recently demonstrated in a few seagrass species (Nguyen et al., 2020), including P. oceanica seedlings (Pazzaglia et al., 2022).

The biochemical composition of the leaves can be good predictor of seagrass stress (Ceccherelli et al., 2018) and the analysis of metabolites in marine plants is used to understand how marine plants respond to dynamic environmental conditions (Kumar et al., 2016). The results of this study also indicate that the content of the largest classes of organic compounds, in particular carbohydrates and lipids, varied significantly depending on the origin of the plant and the temperature. Both in the mensurative and manipulative experiments, carbohydrate content had a late response (T3), with plants having an increase with heat especially for the cold site (current MHW experiment) and cold origin (future MHW experiment), showing again the importance of the origin of the plant. Seagrasses are known to have a high pool of carbohydrates used to buffer the daily and seasonal fluctuations in light availability, so that the variability of carbohydrate content is typically reflected most at a seasonal scale (Alcoverro et al., 2001). The results suggest that the carbohydrate pool could also buffer fluctuations in temperature at least at a short-time scale, so that carbohydrates would deserve to be investigated as early warnings of seagrass resistance ability to summer MHWs. In fact, the warming-induced sugar starvation described in heat-sensitive, but not heat-tolerant P. oceanica plants after a simulated MHWs, further support the importance of carbohydrate metabolism and signaling in the heat-stress response of seagrasses, and hence in their potential as early warning signals (Marín-Guirao et al., 2019). Conversely, recent studies showed how different categories of fatty acids vary when temperature changes: the amount of saturated fatty acid increases and the number of unsaturated fatty acids decreases, maintaining membrane fluidity in response to thermal stress (Beca-Carretero et al., 2018, 2021). Here, a prolonged heat event caused a decrease in the total amount of lipids (consistently in the current and future MHW experiments) that was proportional to the heat intensity (Supplementary Figure 2), thus suggesting that lipids also can be good predictors of heat stress, even as a whole category. Conversely, proteins are likely not a reliable warning of heat stress, because although the interactive effect between temperature and origin was found in both times of the future MHWs experiment (but not in the current ones) a unique direction of the change could not be identified. Similar observations on the protein content were recently found in other seagrasses (Beca-Carretero et al., 2021) and, likely for the intrinsic variability of protein contents, a much higher replication would be needed to define the scale of reliability of such variable. Overall, the change of the biochemical content in primary producers may have an important ecological role (Hernán et al., 2016) and needs to be further investigated also by identifying how specific groups of these biochemical contents may change in relation of heat events. In fact, the change of the components may not only anticipate the changes in the morphological variables after or during a stress, but it can also predict how the trophic interactions will change when nutritional values vary according to the heat, since they can affect the nutritional preferences of seagrass herbivores (Carmen et al., 2012; Hernán et al., 2016) and, consequently, the impact the trophic interactions between producers and consumers.

Overall, morphological changes here induced by MHWs were consistent with expectations formulated on other seagrass species, such as that increased temperature triggers higher necrotic leaf portions (Beca-Carretero et al., 2020) and reduces leaf formation rates (Olsen et al., 2012). The novelty of these findings stands on the high discrepancy between P. oceanica shoots performance of different thermal origin (including the unexpected higher necrosis in warm-origin plants) highlighting the intimate reaction of plants to current and future summer MHWs at least in the short-term after. P. oceanica was suggested to have low thermal optimum and lethal limits compared to other Mediterranean seagrasses (Savva et al., 2018), such as Cymodocea nodosa (Marín-Guirao et al., 2016). In general, seagrass meadow declines are likely linked to the intensity and duration of the MHWs (Marbà and Duarte, 2010; Arias-Ortiz et al., 2018; Smale et al., 2019; Strydom et al., 2020) and these results provided evidence that the resistance to the same event will depend much on the context the seagrass lives or comes from. Even if further studies on the resistance to MHWs are necessary to define the thermal tolerance of P. oceanica and distinguish between acclimation and adaptation process, results are enough to suggest considering the local thermal conditions in restoration efforts (Pazzaglia et al., 2021b), since they can influence the performance of the transplanted plants and thus affect the success of the whole actions.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

PS, GC, GP, and LM-G conceived the ideas and designed methodology. PS, APa, and FP collected the data for the morphological analyses. APu and SS collected the data for the biochemical analysis. PS and FP analyzed the data. PS and GC led the writing of the manuscript. All authors have contributed critically to the drafts, gave final approval for publication, and agreed to be accountable for all aspects of the work.

The study has been funded by Italian Ministry of Education and Research PRIN 2017 (MHHWBN) Marine Habitats restoration in a climate change-impaired Mediterranean Sea (MAHRES) and by PON—National Operational Programme—Research and Innovation 2014–2020—PhDs and research contracts on innovation-related topics.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.