Response of large benthic foraminifera to climate and local changes: Implications for future carbonate production

Large benthic foraminifera are major carbonate components in tropical carbonate platforms, important carbonate producers, stratigraphic tools and powerful bioindicators (proxies) of environmental change. The application of large benthic foraminifera in tropical coral reef environments has gained considerable momentum in recent years. These modern ecological assessments are often carried out by micropalaeontologists or ecologists with expertise in the identification of foraminifera. However, large benthic foraminifera have been under‐represented in favour of macro reef‐builders, for example, corals and calcareous algae. Large benthic foraminifera contribute about 5% to modern reef‐scale carbonate sediment production. Their substantial size and abundance are reflected by their symbiotic association with the living algae inside their tests. When the foraminiferal holobiont (the combination between the large benthic foraminifera host and the microalgal photosymbiont) dies, the remaining calcareous test renourishes sediment supply, which maintains and stabilizes shorelines and low‐lying islands. Geological records reveal episodes (i.e. late Palaeocene and early Eocene epochs) of prolific carbonate production in warmer oceans than today, and in the absence of corals. This begs for deeper consideration of how large benthic foraminifera will respond under future climatic scenarios of higher atmospheric carbon dioxide (pCO2) and to warmer oceans. In addition, studies highlighting the complex evolutionary associations between large benthic foraminifera hosts and their algal photosymbionts, as well as to associated habitats, suggest the potential for increased tolerance to a wide range of conditions. However, the full range of environments where large benthic foraminifera currently dwell is not well‐understood in terms of present and future carbonate production, and impact of stressors. The evidence for acclimatization, at least by a few species of well‐studied large benthic foraminifera, under intensifying climate change and within degrading reef ecosystems, is a prelude to future host–symbiont resilience under different climatic regimes and habitats than today. This review also highlights knowledge gaps in current understanding of large benthic foraminifera as prolific calcium carbonate producers across shallow carbonate shelf and slope environments under changing ocean conditions.


INTRODUCTION
Symbiont-bearing large benthic foraminifera today and in the past Globally, foraminifera play significant functional roles, from meiobenthic nutrient cycling (Enge et al., 2016;Wukovits et al., 2018) to global geochemical cycles (Hallock, 1981;Langer et al., 1997;Langer, 2008). Their broad geographic, taxonomic and morphological diversity is evidence of their ability to adapt and prevail in almost all marine ecosystems (Förderer et al., 2018). This aspect alone allows them to be used in interpretation and reconstruction of modern and palaeoenvironments (Duchemin et al., 2005;Drinia, 2009;Reymond et al., 2011bReymond et al., , 2013bNarayan et al., 2015;de Jesus et al., 2020), thus lending themselves as ideal vessels of geochemical proxies (Curry & Matthews, 1981;Rae, 2018) and bioindicators of coastal pollution (Frontalini & Coccioni, 2011;Pati & Patra, 2012;Suokhrie et al., 2017;Ben-Eliahu et al., 2020). Specifically, the relative proportion of functional groups [large benthic foraminifera (LBF), small heterotrophic and opportunistic] of foraminifera has been established as a powerful bioindicator tool (for example, the FORAM Index) for monitoring the health of coral reefs in terms of nutrient loading (Hallock, 2000a(Hallock, , 2012Prazeres et al., 2020a). Similarly, the ecophenotypic characteristics (Boltovskoy et al., 1991), i.e. in LBF Amphistegina spp., has been shown to alter its test curvature to optimise light exposure for photosymbionts and thereby indicate palaeowater depths (Mateu-Vicens et al., 2009). Overall, foraminifera might appear microscopic on an individual scale compared to other marine calcifies, but their prolific abundance and significance is difficult to overlook.
These highly specialized assemblages were apparently poorly adapted to survive a shift to an 'icehouse' world of cooler, high-nutrient surface waters, upwelling and increased bottom circulation, and falling dissolved CO 2 (Ca + and Ca/ Mg) levels, which resulted in a few genera surviving into the early-middle Miocene (Adams, 1983;Hallock, 1985;Hallock & Glenn, 1986;Renema & Trolestra, 2001;Renema et al., 2008) to the present (for example, Cycloclypeus). The shift from LBF to coral-dominated carbonates occurred around the Oligo-Miocene boundary (Wilson, 2008). The early Miocene, for example, of SE Asia, saw the increased speciation of scleractinian corals and the build-up of reef framework carbonates consisting of corals and coralline algae, accompanied by benthic foraminifera (including LBF), echinoderms, molluscs and Halimeda, typical, for example, for modern Indo-West Pacific reefs (Wilson & Rosen, 1998;Wilson, 2008Wilson, , 2012. Fewer and smaller sized LBF lineages (Hallock, 1981) continued to form along carbonate platforms and ramps in parts of the world (for example, Balearic Islands, Western Mediterranean) where warm, oligomesotrophic conditions prevailed into the Upper Miocene (Pomar et al., 2004;Mateu-Vicens et al., 2008. However, they never became the dominant producer in large-scale carbonate systems again. During the Cenozoic period, global CO 2 levels and regional oceanographic change provide an appropriate analogue for the near-future climate change projections (Wilson, 2008). From these studies, the authors expect that certain modern LBF lineages that show signs of adaptation to differing conditions, for example, persistence in thermally polluted coastal waters of the Eastern Mediterranean and Gulf of Aqaba , deep mesophotic (Renema, 2006a(Renema, , 2018(Renema, , 2019, or in turbid inner shelf (inshore) fringing patch reefs (Renema & Trolestra, 2001;Renema, 2006aRenema, , 2018Narayan & Pandolfi, 2010), may be able to withstand ongoing climatic upheavals in these potential climate refugia (Keppel et al., 2012;Renema, 2019). However, in the context of current rapid increases in anthropogenic CO 2 levels, combined effects of stressors and increasing local impacts, it is speculative whether LBF could once again dominate carbonate production (Lee & Hallock, 1987;Hallock, 2005). The multiple dimensions of the ecological niche that shape LBF species distribution patterns and adaptation, including the diversity and stability of symbiotic partnerships with different algal groups, microbiome characteristics and genetic differentiation influencing high species dispersal, needs further consideration and research (Lee, 2004;Webster et al., 2016;Prazeres et al., 2017aPrazeres et al., , 2020b. Thus, the study of past and present LBF populations offers opportunities for integration and linkages between ecology and evolution or eco-evolutionary ('eco-evo') dynamics (Pelletier et al., 2009) across multiple scales.
This review focuses on LBF, as a vital tropical, shallow-water CaCO 3 -producing group (Zohary et al., 1980;Hallock, 1981;Tudhope & Scoffin, 1988;Langer et al., 1997;Hohenegger, 2002;Langer, 2008), that has seen underwhelming application and representation in global assessments of reef CaCO 3 budgets. While overwhelming 'coral-centric' approaches have been suggested (Vroom, 2011), it is important to note that corals and LBF occupy overlapping, but also different reef-associated environments, with reef framework production by corals mainly occurring in the reef flat, crest to reef slope (upper photic). Large benthic foraminifera production, on the other hand, is species-specific and generally occurs across broad carbonate environments (Hallock, 1984;Yamanouchi, 1998;Hohenegger et al., 1999;Fujita et al., 2009;Renema, 2018) (Figs 1 and 2). In light of increasing global stressors, including: (i) terrestrial inputs and eutrophication; (ii) thermal stress; (iii) ocean acidification (OA); and (iv) sea-level rise (SLR) (Fig. 2), there is strong interest in understanding reef carbonate dynamics, including the contribution and rates of CaCO 3 production by different producers to the global carbonate budget and how this budget may be altered (Lange et al., 2020). Based on these stressors, the potential role of LBF in future reef carbonate production and stabilization, other possible functions, and the implications for further research directions on these ubiquitous reef CaCO 3 producers are discussed.

DISCUSSION
Tropical carbonate factories and the contribution of LBF to carbonate sediments and production Tropical coral reefs are one of the most biologically diverse and productive (Odum & Odum, 1955;Connell, 1978;Reaka-Kudla, 1997) carbonate environments that form complex, three-dimensional, wave-resistant structures, typically dominated by hermatypic, scleractinian corals. As products of long-term accretionary processes, they are facilitated by several other calcium carbonate (CaCO 3 ) producing, functional groups. These carbonate 'engineers' are the skeletons of corals, coralline red algae, calcifying green algae (Halimeda), large benthic foraminifera (LBF) and other calcifiers, that can create, maintain or significantly modify habitats. In some environments they can be regarded as ecosystem engineers (Wilby, 2002). They contribute to reef sediment production, accretion, stabilization and maintenance (Wilson, 2008;Perry et al., 2011;James & Jones, 2015;Janßsen et al., 2017). Carbonate 'engineers' contribute to the CaCO 3 budget through a range of biological, physical and chemically-mediated production and erosion processes (Perry et al., 2008(Perry et al., , 2015Montaggioni, 2009;Lange et al., 2020).
After corals (primary) and coralline red algae (secondary), LBF have been considered as the third most prominent CaCO 3 -producer and carbonate 'engineer' in coral reefs worldwide (Wells, 1957;Hallock, 1981;Harney et al., 1999). Carbonate production rates by LBF are known to exceed that seen by coralline algae in the Great Barrier Reef (GBR) (Tudhope & Scoffin,   Fig. 2. A conceptual diagram of the current state of knowledge of large benthic foraminifera (LBF) (A) Potential adaptative strategies of LBF to future conditions are highlighted for the different reef zones that LBF can migrate to in response to changing physico-chemical conditions. Pink-coloured arrows and text indicate possible climateinduced habitat shifts. Blue-green arrows and text highlight the need for ecological-evolutionary (eco-evo) feedbacks (for example, the ability for LBF to modulate their evolutionary trajectory through ecological interactions), that need to be better understood. This study highlights the various methods to study these acclimatization mechanisms including: acclimatization potential through interspecies-interactions and symbiont shuffling/switching. (B) Current LBF contributions to reefs. The blue, positive/negative symbols indicate the importance of LBF for carbonate production in these particular reef zones; pink symbols indicate the known biodiversity of associated reef zones, and black symbols indicate the number of studies on LBF. (C) The impact of stressors on carbonate production. Primary stressors include: sea-level rise (SLR); ocean acidification (OA); ocean acidification variability (OAV); ocean warming (OW); ocean warming variability (OWV); eutrophication (EU); increased turbidity (TU); and increased light (LI). Blue arrows indicate the expected change in carbonate production in each reef zone, in response to the influence of these primary stressors. 1988). Although LBF themselves do not construct reef frameworks, accumulations of their tests support the stability and maintenance of carbonate habitats and structures (Sakai & Nishihira, 1981) by: filling in voids in the framework matrix; encrusting substrates; in large accumulations, baffling loose sediment; contributing carbonate detrital material to the formation of reefassociated landforms (Stearn et al., 1977;Ford & Kench, 2012;Dawson et al., 2014); as well as consolidating sediments into reef or beach rock (Jell et al., 1965;Woodroffe & Morrison, 2001). Large, deep-dwelling taxa contribute to CaCO 3 accumulations on the outer carbonate shelf to seafloor, creating habitat, and often outweighing accumulation in shallow-water environments (Renema, 2018).
While the role of benthic foraminifera as a major component of carbonate sands is wellestablished (Hallock, 1981;Langer et al., 1997;Langer, 2008;Fujita et al., 2009;Doo et al., 2012Doo et al., , 2014Doo et al., , 2017Dawson et al., 2014), their overall contribution to reef CaCO 3 budgets relative to other carbonate components, remains largely unquantified (Harney & Fletcher, 2003;Vroom, 2011;Lange et al., 2020), with the exception of the work of Reijmer et al. (2012) and Reymond et al. (2014), or it is considered to be relatively minor with respect to other macrocalcifiers (Andersson & Gledhill, 2013). In a recent review of carbonate budget estimates, as indicators of functional reef health, a call was made to incorporate the CaCO 3 contributions of different (non-framework building) reef communities into the global reef budget (Lange et al., 2020). Yet, following extensive review of the literature on reef carbonate budgets, the authors made no mention of LBF contributions. In the light of increasing reef-scale research, LBF research is likely deprioritized by reef scientists for various reasons.
Current estimates suggest that LBF contribute approximately 3.9 to 5.4% to the global carbonate reef budget (Langer, 2008;Doo et al., 2017), and generate an estimated 34 million tonnes of CaCO 3 annually (Langer, 2008). The estimated carbonate production by benthic foraminifera, including the contribution from small heterotrophic and opportunistic (non-symbiotic) taxa, ranges between 150 g and 2800 g CaCO 3 m −2 yr −1 , with LBF contributing amounts in excess of 1000 g CaCO 3 m −2 yr −1 (Hallock, 1981;Langer et al., 1997;Langer, 2008). It is important to note that these budget estimates are very likely skewed by the over-representation of shallow-water reef habitats.
Calcium carbonate production by LBF has not been widely documented in reef budget estimates (Table 1). Complexities and inconsistencies in the methodologies for determining carbonate production rates, including standardized estimation methods, present a considerable problem. A common method of quantifying production rates is the census-based approach, which assesses population densities of individual species, the primary factor controlling production rates Fujita et al., 2016). Census-based estimates have been based Table 1. Reported estimates for: standing crop (individuals/m −2 ) -(unbolded); biomass (g m −2 ) -(blue); and carbonate production rates (g m −2 yr −1 ) -(bolded) for the known large benthic foraminifera (LBF) species from reef habitats. In studies where standing crop and turnover rates (τ) are specified (i.e. based on Hallock, 1981), the annual production rate was divided by τ to obtain the biomass. In reference to the test chemistry, high magnesium calcite (HMC) tests are considered to contain >4 mol % MgCO 3 , while low magnesium calcite (LMC) <4 mol % MgCO 3 . Biomass estimates were provided for studies that reported estimated production (by dividing reported τ by the estimated production rate). on: (i) life history tables, i.e. average population size, size-specific mortality, fecundity and average species life span (Muller, 1974;Zohary et al., 1980;Hosono et al., 2013;Fujita et al., 2016); (ii) growth increment rates, i.e. daily growth rates of cultured individuals (g 3 CaCO 3 m −2 yr −1 ) (Hosono et al., 2013;Fujita et al., 2016); (iii) monthly or biannual biomass estimates (g m −2 ) (Fujita & Fujimura, 2008;Fujita et al., 2016;Doo et al., 2017); and (iv) standing crop estimates, i.e. test density (# per m 2 ) and turnover rates (Hallock, 1981). Of these methods, the last approach listed (iv), also known as the 'simple method', has been the most common (Table 1) method used (Hallock, 1981;Doo et al., 2012;Dawson et al., 2014). In the simple method (iv), carbonate production rates are discussed in units of g CaCO 3 m −2 yr −1 , and carbonate production by a population is divided by the average standing crop (∑ i N i m i ) to obtain the population turnover rate [T ¼ P=∑ i N i m i , where N is the average number at size 'i' and m i is the mass (g) at size i] (Hallock, 1981). Carbonate production rates of species are estimated using: P = Nτm, where N is density (number of individuals m 2 ), τ is the turnover rate (yr −1 ) and m is the mass (g) of a typical LBF individual (Hallock, 1981). A total standing crop of greater than 10 6 tests m −2 has been the benchmark for high productivity, whereas less than 10 4 tests m −2 is considered very low (Murray, 1967;Zohary et al., 1980;Hallock, 1981). The simplicity of this method allows for quick production estimates, assuming that conditions are typically stable, and the effects of seasonality (i.e. length of day and amount of light) and physiochemical conditions are limited (Zohary et al., 1980), as found in tropical coral reefs. Also, it is possible to estimate long-term carbonate production from time-averaged sedimentological data using this method (Dawson et al., 2014). Fujita et al. (2016) studied the carbonate production rate of Baculogypsina sphaerulata based on its population dynamics over a two-year period in an intertidal, turf algal zone in the Funafuti Atoll, South Pacific. Those authors estimated production rates using the four methods -(i) to (iv)listed above (Table 1). In comparison, annual production rates of three -(i) to (iii)of these four methods ranged between 165 and 669 g CaCO 3 m −2 yr −2 . Life history tables [method (i)] and growth increments [method (ii)] were three-fold to four-fold higher than that of monthly biomass estimatesmethod (iii). The simple method, on the other hand, showed an exceptionally high production rate of 1818 g CaCO 3 m −2 yr −2 . One reason for this is that, to date, knowledge of LBF species population dynamics is limited and based on studies of a few species in laboratory cultures (Muller, 1974;Hallock, 1981;Fujita et al., 2000). Therefore, one caveat of the simple method is that it assumes the turnover rates of B. sphaerulata based on the population dynamics of similar, co-occurring species such as Calcarina spengleri (Hallock, 1981;Fujita et al., 2016), as well as the assumption that species have similar turnover rates across their distribution range. Thus, the production rates of B. sphaerulata reported in literature (Table 1), including estimates of 378 g CaCO 3 m −2 yr −2 from Palau, in the West Pacific (Hallock, 1981); 144 AE 122 g CaCO 3 m −2 yr −2 from Green Island ; ca 2860 g CaCO 3 m −2 yr −2 from One Tree Island (Doo et al., 2012); and 1607 AE 1084 g CaCO 3 m −2 yr −2 from Raine Island in the GBR, are likely overestimated by an order of three-fold to ten-fold, due to the employment of the simple method (Fujita et al., 2016). Population dynamic (i.e. life history tables) studies (Muller, 1974;Zohary et al., 1980;Hosono et al., 2013;Fujita et al., 2016) are therefore suggested by Fujita et al. (2016) to provide the most reliable method for estimating annual carbonate production rates. Nonetheless, the lack of life history information is an important knowledge gap in estimating carbonate production. While more assessments of population dynamics and turnover rates, for a greater number of species are needed, it may not always be possible to collect life history information in situ, in which case, estimations that include at least two of the four methods may provide more robust estimates and a means for comparison.
Future estimates of carbonate production rates, in the least, should incorporate seasonal measurements (Zohary et al., 1980;Fujita et al., 2016;Doo et al., 2017) to better understand the extent to which LBF physiology, algal-host symbiosis and population dynamics (growth, fecundity, life-span and population turnover rates) are being impacted by the effects of local environmental and global climate change. The few studies that have documented significant in situ population dynamics suggest seasonal fluctuations in densities and carbonate production rates among species living in different carbonate habitats, i.e. as epiphytes on seagrass or algal turf (Zohary et al., 1980;Hohenegger, 2006;Baker et al., 2009;Fujita et al., 2016;Doo et al., 2017). Also, it should not be assumed that areas of low nutrient and primary productivity conditions will result in high carbonate production rates, without knowledge of seasonal or long-term variability. Low carbonate production rates may reflect differences in species annual life stages/cycles, life-spans, turnover and growth rates, which may be due to mechanisms surrounding host-algal metabolism (Zohary et al., 1980).
In other low-photic environments, where water quality may be low, such as in turbid, mesotrophic inshore patch reefs that occur close to river deltas and within estuarine embayments, it is not uncommon to find low to moderate coral cover, diversity, and stress tolerant species of massive (favid) corals (Perry & Larcombe, 2003;Lybolt et al., 2011) and proportionally low to high density accumulations of LBF (Renema & Trolestra, 2001;Renema, 2006bRenema, , 2008Renema, , 2018Narayan & Pandolfi, 2010;Reymond et al., 2013b;Fajemila et al., 2015;Narayan et al., 2015;Johnson et al., 2017;Humphreys et al., 2019). Higher LBF densities may occur in association with algal turf and seagrass species (Renema, 2006b;Narayan & Pandolfi, 2010;Dawson et al., 2012). In these relatively degraded environments, LBF have shown persistence, consistently over millennial timescales (Reymond et al., 2013b;Narayan et al., 2015). Species composition and abundance are generally linked to long-term adaptation to low-photic conditions, as a result of variable but consistent terrestrial inputs over time, by both coral and LBF populations (Reymond et al., 2013b;Narayan et al., 2015;Johnson et al., 2017). Similar to deep water, oligophotic habitats, these environments may also act as potential climate change (thermal) refugia for escape from bleaching and thermal stress (Cacciapaglia & van Woesik, 2016;Sully & van Woesik, 2020). Broad scale spatial studies across different carbonate environments are needed to account for the missing contribution of these LBF populations to reef-wide carbonate budget estimates. Within a structurally complex reef-framework, the dynamic production of carbonate sands by LBF leads to the sustained formation, maintenance and stability of tropical low-lying islands, coral cays and carbonate beaches (Stoddart & Steers, 1977;Yamano et al., 2005). The transport of empty foraminifera tests by strong and constant wave and current movement towards beaches have been known to lead to relatively high concentrations, particularly of one to two species of LBF (Yamanouchi, 1998;Harney et al., 1999;Hohenegger, 2002). For example, it was found that 27% of the total volume of beach sands was composed of LBF components in Oahu, Hawaii (Moberly Jr. & Chamberlain, 1964;Harney et al., 1999). Of this proportion, 80% consisted of Amphistegina spp. tests alone (Harney et al., 1999). In Okinawa, Japan, the proportion of LBF in beach sands ranged between 20% and 95%, with species such as Ampistegina lobifera, Baculogypsina spharerulata and Calcarina spp. contributing significant proportions to beach deposits (Sakai & Nishihira, 1981;Hohenegger, 1994Hohenegger, , 2002Yamanouchi, 1998). Other examples of beaches that reflect deposition of high densities of monospecific LBF to low diversity foraminifera, are found in Palau (Hallock, 1981), Tuvalu (Collen & Garton, 2004;Hosono et al., 2013), One Tree Island, GBR (Doo et al., 2012), Raine Island, GBR (Dawson et al., 2014) and the Bali-Gili-Lombok (Indonesian) islands (Hoeksema & Tuti, 2001).
The majority of tests produced in reef habitats likely do not accumulate on beaches (Resig, 2004). Furthermore, if test transport and deposition is disrupted by habitat destruction and/or by the diversion of water flow (for example, in the construction of a coastal seawall), test accumulation and abundances in beach deposits may change drastically (Hohenegger, 2006), even though in situ production may remain unchanged. Radiocarbon dates derived from Amphistegina spp. tests revealed ages of 1500 years or more for Hawaiian beach sands, indicating longer sand turnover rates (Resig, 2004). Whereas, in Raine Island, GBR, B. sphaerulata showed minimal storage time, rapid test transport (within 60 years or less) and persistence in deposits for a shorter (<100 years) period of time (Dawson et al., 2012(Dawson et al., , 2014. This was likely due to the low durability of B. sphaerulata skeletal components (Ford & Kench, 2012). Also, the close proximity to highly productive (>1 × 10 6 individual m −2 ) algal turf-dominated substrates and lagoonal habitats, may result in higher beach and coral cay accumulations and indicate a critical source for sand replenishment (Hohenegger et al., 1999;Fujita et al., 2009;Dawson et al., 2014). Consequently, some islands are likely only able to persist due to the constant supply and transport of LBF tests (Yamano, 2000;Collen & Garton, 2004;Fellowes et al., 2016). These studies highlight the importance of preservation potential and diagenetic effects, because some species are more prone to mechanical abrasion and chemical dissolution, for example B. sphaerulata and M. vertebralis, than others, for example Amphistegina spp. and Calcarina spp. (Kotler et al., 1992;Ford & Kench, 2012).
While the distribution of LBF tests is linked to in situ carbonate productivity (Dawson et al., 2014), beach deposits (represented by high proportions of one to two species), do not necessarily reflect high in situ reef carbonate-production potential (Harney et al., 1999), nor high coral cover (Renema, 2018). For example, algal and seagrass habitats that occur peripheral to reefs, have been found to promote high total carbonate storage potential for LBF (Borowitzka & Larkum, 1978), especially during the summer season, suggesting inter-annual (seasonal) and spatial variability in storage and redistribution in different reef-associated habitats (Dawson et al., 2014;Doo et al., 2017). Current understanding of the longterm, source to sink, continuum of autochthonous in situ carbonate production and allochthonous deposition and accumulation, still remains insufficient (Yamanouchi, 1998). As to whether beach accumulations reflect a healthy reef status (i.e. high coral productivity and cover) and active carbonate production potential of reefs is not well-quantified. These are topics that can benefit from further investigation, because they could have implications for interpreting how carbonate factories will be altered by coastal habitat destruction (Hohenegger, 2006) and/or climate change, resulting in implications for shoreline renourishment (Dawson & Smithers, 2010).
Large benthic foraminifera response to changing environmental conditions Below, LBF response to four major groups of global stressors is discussed; water quality, temperature, ocean acidification and sea-level rise. This literature review includes a detailed look into the environmental conditions that have shaped LBF communities and their response to single or combined stressors. Extensive work done to date is summarized (in Tables 2 to 5) and the current trends observed from this review are outlined in conceptual Fig. 2.
In contrast, studies of the effects of coastal pollutants on tropical LBF ecology, morphology and physiology are relatively few (Prazeres et al., 2012;van Dam et al., 2012;Ross & Hallock, 2014;Youssef, 2015;Marques et al., 2017;Akther et al., 2020;Ben-Eliahu et al., 2020). Chronic heavy metal contamination from anthropogenic sources has been associated with low population densities and high rates of bleaching, as seen in studies that reported on two amphisteginids (Amphistegina gibbosa and A. lessoni) from fringing reefs in north-eastern Brazil (Prazeres et al., 2012;Marques et al., 2017), two amphisteginids (A. lessoni and A. lobifera) and a soritid (Sorites orbiculus) from the eastern Mediterranean Sea (Ben-Eliahu et al., 2020), and a soritid (S. marginalis) and peneroplid (Peneroplis planatus) from a sheltered Bay in the Red Sea, next to the major city of Jeddah, Saudi Arabia (Youssef, 2015). It was determined that the exposure (and addition) of the heavy metals Cd, Cu and Zn, over short or prolonged periods, inhibited Ca 2+ -ATPase enzyme activity, which consequently weakened tests, making individuals more susceptible to threats arising from multiple stressors (Prazeres et al., 2012;Marques et al., 2017), and increased bleaching frequency when combined with OA effects (Marques et al., 2017). The exposure to Cd, Cu and Pb resulted in: a species-specific and variable (up to 30%) decrease in growth rates; a negative, non-fatal, effect of Pb on algal symbionts (more than the host); a negative effect of Cu on both the host and symbionts; and a negative effect of Cu on growth, more so than Cd and Pb (Ben-Eliahu et al., 2020). Concentrations of Cr, Fe, Mn, Ni, Zn, Cd, Pb and Cu were measured, and significantly high concentrations of Fe, Mn, Pb and Cu resulted in abnormalities in the shape of the chambers (for example, extreme compression and branched last chamber) and the apertures of P. planatus and S. marginalis (Youssef, 2015). The LBF were present in lagoonal environments. High percentages of opportunistic taxa (Ammonium and Elphidium) were recorded and high terrigenous sediments, in the port area, were noted. It appears that the test structure of miliolids (porcelaneous) is more susceptible to test deformations from environmental pollutants (Samir & El-Din, 2001;Youssef, 2015;Ben-Eliahu et al., 2020) than that of rotalids. Table 2. A summary of eight laboratory experimental and one field study highlighted in this review and the response of large benthic foraminifera (LBF) to terrestrial inputs and water quality (and combined effects) as discussed in this paper. The majority of the studies were conducted in laboratory-controlled aquariums or mesocosms using cultured LBF. References cited: 1.  μM, micromoles; Table 3. A summary of six laboratory experimental studies highlighted in this review and the response of large benthic foraminifera (LBF) to elevated temperature and the combined effects of other stressors (nutrients, herbicides). Temperature interactions with ocean acidification (OA) will be found under combined effects in Table 4. The majority of the studies were conducted in laboratory-controlled aquariums using cultured LBF. Only a few studies compared field (in situ) to laboratory studies. References cited: 1.  μM, micromoles. Table 4. A summary of seven laboratory experimental studies highlighted in this review and the response of large benthic foraminifera (LBF) to ocean acidification (OA) and the combined effects of other stressors (eutrophication and temperature). The studies shown here were conducted in laboratory-controlled aquariums using cultured LBF. A broader review, including natural laboratory (field studies) can be found in the paper. References cited: 1. Fujita et al.  were detected (area size increasing with decreasing calcite saturation state); dissolution also in offspring produced at 2000 μatm.

(−)
A. lessonii showed significant decrease in buoyant weight, a reduction in the density of inner skeletal chambers, an increase of Ca-ATPase and Mg-ATPase activities at pH 7.6 when compared with ambient conditions of pH 8.1.

(+)
M. vertebralis showed reduced growth when incubated in isolation; calcification rates were lowest in the high temp./low pH treatments; when incubated with marine algae L. intricata, growth General response/effects (+) positive, (−) negative, (AE) neutral/no effect and calcification rates were similar to ambient; total chl-a decreased and maximum photochemical efficiency increased in ambient conditions; net production remained constant in isolated and associated; both production and respiration rates were significantly higher when associated μM, micromoles. Table 5. A summary of eleven studies highlighted in this review that discuss the relevance of large benthic foraminifera (LBF) carbonate production to low-lying reef islands and associated landforms, which may be impacted by sea-level rise/change. A broader discussion with additional references can be  6 (AE) Rapid sea-level fall during the Last Glacial Maximum (LGM) resulted in sediment accumulation rates that were higher than or similar to those during the Holocene highstand, resulting in the formation of reef flat and backreef environments and increased LBF production Agricultural herbicides entering coral reefs through catchment areas are an increasing issue (Ben-Eliahu et al., 2020). Van Dam et al. (2012) found that the combined effects of elevated temperature (>30°C) and herbicides (diuron) are generally higher than those caused by individual stressors in inhibiting the photosynthetic efficiency in different LBF species. There was a significant linear correlation between reduced photosynthetic efficiency and loss of chlorophylla (van Dam et al., 2012). In addition, chemical pollutants derived from oil dispersants (propylene glycol and 2-buthoxyethanol) resulted in a significantly high incidence in bleaching in A. gibbosa over short-term (48-hours) exposure (Ross & Hallock, 2014). The study showed that some individuals recovered following exposure and removal into clean water. Short-term exposure to waste water was found to have beneficial components, such as temporarily enhanced photosynthetic efficiency, as well as harmful components, such as decreased LBF biodiversity and epibiont infestations, in two species of LBF (Akther et al., 2020). Nonetheless, more morphophysiological studies on the effects of agrochemical herbicides, chemical pollutants, trace elements and sewage effluent on LBF ecology and physiology are needed.
Large benthic foraminifera persistence has also been recorded in long-term sedimentary archives from turbid inshore reefs, prior to and during early European settlement that saw intense changes in land-use practices in Eastern Australia (Lybolt et al., 2011;Reymond et al., 2013b;Narayan et al., 2015;Johnson et al., 2017Johnson et al., , 2019Fujita et al., 2020).
The nutrient thresholds that are considered to promote macroalgal blooms, i.e. dissolved inorganic nitrogen (DIN) and phosphate (DIP), and that are associated with the onset of mesotrophic or eutrophic conditions in coral reefs are considered to be approximately 1.0 μM for DIN and about 0.2 to 0.3 μM for DIP (Hallock & Schlager, 1986;Bell, 1992;Lapointe et al., 2004). High nutrient concentrations may release photosymbionts from nutrient limitation, which could decrease photosymbiont translocation of organic carbon (photosynthate) to the LBF host and thereby reduce energetically expensive mechanisms regulating test calcification (Muscatine et al., 1984;Uthicke & Altenrath, 2010;Reymond et al., 2013a). Hence, studies have shown that increasing concentrations of dissolved inorganic nutrients from pollution and runoff can impair reef growth (Uthicke & Altenrath, 2010;Reymond et al., 2011b), particularly in inshore reefs, which are exposed to higher runoff than offshore reefs Uthicke & Altenrath, 2010). The interactive effects of OA and natural eutrophication (i.e. flood plume) inhibited growth rate at pH levels of 7.6; however, increased abundance of algal cells suggested release from nitrogen limitation in the dinoflagellate-bearing taxon Marginopora sp. (Reymond et al., 2013a).
Differential tolerance to nutrients inputs (Lee & Hallock, 1987;Lee, 2006) can be found across a broad range of inner to outer-shelf gradients (Lee & Hallock, 1987;Lee, 2006;Prazeres et al., 2016), including sites of seasonal upwelling (Renema, 2018) and mesotropic to eutrophic marginal reefs (Richardson, 2006;Prazeres et al., 2016;Humphreys et al., 2019). Prazeres et al. (2016), studied the potential influence of habitats on populations of diatom-bearing Amphistegina lobifera exposed to high nutrients (1.5 and 4.5 µM) and variable temperatures (26 and 29°C) in three different reef site locations (inner, mid and outer shelf) in the northern GBR. Here, the interaction between reef location with either elevated temperature or nutrient concentration, resulted in a greater negative effect on growth rate and survivorship, in the sensitive mid to outer shelf populations than in the more tolerant inner-shelf populations of A. lobifera (Prazeres et al., 2016(Prazeres et al., , 2017a. In contrast, Schmidt et al. (2011) did not find any effect of elevated nitrate concentrations (0.5 to 1.4 µM) on three LBF taxa. Nutrient concentrations alone (at least below threshold levels), do not strongly drive physiological changes in LBF populations. Overall, more studies are required to understand population variability across a broad range of LBF species and water pollution regimes to fully understand response thresholds.
In summary (Table 2), the effects of coastal pollution on LBF results in: 1 Inhibition of Ca 2+ -ATPase enzyme activity, weakened tests (Prazeres et al., 2012;van Dam et al., 2012;Marques et al., 2017), species-specific and variable changes in test growth (Ben-Eliahu et al., 2020), and test abnormalities, the latter being more evident in miliolids with porcelaneous test structures (Youssef, 2015;Ben-Eliahu et al., 2020). 2 Reduced photosynthetic efficiency, which is compounded by the combined effects with other stressors (i.e. elevated temperature) (van Dam et al., 2012;Prazeres et al., 2016Prazeres et al., , 2017a. 3 Reduced long-term adaptation in habitats with more sensitive and stable water quality (mid to outer shelf) conditions than in fluctuating inner-shelf conditions (Prazeres et al., 2016). 4 Potentially beneficial, to no effects, from short-term exposure to high nutrients (Schmidt et al., 2011;Akther et al., 2020).
Bleaching or a loss of photosymbionts or pigments, as a result of photo-oxidative stress, was first recorded in Amphistegina gibbosa from the Florida Keys (Hallock et al., 1993(Hallock et al., , 2006bHallock, 2000b). When this species was experimentally exposed to elevated temperatures (>31°C) in the laboratory, cytological studies revealed temperature accelerated bleaching, deterioration and digestion of the diatom symbionts, and deterioration of the host cytoplasm, similar to that seen in field-stressed specimens (Talge & Hallock, 1995. An experimental study demonstrated that elevated temperatures (>30°C) resulted in bleaching, and decreased photosynthetic efficiency [measured with pulse amplitude modulated (PAM) fluorometry], chlorophyll-a concentration, an indicator of algal symbiont biomass, and growth in two diatom-bearing species (Schmidt et al., 2011). However, two calcarinid diatom-bearing species originating from two different locations in the GBR showed greater but species-specific responses to thermal stress (Schmidt et al., 2011). Stuhr et al. (2017) showed that exposure to chronic thermal stress (at 32°C for 30 days) had negative effects, causing reduced motility, reduced growth (by 50%), elevated antioxidant capacities and gradual bleaching. Single or pulsed thermal stress events in this experiment, in contrast, were associated with high motility and growth but potentially suppressed reproduction (Stuhr et al., 2017).
Likewise, experimental studies in natural laboratory settings have demonstrated high thresholds of LBF to thermal stress in nearshore habitats, which experience variable conditions (Doo et al., 2012;Engel et al., 2015;Weinmann & Langer, 2017). Common and abundant Baculogypsina sphaerulata and M. vertebralis found in the intertidal reef flats of One Tree Reef, GBR, contribute greatly (97%) to local CaCO 3 production (Doo et al., 2012). They commonly experience pulsed temperature increases of 2 to 4°C above the ambient SST (25 to 29°C) during low tides. In one study, both taxa demonstrated remarkable resilience to bear temperature increases of up to +4°C (Doo et al., 2012). However, significant negative effects (dissolution) were recorded for growth rates in the latter species, when temperatures reached +6°C (Doo et al., 2012). Similarly, a variety of small non-LBF and LBF, particularly perforate taxa (Neorotalia calcar and Amphistegina spp.), occurring in shallow tide pool samples in East Africa, showed the ability to withstand extreme ranges (35 to 40°C) in diurnal temperature fluctuations (Weinmann & Langer, 2017). Short-term thermal tolerance of diatom-bearing species (Calcarina defrancii and A. lessonii) is indicated in a study where they were exposed to intermittent temperature extremes of up to 40°in shallow-water (10 m) habitats, associated with algal biofilms attached to volcanic rocks and exposed directly to hot vent CO 2 seeps in Papua New Guinea (Engel et al., 2015).
Latitudinal range expansions of amphisteginids from the northern Red Sea region demonstrate surprisingly high thermal tolerances, which were likely retained during their expansion from the Red Sea into the Eastern Mediterranean . Amphistegina lobifera showed reduced photosysnthetic and growth rates at 32°C which is above current summer maxima in both regions Pinko et al., 2020). Corals from the Red Sea have extremely high bleaching thresholds when compared with other coral species worldwide and were described based on their invasive traits as positively 'thermally filtered' due to recolonization from the Indian Ocean over thousands of years (Fine et al., 2013). These temperature pre-conditioned populations maybe crucial for the survival of LBF under global warming, a hypothesis that remains to be tested also on other calcifying organisms. In summary, these studies demonstrate that extreme environments such as shallow tidal pools, intertidal flats and CO 2 seeps are areas for potential acclimatization and adaptation of foraminiferal populations (Prazeres et al., 2020b), which may become important sources and suppliers of resilient populations of marine calcifiers under future warmer ocean conditions. Species Distribution Models (SDMs) suggest that areas of higher suitability will expand poleward, with increasing SSTs (Weinmann et al., 2013a). In turn, ranges of prolific LBF (Amphistegina, Archaias and Calcarina) are also expected to increase globally (Langer et al., 2013;Weinmann et al., 2013a;Prazeres et al., 2020b), thus, increasing the potential for CaCO 3 production in high-latitude shelf environments (Langer et al., 2012). For example, the northward range expansion of amphisteginids into the Mediterranean region has resulted in an increase in CaCO 3 production (Langer et al., 2012;Weinmann et al., 2013a,b;Schmidt et al., 2015). Also, their current, rapid expansion southward along the East African coastal waters into South Africa (30.84°S) is attributed to the influence of the warm surface waters of the Aguilas Current and the availability of suitable habitats there (Langer et al., 2013;Weinmann et al., 2013a).
Recently, there has been increased discussion on the influence of internal waves (IW) in shaping benthic carbonate communities (Pomar et al., 2012b;Wall et al., 2015;Reid et al., 2019;Pomar, 2020). Internal waves create strong, bottom-current pulses that influence nutrients and thermal variation by vertically displacing the thermocline (Garrett & Munk, 1979;Pomar et al., 2012b;Pomar, 2020). They travel long distances, transporting cool, nutrient-rich waters to coastal shelf habitats before breaking on beaches (Leichter et al., 1998;Alford et al., 2015;Reid et al., 2019). The occurrence of metazoan buildups in lower-photic, mid to outer carbonate ramp environments, throughout the Phanerozoic has been interpreted to be the result of IWs (Pomar et al., 2012b;Wall et al., 2015). Recent studies show that internal waves cooled SST beneficially during the summer months in the South China and Andaman seas (Wall et al., 2015;Reid et al., 2019). Thus, shoaling IWs may play a critical role in ameliorating metabolic and thermal stress in tropical, carbonate reefs, under changing climatic conditions (Roder et al., 2010;Wall et al., 2015;Reid et al., 2019).
These responses to thermal stressors are likely to vary due to: • species-specific responses, which might be dependent on the type of photosymbionts housed, for example, as seen in the diversity and flexibility of the algal symbiont community in the globally distributed amphisteginids (Schmidt et al., 2011Fujita et al., 2014;Stuhr et al., 2018b;Stuhr et al., 2021); • differences in physico-chemical (local) conditions and the ability for acclimatization in different cross-shelf habitats (for example, shallow tidal pools) (Doo et al., 2014;Fujita et al., 2014;Prazeres et al., 2016Prazeres et al., , 2017aStuhr et al., 2017); or • evolutionarily high adaptation by their historical contingency and inherited traits Stuhr et al., 2021).
Most LBF tests are composed of either low magnesium calcite (LMC) or high-magnesium calcite (HMC) (Kinard, 1980), with the latter having a higher solubility in seawater (Raja et al., 2005). It is expected that LBF taxa possessing non-lamellar, porcelaneous-imperforate HMC tests will be more vulnerable to OA (Berner, 1975;Morse et al., 2006). Porcelaneous taxa have a smooth, often translucent, outer test layer that extends over chambers and chamberlets to allow light penetration (Erez, 2003;Fujita et al., 2011). If it dissolves, it leaves the chambers exposed (Cottey & Hallock, 1988;Engel et al., 2015). Alternatively, taxa with hyaline, perforate, lamellar-walled LMC, are expected to show relatively greater resilience to dissolution in reduced pH conditions (Cottey & Hallock, 1988;Engel et al., 2015). Moreover, hyaline taxa can secrete new layers of calcite over their test, thereby thickening the test and allowing for test repair following damage (Erez, 2003;Fujita et al., 2011). The production of these two calcite polymorphs is primarily the result of the differential biomineralization pathways used by foraminifera. The HMC porcelaneous tests are a byproduct of dissolved inorganic carbon (DIC) that is transported directly from seawater into the site of calcification, whereas LMC hyaline taxa benefit from pumping protons in exchange for calcium ions into large intracellular vacuoles, which increases the saturation states and thus facilitates calcification (ter Kuile & Erez, 1991;Glas et al., 2012;de Nooijer et al., 2014;Engel et al., 2015;Toyofuku et al., 2017).
The increase in atmospheric CO 2 concentrations is expected to increase dissolved inorganic carbon (DIC) availability, releasing algal symbionts from DIC limitation ; hence, enhancing photosynthetic rates (Hanson & Dalberg, 1979;ter Kuile et al., 1989). Such CO 2 -fertilization of algal endosymbionts may increase the supply of energy-rich carbohydrates (photosynthate) to the host and enhance growth and calcification, or as is usually the case, is not sufficient enough to counter the need to remove protons from the site of calcification (Bentov et al., 2009;de Nooijer et al., 2009;Glas et al., 2012;Martinez et al., 2018). It is fundamental to also consider the diurnal changes in the biological activity of the algal symbionts and the host foraminifer, i.e. increased CO 2 induced by dark respiration (Köhler-Rink & Kühl, 2005) or diurnal recycling of carbon (Müller, 1978). Additionally, it is worth noting the ability of the photosymbiont to control its location, and not necessarily be at the site of calcification, for example as seen in planktonic species (Hastigerina sp., Globigerinoides sp. and Orbulina sp.), which create a rhizopodial network where the algae can be found during the day (Bé et al., 1977), or the movement within the lacunary system amongst imperforated LBF (Leutenegger, 1977). The intracellular symbionts of Marginopora vertebralis show phototactic behaviour and are able to relocate (through hostmediated mechanisms) into deeper cavities within the test (Petrou et al., 2017). Such behavioural studies demonstrate the importance of host photoprotection through host-symbiont signalling (Petrou et al., 2017).
On local scales, the potential for studying the adverse effects of OA can be found in studies of natural systems such as shallow water CO 2 -seeps (Dias et al., 2010;Uthicke et al., 2013;Engel et al., 2015;Pettit et al., 2015), groundwater springs (Martinez et al., 2018) and/or in association with marine vegetation (Pettit et al., 2015;Doo et al., 2020). Uthicke et al. (2013) found a steep decline in calcifying foraminifer abundance, corrosion of test walls, an absence of carbonate accumulation and no shift to mixotrophy , in the vicinity of the CO 2 -seeps (pH < 7.9 and pCO 2 > 700 μm) of Papua New Guinea (PNG). Variable symbiont colour was retained in specimens with limited test dissolution under extreme conditions (40 to 60°C, and pH between 5.9 and 7.4) near shallow CO 2 -seeps in PNG (Engel et al., 2015). Differences between these cases, possibly species-specific, linked shell structure and composition and likely substrates (i.e. surficial substrates in the former versus hard substrates in the latter). Martinez et al. (2018) found that total abundance of foraminifera was reduced (similar to Uthicke et al., 2013). However, the relative abundance of LBF and agglutinated foraminifera was higher in the low pH, low calcite conditions compared to control sites, suggesting that the non-symbiont-bearing, heterotrophic calcareous foraminifera were more sensitive to the effects of OA in the submarine groundwater springs in the Caribbean. Pettit et al. (2015) found that brown seaweed (Padina pavonica) associated with CO 2 -seeps in the Mediterranean Sea, could not mitigate the effects of OA on epiphytic foraminifera. Instead, as the calcium carbonate saturation state fell, there was a reduction in the number of species, and the assemblage shifted from one dominated by calcareous taxa (pH ca 8.19) to one dominated by an agglutinated assemblage (pH ca 7.71); this is similar to other findings (Dias et al., 2010). However, Doo et al. (2020) found that the stress effects of reduced growth and calcification rates (with increased warming), for the epiphytic M. vertebralis were mitigated when associated with non-calcifying macroalgae (Laurencia intrica), from the GBR, under end-of-century OA (pH 7.7) scenarios. It seems that, in certain regions, algal (Doo et al., 2020) and seagrass Fabricius et al., 2011) substrates may play a role in buffering seawater carbonate chemistry; however, varying results suggest different species-specific interactions between foraminifera and require further exploration.
It was found that the strongest stress response occurred with the interaction of two stressors, i.e. warming and OA. Multi-stressors resulted in: • reduced photosynthetic efficiency (Vogel & Uthicke, 2012;Schmidt et al., 2014;Stuhr et al., 2021); • reduced chlorophyll-a and respiration rates (Schmidt et al., 2014;Stuhr et al., 2021); and • micro-structural alterations to the test including decreased pore size and increased breakage frequency (Stuhr et al., 2021).
Natural laboratories offer opportunities to examine the effect of elevated pCO 2 , also in combination with elevated, or even extreme, temperature effects. Few studies have investigated the response of LBF in shallow water CO 2vents Dias et al., 2010;Uthicke et al., 2013;Engel et al., 2015) or other in situ habitats (Martinez et al., 2018;Doo et al., 2020). Findings from these studies reveal: 1 Significant decreases in calcification rates and elevated respiration rates (net oxygen production) , i.e. in M. vertebralis. 2 Either signs of corrosion (pitting) of test walls (Uthicke et al., 2013) or limited dissolution (Engel et al., 2015). 3 Decline in the abundance of calcifying taxa (Uthicke et al., 2013;Pettit et al., 2015); and a shift to non-calcifying, agglutinated taxa (Dias et al., 2010;Pettit et al., 2015;Martinez et al., 2018); or higher relative abundances of symbiont-bearing and agglutinated taxa, compared to non-symbiont-bearing, heterotrophic taxa (Martinez et al., 2018). 4 Remarkable short-term tolerance of LBF exposed to low pH conditions and intermittent extremes in temperature, likely associated with substratum type (Engel et al., 2015). 5 Either a reduction in the number of species and a shift from a calcareous to agglutinated-dominated assemblage in association with (non-calcareous) macroalgae (Dias et al., 2010;Pettit et al., 2015) or a positive mitigation against the negative of OA on growth and calcification rates with interaction with macroalgal substratum (Doo et al., 2020).
In an attempt to evaluate the ubiquitous effect of OA on LBF, the authors came across a wide range of responses from: severely damaging (to the algal symbionts), to the opposite response of slightly positive effects. Further in situ and cultured studies are necessary to constrain these different trends among different species across broad spatial scales, and to determine how the influence of, for instance: life stages, biochemical pathways of test calcification and experimental parameters (dose levels), will affect LBF response to future impacts. The impact of sea-level rise on LBF carbonate production Global mean sea-level rise (SLR) has been occurring at an accelerated rate since the 1960s, due to thermal expansion (Dangendorf et al., 2019), averaging at 3 mm yr −1 (Bamber et al., 2019), with a projected rise of at least 0.3 m by 2100 (Bamber et al., 2019;Lindsey, 2019). This is, at best, a conservative estimate, based on a newly defined sea-level sensitivity metric (Grinsted & Christensen, 2021). A very rapid rise in coastal seawater level is expected to increase turbidity and runoff (Rogers, 1990;Fabricius, 2005), drown-out shallow reefs with dire consequences for sessile, photosymbiont-bearing organisms dependent on algal photosynthesis (Edwards & Wright, 2015;Perry et al., 2018). However, the effects of SLR may be mitigated in areas where reef growth and carbonate production rates 'keep-up' with SLR, (Buddemeier & Hopley, 1988;Woodroffe & Webster, 2014).
The use of intertidal benthic foraminifera taxa from post-glacial sedimentary deposits on carbonate shelves has been well-established as indicators of sea-level change in temperate regions (Scott & Medioli, 1978Gehrels, 1994;Horton & Edwards, 2003;Cann et al., 2006;Southall et al., 2006;Sifat & Saha, 2019). Similarly, in the tropics, foraminifer taxa from nearshore mangrove and marsh (Horton et al., , 2007, as well as from reef habitats (Horton et al., 2007;Reymond et al., 2011a;Narayan et al., 2015;Johnson et al., 2017;Fujita et al., 2020), show a strong, positive correlation with water depth, and have been used as tools for reconstructing Holocene sea-level change (Frost & Langenheim, 1974;Horton et al., 2005). Such studies provide an analogue for assessing the effect of future predicted SLR on carbonate shelf and slope/ramp environments, including coral reef areas.
It has been suggested that predicted sea-level rise will impact habitats of LBF by reducing current populations to shallow reef crests, thereby reducing test accumulation in back-reef habitats and exacerbating shoreline erosion (Yamano, 2000;Yamano et al., 2007). The impact of sealevel rise on LBF carbonate production dynamics has been assessed through reconstructing the initiation and development of reefs and reef islands, in the light of post-glacial, Holocene sea-level change (Reymond et al., 2011a;Fujita et al., 2020). However, there has been geographical variability of Holocene sea-level in equatorial basins, resulting from regional differences in the solid Earth response to water loading (Camoin & Webster, 2015;Horton et al., 2018;Mann et al., 2019a,b). Accordingly, reconstructions of Holocene LBF carbonate production and their importance for reef island development and maintenance are likely to differ between sites (Kayanne et al., 2002;Fujita et al., 2020).
A recent reconstruction of reef island formation in the Maldives shows that the onset of island formation occurred during a sea-level highstand in the Central Indian Ocean about 2.5 ka BP. During this early phase of island development, the sediment facies was coral dominated, pointing to coral reefs as the most important sediment source (Kench et al., 2020). Following its incipient formation, the shoreline prograded rapidly over the reef flat between 2.1 to 1.5 ka BP, leading to a significant expansion of island size and volume. This major phase of island expansion coincided with a short-term sea-level drop that resulted in adjustments of the surrounding reef flat ecology and concurrent sediment production through a pronounced increase of LBF test accumulation along the shoreline. Consequently, the interplay between future sea-level rise and variable 'keep-upcatch-up' modes of reef platform development, might result in episodic shifts in reef flat ecology, related to water depth, and subsequent pulses of carbonate sediment production through the proliferation of LBF and other taxa (Woodroffe & Webster, 2014;Camoin & Webster, 2015;Yokoyama et al., 2018;Fujita et al., 2020).
A recent study by Fujita et al. (2020), found that LBF carbonate sediments accumulated at high rates during the last glacial lowstand, i.e. the Last Glacial Maximum (LGM), at rates similar to or even higher than those accumulated during the Holocene highstand. These deposits were associated with high proportions of fossil Baculogypsina and Calcarina tests, indicating an earlier development of geomorphically-mature fringing reefs on the modern shelf-edge of the GBR . Vertical reef accretion and subsequent development of reef flat and lagoon habitats followed LGM 'keep-up-catchup' mode of rapid SLR. Thus, geological studies of LBF populations from fossil and submerged ('drowned') reefs, across broad carbonate shelf and ramp environments, can help to improve current understanding of how long-term sealevel fluctuations control the architecture and the potential for accumulation in carbonate depositional systems. The above observation of high accumulation rates during rising sea levels, for example, also corresponds with the behaviour of Miocene reef systems (Pomar, 1991;Pomar et al., 2004;Mateu-Vicens et al., 2012).
Detrimental effects of SLR may therefore be mitigated in areas where active reef growth and carbonate production will be maintained and stable, to allow the reef to keep-up with SLR (Buddemeier & Hopley, 1988;Webb & Kench, 2010;Woodroffe & Webster, 2014). This has been documented in reconstructions of postglacial Holocene (transgressive) SLR, resulting in reef initiation, development and persistence of reef communities over time (Reymond et al., 2011a(Reymond et al., , 2013bFujita et al., 2020). It is evident that reef ecosystems and low-light adapted carbonate producers (Renema & Trolestra, 2001;Mateu-Vicens et al., 2012;Renema, 2019) can occur under mesophotic conditions and at greater depth ranges than previously thought (Beaman et al., 1994;Woodroffe & Webster, 2014). However, a loss of functionality, due to chronic bleaching or anthropogenic degradation for instance, may not allow reef ecosystems as carbonate depositional systems to keep pace with SLR (Woodroffe & Webster, 2014). The combination of sea-level rise and expanding subtropical climatic zones into higher latitudes, could also create additional, suitable shoreline habitats for carbonate deposition by LBF (Hallock, 2005). This could particularly be the case for a few taxa, such as Amphistegina spp., which show a wide depth range and temporal latitudinal range expansion (Langer et al., 2012(Langer et al., , 2013Weinmann et al., 2013a), and colonization of tropical species facilitated by increasing temperatures in regions such as the Mediterranean (Prazeres et al., 2020b). Furthermore, increasing temperatures at higher latitudes and the availability of new habitats, could trigger radiation and diversification in LBF (Hallock et al., 1991;Prazeres et al., 2020b).
As a possible countermeasure against shoreline erosion on reef islands, and to enhance the self-sustaining mechanism in reef systems, researchers from Japan and the tropical South Pacific Island of Tuvalu have been mass-culturing Baculogypsina sphaerulata (Hosono et al., 2013). This species typically lives in shallow waters (less than 5 m depth) and is thus adapted to the high light intensity and temperature conditions on exposed reef crests at low tide. Such efforts highlight the importance of ecosystem services provided by LBF and the potential for sustainable, high efficacy ecoengineering solutions with low environmental impacts Fujita et al., 2016).

FUTURE DIRECTIONS AND SUMMARY
Large benthic foraminifera (LBF) diversification and carbonate production have been intricately linked to ocean-climate conditions. In the 'hothouse world' of the early Cenozoic Era, vast nummulitic carbonate accumulations, consisting of exceptionally large, flattened and possibly long-lived (Ferrández-Canadell et al., 2014;Hallock & Seddighi, 2021) LBF taxa, developed steadily over time, and were accompanied by warm temperatures, extremely depleted nutrients, reduced thermal stratification and ocean circulation, in possibly deep-water oligophotic settings (Racey, 2001;Hallock & Pomar, 2009;Pomar et al., 2017;Hallock & Seddighi, 2021). Modern analogues are rare. The closest extant, largest and deepest-dwelling nummulitid LBF, Cycloclypeus carpenteri, is limited today to deep-waters in the Indo-west Pacific (Hohenegger, 1994;Beavington-Penney & Racey, 2004). It does not form similar carbonate deposits as in ancient times, largely due to dissimilarity in environmental conditions (Beavington-Penney & Racey, 2004;Hallock & Seddighi, 2021). Under rapidly changing oceanclimate conditions, the question remains as to whether LBF will have the capacity to acclimatize and maintain active carbonate production at current or higher production rates. In order to reduce under and/or overestimation, this review highlights the need for reef carbonate budget estimates, based on LBF, to encompass broader carbonate-producing reef habitats; and increase knowledge of life-history, growth and turnover rates, of a greater number of LBF species.
While LBF have had successful and widespread application in long-term, ecological studies, it appears that their underwhelming representation in modern coral reef studies lags, due to a lack of awareness of LBF relevance as carbonate producers and/or the impression that their application is too traditional and time-consuming. However, as greater collaborative efforts among reef scientists continue, it will encourage the application and further refinement of foraminifer bio-indicator tools, because they are costeffective ways to sample, monitor and assess reef health, with a minimal environmental footprint Prazeres et al., 2020a). Prioritizing the use of comparative, comprehensive and complementary micropalaeontological and ecological research will benefit future pursuits in understanding carbonate ecosystems. As a way forward, greater inter-disciplinary approaches that integrate established, standardized protocols with relevant, new technologies, for example, remote sensing and habitat mapping (Doo et al., 2017;Förderer et al., 2018), and field-based ecological approaches, will provide valuable insights into the magnitude of species diversity and carbonate production across broader spatial and temporal scales. Such inter-disciplinary approaches may inspire increased cooperation among reef scientists, to draw a clearer picture of carbonate dynamics under changing conditions and potentially how to protect vital reef habitat areas.
Globally, in addition to warmer ocean temperatures, increasing local anthropogenic impacts such as sewage effluents, oil spills and overharvesting, directly diminish the potentiality for the existence of such refugia. Therefore, the urgent identification and protection of carbonate/climate refugia (Riegl & Piller, 2003;Bongaerts et al., 2010;Keppel et al., 2012;Beckwith et al., 2019) is an important first step to supporting and safeguarding current resilient stocks (Carter et al., 2020). Reef ecosystems experiencing high anthropogenic or climatic pressures may still contain conditions (for example, exposure hotspots) to allow for adaptive behaviours and suitable microhabitats that support resilient LBF taxa. It is likely that suitable refugia for LBF may include environments that show: high alkalinity and seawater buffering potential, for example: among non-calcifying macroalgal (Doo et al., 2020) and/or seagrass beds; high habitat connectivity; suitable hydrodynamics including vigorous water circulation (Riegl & Piller, 2003), currents, internal waves (Reid et al., 2019) and low sedimentation rates. They may include areas where species show a capacity for natural acclimatization to high thermal tolerance thresholds or past bleaching events, such as in the high-latitude reefs of the Persian Gulf (Riegl, 2003;Riegl et al., 2011), Caribbean (Castillo et al., 2012), Kiribati (Carilli et al., 2012) and Hawaii (Coles et al., 2018). Further work is needed to better understand the eco-evo dynamics (Pelletier et al., 2009), in which ecological (habitat) associations and interactions can affect evolutionary trajectories of species to promote adaptation (Fig. 2).
In summary, this review highlights key contributions to current understanding of how symbiont-bearing LBF may respond to current and future predicted environmental and climatic change, and its implications for carbonate production. Knowledge gaps exist, and the future of LBF research may benefit from: (i) understanding of the role that species-specific differences play in LBF resilience; (ii) refinement and standardization of the methods for estimating carbonate production rates; (iii) understanding the dynamic continuum between LBF from source to sink, and how this may be critically linked to the maintenance of coastal shorelines and landforms, as well as ecological reef (health) status; and (iv) increased assessment of LBF distribution and carbonate production in under-represented carbonate environments, LBF long-term persistence in these environments and their potential for acting as carbonate refugia.
Large benthic foraminifera appear to be an adaptive, robust and resilient group of CaCO 3producers in the face of stressors in recent times. Several studies presented in this review provide support for the potential success of LBF, while others showed opposite and detrimental effects, under future predicted climate change scenarios. Overall, these studies support their circum-global distribution; latitudinal range expansions into high-latitude reefs (amphisteginids); flexibility in their host-diatom partnerships to suit different environmental conditions; persistence under varying conditions; and their relative tolerance for episodic thermal stress in nearshore environments. The studies that support the hypothesis that LBF have a high acclimatization capacity to short-term environmental stress, touch upon the idea that LBF may potentially thrive in a 'hothouse world' over other calcifiers such as corals and calcareous coralline algae. However, knowledge gaps that emerge suggest that further studies are needed in the understanding of: (i) host-symbiont acclimatization and multi-generational adaption potential, an area that is lacking or still in its infancy; and (ii) the differences in responses across temporal and spatial scales amongst a wider variety of LBF taxa and across broader oligotrophic to mesotrophic, shelf to slope carbonate environments.