DOM is increasingly recognized for its important role in the element cycling and the generalecosystem functioning of tropical coral reefs (e.g., Rohwer and Youle 2010; Barott and Rohwer2012; De Goeij et al. 2013; Haas et al. 2013b). The DOM pool on coral reefs is mainly fueled bythe DOM release of benthic algae, benthic cyanobacterial mats (BCMs) and scleractinian corals(Haas et al. 2010a; Haas et al. 2011; Van Duyl et al. 2011; Haas et al. 2013b; Brocke et al. 2015).Yet, while the majority of coral-derived DOM is mineralized by specialized microbes in coralsands, algal DOM is mainly degraded by a planktonic microbial community (Wild et al. 2004a;Haas et al. 2011; Haas et al. 2013b). Algal-DOM is considered to play a crucial role in coral-algalinteractions by promoting the growth of opportunistic microbes in the coral-algal interface (Barottand Rohwer 2012). Energy and nutrients stored in DOM are cycled via two important pathways tohigher trophic levels: i) the conventional microbial loop (Azam et al. 1983) and ii) the sponge loopthrough DOM-feeding and subsequent detritus production by cavity and open reef sponges (DeGoeij et al. 2013).In this thesis, the effect of light on DOC release of benthic primary producers was investigated onthe coral reefs of Curaçao (Fig. S; II?. Incubation experiments at three light levels (0; 30-80 and200-400 µmol photons m-2 s-1) revealed a positive relation between the DOC release rates of thecommon Caribbean reef algae Cladophora sp., Dictyota menstrualis and Lobophora variegata. Incontrast, the DOC release of the two tested scleractinian coral species Madracis mirabilis andOrbicella annularis was either minimal or DOC uptake occurred independent of light availability.Additionally, in situ DOC concentrations were measured in close proximity to seven benthicprimary producers, sediment and the water column at nine sites evenly distributed along theleeward coast of Curaçao. In situ DOC concentrations were positively correlated with lightintensity, although the magnitude of the correlation differed among species and bottom typestested. Furthermore, average DOC concentrations were twice as high during the sampling in May-June compared to the sampling in November-December. This can at least be partly explained by asignificantly higher light availability in May-June.Given the positive effect of light availability on the DOC release rates of reef algae and in situDOC concentrations, it was tested whether light availability can be used to predict the occurrenceof elevated DOC concentrations near the common reef alga Dictyota sp. (Fig. S; III). Therefore, insitu DOC concentrations in close proximity to Dictyota sp. were investigated along a natural lightgradient from 5 to 20 m depth. Additionally, the change in DOC concentration near Dictyota sp.was determined from 9:00 till 18:00 hrs at 11 m depth and compared to diurnal changes in lightavailability. Elevated DOC concentrations near Dictyota sp. were only observed at 10 m, but notat 5 and 20 m depth. Moreover, DOC accumulated in the vicinity of Dictyota sp. relative tobackground concentrations during the course of the experiment. These findings suggest that lightdrivenDOC release is an important factor in the occurrence of diurnal and depth-related variationsin DOC concentrations near Dictyota sp. relative to concentrations in the overlying water column.However, other factors, such as photoinhibition and water exchange also need to be taken intoaccount to adequately explain the occurrence of elevated DOC concentrations.While several studies also reported a positive effect of light availability on the DOC release ofaquatic primary producers (e.g., Fogg 1983; Wood and Van Valen 1990; Cherrier et al. 2014),others did not find such a relationship (e.g., Bjørnsen 1988; Marañón et al. 2004; Marañón et al.2005). This discrepancy could be based on the presence of two different DOC release mechanisms:i) a light-dependent mechanism, where DOC is actively released in an overflow mechanism, andii) a light-independent mechanism, where DOC passively diffuses through the cell membrane.These two mechanisms are certainly not mutually exclusive and varying environmental factors(e.g., nutrient availability) are proposed to determine which mechanism dominates (Carlson 2002;Borchard and Engel 2015). Incubation experiments with turf algae were performed to test theinteractive effects of light and nutrient availability on DOC release (Fig. S; IV). Turf algae wereexposed to two light levels (full and reduced light) and two nutrient treatments (natural seawaterand enriched seawater). DOC release by turf algae increased with increasing light availability undernaturally occurring nutrient concentrations. Addition of nutrients resulted in the disappearance ofthe positive relationship with light availability. DOC release rates in these nutrient repletetreatments were comparable at full and at reduced light levels. These results indicate that low lightin combination with low nutrient availability reduces the release of DOC by Caribbean turf algaeand that light and nutrient availability simultaneously and interactively affected this release.BCMs are becoming more abundant on coral reefs worldwide and dominate the benthiccommunities on some reefs. Despite the high abundance, their contribution to the local DOC poolis virtually unknown. DOC and oxygen fluxes of BCMs were determined in situ using benthicchambers and compared to those of other benthic primary producers (Fig. S; V). BCMs releasedapproximately 16% of their photosynthetically fixed carbon as DOC. However, DOC release atnight was more than twice that high. This DOC release in the dark is proposed to be the result ofincomplete organic matter degradation and fermentation under anoxic conditions. Whenintegrating the DOC release in light and dark conditions over a 24 hrs diurnal cycle at the studysite on Curaçao, BCMs (benthic cover: 24%) were estimated to release four times more DOC thanmacroalgae and turf algae combined (benthic cover: 17 and 19%, respectively). BCMs therebydisplay the highest contribution to the benthic DOC production (80%).Coral-excavating sponges are the most important bioeroders on Caribbean reefs. If they, similar toother non-excavating sponges, can directly feed on DOC, they may benefit from an increase inDOC production due to ongoing coral-algal phase shifts. Therefore, the potential DOC andbacterial uptake of the common excavating sponges Siphonodictyon sp. and Cliona delitrix wereinvestigated in situ (Fig. S; VI). Despite bacterial retention efficiencies of 72 and 87% forSiphonodictyon sp. and C. delitrix, respectively, both species relied mainly on the uptake of DOCto meet their carbon demand (82 and 76%, respectively). This utilization of DOC indicates thatexcavating sponges might i) benefit from an increase in DOM production as result of coral-algalphase shifts and ii) may also participate in the sponge loop.In conclusion, the work in this thesis has shown that under natural nutrient conditions (nutrientlimitation), light availability positively affects the DOC release of reef algae. Yet, also otherfactors, such as photoinhibition due to excessive light and DOC removal processes (e.g., waterexchange, diffusion and consumption by heterotrophic organisms) have to be taken into account tomake predictions about the occurrence of elevated DOC concentrations in close proximity to reefalgae. Next to the light-driven DOC release under natural nutrient conditions, light-independentDOC release can occur in Caribbean turf algae under nutrient replete conditions. Thus, nutrientavailability might determine whether light-dependent or light-independent DOC release prevailsand a third DOC release mechanism was proposed that operates in BCMs in the absence of light.Lastly, the ability of coral-excavating sponges to use DOC as food source suggests that they maybenefit from current coral-algal phase shifts and increases in BCMs and could participate in thesponge loop.Fig. S Simplified scheme of DOM dynamics on tropical coral reefs. Roman numbers represent investigatedprocesses and corresponding chapters in this thesis. Red arrows indicate competitive species interactions. |
