Nitrogen enrichment offsets direct negative effects of ocean acidification on a reef-building crustose coralline alga

Abstract

Ocean acidification (OA) and nutrient enrichment threaten the persistence of near shore ecosystems, yet little is known about their combined effects on marine organisms. Here, we show that a threefold increase in nitrogen concentrations, simulating enrichment due to coastal eutrophication or consumer excretions, offset the direct negative effects of near-future OA on calcification and photophysiology of the reef-building crustose coralline alga, Porolithon onkodes. Projected near-future pCO₂ levels (approx. 850 µatm) decreased calcification by 30% relative to ambient conditions. Conversely, nitrogen enrichment (nitrate + nitrite and ammonium) increased calcification by 90–130% in ambient and high pCO₂ treatments, respectively. pCO₂ and nitrogen enrichment interactively affected instantaneous photophysiology, with highest relative electron transport rates under high pCO₂ and high nitrogen. Nitrogen enrichment alone increased concentrations of the photosynthetic pigments chlorophyll a, phycocyanin and phycoerythrin by approximately 80–450%, regardless of pCO₂. These results demonstrate that nutrient enrichment can mediate direct organismal responses to OA. In natural systems, however, such direct benefits may be counteracted by simultaneous increases in negative indirect effects, such as heightened competition. Experiments exploring the effects of multiple stressors are increasingly becoming important for improving our ability to understand the ramifications of local and global change stressors in near shore ecosystems.

1. Introduction

Ocean acidification (OA) has the potential to profoundly change marine ecosystems by altering biological processes such as calcification and photosynthesis. The changes in carbonate chemistry that occur with OA are predicted to impair calcification in many marine calcifiers [1], though there are taxa-specific sensitivities [1,2]. Local stressors can have similarly profound effects on benthic ecosystems. Eutrophication from agricultural run-off and sewage effluent is one of the leading causes of degradation in marine ecosystems across the globe. The associated increases in inorganic nitrogen, particularly nitrite + nitrate and ammonium , can supply primary producers with limiting nutrients, thus stimulating photosynthesis [3] and increasing macroalgal biomass [4]. At local scales, fish and invertebrate excretions also can increase nitrogen concentrations [5,6]. Such processes of nutrient enrichment are occurring simultaneously with OA. Understanding the combined effects of these stressors is critical to determining direct organismal responses to local and global change, particularly in eutrophic environments.

Coral reefs are among the most sensitive ecosystems to OA and eutrophication [4], partly because they are built by calcifiers and are typically oligotrophic. One taxonomic group that is essential to building the reef carbonate platform is crustose coralline algae (CCA). CCA secrete the most soluble polymorph of CaCO₃ (high-Mg calcite), and are highly sensitive to seawater changes associated with OA [7,8]. The reef-building CCA, Porolithon onkodes (formerly Hydrolithon onkodes), has demonstrated lower calcification rates under near-future OA conditions [9,10]. Conversely, nutrient enrichment has stimulatory effects on algal photophysiology (electron transport rates and photosynthetic pigments) and photosynthetic rates [3]. Red algae, such as CCA, respond readily to increased availability of nitrogen because they can actively store luxury nitrogen in their accessory phycobilin pigment complexes [11]. Despite the potential synergistic effects of OA and nutrient enrichment, relatively little is known about how CCA respond to these simultaneous stressors (except see [12]). The goal of this study was to quantify the combined effects of OA and nitrogen enrichment on calcification and photophysiology of P. onkodes. Understanding how nutrient enrichment influences direct biological responses to OA will allow us to better predict organismal responses to global change in the framework of local habitat variability.

2. Material and methods

We exposed P. onkodes cores to a factorial combination of ambient (AC) and high pCO₂ (HC) (approx. 400 and 850 µatm) crossed with ambient (AN) and high nitrogen (HN) concentrations for two weeks using established methods [9]. The high pCO₂ treatment represents future ocean conditions projected for the end of the century [13]. The nitrogen enrichment treatment simulated elevated concentrations associated with eutrophication or local consumer excretions [4–6] (see the electronic supplementary material for details). We quantified effects of treatment on coralline net calcification by assessing changes in buoyant weight. We quantified effects on photophysiology with PAM fluorometry to determine relative electron transport rates (rETR) and extractions to determine photosynthetic pigment concentrations. Treatment effects on response variables were analysed with separate linear mixed effects models using the package lme4 in R v. 3.4.2 [14], with pCO₂ and nitrogen as fixed factors and tank as a random nested factor. All methods and statistical analyses are described in detail in the electronic supplemental material. Raw data are archived at Pangaea (https://doi.pangaea.de/10.1594/pangaea.887917).

3. Results

(a). Environmental parameters

CO₂ manipulation decreased mean ambient total scale pH (pH_T) by approximately 0.20 units in the high pCO₂ treatment, simulating the increase in acidity projected for the end of the century [13]. Nitrogen enrichment increased ambient nitrate + nitrite concentrations by approximately 50% and ammonium concentrations by approximately 270% (table 1), simulating elevated nutrient levels associated with coastal eutrophication [4] or localized consumer excretions [5,6] on coral reefs. There were no substantial differences in environmental parameters across replicate tanks within treatments (electronic supplementary material, tables S1 and S2).

Table 1.

Mean (±s.e.) treatment parameters. Temperature, salinity, light, water flow and pH_T (total scale pH) were measured daily (N = 14). Replicate tank values (N = 2 per treatment) were averaged for each day, and then averaged to yield overall daily treatment means. Total alkalinity (A_T) was measured every third day (N = 5). pCO₂ and Ω_c (the saturation state of calcite) were derived from measured values of A_T, salinity, temperature and pH_T using CO₂SYS with corrections for nitrogen enrichment [15]. Water was sampled twice daily for nitrogen concentrations every 3–4 days (N = 4). Treatment combinations were ambient pCO₂ x ambient nitrogen (ACAN); ambient pCO₂ × high nitrogen (ACHN), high pCO₂ ×ambient nitrogen (HCAN), high pCO₂ × high nitrogen (HCHN). Light^† = photosynthetically active radiation (PAR, µmol quanta m⁻² s⁻¹), , .

treatment	T (°C)	salinity (PSU)	light†	flow (ml min−1)	pHT‡	AT (µmol kg−1)	pCO2 (µatm)	Ωc	(µM)	(µM)
ACAN	28.4 ± 0.07	36.1 ± 0.07	334 ± 16	139 ± 2	7.99 ± 0.003	2342 ± 7	464 ± 4	5.30 ± 0.02	0.66 ± 0.02	6.34 ± 0.61
ACHN	28.4 ± 0.03	36.1 ± 0.06	329 ± 13	138 ± 2	8.00 ± 0.006	2326 ± 4	454 ± 8	5.32 ± 0.05	0.99 ± 0.05	19.08 ± 2.27
HCAN	28.5 ± 0.07	36.2 ± 0.08	336 ± 13	140 ± 2	7.78 ± 0.017	2338 ± 2	843 ± 34	3.62 ± 0.12	0.77 ± 0.09	6.05 ± 0.56
HCHN	28.3 ± 0.04	36.1 ± 0.08	329 ± 8	138 ± 2	7.79 ± 0.013	2329 ± 5	800 ± 26	3.66 ± 0.09	0.84 ± 0.08	22.45 ± 1.55

(b). Net calcification

There was no interactive effect of pCO₂ and nitrogen enrichment on net calcification of P. onkodes (p = 0.965), but there were significant negative effects of pCO₂ (p = 0.035) and positive effects of nitrogen enrichment (p = 0.001) (table 2). High pCO₂ alone (HCAN) decreased net calcification of P. onkodes by 30% relative to the ambient treatment (ACAN) with average rates (±s.e.) of 2.7 (±0.5) and 3.9 (±0.5) mg CaCO₃ cm⁻², respectively (figure 1a). Nitrogen enrichment increased net calcification by 90% and 130% in the ACHN and HCHN treatments, respectively, relative to the respective ambient nitrogen (AN) treatments, with average rates of 7.4 (±0.5) and 6.2 (±0.6) mg CaCO₃ cm⁻².

Table 2.

Results of mixed model fixed effects. Significance at p < 0.05 is noted in italic.

treatment	source	d.f.	F	p-value
net calcification	pCO2	1	4.582	0.035
	nitrogen	1	41.16	<0.001
	pCO2 × nitrogen	1	0.002	0.965
rETR	pCO2	1	2.096	0.151
	nitrogen	1	22.46	<0.001
	pCO2 × nitrogen	1	30.75	<0.001
chlorophyll a	pCO2	1	0.442	0.510
	nitrogen	1	15.75	<0.001
	pCO2 × nitrogen	1	2.843	0.100
phycocyanin	pCO2	1	0.447	0.508
	nitrogen	1	21.08	<0.001
	pCO2 × nitrogen	1	0.040	0.843
phycoerythrin	pCO2	1	0.038	0.856
	nitrogen	1	28.86	0.006
	pCO2 × nitrogen	1	0.269	0.632

Figure 1.

Effects of pCO₂ and nitrogen enrichment on Porolithon onkodes (a) net calcification normalized to core surface area and (b) relative electron transport rate (rETR). Blue represents ambient pCO₂ (AC) and red represents high pCO₂ (HC). Light colour represents ambient nitrogen (AN) and dark colour represents high nitrogen (HN). Values are mean ± s.e.

(c). Relative electron transport rates

There was a significant interactive effect of pCO₂ and nitrogen on rETR of P. onkodes (p < 0.001, table 2). Nitrogen enrichment increased rETR, but only in the high pCO₂ treatment where it was 19–37% higher than the other three treatments (figure 1b).

(d). Photosynthetic pigments

Nitrogen enrichment significantly increased the concentrations of chlorophyll a (p < 0.001), phycocyanin (p < 0.001) and phycoerythrin (p = 0.006), and there was neither an effect of pCO₂ nor a significant interaction (table 2). Nitrogen enrichment increased chlorophyll a by 84–323% (figure 2a), phycocyanin by 207–379% (figure 2b) and phycoerythrin by 239–445% (figure 2c) in the ambient and high pCO₂ treatments relative to the respective AN treatments.

Figure 2.

Effects of pCO₂ and nitrogen enrichment on Porolithon onkodes (a) chlorophyll a, (b) phycocynanin and (c) phycoerythrin concentrations normalized to fresh weight (FW) of extracted tissue. Blue represents ambient pCO₂ (AC) and red represents high pCO₂ (HC). Light colour represents ambient nitrogen (AN) and dark colour represents high nitrogen (HN). Values are mean ± s.e.

4. Discussion

Our study demonstrates that nitrogen enrichment can mediate the direct effects of simulated OA on biological responses of a crustose coralline alga. As expected, high pCO₂ alone decreased coralline calcification, which agrees with established trends showing negative effects of OA on calcification of tropical CCA [8–10]. The combined effect of nitrogen enrichment significantly increased calcification rates in both ambient and high pCO₂ conditions. Notably, this stimulatory effect partially offset the direct negative effects of simulated OA alone on net calcification. Nitrogen enrichment and pCO₂ interactively affected P. onkodes photophysiology, with highest rETR under high nitrogen and high pCO₂. Nitrogen enrichment also had a significant stimulatory effect on P. onkodes photosynthetic pigment content, and consistently increased chlorophyll a, phycocyanin and phycoerythrin content by 1–4-fold, regardless of pCO₂ treatment. Nitrogen enrichment increased ammonium concentrations by approximately 270%, and ambient nitrite + nitrate concentrations by up to approximately 50%. Ammonium is a biologically available form of nitrogen readily taken up by red algae [11], thus the response of P. onkodes to enrichment was likely driven by elevated ammonium concentrations.

The increases in photophysiology (rETR, pigments) in response to nitrogen enrichment indicates a potential for higher photosynthetic rates, and corroborates extensive literature documenting a positive relationship between nutrient availability and both algal photophysiology and photosynthesis [3]. Previous studies have similarly found stimulatory effects of nitrogen enrichment on photosystem II [16] (represented in our measurement of rETR) and on the concentrations of primary (i.e. chlorophyll a) and accessory (i.e. phycocyanin, phycoerythrin) photosynthetic pigments in algae [17]. Excess availability of nitrogen, and the potential for luxury nitrogen storage [11], thus may have facilitated production of pigment molecules and stimulated photosystem and pigment activity under nitrogen enrichment. However, identifying the exact mechanisms driving the interactive effect of pCO₂ and nitrogen on rETR are beyond the scope of this study. Algal photosynthesis and calcification are tightly coupled [18], thus higher rates of photosynthesis in response to nitrogen enrichment may be an underlying mechanism for the corresponding increase in P. onkodes net calcification.

The levels of nitrogen enrichment used in this study represent nutrient concentrations documented on coral reefs due to coastal eutrophication [4] or consumer excretions [5,6]. These results indicate that such local environmental factors may mediate the direct biological responses of some organisms to OA. For example, CCA from adjacent reef environments, such as an exposed fore reef versus a nearby fringing reef, may respond differently to OA due to organismal acclimatization to environmental parameters other than carbonate chemistry [9]. How potential direct benefits of nutrient enrichment will manifest at the community-scale are unclear, because nutrient enrichment may simultaneously increase indirect effects. Nitrogen enrichment and increased availability of dissolved CO₂ associated with OA can facilitate growth of macroalgae [2,4], which compete with and can overgrow coralline algae. An increase in negative indirect effects may thus counteract any direct benefit of nitrogen enrichment to coralline calcification under OA. Further, the impact of interacting global and local stressors may vary across ecosystems, as evidenced by the negative effect of high nutrients combined with high pCO₂ on temperate, understory corallines in Southern Australia [12]. Using a multi-stressor approach in global change experiments that incorporates local-scale variability and ecological responses will allow us to better predict the impacts of simultaneously increasing local and global stressors on near shore ecosystems.

Ethics

Research was conducted under a Protocole D'accueil (Scientific Research Permit) to R.C.C. from the Delegacion a la Recherche de la Polynesie Francais (unnumbered).

Data accessibility

Data are available at Pangaea (https://doi.pangaea.de/10.1594/pangaea.887917).

Authors' contributions

Both authors conceived the study, conducted the experiment, wrote and revised the manuscript, approved the final version and agree to be held accountable for the content therein.

Competing interests

The authors have no competing interests to declare.

Funding

This research was supported by funding from the National Science Foundation to the Moorea Coral Reef Long Term Ecological Research program (OCE- 04 17412).