Nutrient enrichment of the coastal zone places intense pressure on marine communities. Previous studies have shown that growth of intertidal mangrove forests is accelerated with enhanced nutrient availability. However, nutrient enrichment favours growth of shoots relative to roots, thus enhancing growth rates but increasing vulnerability to environmental stresses that adversely affect plant water relations. Two such stresses are high salinity and low humidity, both of which require greater investment in roots to meet the demands for water by the shoots. Here we present data from a global network of sites that documents enhanced mortality of mangroves with experimental nutrient enrichment at sites where high sediment salinity was coincident with low rainfall and low humidity. Thus the benefits of increased mangrove growth in response to coastal eutrophication is offset by the costs of decreased resilience due to mortality during drought, with mortality increasing with soil water salinity along climatic gradients.
Citation: Lovelock CE, Ball MC, Martin KC, C. Feller I (2009) Nutrient Enrichment Increases Mortality of Mangroves. PLoS ONE 4(5): e5600. https://doi.org/10.1371/journal.pone.0005600
Editor: Ross Thompson, Monash University, Australia
Received: March 1, 2009; Accepted: April 9, 2009; Published: May 19, 2009
Copyright: © 2009 Lovelock et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: Funding for this work was provided by NSF Biocomplexity award (DEB-9981535; http://www.nsf.gov/), AAAS WISC award (http://www.aaas.org/programs/international/wist/), Smithsonian Marine Science Network (http://www.si.edu/marinescience/) and the Australian Research Council awards (LP0561498 and DP0774491; http://www.arc.gov.au/). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Nutrient enrichment is one of the most serious threats to near shore coastal ecosystems , . The consequences of nutrient enrichment include algal blooms , coral reef degradation , , loss of diversity and ecosystem resilience ,  and, in extreme cases, the development of “dead” zones” . These negative consequences contrast with observations that marine plant growth, including that of tropical mangroves is enhanced with nutrient enrichment , –. Indeed, for decades mangroves have been proposed as suitable for use in sewage and aquaculture effluent treatment –, although assessment of long-term consequences of nutrient enrichment for mangrove ecosystems is lacking.
There is reason to expect the addition of nutrients will have detrimental effects on mangrove forests. As nutrient availability increases, plants invest less in roots and more in aboveground parts, thereby further enhancing growth rates –. However, it follows that plants exposed to high levels of nutrient availability must have greater susceptibility to environmental stressors, such as drought, that require large investment in roots for tolerance . Mangroves characteristically increase allocation of carbon to growth of roots relative to shoots with increase in salinity, with this pattern being amplified with decreasing humidity , . These observations invite two hypotheses: 1) that mangroves exposed to high nutrient availability, with their relatively lower investment in roots and greater investment in canopies, should suffer greater mortality during drought, and 2) that nutrient-induced mortality should be particularly high in sites subject to periods of low and fluctuating rainfall and humidity and high sediment salinity.
Here we report on the long term survivorship of mangrove trees that have been experimentally fertilized at 12 sites for 3 to 12 years (Table S1). Our sites ranged over two distinct biogeographic provinces. They were located within the Caribbean basin and in the Indo Pacific region, where they were situated in New Zealand and on the west and east coasts of Australia. Within sites, we fertilized trees in two or more distinct environments and where possible we fertilized multiple species. Trees were fertilized in seaward fringing forests that are inundated frequently (9 experiments) and also in scrub forests that are less frequently inundated and where soils are often hypersaline (16 experiments). The mean annual rainfall ranged over our sites from 0.3 m in Exmouth, Western Australia to >4 m in Bocas del Toro, Panamá. We have previously shown that nutrient enrichment increases growth over our wide range of sites (12, 14–19, Table S1) and here we show that tree mortality is also increased with nutrient enrichment.
There was a marked difference between tree mortality in seaward fringing forests and landward scrub forests. No unfertilized trees died within the fringe forest where soil water salinities were similar to those of ambient tidal flood water, averaging 37 ppt over all sites. Similarly, no tree death occurred in response to nutrient additions in fringe forests. In contrast, death of unfertilized trees occurred in landward scrub forests where hypersaline soils developed at sites with low annual rainfall (Fig S1). This pattern was accentuated by the addition of nitrogen (N) fertilizer, but not P fertilizer, with tree death occurring in response to N fertilization at sites with lower annual rainfall than other treatments. Time since fertilization began (years of experimental fertilization) did not influence tree mortality. Over all scrub forest sites, tree death was significantly correlated with the mean salinity of soil porewater (Kendall's Tau 0.36, P = 0.035) and negatively correlated with mean annual rainfall (Kendall's Tau −0.34, P = 0.039). The probability of tree survival was significantly lower in N-fertilized than P-fertilized or unfertilized trees (Fig S2).
A period of particularly low rainfall on the Queensland coast of northeastern Australia provided an opportunity to assess canopy loss over treatments in scrub forests at two sites differing in soil water salinity (Fig S3). Soil water salinity was higher at the Port Douglas site than at the Hinchinbrook Channel site (75 vs 60 ppt). Fertilization with N resulted in greater canopy loss than observed in unfertilized trees or those fertilized with P. Averaged over all treatments, trees at the more saline Port Douglas site suffered greater canopy loss than those at the less saline Hinchinbrook Channel site (35% compared to 22%).
Our results show that increasing nutrient availability introduces an instability into mangrove forests that lowers their resilience to environmental variability. The instability arises because nutrients, particularly N, stimulate growth of shoots relative to roots , thereby enhancing productivity during favourable periods but increasing vulnerability to water stress during drought. Such stress can become acute where hypersaline soils limit the capacity for water uptake by roots while low atmospheric humidity exacerbates rates of water loss by shoots during periods of low rainfall. Consequently, N fertilization enhances the probability of tree death along natural gradients of increasing soil salinity and aridity. The emergence of this pattern across species and biogeographic regions underscores the importance of climatic interactions with the intertidal landscape in determining how coastal ecosystems respond to eutrophication.
At many sites fertilization with N enhanced tree mortality in scrub forests, but not in seaward fringing forests. These differences are probably associated with tidally maintained differences in sediment salinities and their variability. Seaward fringing forests are buffered against negative impacts of nutrient enrichment because regular tidal inundation limits development of hypersalinity in sediment pore water. In contrast, scrub forests are less frequently inundated by tides, and, consequently, are exposed to highly variable soil salinities which can become hypersaline during dry periods. Under these conditions, fertilization with N was associated with increased mortality (Fig S1). There was however variation in the influence of fertilizer among sites indicating factors other than nutrient enrichment may enhance mortality of mangroves during droughts or ameliorate the effects of drought. For example, high mortality in Exmouth in unfertilized and fertilized trees occurred in 2003 during a period of reduced tidal inundation, while low mortality in scrub forests in Belize may attributed to the high frequency and intensity of inundation in scrub forests.
Trees fertilized with P tended to have higher probability of survival than those fertilized with N (Fig S2) and had lower levels of canopy loss than N fertilized trees or unfertilized trees (Fig S3). Positive effects of P fertilizer on the water status of leaves, photosynthetic water use efficiency ,  and hydraulic conductivity of stems , ,  indicate that added P improves water relations under high salinity conditions, possibly through enhancing aquaporin function . Thus, the ratio of available N:P at sites enriched in nutrients may be important in determining the resilience of mangrove forests to periodic development of hypersaline soils under conditions of low rainfall and humidity.
Although our study focused on mangroves, our results may be applicable to other ecosystems where anthropogenic nutrient enrichment occurs. For example, McCarthy et al (2006)  reported no adverse effects of N fertilisation on forest trees at a site where hot summers were accompanied by high rainfall and high humidity. In contrast, the combination of high N availability followed by drought has been observed to decrease productivity and enhance mortality in tree species of genera as distantly related as Pinus  and Eucalyptus . Similarly, van Herwaarden et al. (1998)  found that application of high levels of N fertilizers negatively affected the productivity and yield of wheat when rainfall was low. The occurrence of similar results in such disparate systems is strong evidence of a general pattern in plant responses to high levels of N fertilization with variation in drought.
Fluctuations in rainfall have been strongly implicated in causing changes in ecosystem state from mangrove to salt marsh and cyanobacterial mat in arid regions ,  indicating that die-back of mangroves may be a frequent feature of arid zone intertidal landscapes. Given the strength of the effect of N enrichment on mortality of mangroves, die-back of mangrove forests would be expected to occur in coastal areas that receive large nutrient influxes from anthropogenic sources and are subject to pronounced seasonal or inter-annual variation in rainfall and evaporative demand. Indeed, die-back of mangrove forests has been documented in many regions affected by runoff from agricultural lands. Although these dieback events have often been attributed to a wide range of causes, including herbicides  and diseases , they have also been associated with periods of low rainfall  and high soil salinity . Our study shows that the die-back could also be exacerbated by adverse effects of growth under high nutrient availability on the capacity to survive development of hypersaline soil conditions during droughts.
Our results indicating enhanced instability with coastal eutrophication has far reaching consequences for many aspects of mangrove ecosystem function under contemporary and future climatic conditions. Enhancement of canopy loss and tree mortality would reduce benefits to productivity due to increasing atmospheric [CO2], and increase the disparity in productivity between fringe and scrub forest types, with the differences becoming greater along climatic gradients of increasing aridity. This could have greater relative impact on ecosystem function in drier areas where scrub forests account for the majority of mangrove forest cover .
Our results show that mangroves exposed to high nutrient availability suffer greater mortality during drought, and that nutrient-induced mortality is greater in sites subject to periods of low rainfall, low humidity and high sediment salinity. However, it would be incorrect to assume that fertilization would have no adverse effects in forests where hot summers are accompanied by high rainfall and humidity. Stimulation of shoot growth relative to root growth by fertilization could make these forests more vulnerable to windthrow and waves associated with intense tropical storms and tsunamis . Thus the benefits of increased mangrove growth in response to coastal eutrophication will be offset by the costs of lower resilience of mangrove forests when exposed to increasing aridity and to other disturbances.
Materials and Methods
There were 12 study sites where mangrove trees have been fertilized (Table S1). Within each site x forest zone (seaward fringing or landward scrub) replicate trees (6–9 replicates) were fertilized either annually or biannually by inserting 200–300 g of urea (nitrogen, N) or triple superphosphate (Phosphorus, P) into 30 cm deep holes cored on either side of the main stem of the tree. Holes were then sealed with a portion of the extracted sediment core. Control trees were cored but not fertilized. Tree growth was measured as extension of 5 replicate twigs per tree in sun lit positions in the canopy either biennially or annually for a minimum of 2 years. Fertilization continued after growth measurements were completed. All sites have been fertilized for at least 3 years. Climatic summaries for each site were obtained from within country government meteorological services. Porewater was extracted from soil beneath each tree using a suction device and salinity measured with a handheld refractometer . Mortality of trees was recorded at each site during the course of the experiments. Tree mortality (number of trees that were live and dead) over the fertilization treatments was assessed using logistic regression with salinity and rainfall as covariates in the model. Differences in survivorship among treatments were assessed using logits and Chi squared tests. We assessed correlation between % mortality and characteristics of the sites (rainfall and porewater salinity) and the time since fertilization using Kendall's Tau.
At two of the sites (Hinchinbrook Channel and Port Douglas) we observed significant canopy loss during July 2007 which was associated with a prolonged drought in the region (Australian Bureau of Meteorology 2007). We estimated canopy loss as the proportion of twigs that had died, but were still attached to the tree. The effect of nutrient enrichment on the proportion of canopy loss in 2007 at Hinchinbrook Channel and Port Douglas was assessed using analysis of variance (ANOVA) where site was a random effect in the model and fertilization treatment a fixed effect. Data were log transformed prior to analysis.
Death of experimental mangrove trees over fertilization treatments and over rainfall and porewater salinity gradients. Mortality increased with decreasing average annual rainfall and increasing porewater salinity. Trees fertilized with Nitrogen (+Nitrogen) show more pronounced sensitivity to rainfall and salinity than trees fertilized with phosphorus (+Phosphorus) and trees that were not fertilized (Control). Analysis by logistic regression indicates significant fertilization treatment x rainfall (P = 0.001) and fertilization treatment x salinity (P = 0.025) effects on mortality.
(8.71 MB TIF)
The probability of survivorship of scrub mangrove trees over fertilization treatments. The probability of survivorship was lower with nitrogen fertilization (+N) than for unfertilized control trees (C) or for phosphorus fertilized (+P) trees (χ2 = 9.73, P = 0.008). Bars are means and standard errors. Data are means and standard errors from 16 scrub mangrove sites. Different letters above the bars indicate significantly different means (P<0.05).
(7.93 MB TIF)
Canopy loss during a drought episode over fertilization treatments at two mangrove sites in north Queensland. Canopy loss was greater in nitrogen fertilized trees (+N) compared to phosphorus fertilized (+P) or unfertilized control trees (C) trees at two sites with varying salinities; Port Douglas (closed bars, 75 ppt) and Hinchinbrook Channel (hatched bars, 60 ppt) in north Queensland. Fertilizer treatment effect was significant (F2,2 = 73.06, P = 0.0135). Canopy loss was significantly greater at Port Douglas than Hinchinbrook Channel (F1,45 = 6.01, P = 0.018). Data are means and standard errors for 9 trees per treatment at each site.
(8.70 MB TIF)
Locations and characteristics of mangrove fertilization experimental sites. Locations and characteristics of mangrove fertilization experimental sites, including climatic variables (average annual temperature and rainfall), tidal range, soil type, species included in the experiment, forest type (seaward fringe or scrub forest), canopy height, number of trees included in each experiment and duration of the experimental observations. Human influences on the site are also indicated and the nutrient that limits growth with the magnitude of the growth enhancement above non-fertilized controls appears in parenthesis.
(0.06 MB DOC)
The authors thank Eldon Ball, Graham Farquhar, Kyra Hay, Ove Hoegh-Guldberg, John Pandolfi and Ruth Reef for thoughtful comments on the manuscript and Simon Blomberg for statistical advice.
Conceived and designed the experiments: CEL MCB KCM ICF. Performed the experiments: CEL MCB KCM ICF. Analyzed the data: CEL MCB KCM ICF. Contributed reagents/materials/analysis tools: CEL MCB ICF. Wrote the paper: CEL MCB KCM ICF.
- 1. Downing JA, McClain M, Twilley R, Melack JM, Elser J, et al. (1999) The impact of accelerating land-use change on the N-Cycle of tropical aquatic ecosystems: Current conditions and projected changes. Biogeochemistry 46: 109–148.
- 2. Cloern JE (2001) Our evolving conceptual model of the coastal eutrophication problem. Mar Ecol Prog Ser 210: 223–253.
- 3. Paerl HW (1997) Coastal eutrophication and harmful algal blooms: Importance of atmospheric deposition and groundwater as “new” nitrogen and other nutrient sources. Limnol Oceanogr 42: 1154–11.
- 4. Lapointe BE (1997) Nutrient thresholds for bottom-up control of macroalgal blooms on coral reefs in Jamaica and southeast Florida. Limnol Oceanogr 42: 1119–1131.
- 5. Hughes TP, Baird AH, Bellwood DR, Card M, Connolly SR, et al. (2003) Climate change, human impacts and the resilience of coral reefs. Science 301: 929–933.
- 6. Levine J, Brewer S, Bertness M (1998) Nutrients, competition and plant zonation in a New England salt marsh. J Ecol 86: 285–292.
- 7. Scheffer M, Carpenter S, Foley JA, Folke C, Walker B (2001) Catastrophic shifts in ecosystems. Nature 413: 591–596.
- 8. Rabalais NN, Turner RE, Wiseman WJ Jr. (2002) Gulf of Mexico hypoxia, AKA “the dead zone”. Ann Rev Ecol Syst 33: 235–263.
- 9. Boto KG, Wellington JT (1983) Nitrogen and phosphorus nutritional status of a northern Australian mangrove forest. Mar Ecol Prog Ser 11: 63–69.
- 10. Naidoo G (1987) Effects of salinity and nitrogen on growth and relations in the mangrove, Avicennia marina (Forsk.) Vierh.. New Phytol 107: 317–325.
- 11. Lin G, Sternberg LSL (1992) Effect of growth form, salinity, nutrient and sulfide on photosynthesis, carbon isotope discrimination and growth of red mangrove (Rhizophora mangle L.). Aust J Plant Physiol 19: 509–517.
- 12. Feller IC (1995) Effects of nutrient enrichment on growth and herbivory of dwarf red mangrove (Rhizophora mangle). Ecol Monogr 65: 477–505.
- 13. Koch MS (1997) Rhizophora mangle L. seedling development into the sapling stage across resource and stress gradients in subtropical Florida. Biotropica 29: 427–439.
- 14. Feller IC, Whigham DF, McKee KL, O'Neill JP (2002) Nitrogen vs. phosphorus limitation across an ecotonal gradient in a mangrove forest. Biogeochemistry 62: 145–175.
- 15. Feller IC, Whigham DF, McKee KL, Lovelock CE (2003) Nitrogen limitation of growth and nutrient dynamics in a mangrove forest, Indian River Lagoon, Florida. Oecologia 134: 405–414.
- 16. Lovelock CE, Feller IC, McKee KL, Engelbrecht BM, Ball MC (2004) The effect of nutrient enrichment on growth, photosynthesis and hydraulic conductance of dwarf mangroves in Panama. Funct Ecol 18: 25–33.
- 17. Lovelock CE, Feller IC, Ellis J, Hancock N, Schwarz AM, et al. (2007) Mangrove growth in New Zealand estuaries: The role of nutrient enrichment at sites with contrasting rates of sedimentation. Oecologia 153: 633–641.
- 18. Lovelock CE, Feller IC, Ball MC, Ellis J, Sorrell B (2007) Testing the Growth Rate vs. Geochemical Hypothesis for latitudinal variation in plant nutrients. Ecology Letters 10: 1154–1163.
- 19. Martin KC (2007) Interactive effects of salinity and nutrients on mangrove physiology: implications for mangrove forest structure and function. PhD thesis, Australian National University, Canberra.
- 20. Clough BF, Boto KG, Attiwill PM (1983) Mangrove and sewage: a re-evaluation. In: Teas HJ, editor. Biology and Ecology of Mangroves. Tasks for Vegetation Science Series, Vol. 8. Lancaster, UK: Dr W Junk Publishers. pp. 151–162.
- 21. Robertson AI, Phillips MJ (1995) Mangroves as filters of shrimp farm effluent: Predictions and biogeochemical research needs. Hydrobiologia 293: 311–319.
- 22. Wong YS, Tam NFY, Lan CY (2007) Mangrove wetlands as wastewater treatment facility: a field trial. Hydrobiologia 352: 49–59.
- 23. Grime JP (1979) Plant Strategies and Vegetation Processes, Wiley, Chichester, USA
- 24. Chapin FS (1980) The mineral nutrition of wild plants. Ann Rev Ecol Syst 11: 233–260.
- 25. Tilman D (1991) Relative growth rate and plant allocation patterns. Am Nat 138: 1269–1275.
- 26. Lambers H, Poorter H (1992) Inherent variation in growth rate between plants: a search for physiological causes and ecological consequences. Adv Ecol Res 23: 187–261.
- 27. Chapin FS (1991) Integrated responses of plants to stress: a centralized system of physiological responses, BioScience 41: 29–36.
- 28. Ball MC (1988) Salinity tolerance in the mangroves, Aegiceras corniculatum and Avicennia marina. I. Water use in relation to growth, carbon partitioning and salt balance. Austr J Plant Physiol 15: 447–464.
- 29. Ball MC, Cochrane MJ, Rawson HM (1997) Growth and water use of the mangroves, Rhizophora apiculata and R. stylosa, in response to salinity and humidity under ambient and elevated concentrations of atmospheric CO2. Plant Cell Environ 20: 1158–1166.
- 30. Naidoo G (2009) Differential effects of nitrogen and phosphorus enrichment on growth of dwarf Avicennia marina mangroves. Aqua Bot 90: 184–190.
- 31. Lovelock CE, Feller IC (2003) Photosynthetic performance and resource utilization of two mangrove species coexisting in hypersaline scrub forest. Oecologia 134: 455–462.
- 32. Lovelock CE, Ball MC, Choat B, Engelbrecht BMJ, Holbrook NM, et al. (2006) Linking physiological processes with mangrove forest structure: Phosphorus deficiency limits canopy development, hydraulic conductivity and photosynthetic carbon gain in dwarf Rhizophora mangle. Plant Cell Environ 29: 793–802.
- 33. Lovelock CE, Ball MC, Feller IC, Engelbrecht BMJ, Ewe ML (2006) Plant function in nitrogen and phosphorus limited mangrove ecosystems. New Phytol 172: 514–522.
- 34. Clarkson DT, Carvajal M, Henzler T, Waterhouse RN, Smyth AJ, et al. (2000) Root hydraulic conductance: diurnal aquaporin expression and the effects of nutrient stress. J Exp Bot 51: 61–70.
- 35. McCarthy HR, Oren R, Finzi AC, Johnsen KH (2006) Canopy leaf area constrains [CO2]-induced enhancement of productivity and partitioning among aboveground carbon pools. Proc Natl Acad Sci 103: 19356–19361.
- 36. Linder ML, Benson B, Myers J, Raison RJ (1987) Canopy dynamics and growth of Pinus radiata. I. Effects of irrigation and fertilization during a drought. Can J For Res 17: 1157–1165.
- 37. Atwell B, Henery ML, Ball MC (2009) Does soil nitrogen influence growth, water transport and survival of snow gum (Eucalyptus pauciflora Sieber ex Sprengel.) under CO2 enrichment? Plant Cell Environ 32: 553–566.
- 38. Van Herwaarden AF, Farquhar GD, Angus JF, Richards RA, Howe GH (1998) ‘Haying-off’, the negative grain yield response of dryland wheat to nitrogen fertilizer I. Biomass, grain yield, and water use. Aust J Agric Res 49: 1067–1081.
- 39. Fosberg FR (1961) Vegetation-free zone on dry mangrove coastline. U S Geol Soc Prof Pap 424 (D): 216–218.
- 40. Cintron G, Lugo AE, Pool DJ (1978) Mangrove of arid environments in Puerto Rico and adjacent islands. Biotropica 10: 110–121.
- 41. Duke NC, Bell AM, Pedersen DK, Roelfsema CM, Bengston-Nash S (2005) Herbicides implicated as the cause of severe mangrove dieback in the Mackay region, NE Australia — serious implications for marine plant habitats of the GBR World Heritage Area. Mar Poll Bull 51: 308–324.
- 42. Wier AM, Tattar TA, Klekowski EJ (2000) Disease of red mangrove (Rhizophora mangle) in Southwest Puerto Rico caused by Cytospora rhizophorae. Biotropica 32: 299–306.
- 43. Jupiter SD, Potts DM, Phinn SR, Duke NC (2007) Natural and anthropogenic changes to mangrove distributions in the Pioneer River Estuary (QLD, Australia). Wetl Ecol Manag 15: 51–62.
- 44. Armentano TV (1995) Analysis of pattern and possible causes of die-back of Florida Bay keys mangroves. Report to the U. S. Army Corps of Engineers National Park Service, Everglades National Park.
- 45. Alongi DM (2008) Mangrove forests: resilience, protection from tsunamis, and responses to global climate change. Estuar Coast Shelf Sci 76: 1–13.