Spatiotemporal dynamics of Phormidium cover and anatoxin concentrations in eight New Zealand rivers with contrasting nutrient and flow regimes
Graphical abstract
Introduction
Toxic planktonic cyanobacterial blooms have been described for many decades (Francis, 1878, Hallegraeff, 1993, Wynne et al., 2010). This has resulted in many studies investigating the physicochemical factors influencing the growth and proliferation of planktonic species (see Oliver et al., 2012). In contrast, toxic benthic proliferations are a more recent but escalating problem in freshwater environments worldwide (Quiblier et al., 2013). Toxin-producing species belonging to the genus Phormidium, have a cosmopolitan distribution and are found in a variety of benthic habitats including eutrophic and oligotrophic lakes, many ponds and in oligotrophic lotic systems (Komárek, 1999, Mez et al., 1998, Puschner et al., 2008, Wood et al., 2012a, Wood et al., 2016). In New Zealand's cobble-bed rivers, Phormidium proliferations (defined as > 20% cover) have become increasingly prevalent over the last decade and have been identified in > 100 rivers nationwide (McAllister et al., 2016). A recent study of wadeable streams in California (USA) also detected Phormidium species in over a third of sites sampled (Fetscher et al., 2015).
Although benthic algae, including cyanobacteria, are an important component of river ecosystems, excessive biomass and dominance by toxin-producing species can be problematic. Knowledge on the drivers of such proliferations is essential for effective management. Light, nutrient supply, hydrological regime and temperature are thought to be the most important abiotic factors determining benthic algal biomass accrual dynamics in lotic systems (Bowman et al., 2007, Clausen and Biggs, 1997, Dodds et al., 2002, Schiller et al., 2007). Non-linear responses to these variables and interactions can be expected, for example Biggs et al. (1998) suggested that, under oligotrophic conditions, increasing flow enhances accrual by increasing the flux of nutrients to the algae by reducing the thickness of the diffusive boundary layer, but that this effect is less evident at high ambient nutrient concentrations (Biggs et al., 1998, Biggs and Hickey, 1994, Borchardt, 1994). As velocity increases, however, greater shear stress can lead to enhanced sloughing and biomass reduction regardless of nutrient status (Biggs and Thomsen, 1995). Phormidium-dominated (hereafter referred to as Phormidium) mats, however, are functionally different, due to their thick, cohesive growth form. In particular it has been shown that biogeochemical conditions, including nutrient concentrations, within the mats can be very different from the overlying water (Wood et al., 2015) and it is thus uncertain whether water-derived nutrients that promote periphyton growth in general are equally relevant to Phormidium mats.
Only recently has research begun to explore relationships between physicochemical variables and Phormidium dynamics. Schneider (2015) found, in a long-term study of two Norwegian rivers, that Phormidium cover increased with rising annual temperatures, but decreased with prolonged high flow events. Similarly, Heath et al. (2011) observed highest Phormidium cover at sites in the Hutt River (New Zealand), when river flows were stable and water temperatures elevated. Heath et al. (2011) noted that there was no correlation between water-column nutrients and Phormidium cover, although nutrient samples were only collected monthly. In contrast, in a multi-site investigation Wood et al. (2017) indicated that Phormidium proliferations predominantly occurred at sites with low dissolved reactive phosphorus (DRP; ca. < 0.005 mg L− 1) and dissolved inorganic nitrogen (DIN) concentrations of < 0.8 mg L− 1. They also found that increasing conductivity and temperature were associated with higher Phormidium cover.
Species of Phormidium are known to produce a variety of cyanotoxins including; anatoxin-a (ATX), homoanatoxin-a (HTX), dihydro-anatoxin-a (dhATX) and dihydro-homoanatoxin-a (dhHTX; Faassen et al., 2012, Fetscher et al., 2015, Gugger et al., 2005, Heath et al., 2010, Wood et al., 2007). Phormidium has been implicated in animal toxicosis events in France, Netherlands, New Zealand, Scotland, Switzerland and the United States of America (Edwards et al., 1992, Faassen et al., 2012, Gugger et al., 2005, Hamill, 2001, Mez et al., 1997, Puschner et al., 2008, Wood et al., 2007). The effects of Phormidium proliferations on ecosystem health and aquatic organisms are largely unknown. In Spanish streams, toxic benthic cyanobacterial mats have been associated with decreased macroinvertebrate diversity (Aboal et al., 2000, Aboal et al., 2002). Despite the apparent health risk and potential ecosystem effects, few studies have investigated relationships between toxin concentrations within mats and physicochemical variables.
Phormidium mats vary substantially in their toxin content both spatially and temporally (Heath et al., 2010, Wood et al., 2007). This variability is likely to be at least partially explained by the presence of both non-toxic and toxic genotypes, which co-occur within mats (Wood et al., 2010, Wood et al., 2012b). Wood et al. (2012b) found that anatoxin concentrations varied 100-fold between Phormidium strains isolated from the same mat and that there were differences in the 16S ribosomal RNA gene sequences of at least 17 nucleotides between anatoxin and non-anatoxin producing strains. The abundance of different strains may therefore also influence the toxin content of a mat. Both observational-based and culture-based laboratory studies have begun to explore factors that regulate the abundance of non-toxic and toxic strains and the amount of anatoxin produced. Heath et al. (2011) suggested temperatures exceeding 13.4 °C favor toxic over non-toxic strains. However, subsequent studies have found no relationship between water temperature and toxin concentrations (Wood et al., 2017). They suggested that understanding the relationships between physicochemical factors and toxin concentrations could be enhanced through the development of molecular techniques, which would allow differentiation of toxic and non-toxic strains.
In the present study we surveyed Phormidium weekly for 30 weeks, at a single site on eight rivers in the Canterbury region in New Zealand. The aims were to describe the spatial and temporal variability of Phormidium cover and total anatoxin concentrations and to investigate relationships between physicochemical variables and Phormidium cover and total anatoxin concentrations in the mats. The following hypotheses were tested: (1) water chemistry plays a significant role in determining Phormidium cover; (2) as river flow decreases, Phormidium cover will increase; and (3) there will be no relationships between total anatoxin concentrations in the mats and measured physicochemical variables.
Section snippets
Study area and site selection
All study sites were located within the Canterbury region in the South Island of New Zealand, where catchment land-use is dominated by intensive agriculture (Table 1). A single site on each of the eight predominately gravel and cobble-bed rivers were chosen for this study (Fig. 1). Sites were selected to include areas with and without historical records of Phormidium proliferations. Catchment land-use data were retrieved from the Freshwater Ecosystems of New Zealand (FENZ) database (Leathwick
Spatiotemporal variability in Phormidium cover, and physicochemical characteristics
The dominant species within mats was microscopically identified as Phormidium autumnale, and is hereafter referred to as “Phormidium”. It was present at all sites, but cover was highly variable among sites (Fig. 2). Five sites (Site 2, 5, 6, 7 and 8) exceeded the “bloom” threshold with a mean Phormidium cover of > 20% on multiple occasions, but the timing and persistence of highest cover varied among sites. Cover at Sites 2, 3, 7 and 8 in early summer (November to January), whereas at Sites 1,
Relationships among nutrients and flow and Phormidium cover
Field investigations (Heath et al., 2011, Wood et al., 2015, Wood et al., 2017) and laboratory culture experiments (Heath et al., 2014, Heath et al., 2016) have begun to explore the roles of physicochemical parameters in determining Phormidium cover and toxin production. These studies have generally examined a limited number of variables in only a few rivers and have lacked high temporal resolution. They have highlighted the importance of sediment, water temperature, chemistry and river flow as
Conclusion
This study demonstrates that in the absence of varied hydrology, Phormidium accrual was influenced by the site, time of the year and water chemistry. There was a strong, positive relationship between conductivity and Phormidium cover. Further investigations into what specific ions underpin this relationship are required. There was no effect of accrual DIN between 0.02 and 0.2 mg L− 1, however, a strong, negative influence was observed beyond this, though this depends on data from a single site.
Acknowledgements
The authors thank Renan Thiesen, Olivia Rowley, Oliver Gooday, Ruigu Mwagi, Emma Mackenzie, Channell Thoms, Anna Henderson, Matthew Fraser, Sarah Hayman, Phil Clunies-Ross, Peter Taylor and Simona Seizova for assisting with the collection of this data. The authors acknowledge Andrew Neverman (Massey University) for measuring substrate stability, using his bed stability index (paper in prep.), at each of our sites. TGM thanks Environment Canterbury and the Ngāi Tahu Research Centre for funding.
References (63)
- et al.
Identification of anatoxin-a in benthic cyanobacteria (blue-green algae) and in associated dog poisonings at Loch Insh, Scotland
Toxicon
(1992) - et al.
First report of (homo) anatoxin-a and dog neurotoxicosis after ingestion of benthic cyanobacteria in The Netherlands
Toxicon
(2012) - et al.
Wadeable streams as widespread sources of benthic cyanotoxins in California, USA
Harmful Algae
(2015) - et al.
First report in a river in France of the benthic cyanobacterium Phormidium favosum producing anatoxin-a associated with dog neurotoxicosis
Toxicon
(2005) - et al.
Effects of nitrogen and phosphorus on anatoxin-a, homoanatoxin-a, dihydroanatoxin-a and dihydrohomoanatoxin-a production by Phormidium autumnale
Toxicon
(2014) - et al.
The rise of toxic benthic Phormidium proliferations: a review of their taxonomy, distribution, toxin content and factors regulating prevalence and increased severity
Harmful Algae
(2016) Greener rivers in a changing climate?—effects of climate and hydrological regime on benthic algal assemblages in pristine streams
Limnologica-Ecol. Manag. Inland Waters
(2015)- et al.
First report of homoanatoxin-a and associated dog neurotoxicosis in New Zealand
Toxicon
(2007) - et al.
Development of solid phase adsorption toxin tracking (SPATT) for monitoring anatoxin-a and homoanatoxin-a in river water
Chemosphere
(2011) - et al.
Relationship between macroinvertebrate diversity and toxicity of Cyanophyceae (cyanobacteria) in some streams from eastern Spain
Proc. Int. Assoc. Theor. Appl. Limnol.
(2000)