Predicting and Managing the Mass Occurrence of Lyngbya sensu lato in Marine and Freshwater Environments: Current Knowledge, Challenges, and Opportunities

Abstract

Significant cyanobacterial proliferations dominated by Lyngbya have been increasingly reported since the 2000s, posing environmental, economic, and human-health risks. This review synthesizes their distribution, predictors, toxicity, and management strategies of all identified Lyngbya species, including species historically classified as Lyngbya despite later taxonomic changes. Research has focused mainly on Lyngbya majuscula and Lyngbya wollei. For L. majuscula, bloom initiation is driven by proximate abiotic factors such as nutrients, light, and temperature; while broader conditions, including bottom currents, sediment nutrients, rainfall, and land use, set the stage for proliferation. Toxin production appears related to nutrient levels and temperature, although mechanisms remain poorly understood. Management of L. wollei commonly relies on copper-based chelated algaecides, despite their risks to non-target organisms, highlighting the need for more sustainable tools used in managing other cyanobacteria. Existing predictive models for Lyngbya proliferation show limited accuracy, partly due to insufficient in situ data. This review argues that novel monitoring approaches could provide the data needed to strengthen predictive models, also offering insights into a new modeling approach, supporting more proactive and effective Lyngbya bloom management. It is particularly valuable for research in water resource management and environmental science, as it synthesizes current knowledge essential for advancing management strategies.

1. Introduction

Cyanobacteria, also known as blue-green algae, are a concern for governments and water utilities because they can lead to harmful effects on water resources. These effects become more significant during cyanobacteria harmful bloom (cyanoHABs) events, which are increasing worldwide, in both freshwater and marine systems [1]. It is expected that these events will increase in both frequency and magnitude in the future [2,3], because cyanoHABs events are influenced by factors associated with climate change, such as global warming and altered precipitation patterns [4,5]. Additionally, the increased nutrients associated with urban, industrial, and agricultural development are other reasons for the worldwide increase in these blooms [6].

One genus that has raised growing concern in recent decades and has been observed with increasing frequency since its first identification is Lyngbya. Lyngbya spp., belonging to the family Microcoleaceae, have undergone recent taxonomic revisions based on molecular data [7]. They are prokaryotic, photosynthetic cyanobacteria that form unbranched filaments of cells covered by a protective sheath. They prefer moderate temperatures and do not possess heterocysts to fix nitrogen [8]; however, they convert atmospheric nitrogen into ammonia, a bioavailable form of nitrogen [9]. Lyngbya is primarily a benthic cyanobacterium, that can form dense biphasic mats [10], growing on the bottom surfaces of aquatic environments, such as sediments, rocks, or submerged vegetation [9,11]. This benthic nature allows Lyngbya to access nutrients from sediments and anchor itself to the substrate, contributing to its persistence and ability to form extensive proliferations. Lyngbya can also form floating mats [10]. Those mats alter food web structure, cause asphyxia from microbial decomposition of biomass, displace desirable plants and algal species, and are visually unpleasant [10,12]. Importantly, they can produce a broad spectrum of cyanotoxins. These toxins include hepatotoxins such as cylindrospermopsins, neurotoxins such as saxitoxins, and dermatotoxins such as lyngbyatoxin-A [13,14]. Cylindrospermopsins typically affect liver function [15,16], while saxitoxins can affect multiple systems in the human body [17]. Lyngbyatoxin-A (LA) is responsible for acute dermatologic reactions in humans, as well as eye and respiratory irritation [18].

Lyngbya species, including L. majuscula, L. wollei, L. bouillonii, L. putealis, and L. magnifica, occur in a variety of aquatic environments, such as estuaries, embayments, freshwater systems, and reefs [19,20,21]. Each Lyngbya species has its own preferred habitat. Proliferations of L. majuscula usually occur in marine or estuarine waters [19], whereas in the papers analyzed in this review, L. bouillonii blooms were reported exclusively in marine waters [22]. In contrast, L. wollei proliferations are common in freshwater environments such as rivers, lakes, ponds, and reservoirs [20,23], as are L. putealis [24] and L. magnifica [12]. These species reproduce asexually through hormogonia, small fragments of filaments that can move, settle, and grow into new filaments [25,26]. Lyngbya proliferations are often associated with nutrient enrichment and environmental conditions such as elevated water temperatures and calm weather [27,28].

Additionally, Lyngbya blooms pose a threat to water reservoirs [29,30], disrupting both ecological functions and human recreational activities, such as fishing, boating, and swimming. Potable water use, industrial use, and irrigation are also affected [30,31]. Lyngbya blooms also have economic impacts. When L. majuscula detaches from the benthos, its biomass frequently deposits on foreshore beaches, disrupting tourism activities and requiring government agencies to allocate significant resources for its removal and disposal [32]. Steffensen [33] stated that Australia spends 180–240 million Australian dollars each year because of cyanoHABs. These costs include management and additional water-treatment expenses, along with financial losses for farmers who depend on river or reservoir water for agriculture. The estimate also accounts for broader economic impacts on tourism, recreation, and the environment. Recent events in subtropical Australia further highlight the growing socio-ecological significance of Lyngbya proliferations. In Moreton Bay, Queensland, extensive Lyngbya blooms have been linked to sediment erosion and nutrient mobilization, with reported impacts on oyster aquaculture and coastal ecosystems, prompting management concern at state government and industry levels [34]. In Canada, Smith et al. [35] projected that, if no action is taken, algal blooms will cost approximately $272 million per year (in 2015 Canadian dollars) over the next 30 years.

Besides Australia and Canada, cyanoHABs, including Lyngbya spp., are recognized as a global issue affecting freshwater and coastal systems worldwide [36]. In the United States, L. wollei blooms in reservoirs have resulted in the degradation of water quality, impairment of recreational activities, and the production of cyanotoxins. These blooms also require intensive, multi-year management programs involving repeated treatments and monitoring, representing substantial long-term management costs [31]. Similar impacts have been reported in Europe (e.g., Spain [37]) and Asia (e.g., Saudi Arabia [38]), where Lyngbya blooms affect water quality and ecosystem functioning, with potential implications for socio-economic activities. Management strategies have been implemented to address Lyngbya issues, such as the use of algaecides [30,31] and the development of models to understand/predict bloom events [39]. According to Rousso et al. [40], the selection of management strategies for both long-term mitigation and short-term response to risky events should rely on precise tools such as process-based and data-driven models, as they can enhance management effectiveness while minimizing the risks and costs associated with cyanoHABs. Effective management of Lyngbya is particularly important because of its wide-ranging impacts across multiple domains. This genus produces toxins with significant human health implications, such as dermatitis and respiratory irritation; its benthic, mat-forming nature makes monitoring especially challenging and requires approaches different from those used for planktonic cyanobacteria; it causes substantial structural damage to ecosystems, smothering seagrass, altering ecosystem structure, reducing fish abundance; and its blooms can be persistent. Together, these consequences mean that Lyngbya generates multiple categories of hazards simultaneously, namely ecological, health, economic, management, and social: Making it a high-risk cyanobacterial genus.

To ensure more accurate prediction and effective management of Lyngbya proliferation events, it is essential to gain a comprehensive understanding of risks and adaptation strategies based on English language available published scientific evidence. Therefore, this review synthesizes current scientific knowledge on the characteristics, drivers, risks, detection, prediction, and management of Lyngbya blooms. By integrating these aspects, the paper aims to support the development of more effective monitoring, forecasting, and mitigation strategies for Lyngbya blooms. The primary discussion topics are delineated across six sections: (i) data on relevant publications and their location, (ii) bloom predictors, (iii) toxin production, (iv) monitoring Lyngbya, (v) modeling approaches, (vi) Lyngbya control management.

Because Lyngbya taxonomy is complex and has undergone multiple reclassifications over the years, for instance, some studies suggest that L. majuscula may belong to the genus Moorena [41]. For the purposes of this study, we refer to this species as Lyngbya, following the taxonomic usage in the analyzed papers. The updated classification names are presented in

Table 1. Summary of taxonomic revisions of Lyngbya species, highlighting previously reported names, currently accepted nomenclature, taxonomic status (including reclassification), and supporting references.

Previously Reported Name (Lyngbya)	Current Accepted Name	Taxonomic Status/Notes	References
Lyngbya majuscula Harvey ex Gomont (1892)	Lyngbya majuscula (currently accepted, though often reassigned to Moorea in recent literature)	Still formally accepted, but widely discussed as Moorea in modern taxonomy	[42,43,44]
Lyngbya wollei (Farlow ex Gomont) Speziale & Dyck (1992)	Microseira wollei (Farlow ex Gomont) G.B. McGregor & Sendall ex Kenins	Reclassified to Microseira	[45]
Lyngbya putealis Montagne ex Gomont (1892)	Phormidium puteale (Montagne ex Gomont) Anagnostidis & Komárek	Reclassified to Phormidium; Lyngbya name is now a synonym	[46]
Lyngbya bouillonii Hoffmann & Demoulin (1991, nom. inval.)	Moorena bouillonii (Hoffmann & Demoulin) Engene & Tronholm	Transferred to Moorena; original name invalid	[41]
Lyngbya magnifica Gardner (1927)	Microseira wollei (Farlow ex Gomont) G.B. McGregor & Sendall ex Kenins	Reclassified to Microseira	[45]

.

2. Materials and Methods

The process of selecting the papers included in this review begins by defining which academic databases (e.g., ScienceDirect, Scopus) are used to locate relevant papers. Next, search criteria are formulated, including inclusion and exclusion criteria (e.g., keywords). These criteria guide the selection of papers, after which the data of interest are extracted, and the findings summarized. The final sample is determined based on the authors’ criteria outlined in Section 2.1, developed to address the specific aims of the review. Figure 1 provides an overview of the review design process.

2.1. Data Source and Search Criteria

Before identifying the data source, five topics were defined to structure the content of the review (Figure 1). Subsequently, a search for papers on Lyngbya was conducted up to early 2024 using Science Direct, Springer and Scopus databases. The search was conducted similarly across all databases, using the keywords: Lyngbya, Blue-Green Algae and Benthic cyanobacteria. These general terms were chosen to capture as many Lyngbya-related papers as possible. Each term was used first individually, then in combination with Boolean operators AND and OR. Based on these five final topics, the inclusion criteria were established. The first inclusion criterion required papers to focus on Lyngbya in the context of freshwater or marine environments, including transitional estuarine waters. Papers with other focuses, such as investigations of chemical and biological properties, or pharmaceutical applications, were excluded. This inclusion criterion was applied primarily during the title evaluation stage. The selection of papers written in English served as the second inclusion criterion; as such, papers written in languages other than English were excluded from the analysis. However, non-English papers whose abstracts were provided in English are listed in the Supplementary Material [47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79]. No restrictions were applied regarding the publication year of the papers; nevertheless, as noted above, the selected papers extend through early 2024. The next phase consisted of evaluating the general content of the papers by reviewing their abstracts. To be considered, the papers needed to address the causes, consequences, or different approaches to managing the Lyngbya issue. This phase was responsible for removing a large number of papers, leaving 69 for the next phase. Finally, the remaining papers were evaluated in depth through a full-text reading, and 18 were excluded because they did not provide relevant information regarding the five topics established at the beginning of the analysis. After that, the papers were categorized into their respective topics. Therefore, the analysis for selecting the papers was done following three subsequent screening phases: (i) title, (ii) abstract, (iii) full text. A total of 51 papers were analyzed (Figure 1), and their findings were discussed in this review within the relevant topic. Additionally, other papers were also referenced here to supplement or validate the information.

2.2. Data Extraction and Summary

A qualitative assessment of study reliability was conducted based on study design, sample size, and methodological robustness. The relevant information and data from the selected papers were analyzed in depth in this review and clustered into six different topics:

i.

Proximate Factors: The factors that are directly related to Lyngbya bloom.

ii.

Underlying Factors: The factors that enable proximate factors to occur.

iii.

Predominant toxins produced by Lyngbya.

iv.

Different approaches to monitor Lyngbya.

v.

Different modeling approaches applied to Lyngbya issue.

vi.

Approaches for managing Lyngbya control.

These topics were defined to encompass the causes, consequences and strategies for addressing Lyngbya-related issues in marine, freshwater, and estuarine waters. Once the categories were defined and the papers assigned to them, an Excel (Version 2604, Built 19929.20172) file was created for each topic to compile their key findings. The findings were compared, and a discussion was developed based on those comparisons. Furthermore, possible explanations were given for the results found in the literature.

3. Results

3.1. Overview of Analyzed Literature

3.1.1. Chronological Order of Publication

Most of the papers analyzed on Lyngbya focus on L. majuscula and L. wollei: 31 on L. majuscula and 17 on L. wollei. Regarding L. majuscula, the first paper was published in 1985 and discusses toxin production. There are earlier papers focusing on the chemicals extracted from L. majuscula [80], but they fall outside the scope of this review. The papers about L. majuscula became more common during the 2000s, and the peak of publications on this species occurred in 2007, but the number of related papers has plateaued or declined over the last decade. The majority of papers—17 in total—have focused on toxin production: eight addressed L. majuscula, another eight focused on L. wollei, and one on L. bouillonii (Figure 2).

Seventeen papers focused on the species L. wollei, representing a smaller portion compared to L. majuscula. They primarily address bloom predictors, control management, and toxin production. The earliest paper considered here was published in 1994. Although some earlier paper on L. wollei exists, such as its taxonomic description by Speziale and Dyck [10], it falls outside the scope of this review. Papers on L. wollei became more common after 2012 (Figure 2). Besides L. majuscula and L. wollei, there were also papers of L. bouillonii, L. magnifica and L. putealis, but they are less common. This review identified just one paper about L. bouillonii, addressing toxin production, published in 2020 [22]. The species was first documented and formally described in 1991 [81]; however, that earlier paper was not included in this review because its primary focus was taxonomic rather than aligned with the review’s criteria. Concerning L. magnifica and L. putealis, only one study was identified for each of these species, both addressing the control management topic and were published in 2011.

3.1.2. Geographical Distribution of Published Research

The countries are categorized based on the identified Lyngbya species, representing the location of the case study (Figure 3). Australia had the highest number of papers, mainly on L. majuscula (n = 21), followed by L. wollei (n = 2). The United States (USA) ranked second with 16 papers, 13 of which focused on L. wollei; however, L. majuscula (n = 2) and L. magnifica (n = 1) were also identified there. India had three papers on L. majuscula, L. putealis, and L. bouillonii, while Canada had two papers focused on L. wollei. Other countries reported one paper each, all related to L. majuscula. Overall, L. majuscula was the most widely documented species, appearing in papers from eight out of ten countries.

In general, the focus of the papers has been on areas where humans interact with the coastal environment, e.g., fishing, boating and swimming areas (e.g., Moreton Bay, Australia [19]), and drinking water reservoirs, e.g., three different reservoirs located in AL, USA [30].

3.2. Bloom Predictors

3.2.1. Proximate Factors

L. majuscula has a flexible metabolism for nutrient acquisition in shallow benthic environments, enabling adaptation to frequent nutrient changes and rapid growth from nutrient pulses [82], a typical characteristic of cyanobacteria globally [83]. It contains phycoerythrin, which is a photosynthetic pigment that increases rapidly with nutrient addition [82]. Nutrient inputs, particularly iron (Fe), phosphorus (P), and nitrogen (N), enhance its distribution and proliferation [84,85]. In Moreton Bay, Australia, proliferation-period averages were 2.3–7.1 µM Fe, 0.6–0.9 µM P, and 11.4–13.8 µM N [28]. No clear nutrient threshold was defined to avoid proliferation, but these ranges may serve as reference values.

Usually, L. majuscula possesses nitrogen-fixing abilities, meaning that N is a non-limiting factor for growth [84,86]. However, depending on light and oxygen conditions, it may be limited by the availability of dissolved inorganic N; under high-light conditions, though, L. majuscula can acquire N directly from the water column [85]. Nitrogen-fixing cyanobacteria require more Fe and P for nitrogenase synthesis and activation than algae [87], for that reason, they are more vulnerable to low Fe availability, especially when Fe is in its insoluble form [88]. As such, Fe and P have been shown to stimulate N fixation, and these three elements collectively contribute to regulating photosynthesis and growth [11].

Papers have shown that Fe solubility can increase with organic ligands, and high concentrations of dissolved organic carbon (DOC) can make Fe more bioavailable [28]. In one paper, DOC and soluble Fe concentrations were shown to be strongly correlated, suggesting that organic matter influences Fe solubility [19]. Additionally, in Moreton Bay, Australia, limitations on the organic ligands supply, rather than Fe absence, may restrict L. majuscula growth [87]. The chelating agent EDTA is important because it forms soluble Fe complexes available for L. majuscula uptake [87]. It is common in runoff and wastewater, and land uses near proliferation areas could enhance its growth through EDTA inputs [19,89].

P is also essential for growth, with benthic release of P contributing to proliferation of L. majuscula [90]. Nevertheless, hypoxic conditions are necessary for the benthic release of phosphate (PO₄³⁻) and ferrous iron (Fe²⁺) to occur, i.e., when the dissolved oxygen in the water column drops below 3 mg L⁻¹ [90]. In Cocos Lagoon, Guam, growth of L. majuscula and two other cyanobacteria increased with PO₄³⁻ enrichment, but with not soluble Fe [86]. Some authors find P less important than Fe in L. majuscula growth [84,85], while others report that the growth of L. majuscula is controlled by P availability [86,90]. Papers may also differ in their findings at the same location over time, e.g., Elmetri and Bell [91] saw PO₄³⁻ boost L. majuscula growth, but years later highlighted Fe bioavailability as a key limitation due to lack of organic ligands [87].

In summary, seven out of the 11 papers indicate that bioavailable Fe is the primary limiting factor for L. majuscula growth, whereas the remaining four identify P as the limiting nutrient. L. majuscula has access to nutrients from both sediments and the water column [92]. Furthermore, of the six papers that investigated the main drivers of L. majuscula blooms, three considered available nutrient pools, and all three highlighted their direct role in initiating bloom events [39,92]. Although nutrient inputs from sediments contribute to L. majuscula proliferation, nutrients delivered by runoff may also play an important role when environmental conditions are favorable [85]. Different authors proposed that sediments may not supply all needed nutrients; floods provide P crucial for L. majuscula growth [90].

Nutrient availability represents just one component of the complex array of factors contributing to Lyngbya proliferation [93]. Other factors include bottom currents, light, temperature, and wind [92,94]. L. majuscula has been shown to grow across a range of temperatures, from tropical Moreton Bay (water temperatures between 24 and 30 °C) [28] to temperate Canary Islands (sea surface temperatures between 15 and 26 °C) [37], where proliferation persists year-round, even in the coldest winter months. Hamilton et al. [94] found that the average minimum monthly temperature best predicted L. majuscula proliferation compared with variables including solar exposure and rainfall. Six of seven reviewed studies highlighted temperature as a critical factor for L. majuscula proliferation, though optimal conditions vary by location. Kehoe et al. [93] linked growth to rising water temperatures (18–26 °C) and wind changes, suggesting proliferation might result from multiple factors rather than a singular one.

Light availability, along with temperature, has been underscored in six out of six papers as a critical factor for L. majuscula proliferation, as with all cyanobacterial species. In Moreton Bay, high light levels (~44.7 mmol-quanta m⁻² day⁻¹) marked proliferation onset [28], while in Deception Bay, light ranged from 100 to 640 mmol-quanta m⁻² s⁻¹ during bloom initiation [93]. Photoinhibitory effects have been reported at irradiances exceeding 700 mmol-quanta m⁻² s⁻¹ [28]. In conclusion, nutrient pools, together with light and water temperature, are unsurprisingly, key bloom predictors [39,92,95].

3.2.2. Underlying Factors

Multiple studies indicate that hydrodynamic stability is a critical precondition for L. majuscula proliferation. Specifically, four out of six relevant papers suggest that calm conditions are necessary for L. majuscula proliferation. Johnson et al. [95] quantified this relationship, showing that the probability of L. majuscula proliferation increased to 43% under low bottom-current velocities but decreased to 15% under high-current conditions. Rainfall events have also been identified as important indirect drivers of L. majuscula blooms, as runoff delivers nutrients essential for growth. Two out of six studies linked bloom occurrence to rainfall, with proliferations typically developing after runoff events, particularly when followed by clear, warm, and low-energy conditions [28,94].

The lag between rainfall events and bloom onset is variable and site-specific, even within the same system [28]. For example, in Deception Bay, Australia, L. majuscula blooms were observed approximately one month after a 270 mm rainfall/runoff event and two weeks after an 86 mm event [11,28]. In this catchment, sediment and soil extracts enriched in iron, phosphorus, and organic carbon were shown to enhance growth [19], while aquaculture and poultry production were identified as land uses most likely to contribute nutrient inputs conducive to proliferation [39]. On the other hand, Ramesh et al. [21] documented L. majuscula blooms in an area lacking nearby rivers or obvious land-based runoff sources, indicating that alternative nutrient pathways or in situ processes may also support bloom development.

Regarding L. wollei, light availability is also a key requirement, although its minimum light threshold is lower than that of many other benthic cyanobacteria, allowing growth at irradiance levels of 18–53 μE/m²/s. Maximum biomass has been reported at depths of 1.5–3.5 m [96]. Similar to L. majuscula, L. wollei grows loosely attached to substrates, though avoiding very hard surfaces [96]. Growth is promoted under nutrient-rich conditions, particularly elevated phosphorus and nitrogen concentrations [96]. Calm hydrodynamic conditions are again required, together with water temperatures in the range of approximately 23–26 °C [97]. Optimal growth occurs at neutral to alkaline pH, with a preference for pH ≥ 7 and peak growth near pH 8, and L. wollei is restricted to freshwater systems with low salinity.

Overall, the literature shows a strong imbalance in research focus. Among the papers analyzed in this review, the majority (12 of 14) examined factors influencing L. majuscula bloom initiation, whereas only two addressed L. wollei, and no studies explicitly considered other Lyngbya species. Consequently, the range of environmental drivers associated with L. majuscula bloom initiation is synthesized in Figure 4.

3.3. Toxin Production

Seventeen out of the 51 papers provided information on toxin production. Of these 17, eight are related to L. majuscula, eight to L. wollei, and one to L. bouillonii.

Table 2. Overview of toxin types, their associated symptoms/characteristics, and key references, categorized by Lyngbya species.

Species	Toxins	Symptoms/Characteristics	References
L. majuscula	Lyngbyatoxin A (LA)	- Contact dermatitis in humans. - Produce erythema, blisters, and necrosis. - Gastrointestinal effects. - Swelling in mice’s ears.	[18,80,98,99,100,101]
L. majuscula	Debromoaplysia-toxin (DAT)	- Contact dermatitis in humans. - Produce erythema, blisters, and necrosis. - Gastrointestinal effects. - Swelling in mice’s ears.	[17,18,39,58,59,60,61,80,98,99,100,101]
L. majuscula	Aplysiatoxin (AT)	- Together with DAT, promote tumors in mice. - Produce erythema, blisters, and necrosis. - Gastrointestinal effects (bleeding, diarrhea, burning).	[18,80,102]
L. majuscula	Manauealides (derived from DAT & AT)	- Burning sensations in the mouth/throat (15–90 min after ingestion). - Diarrhea (dose-dependent).	[18]
L. wollei	Saxitoxin (STX) & Paralytic Shellfish Toxins (PSTs)	- Neurotoxic effects - Toxin concentrations are influenced by site, temperature, and filament chlorophyll a.	[23,103]
L. wollei	L. wollei toxins (LWTs 1–6)	- Dermatotoxicity (mainly LWTX-1). - LWTX-1 production is influenced by environmental conditions. - Toxicity varies from highly toxic (LWT5) to non-toxic analogues (LWTs 1–4,6). - Breakdown products are toxic too.	[13,20,29,104]
L. wollei	Cylindrospermopsin (CYN)	- Identified in the Australian population.	[105,106]
L. wollei	Deoxy-cylindrospermopsin (deoxy-CYN)	- Identified in the Australian and EUA population.	[105,106]
L. bouillonii	Unkown	- Ecotoxicological effects (LC50 = 0.81 mg/mL in brine shrimp); - Genotoxicity (DNA damage in human lymphocytes, even at 1 ng/mL)	[22]

summarizes the main toxins found in the three Lyngbya species, their characteristics, associated symptoms, and key references.

About L. majuscula, it can produce various secondary metabolites, with over 70 compounds that exhibit biological activity [18,80]. Records of health consequences following exposure to L. majuscula have been documented since the 1950s in various locations around the world. In Hawaii and Japan, individuals reported dermatitis after exposure to L. majuscula [18,80,98]. In Australia, Moreton Bay residents reported symptoms ranging from simple skin itching to severe conditions requiring medical attention from health professionals [107]. Among the secondary metabolites produced by L. majuscula, Lyngbyatoxin A (LA) and Debromoaplysia-toxin (DAT) are the most commonly studied, and these toxins are discussed in seven of the eight papers on L. majuscula toxin production analyzed in this review.

Three studies indicate that the presence and concentration of LA and DAT in L. majuscula vary across time and location. For example, a bloom sample from Shoalwater Bay, Australia, contained low LA levels (26 µg kg⁻¹ dry weight) and no DAT [100], whereas in Deception Bay, DAT concentrations exceeded LA [101]. At Adams Beach in Moreton Bay, only LA was detected during the experiment, although both toxins had previously been reported at this site [101]. Samples of L. majuscula with no detectable levels of LA and DAT were also found to trigger an inflammatory response in the mouse ear [99]. In the literature, limited information is available on the environmental conditions that promote toxin production. Only one of the eight papers analyzed reported that L. majuscula toxin cell quota was highest in phosphorus-depleted cultures [38]. Also, their color may be related to the toxin content, i.e., toxic varieties were red [80,98], rather than olive.

Regarding L. wollei cultures, maximum STX and biomass production occurred at warmer temperatures (26 °C), higher ${\mathrm{P}\mathrm{O}}_4^{3-}$ (ranging from 0.55 to 5.5 mg P L⁻¹) and nitrate (83 mg N L⁻¹) concentrations, and lower light intensity (11–22 µmol m⁻² s⁻¹) [103]. In terms of concentrations, PSTs—a toxin group that includes STX—were detected in Butterfield Lake, NY, USA, with an average of 8.45 ± 2.30 µg STX eq./g dry weight. For LWTX-1, the average concentrations detected in samples from two Canadian lakes were 51 ± 40 µg/g DM (Lake Saint Louis, MO, USA) and 25 ± 30 µg/g DM (Lake Saint-Pierre, QC, Canada) [13]. In the St. Lawrence River, LWTX-1 was detected in L. wollei benthic mats (75.29–103.26 ng/mg) and overlying water (3.01–11.03 ng/L), showing strong correlations with dry biomass (r = 0.94), dissolved organic carbon (r = 0.74), and nitrogen content in filaments (r = 0.52).

Moreover, LWTX toxins from degraded L. wollei biomass break down into other compounds that may also be toxic. In Lake Wateree, South Carolina, LWTXs toxins were released solely after the organisms had died and completely dried, degrading faster at pH ≥ 8 into other toxic compounds, such as dicarbamoyl STX and LWTXs 4. Despite toxin breakdown, overall biomass toxicity increased compared to STX [29]. In general, for L. wollei, six of eight papers address PSTs, being two focusing mainly on LWTX-1, one on LWTX toxins, and three on other PSTs. CYN and deoxy-CYN were the central point of two out of the eight papers. Concerning the latter two, Seifert et al. [106] observed that deoxy-CYN occurred at considerably higher concentrations than CYN, suggesting that the primary compound produced by L. wollei is deoxy-CYN.

To conclude, just one paper addressed the less well-researched species, L. bouillonii. In the Andaman and Nicobar Islands, it was found that toxin extracts exhibited significant ecotoxicological and genotoxic effects. The crude cell extracts were highly toxic to brine shrimp, with a lethal concentration (LC50) of 0.81 mg/mL at 48 h. In human beings, the assay showed a dose-dependent, statistically significant increase in DNA damage in human lymphocytes in vitro. Even the lowest extract concentration, 1 ng/mL, presented toxicity. In contrast to L. majuscula and L. wollei, the compounds or causative effects responsible for inducing genotoxic effects are unknown [22].

3.4. Monitoring Lyngbya

Among the five thematic areas addressed in this review, monitoring of Lyngbya blooms is the least developed, with only two studies identified in the literature, both focusing exclusively on L. majuscula. One of the few available studies documented the spatiotemporal progression of an L. majuscula bloom in northern Deception Bay, Australia, demonstrating the capacity for rapid expansion over short time scales. The bloom initially covered approximately 49 ha with a mean biomass density of 7.9 g wet weight m⁻². Within less than one month, the affected area expanded to 329 ha, accompanied by an increase in biomass density to 22.0 g wet weight m⁻². A further 14 days later, bloom extent reached 529 ha, while biomass density increased markedly to 97 g wet weight m⁻² [11].

The second study focused on integrating satellite imagery from Landsat 7 ETM with field mapping conducted by boat to monitor L. majuscula in Moreton Bay, Australia. A boat-based GPS survey was conducted to map the spatial extent of L. majuscula, timed with a Landsat 7 ETM+ overpass. A cloud-free Landsat image was then acquired for dates when boat mapping indicated over 25% L. majuscula cover in the study area [108]. As a conclusion, satellite-based mapping (Landsat 7 ETM+) successfully captured 100% of the bloom’s surface area but achieved only 58% accuracy in detecting L. majuscula. In contrast, the field survey reached 100% accuracy but covered just 0.5% of the study area. Additionally, Landsat imagery effectively mapped areas where L. majuscula covered more than 30% of the surface [108]. Although L. wollei is not mentioned in this section, its images, along with those of L. majuscula, can be seen in Figure 5.

3.5. Modeling Approches

A total of seven papers cover the modeling approach topic, and all of them are related to L. majuscula. There have been different approaches applied to understand and manage its proliferation

Table 3. Modeling approaches applied in publications on L. majuscula, including research goals and sources.

Modeling Approach	Research Goal	References
Bayesian Network	- Understanding the major pathways that trigger L. majuscula bloom. - Identify gaps, prioritize future research and evaluate management options.	[39,92,95]
Bayesian Network + Probit Time Series Model	- Bloom prediction.	[94]
Model based on the Classic System Dynamics Model of Eutrophication Lakes	- Scoping and consensus-building model for developing research directions.	[111]
Random Forest Algorithm	- Demonstrating the influence of benthic light flux on bloom initiation.	[93]
Maxent Distribution Model	- Modelling the potential spread of the existing bloom and its possible impact on other local communities.	[37]

.

Bayesian Networks are the most commonly used approach for predicting and addressing L. majuscula issues. They incorporate major drivers [92,95], identify knowledge gaps, prioritize future research, and evaluate management options [39]. Nevertheless, other modeling approaches have also been applied. For example, based on an existing system dynamics model of lake eutrophication, a “new”/modified model was developed as a scoping and consensus-building tool for L. majuscula [111]. This modified model was formulated under the hypothesis that L. majuscula is Fe-limited, and its bloom is initiated and sustained in response to increased bioavailable Fe concentrations. The results showed that the occurrence of an L. majuscula bloom is sensitive primarily to the quantity and duration of organically complexed Fe entering the water, as well as sediment bioturbation. These findings are in line with outcomes from other studies that applied different modeling approaches.

In addition, a Random Forest algorithm and Maximum Entropy Model (Maxent) were also employed to address the L. majuscula issue from different perspectives. The first demonstrated the influence of benthic light availability on the L. majuscula bloom initiation at shallow coastal sites. The authors stated that the causes behind the initiation of major blooms in Deception Bay were an increase in benthic light intensities, daily average water temperature, and changes in wind direction [93]. The second one, Maxent model, provided insights into the potential distribution of L. majuscula on a regional scale at a fine resolution (5 m) and its possible impact on other aquatic communities residing at the same spot, such as meadows of the seagrass, Cymodocea nodosa, in a Marine Reserve located in the Canary Islands [37].

3.6. Lyngbya Control Management

A variety of control management options have been applied around the world for different Lyngbya species (Figure 6). Yet the papers analyzed in this review have focused primarily on metal-based products, whilst acknowledging that a range of other algaecides is also available. Copper-based algaecides are frequently applied to control algal blooms in surface waters [112]. As such, controlling Lyngbya blooms has been studied through various approaches, including the processes of copper adsorption, internalization, and desorption following algaecide application [113], short-term copper sorption kinetics and its impact on Lyngbya vitality [112], as well as overall Lyngbya’s responses to copper algaecide exposures [114,115]. L. wollei is the main species analyzed in most of the papers, seven of 11, and four papers emphasized the effectiveness of treatments using the formulation-enhanced Ethanolamine Chelated Copper algaecide (Captain XTR) in controlling this species [30,31,113,116], whether applied alone or in combination with others.

The treatment using chelated copper revealed significantly more internalized copper without desorbing too much, which is advantageous, given that desorbed copper presents an elevated risk to non-target organisms [113]. Additionally, chelated copper was shown to be less toxic to non-target freshwater fish, such as Salvelinus fontinalis [117]. To further increase the safety margins for non-target species and reduce the estimated copper mass loading to sediments, alternative algaecide treatment using Phycomycins-SCP (sodium carbonate peroxyhydrate), followed by Cide-Kick II and Algimycins- PWF (a blend of gluconate and citrate chelated copper), was proposed [118]. Phycomycin-SCP was demonstrated to be efficient in controlling L. magnifica too [12].

Regarding other species, copper was also tested against L. putealis, but this time alongside cobalt. In a single-metal system (either copper or cobalt), the tolerance of L. putealis was tested, and in general, it demonstrated more tolerance to copper compared to cobalt [24]. Consequently, treatments based on copper may not be as effective as they are on L. wollei. In the case of L. majuscula, the use of Octolig, a commercial chelating material bound to high–surface-area silica gel, was investigated to remove key nutrients essential for the growth of L. majuscula. Although it was not possible to identify the removal action of P and N, differences were observed in the growth and appearance of L. majuscula samples treated with Octolig, when compared to non-treated ones [119]. Additionally, emerging nanotechnology-based approaches have been tested primarily under laboratory conditions. For example, the effects of different magnesium oxide nanoparticle (MgONPs) concentrations on L. majuscula growth have been investigated, showing that increasing MgONPs concentrations led to a reduction in L. majuscula across most tested treatments [120]. Similarly, biosynthesized copper nanoparticles (Cu-NPs) have been evaluated as an inhibitory agent against L. majuscula, achieving up to 90% biomass reduction in a concentration-dependent manner [121]. However, their potential ecological impacts and applicability at the field scale remain unknown. Hydrogen peroxide has also been tested at the laboratory scale and found to be effective at controlling L. majuscula at a concentration of 4 mg L⁻¹, i.e., a 55% reduction in biomass [122].

4. Discussion

4.1. Geographic Expansion and Reporting of Lyngbya Occurrences

The concentration of Lyngbya studies in Australia and the United States likely reflects the greater monitoring capacity, research funding, and publication output of these countries rather than the true global distribution of Lyngbya blooms. While robust evidence for a quantified global expansion is limited, recent reports of Lyngbya occurrence have emerged from additional regions, including China, Cameroon, and Russia [47,50,123], suggesting that this genus is present across a broader range of geographic settings than is reflected in the English-language literature. These studies are summarized in Supplementary Material, as they were excluded from the full analysis due to language restrictions.

Furthermore, the apparent geographic bias in the literature likely results in systematic under-reporting of Lyngbya blooms in regions with limited monitoring infrastructure, management capacity, or academic publishing resources, particularly in tropical and subtropical areas where environmental conditions such as elevated water temperature and high solar irradiance are favorable to cyanobacterial growth. Australia represents a well-documented case, as it includes tropical and subtropical regions that frequently experience Lyngbya blooms. The recurrence and impacts of these blooms, particularly those of L. majuscula, have driven sustained investment in monitoring programs, management interventions, and scientific research. As a result, Australia accounts for the largest number of published studies on Lyngbya compared with other regions. Therefore, the absence of published evidence from regions with climatic conditions more favorable for Lyngbya occurrence should not be interpreted as evidence of absence. Taken together, the available literature supports the interpretation that Lyngbya blooms are more widely distributed than currently documented, while also highlighting the need for improved monitoring and reporting frameworks to better assess their spatial extent and temporal trends at a global scale.

Importantly, this observed geographic bias may influence the universality of current management strategies. For example, management strategies, including modeling approaches and mitigation measures such as algaecide use, have largely been developed based on site-specific conditions. While these approaches are broadly applicable, their effective implementation requires adaptation to local environmental, ecological, and socio-economic contexts.

4.2. Bloom Predictors and Critical Knowledge Gaps in Lyngbya Proliferation

Available evidence indicates that L. majuscula bloom initiation arises from the convergence of multiple environmental drivers, including nutrient availability, temperature, light, and hydrodynamic stability. As discussed in Section 3.2.2, calm conditions play a central role by promoting nutrient accumulation within sediment interstitial spaces, where dissolved nutrients become more accessible for benthic uptake [92]. Conversely, elevated current velocities increase turbidity through sediment resuspension, thereby reducing light penetration and limiting photosynthetic activity, which constrains bloom development. Physical disturbances, such as boat traffic, can further complicate these dynamics by episodically resuspending sediments and releasing nutrients into the water column, potentially stimulating L. majuscula growth under otherwise suboptimal conditions [124]. Importantly, because bloom initiation depends on the simultaneous alignment of several drivers, a consistent or predictable time lag between rainfall events and bloom onset should not be expected. Rainfall, therefore, cannot be interpreted as a direct or universal trigger; rather, it acts as a context-dependent enabling factor, whose influence is mediated by local hydrodynamic, sedimentary, and light conditions.

Despite repeated identification of nutrients as key drivers, no consensus has emerged regarding which nutrient(s) primarily limit L. majuscula proliferation, with phosphorus and bioavailable iron most frequently implicated. Other compounds, including dissolved organic carbon and chelating agents such as EDTA, have also been shown to influence growth. This lack of agreement highlights a major knowledge gap and suggests that nutrient limitation is strongly site-specific, governed by local geochemical conditions rather than a single universal constraint. Crucially, much of the existing evidence is derived from laboratory or mesocosm studies, while in situ investigations remain scarce, limiting our ability to generalize findings across systems. Field responses are further complicated by interactions within benthic communities and environmental gradients, which can modulate nutrient availability, uptake efficiency, and competitive dynamics [86,125]. As a result, the specific conditions required for bloom initiation remain poorly constrained, underscoring the need for high-resolution, field-based monitoring and experimental studies that capture real-world complexity.

In addition, in Moreton Bay, Australia, significant L. majuscula blooms began in 1997, drawing substantial attention from the scientific community [82]. Although publications focusing on L. majuscula became common in the 2000s in this region, its taxonomy in Moreton Bay has only been confirmed using microscopic methods and still requires confirmation through molecular techniques. Therefore, this represents another important opportunity for future research. Finally, the discussion of bloom predictors remains heavily skewed toward L. majuscula. Only two studies addressing L. wollei were identified, and other Lyngbya species are largely absent from the predictor literature. This imbalance represents a significant limitation of current knowledge and restricts the development of transferable predictive frameworks across marine and freshwater systems. Addressing these gaps are essential for advancing both mechanistic understanding and effective management of Lyngbya blooms.

4.3. Monitoring Innovations and Management Implications

Monitoring the origins and progression of Lyngbya blooms is essential for reducing or mitigating their adverse effects, such as water quality degradation, disruption of the ecosystem and biodiversity functioning. For example, seagrass beds were overgrown along at least 18 km of coastline by dense mats of L. majuscula, which likely degraded the quality of seagrass habitats. As seagrass is the primary food source for green turtles, this change in habitat meant that turtles were exposed to the bloom, with evidence showing that they ingested Lyngbya when it was abundant. Although no immediate decline in turtle body condition was observed, a higher proportion of damaged seagrass in turtle diets was reported, as well as reduced blood parameters (such as sodium and glucose), suggesting that turtles experienced a poorer-quality diet during the bloom. Additionally, because L. majuscula is a toxic cyanobacterium, turtles feeding in bloom-affected areas may be exposed to biologically active compounds that are potentially harmful. Therefore, L. majuscula blooms can negatively affect both seagrass ecosystems and green turtles by degrading habitat quality, altering feeding patterns, and increasing exposure to toxins [100].

Among the five thematic categories considered in this review, monitoring Lyngbya had the greatest paucity of studies. This scarcity highlights a substantial and underexplored research gap, particularly given the rapid expansion, persistence, and management challenges associated with benthic cyanobacterial blooms. For example, the study promoted by Ahern et al. [11] illustrate not only the rapid spatial spread of L. majuscula blooms, but also the pronounced escalation in biomass over short periods. Thereby, the need for high-frequency, spatially resolved monitoring approaches capable of detecting early-stage proliferation and informing timely management responses is crucial.

Accurately tracking these proliferations is complex particularly in large waterbodies and estuaries, because cyanobacterial biomass often shows high spatial and temporal heterogeneity [126]. Traditional monitoring methods designed for water column algal or phytoplankton blooms are generally unsuitable for benthic species. For instance, conventional water sampling targets surface or near-surface layers, whereas benthic cyanobacteria grow attached to substrates on the bottom. As a result, these methods may fail to detect benthic mats altogether. Their distribution is also highly patchy, meaning that a single point sample can easily miss dense accumulations [127]. Additionally, water sampling provides only limited spatial coverage and typically represents very small areas rather than whole regions. The required sample collection and subsequent laboratory analyses are also labor-intensive and time-consuming [128,129].

Different approaches are used to monitor benthic species, such as estimating the biomass in quadrats. However, this is time-consuming and much less accurate than sampling methods for phytoplankton. The use of new technologies, or the combination of them with on-ground methods, is becoming an increasingly common practice to make monitoring methods more efficient. Commercial unoccupied aerial systems (UASs) [130,131], optical sensors combined with machine learning techniques [132], and remote sensing combined with predictive modeling [126] are examples of these new monitoring approaches.

Newer satellites have overpass times of less than a week, and as such, future work could focus on developing a larger matching dataset (satellite retrievals-field surveys) to enhance the accuracy of satellite-based detection and enable precise mapping of large water bodies. Nevertheless, it is important to consider that under certain conditions, satellite imagery has limitations, and consequently, its accuracy decreases. In highly turbid waters, the suspended inorganic particles reduce light penetration, altering the surface reflectance and diminishing the accuracy of the final image. Moreover, the most common satellite sensors capture light from the surface or near-surface layer; thereby, the signals of organisms living in deeper waters, such as benthos, may not be captured. In the case of small sites, satellite image resolution is insufficient.

Commercial unmanned aerial vehicles (UAVs) emerge as another monitoring option. They have been successfully deployed to quantitatively map surface water chlorophyll-a (chl-a) distribution in coastal waters [133], and to estimate cyanobacterial abundance using chl-a and phycocyanin-related wavelengths [131]. The authors demonstrated that visible light spectra (red, green, blue) sensors are moderately effective for estimating chl-a. Moreover, UAVs equipped with red, green, blue (RGB) and near-infrared (NIR) sensors were utilized for trophic state mapping of small reservoirs in Taiwan, providing data on average levels of chl-a, total P, and Secchi disk depth [134]. The effectiveness of monitoring cyanobacteria in an urban lake located in Mexico using UAVs combined with in situ data was demonstrated by Aguirre-Gómez et al. [135].

However, UAVs present some disadvantages, such as the inability to function under certain weather conditions and the requirement for a visual line of sight, which limits the speed of conducting aerial surveys. In addition, challenges in identifying algal species and processing images of water presented by UAVs [130] are complex to solve. Therefore, underwater drones, a type of remotely operated vehicle (ROV), seem to offer the most feasible approach for early detection of benthic algae such as Lyngbya, because they can perform high-resolution scans of the seafloor. To the authors’ knowledge, the use of ROVs for monitoring Lyngbya has not been documented, representing a novel approach in this field. These vehicles have been applied for other tasks, such as gathering data to create high-resolution maps of water quality parameters—temperature, salinity, conductivity and chlorophyll fluorescence—to assess the spatial variability of riverine water quality [136].

In short, better monitoring approaches are required to improve the current understanding and predictive capacity of Lyngbya. UAVs offer a viable tool for detecting algal blooms [130]; but they are not the most suitable technology to monitor the initial phase of Lyngbya growth. For that purpose, the utilization of underwater drones appears to be a better alternative. UAVs can detect surface-level blooms, while underwater drones may be better suited for early detection of Lyngbya growth, though both technologies pose cost and training challenges, especially for the less common underwater drones. On the other hand, identifying Lyngbya at the beginning of its cycle would allow more proactive mitigation, preventing Lyngbya’s spread and reducing the costs for later stages of proliferation management and health risks for swimmers/bathers.

In lower-resource contexts, the development of models to predict its appearance may prove a more cost-effective alternative and will provide an early warning of Lyngbya appearance. The utilization of qualitative models, such as Bayesian Networks, can overcome data constraints. Another option for monitoring Lyngbya, although it has lower spatial resolution than UAVs, is the use of satellite imagery. Matching the timing of field surveys with satellite retrieval dates can help with the model calibration and consequently achieve better estimations. In addition, since satellite images are collected every few days, stakeholders can identify spatial changes and trigger further sampling if required. It offers a complementary option, especially for large water bodies, following initial in-situ data collection for model calibration purposes.

4.4. Modeling Approaches, Their Limitations, and Opportunities for Future Work

Ecological problems are generally complex and multifaceted. This is evident in the case of Lyngbya proliferation prediction. Because numerous causative factors can trigger its appearance, predicting the timing, magnitude, and duration of blooms is a significant challenge. The complexity of the proximate and underlying factors affecting the blooms makes the existing models in the literature either site-specific, qualitative/conceptual only, or with large uncertainties. For example, regarding the Bayesian Network approach, Johnson et al. [95] noted a key limitation: the accuracy of its outputs in predicting the initiation of L. majuscula bloom was not assessed due to insufficient data. Furthermore, the limited number of covariates may have been a limiting factor in the research conducted by Hamilton et al. [94], and the absence of longer time series under diverse environmental conditions, which are essential for developing more accurate and validated models.

In the case of the system dynamics model of eutrophication in lakes [111], limitations were also highlighted, such as the lack of reliable parameter values and concerns about the accuracy of the model’s structure in representing the real-world interactions and complexities. For the Maxent model [37], the limitations included the need for a complete water quality database. In conclusion, the literature on using models to predict Lyngbya is lacking in the use of an effective and innovative approach that may enhance the robustness of prediction and, consequently, its management. Additionally, no modeling approaches were found for predicting the occurrence of Lyngbya species other than L. majuscula, and the papers analyzed here focused only on the main factors contributing to L. majuscula blooms’ initiation. Other aspects, such as prediction, growth, biomass, decay, bloom duration, and senescence, remain underexplored in the literature [92,95].

A novel approach is the development of a prediction model that combines a process-based model with a data-driven model [137,138,139]. The process-based model provides well-known knowledge regarding the physical, chemical and biological mechanisms of the system, while the data-driven models apply data mining and statistical techniques to analyze monitoring data, identify patterns and relationships, and then create predictive rules for the system [40]. In other words, process-based models simulate environmental factors such as nutrients, light, and temperature that influence Lyngbya proliferation. They also offer the option to create different scenarios or conditions that have not been observed before, aiding in understanding where and when proliferation might occur. Conversely, data-driven models identify trends in historical data and learn complex relationships that the process-based model may oversimplify or overlook. Thus, data-driven models can refine, calibrate, or validate the process-based model. Together, they provide more reliable simulations for various scenarios, if needed.

The approach of using process-based combined with data-driven models may be effective because it combines well-established knowledge with predictions derived from observed data. Although, based on the findings of this review, this approach has not yet been applied to predict Lyngbya species specifically, it has been successfully used for other objectives in different aquatic systems: for example, to predict algal bloom timing and magnitude using chl-a as a bloom proxy in a lake [139], to estimate spatial chl-a distributions across a reservoir [137], and to predict cyanobacteria cell count categories in a river [138]. Lin et al. [139] applied three different workflows to predict chl-a. The first applied a direct machine-learning model using observed environmental data. The second workflow implemented a two-step machine-learning process that first predicted nutrient concentrations and then combined those predictions with meteorological data to estimate chl-a. The third workflow integrated a hydrodynamic process-based model with machine learning. Of the three, the hybrid model performed best, particularly by improving predictions of algal bloom onset.

In the case of Lyngbya predictions, hydrodynamic models such as MIKE [140] can be used as the process-based model, while a range of machine learning and statistical models can be applied as a data-driven model. Currently, site-specific data limitations are the main difficulty hindering the development of data-driven or process-based models in this space; however, this issue can be addressed through the use of new technologies, such as underwater drones, as discussed in the previous section. As a consequence of in situ data collection, data sources for modeling work were often derived from process-based model simulations and statistical models, empirical data, and expert opinion. Although these studies provide insights into the main factors that influence L. majuscula proliferation initiation, there is still a gap in developing a model based solely on field-collected data. Developing a model that integrates process-based and data-driven models relying on observed data collected through this new technology would enhance the novelty of the model.

For locations with the availability of data and resources, the combination of process-based models and data-driven models could address the limitations of each approach and thereby improve the accuracy and reliability of the model’s predictions. The model effectiveness for specific applications, such as predicting cyanoHAB may be significantly enhanced through the integration of established scientific knowledge of the phenomena being modelled. For instance, the development of models that integrate data-driven and process-based components could help better understand and predict processes affecting Lyngbya blooms at appropriate spatial and temporal resolutions. Several hydrodynamic models already exist for several locations where Lyngbya has been reported. Since hydrodynamics strongly affect crucial processes leading to blooms, these models could set the basis for a hybrid model, where a data-driven component, based on historical proliferations data, combines different data inputs and process-based modeling outputs to produce an estimation of Lyngbya presence.

4.5. Considerations in Using Algaecides for Lyngbya Management

To the authors’ knowledge, there are no studies about the application of algaecide to control L. majuscula, potentially because it usually blooms in estuarine waters. Effective application of algaecides in these open waters can be complex due to highly variable factors, such as currents and tides. There are also no papers on algaecide application to control L. putealis. In the case of L. wollei, it is usually found in freshwater bodies such as lakes and reservoirs, and most authors recommend their control using copper-based algaecides with a chelated formulation. This type of formulation has also been shown to be efficient for L. magnifica in a limited number of papers [12].

The application of algaecides appears to be a viable option since, at the appropriate concentration, it could shorten the time required to restore the functionality of different water sources, such as reservoirs, lakes and rivers [141]. Additionally, it could lower costs and mitigate risks to non-target species [12]. Various factors influence the site-specific algae response to algaecides, affecting the effectiveness of a specific treatment. These factors may be water characteristics, the intrinsic character and sensitivity of the target algae, formulation of the project, initial concentration, and exposure duration [112,114]. Even with the positive outcomes reported for chelated copper treatments in controlling L. wollei [30,31,118], this treatment continues to pose several complications. For example, the amount of desorbed copper decreases when this algaecide is applied; however, the desorption process is not eliminated, and a portion of copper remains in the water, presenting a risk to non-target organisms and accumulating in the sediments.

Furthermore, the freshwater species Daphnia magna presented negative effects such as reduced reproduction, delayed age at first reproduction, smaller body size, and other adverse effects when three chelated copper algaecides were applied. Notably, Captain XTR—documented in four of the seven studies as the most effective chelated copper algaecide for controlling L. wollei—was also associated with these impacts. One of them demonstrated to be more toxic in the long-term exposure than copper sulphate formulations, highlighting that the chelated formulations are not necessarily safer, especially under prolonged exposure [142]. Additionally, in many jurisdictions, copper-based treatments are not permitted in natural waters. Treatments based on copper sulphate spraying are reported to be dangerous in the long term. They may affect both drinking water quality and reservoir management as they involve changes in nutrient concentrations and shifts in the ecosystem. Thus, those treatments initiate a series of events that conclude with a less predictable environment for water quality management, further complicating it [143].

While algaecides are the most commonly reported method for controlling certain Lyngbya species in the reviewed studies, several alternative approaches are also available for managing cyanobacteria. These methods are generally considered safer and may offer more targeted, environmentally sustainable solutions. For example, moderate pre-oxidation techniques—such as ultrasound-based processes that avoid chemical oxidants and can be combined with chemical treatments—show promise [144]. These approaches present significant research opportunities; nevertheless, their application to benthic species such as Lyngbya may be constrained by cost, scalability, and effectiveness when compared with planktonic counterparts. Nanoparticle-based treatments have emerged as a promising approach; however, existing studies are largely limited to laboratory conditions and often emphasize material synthesis and mechanistic toxicity. This highlights the need for further research focusing on feasibility, environmental safety, and applicability in real-world aquatic systems. Therefore, proactive mitigation measures remain the most reliable strategy to prevent blooms from reaching a stage where reactive interventions become necessary.

4.6. A Framework for Proactive Lyngbya Blooms Monitoring, Prediction and Control

The framework presented in Figure 7 outlines ten interconnected strategies to better understand Lyngbya bloom dynamics and consequently improve their management and prevention. At its foundation, the framework highlights the need to fill critical knowledge gaps through in-situ studies of nutrient drivers and bloom triggers, and to expand monitoring approaches that integrate field-based and remote sensing data. Building on these insights, it stresses the importance of using cost-effective predictive models and applying existing hydrodynamic models to strengthen forecasting capacity. The framework also encourages adopting hybrid modeling approaches that combine data-driven and process-based methods to improve accuracy and reliability. From a management perspective, strategies include exploring underwater drones, UAVs, and satellite imagery as monitoring tools, while considering Bayesian networks and other qualitative models where data are scarce. Finally, the framework emphasizes prioritizing affordable and sustainable mitigation strategies and integrating models into decision-support systems to guide proactive and adaptive environmental management. When combined in an iterative, self-improving cycle, these strategies provide a holistic roadmap for advancing predictive capacity, monitoring efficiency, and management effectiveness for Lyngbya blooms.

5. Conclusions

This review has synthesized current knowledge from studies on Lyngbya ecology in marine and freshwater, including estuarine environments. This information is essential for advancing management strategies. Beyond summarizing existing findings, it outlines a range of approaches for controlling and preventing its proliferation, encompassing both resource-intensive alternatives and more affordable ones. The scientific evidence compiled here provides the basis necessary for improving modeling efforts, contributing to the development of more robust and effective predictive frameworks.

The paper also highlights key limitations in existing literature and suggests strategies to address them, such as field data collection using aerial and underwater drones or satellite imagery for resource-poor contexts. Such data would also facilitate the calibration of relevant predictive models, which would then, in turn, help refine future monitoring as well as help with early detection and bloom management optimization. We elaborate these findings and develop a management framework which offers a holistic roadmap for integrating knowledge generation, predictive modeling, and practical management strategies to advance proactive and adaptive environmental management of Lyngbya blooms.